Merge tag 'drm-misc-next-2018-11-21' of git://anongit.freedesktop.org/drm/drm-misc into drm-next
drm-misc-next for v4.21, part 2:
UAPI Changes:
- Remove syncobj timeline support from drm.
Cross-subsystem Changes:
- Document canvas provider node in the DT bindings.
- Improve documentation for TPO TPG110 DT bindings.
Core Changes:
- Use explicit state in drm atomic functions.
- Add panel quirk for new GPD Win2 firmware.
- Add DRM_FORMAT_XYUV8888.
- Set the default import/export function in prime to drm_gem_prime_import/export.
- Add a separate drm_gem_object_funcs, to stop relying on dev->driver->*gem* functions.
- Make sure that tinydrm sets the virtual address also on imported buffers.
Driver Changes:
- Support active-low data enable signal in sun4i.
- Fix scaling in vc4.
- Use canvas provider node in meson.
- Remove unused variables in sti and qxl and cirrus.
- Add overlay plane support and primary plane scaling to meson.
- i2c fixes in drm/bridge/sii902x
- Fix mailbox read size in rockchip.
- Spelling fix in panel/s6d16d0.
- Remove unnecessary null check from qxl_bo_unref.
- Remove unused arguments from qxl_bo_pin.
- Fix qxl cursor pinning.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/9c0409e3-a85f-d2af-b4eb-baf1eb8bbae4@linux.intel.com
diff --git a/.mailmap b/.mailmap
index a76be45f..28fecaf 100644
--- a/.mailmap
+++ b/.mailmap
@@ -159,6 +159,7 @@
Peter Oruba <peter.oruba@amd.com>
Pratyush Anand <pratyush.anand@gmail.com> <pratyush.anand@st.com>
Praveen BP <praveenbp@ti.com>
+Punit Agrawal <punitagrawal@gmail.com> <punit.agrawal@arm.com>
Qais Yousef <qsyousef@gmail.com> <qais.yousef@imgtec.com>
Oleksij Rempel <linux@rempel-privat.de> <bug-track@fisher-privat.net>
Oleksij Rempel <linux@rempel-privat.de> <external.Oleksij.Rempel@de.bosch.com>
diff --git a/CREDITS b/CREDITS
index 5befd2d..84cbec4 100644
--- a/CREDITS
+++ b/CREDITS
@@ -2138,6 +2138,10 @@
D: Soundblaster driver fixes, ISAPnP quirk
S: California, USA
+N: Jarkko Lavinen
+E: jarkko.lavinen@nokia.com
+D: OMAP MMC support
+
N: Jonathan Layes
D: ARPD support
diff --git a/Documentation/ABI/testing/sysfs-class-led-trigger-pattern b/Documentation/ABI/testing/sysfs-class-led-trigger-pattern
index fb3d1e0..1e5d172 100644
--- a/Documentation/ABI/testing/sysfs-class-led-trigger-pattern
+++ b/Documentation/ABI/testing/sysfs-class-led-trigger-pattern
@@ -37,8 +37,8 @@
0-| / \/ \/
+---0----1----2----3----4----5----6------------> time (s)
- 2. To make the LED go instantly from one brigntess value to another,
- we should use use zero-time lengths (the brightness must be same as
+ 2. To make the LED go instantly from one brightness value to another,
+ we should use zero-time lengths (the brightness must be same as
the previous tuple's). So the format should be:
"brightness_1 duration_1 brightness_1 0 brightness_2 duration_2
brightness_2 0 ...". For example:
diff --git a/Documentation/admin-guide/pm/cpufreq.rst b/Documentation/admin-guide/pm/cpufreq.rst
index 47153e6..7eca902 100644
--- a/Documentation/admin-guide/pm/cpufreq.rst
+++ b/Documentation/admin-guide/pm/cpufreq.rst
@@ -150,7 +150,7 @@
a governor ``sysfs`` interface to it. Next, the governor is started by
invoking its ``->start()`` callback.
-That callback it expected to register per-CPU utilization update callbacks for
+That callback is expected to register per-CPU utilization update callbacks for
all of the online CPUs belonging to the given policy with the CPU scheduler.
The utilization update callbacks will be invoked by the CPU scheduler on
important events, like task enqueue and dequeue, on every iteration of the
diff --git a/Documentation/cpu-freq/cpufreq-stats.txt b/Documentation/cpu-freq/cpufreq-stats.txt
index a873855..14378ce 100644
--- a/Documentation/cpu-freq/cpufreq-stats.txt
+++ b/Documentation/cpu-freq/cpufreq-stats.txt
@@ -86,9 +86,11 @@
This will give a fine grained information about all the CPU frequency
transitions. The cat output here is a two dimensional matrix, where an entry
<i,j> (row i, column j) represents the count of number of transitions from
-Freq_i to Freq_j. Freq_i is in descending order with increasing rows and
-Freq_j is in descending order with increasing columns. The output here also
-contains the actual freq values for each row and column for better readability.
+Freq_i to Freq_j. Freq_i rows and Freq_j columns follow the sorting order in
+which the driver has provided the frequency table initially to the cpufreq core
+and so can be sorted (ascending or descending) or unsorted. The output here
+also contains the actual freq values for each row and column for better
+readability.
If the transition table is bigger than PAGE_SIZE, reading this will
return an -EFBIG error.
diff --git a/Documentation/devicetree/bindings/arm/shmobile.txt b/Documentation/devicetree/bindings/arm/shmobile.txt
index f5e0f82..58c4256 100644
--- a/Documentation/devicetree/bindings/arm/shmobile.txt
+++ b/Documentation/devicetree/bindings/arm/shmobile.txt
@@ -27,7 +27,7 @@
compatible = "renesas,r8a77470"
- RZ/G2M (R8A774A1)
compatible = "renesas,r8a774a1"
- - RZ/G2E (RA8774C0)
+ - RZ/G2E (R8A774C0)
compatible = "renesas,r8a774c0"
- R-Car M1A (R8A77781)
compatible = "renesas,r8a7778"
diff --git a/Documentation/devicetree/bindings/cpufreq/arm_big_little_dt.txt b/Documentation/devicetree/bindings/cpufreq/arm_big_little_dt.txt
deleted file mode 100644
index 2aa06ac..0000000
--- a/Documentation/devicetree/bindings/cpufreq/arm_big_little_dt.txt
+++ /dev/null
@@ -1,65 +0,0 @@
-Generic ARM big LITTLE cpufreq driver's DT glue
------------------------------------------------
-
-This is DT specific glue layer for generic cpufreq driver for big LITTLE
-systems.
-
-Both required and optional properties listed below must be defined
-under node /cpus/cpu@x. Where x is the first cpu inside a cluster.
-
-FIXME: Cpus should boot in the order specified in DT and all cpus for a cluster
-must be present contiguously. Generic DT driver will check only node 'x' for
-cpu:x.
-
-Required properties:
-- operating-points: Refer to Documentation/devicetree/bindings/opp/opp.txt
- for details
-
-Optional properties:
-- clock-latency: Specify the possible maximum transition latency for clock,
- in unit of nanoseconds.
-
-Examples:
-
-cpus {
- #address-cells = <1>;
- #size-cells = <0>;
-
- cpu@0 {
- compatible = "arm,cortex-a15";
- reg = <0>;
- next-level-cache = <&L2>;
- operating-points = <
- /* kHz uV */
- 792000 1100000
- 396000 950000
- 198000 850000
- >;
- clock-latency = <61036>; /* two CLK32 periods */
- };
-
- cpu@1 {
- compatible = "arm,cortex-a15";
- reg = <1>;
- next-level-cache = <&L2>;
- };
-
- cpu@100 {
- compatible = "arm,cortex-a7";
- reg = <100>;
- next-level-cache = <&L2>;
- operating-points = <
- /* kHz uV */
- 792000 950000
- 396000 750000
- 198000 450000
- >;
- clock-latency = <61036>; /* two CLK32 periods */
- };
-
- cpu@101 {
- compatible = "arm,cortex-a7";
- reg = <101>;
- next-level-cache = <&L2>;
- };
-};
diff --git a/Documentation/devicetree/bindings/i2c/i2c-omap.txt b/Documentation/devicetree/bindings/i2c/i2c-omap.txt
index 7e49839..4b90ba9f 100644
--- a/Documentation/devicetree/bindings/i2c/i2c-omap.txt
+++ b/Documentation/devicetree/bindings/i2c/i2c-omap.txt
@@ -1,8 +1,12 @@
I2C for OMAP platforms
Required properties :
-- compatible : Must be "ti,omap2420-i2c", "ti,omap2430-i2c", "ti,omap3-i2c"
- or "ti,omap4-i2c"
+- compatible : Must be
+ "ti,omap2420-i2c" for OMAP2420 SoCs
+ "ti,omap2430-i2c" for OMAP2430 SoCs
+ "ti,omap3-i2c" for OMAP3 SoCs
+ "ti,omap4-i2c" for OMAP4+ SoCs
+ "ti,am654-i2c", "ti,omap4-i2c" for AM654 SoCs
- ti,hwmods : Must be "i2c<n>", n being the instance number (1-based)
- #address-cells = <1>;
- #size-cells = <0>;
diff --git a/Documentation/gpu/amdgpu-dc.rst b/Documentation/gpu/amdgpu-dc.rst
new file mode 100644
index 0000000..cc89b0f
--- /dev/null
+++ b/Documentation/gpu/amdgpu-dc.rst
@@ -0,0 +1,68 @@
+===================================
+drm/amd/display - Display Core (DC)
+===================================
+
+*placeholder - general description of supported platforms, what dc is, etc.*
+
+Because it is partially shared with other operating systems, the Display Core
+Driver is divided in two pieces.
+
+1. **Display Core (DC)** contains the OS-agnostic components. Things like
+ hardware programming and resource management are handled here.
+2. **Display Manager (DM)** contains the OS-dependent components. Hooks to the
+ amdgpu base driver and DRM are implemented here.
+
+It doesn't help that the entire package is frequently referred to as DC. But
+with the context in mind, it should be clear.
+
+When CONFIG_DRM_AMD_DC is enabled, DC will be initialized by default for
+supported ASICs. To force disable, set `amdgpu.dc=0` on kernel command line.
+Likewise, to force enable on unsupported ASICs, set `amdgpu.dc=1`.
+
+To determine if DC is loaded, search dmesg for the following entry:
+
+``Display Core initialized with <version number here>``
+
+AMDgpu Display Manager
+======================
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+ :doc: overview
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+ :internal:
+
+Lifecycle
+---------
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+ :doc: DM Lifecycle
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+ :functions: dm_hw_init dm_hw_fini
+
+Interrupts
+----------
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
+ :doc: overview
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
+ :internal:
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+ :functions: register_hpd_handlers dm_crtc_high_irq dm_pflip_high_irq
+
+Atomic Implementation
+---------------------
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+ :doc: atomic
+
+.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+ :functions: amdgpu_dm_atomic_check amdgpu_dm_atomic_commit_tail
+
+Display Core
+============
+
+**WIP**
diff --git a/Documentation/gpu/drivers.rst b/Documentation/gpu/drivers.rst
index 7d2d387..7c16721 100644
--- a/Documentation/gpu/drivers.rst
+++ b/Documentation/gpu/drivers.rst
@@ -5,6 +5,7 @@
.. toctree::
amdgpu
+ amdgpu-dc
i915
meson
pl111
diff --git a/Documentation/gpu/drm-mm.rst b/Documentation/gpu/drm-mm.rst
index e725e84..d0f3c6b 100644
--- a/Documentation/gpu/drm-mm.rst
+++ b/Documentation/gpu/drm-mm.rst
@@ -72,8 +72,8 @@
the kernel itself. The type of this object should be
TTM_GLOBAL_TTM_BO, and its size should be sizeof(struct
ttm_bo_global). Again, driver-specific init and release functions may
-be provided, likely eventually calling ttm_bo_global_init() and
-ttm_bo_global_release(), respectively. Also, like the previous
+be provided, likely eventually calling ttm_bo_global_ref_init() and
+ttm_bo_global_ref_release(), respectively. Also, like the previous
object, ttm_global_item_ref() is used to create an initial reference
count for the TTM, which will call your initialization function.
diff --git a/Documentation/i2c/busses/i2c-nvidia-gpu b/Documentation/i2c/busses/i2c-nvidia-gpu
new file mode 100644
index 0000000..31884d2
--- /dev/null
+++ b/Documentation/i2c/busses/i2c-nvidia-gpu
@@ -0,0 +1,18 @@
+Kernel driver i2c-nvidia-gpu
+
+Datasheet: not publicly available.
+
+Authors:
+ Ajay Gupta <ajayg@nvidia.com>
+
+Description
+-----------
+
+i2c-nvidia-gpu is a driver for I2C controller included in NVIDIA Turing
+and later GPUs and it is used to communicate with Type-C controller on GPUs.
+
+If your 'lspci -v' listing shows something like the following,
+
+01:00.3 Serial bus controller [0c80]: NVIDIA Corporation Device 1ad9 (rev a1)
+
+then this driver should support the I2C controller of your GPU.
diff --git a/Documentation/x86/x86_64/mm.txt b/Documentation/x86/x86_64/mm.txt
index 73aaaa3..804f942 100644
--- a/Documentation/x86/x86_64/mm.txt
+++ b/Documentation/x86/x86_64/mm.txt
@@ -34,23 +34,24 @@
____________________________________________________________|___________________________________________________________
| | | |
ffff800000000000 | -128 TB | ffff87ffffffffff | 8 TB | ... guard hole, also reserved for hypervisor
- ffff880000000000 | -120 TB | ffffc7ffffffffff | 64 TB | direct mapping of all physical memory (page_offset_base)
- ffffc80000000000 | -56 TB | ffffc8ffffffffff | 1 TB | ... unused hole
+ ffff880000000000 | -120 TB | ffff887fffffffff | 0.5 TB | LDT remap for PTI
+ ffff888000000000 | -119.5 TB | ffffc87fffffffff | 64 TB | direct mapping of all physical memory (page_offset_base)
+ ffffc88000000000 | -55.5 TB | ffffc8ffffffffff | 0.5 TB | ... unused hole
ffffc90000000000 | -55 TB | ffffe8ffffffffff | 32 TB | vmalloc/ioremap space (vmalloc_base)
ffffe90000000000 | -23 TB | ffffe9ffffffffff | 1 TB | ... unused hole
ffffea0000000000 | -22 TB | ffffeaffffffffff | 1 TB | virtual memory map (vmemmap_base)
ffffeb0000000000 | -21 TB | ffffebffffffffff | 1 TB | ... unused hole
ffffec0000000000 | -20 TB | fffffbffffffffff | 16 TB | KASAN shadow memory
+__________________|____________|__________________|_________|____________________________________________________________
+ |
+ | Identical layout to the 56-bit one from here on:
+____________________________________________________________|____________________________________________________________
+ | | | |
fffffc0000000000 | -4 TB | fffffdffffffffff | 2 TB | ... unused hole
| | | | vaddr_end for KASLR
fffffe0000000000 | -2 TB | fffffe7fffffffff | 0.5 TB | cpu_entry_area mapping
- fffffe8000000000 | -1.5 TB | fffffeffffffffff | 0.5 TB | LDT remap for PTI
+ fffffe8000000000 | -1.5 TB | fffffeffffffffff | 0.5 TB | ... unused hole
ffffff0000000000 | -1 TB | ffffff7fffffffff | 0.5 TB | %esp fixup stacks
-__________________|____________|__________________|_________|____________________________________________________________
- |
- | Identical layout to the 47-bit one from here on:
-____________________________________________________________|____________________________________________________________
- | | | |
ffffff8000000000 | -512 GB | ffffffeeffffffff | 444 GB | ... unused hole
ffffffef00000000 | -68 GB | fffffffeffffffff | 64 GB | EFI region mapping space
ffffffff00000000 | -4 GB | ffffffff7fffffff | 2 GB | ... unused hole
@@ -83,7 +84,7 @@
__________________|____________|__________________|_________|___________________________________________________________
| | | |
0000800000000000 | +64 PB | ffff7fffffffffff | ~16K PB | ... huge, still almost 64 bits wide hole of non-canonical
- | | | | virtual memory addresses up to the -128 TB
+ | | | | virtual memory addresses up to the -64 PB
| | | | starting offset of kernel mappings.
__________________|____________|__________________|_________|___________________________________________________________
|
@@ -91,23 +92,24 @@
____________________________________________________________|___________________________________________________________
| | | |
ff00000000000000 | -64 PB | ff0fffffffffffff | 4 PB | ... guard hole, also reserved for hypervisor
- ff10000000000000 | -60 PB | ff8fffffffffffff | 32 PB | direct mapping of all physical memory (page_offset_base)
- ff90000000000000 | -28 PB | ff9fffffffffffff | 4 PB | LDT remap for PTI
+ ff10000000000000 | -60 PB | ff10ffffffffffff | 0.25 PB | LDT remap for PTI
+ ff11000000000000 | -59.75 PB | ff90ffffffffffff | 32 PB | direct mapping of all physical memory (page_offset_base)
+ ff91000000000000 | -27.75 PB | ff9fffffffffffff | 3.75 PB | ... unused hole
ffa0000000000000 | -24 PB | ffd1ffffffffffff | 12.5 PB | vmalloc/ioremap space (vmalloc_base)
ffd2000000000000 | -11.5 PB | ffd3ffffffffffff | 0.5 PB | ... unused hole
ffd4000000000000 | -11 PB | ffd5ffffffffffff | 0.5 PB | virtual memory map (vmemmap_base)
ffd6000000000000 | -10.5 PB | ffdeffffffffffff | 2.25 PB | ... unused hole
ffdf000000000000 | -8.25 PB | fffffdffffffffff | ~8 PB | KASAN shadow memory
- fffffc0000000000 | -4 TB | fffffdffffffffff | 2 TB | ... unused hole
- | | | | vaddr_end for KASLR
- fffffe0000000000 | -2 TB | fffffe7fffffffff | 0.5 TB | cpu_entry_area mapping
- fffffe8000000000 | -1.5 TB | fffffeffffffffff | 0.5 TB | ... unused hole
- ffffff0000000000 | -1 TB | ffffff7fffffffff | 0.5 TB | %esp fixup stacks
__________________|____________|__________________|_________|____________________________________________________________
|
| Identical layout to the 47-bit one from here on:
____________________________________________________________|____________________________________________________________
| | | |
+ fffffc0000000000 | -4 TB | fffffdffffffffff | 2 TB | ... unused hole
+ | | | | vaddr_end for KASLR
+ fffffe0000000000 | -2 TB | fffffe7fffffffff | 0.5 TB | cpu_entry_area mapping
+ fffffe8000000000 | -1.5 TB | fffffeffffffffff | 0.5 TB | ... unused hole
+ ffffff0000000000 | -1 TB | ffffff7fffffffff | 0.5 TB | %esp fixup stacks
ffffff8000000000 | -512 GB | ffffffeeffffffff | 444 GB | ... unused hole
ffffffef00000000 | -68 GB | fffffffeffffffff | 64 GB | EFI region mapping space
ffffffff00000000 | -4 GB | ffffffff7fffffff | 2 GB | ... unused hole
diff --git a/Documentation/x86/zero-page.txt b/Documentation/x86/zero-page.txt
index 97b7adb..68aed07 100644
--- a/Documentation/x86/zero-page.txt
+++ b/Documentation/x86/zero-page.txt
@@ -25,7 +25,7 @@
0C8/004 ALL ext_cmd_line_ptr cmd_line_ptr high 32bits
140/080 ALL edid_info Video mode setup (struct edid_info)
1C0/020 ALL efi_info EFI 32 information (struct efi_info)
-1E0/004 ALL alk_mem_k Alternative mem check, in KB
+1E0/004 ALL alt_mem_k Alternative mem check, in KB
1E4/004 ALL scratch Scratch field for the kernel setup code
1E8/001 ALL e820_entries Number of entries in e820_table (below)
1E9/001 ALL eddbuf_entries Number of entries in eddbuf (below)
diff --git a/MAINTAINERS b/MAINTAINERS
index 4caac2f..1026150 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -6620,9 +6620,9 @@
HID CORE LAYER
M: Jiri Kosina <jikos@kernel.org>
-R: Benjamin Tissoires <benjamin.tissoires@redhat.com>
+M: Benjamin Tissoires <benjamin.tissoires@redhat.com>
L: linux-input@vger.kernel.org
-T: git git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid.git
+T: git git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid.git
S: Maintained
F: drivers/hid/
F: include/linux/hid*
@@ -6874,6 +6874,13 @@
S: Maintained
F: drivers/i2c/i2c-core-acpi.c
+I2C CONTROLLER DRIVER FOR NVIDIA GPU
+M: Ajay Gupta <ajayg@nvidia.com>
+L: linux-i2c@vger.kernel.org
+S: Maintained
+F: Documentation/i2c/busses/i2c-nvidia-gpu
+F: drivers/i2c/busses/i2c-nvidia-gpu.c
+
I2C MUXES
M: Peter Rosin <peda@axentia.se>
L: linux-i2c@vger.kernel.org
@@ -8380,7 +8387,7 @@
LIBATA PATA ARASAN COMPACT FLASH CONTROLLER
M: Viresh Kumar <vireshk@kernel.org>
L: linux-ide@vger.kernel.org
-T: git git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata.git
+T: git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
S: Maintained
F: include/linux/pata_arasan_cf_data.h
F: drivers/ata/pata_arasan_cf.c
@@ -8397,7 +8404,7 @@
LIBATA PATA FARADAY FTIDE010 AND GEMINI SATA BRIDGE DRIVERS
M: Linus Walleij <linus.walleij@linaro.org>
L: linux-ide@vger.kernel.org
-T: git git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata.git
+T: git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
S: Maintained
F: drivers/ata/pata_ftide010.c
F: drivers/ata/sata_gemini.c
@@ -8416,7 +8423,7 @@
LIBATA SATA PROMISE TX2/TX4 CONTROLLER DRIVER
M: Mikael Pettersson <mikpelinux@gmail.com>
L: linux-ide@vger.kernel.org
-T: git git://git.kernel.org/pub/scm/linux/kernel/git/tj/libata.git
+T: git git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git
S: Maintained
F: drivers/ata/sata_promise.*
@@ -10797,6 +10804,14 @@
S: Maintained
F: arch/arm/mach-omap2/omap_hwmod.*
+OMAP I2C DRIVER
+M: Vignesh R <vigneshr@ti.com>
+L: linux-omap@vger.kernel.org
+L: linux-i2c@vger.kernel.org
+S: Maintained
+F: Documentation/devicetree/bindings/i2c/i2c-omap.txt
+F: drivers/i2c/busses/i2c-omap.c
+
OMAP IMAGING SUBSYSTEM (OMAP3 ISP and OMAP4 ISS)
M: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
L: linux-media@vger.kernel.org
@@ -10806,9 +10821,9 @@
F: drivers/staging/media/omap4iss/
OMAP MMC SUPPORT
-M: Jarkko Lavinen <jarkko.lavinen@nokia.com>
+M: Aaro Koskinen <aaro.koskinen@iki.fi>
L: linux-omap@vger.kernel.org
-S: Maintained
+S: Odd Fixes
F: drivers/mmc/host/omap.c
OMAP POWER MANAGEMENT SUPPORT
@@ -11743,6 +11758,7 @@
PIN CONTROLLER - INTEL
M: Mika Westerberg <mika.westerberg@linux.intel.com>
M: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
+T: git git://git.kernel.org/pub/scm/linux/kernel/git/pinctrl/intel.git
S: Maintained
F: drivers/pinctrl/intel/
@@ -15449,9 +15465,9 @@
USB HID/HIDBP DRIVERS (USB KEYBOARDS, MICE, REMOTE CONTROLS, ...)
M: Jiri Kosina <jikos@kernel.org>
-R: Benjamin Tissoires <benjamin.tissoires@redhat.com>
+M: Benjamin Tissoires <benjamin.tissoires@redhat.com>
L: linux-usb@vger.kernel.org
-T: git git://git.kernel.org/pub/scm/linux/kernel/git/jikos/hid.git
+T: git git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid.git
S: Maintained
F: Documentation/hid/hiddev.txt
F: drivers/hid/usbhid/
diff --git a/Makefile b/Makefile
index 9fce8b9..ddbf627 100644
--- a/Makefile
+++ b/Makefile
@@ -2,7 +2,7 @@
VERSION = 4
PATCHLEVEL = 20
SUBLEVEL = 0
-EXTRAVERSION = -rc1
+EXTRAVERSION = -rc3
NAME = "People's Front"
# *DOCUMENTATION*
diff --git a/arch/alpha/include/asm/termios.h b/arch/alpha/include/asm/termios.h
index 6a8c53d..b7c77bb 100644
--- a/arch/alpha/include/asm/termios.h
+++ b/arch/alpha/include/asm/termios.h
@@ -73,9 +73,15 @@
})
#define user_termios_to_kernel_termios(k, u) \
- copy_from_user(k, u, sizeof(struct termios))
+ copy_from_user(k, u, sizeof(struct termios2))
#define kernel_termios_to_user_termios(u, k) \
+ copy_to_user(u, k, sizeof(struct termios2))
+
+#define user_termios_to_kernel_termios_1(k, u) \
+ copy_from_user(k, u, sizeof(struct termios))
+
+#define kernel_termios_to_user_termios_1(u, k) \
copy_to_user(u, k, sizeof(struct termios))
#endif /* _ALPHA_TERMIOS_H */
diff --git a/arch/alpha/include/uapi/asm/ioctls.h b/arch/alpha/include/uapi/asm/ioctls.h
index 1e9121c..9713116 100644
--- a/arch/alpha/include/uapi/asm/ioctls.h
+++ b/arch/alpha/include/uapi/asm/ioctls.h
@@ -32,6 +32,11 @@
#define TCXONC _IO('t', 30)
#define TCFLSH _IO('t', 31)
+#define TCGETS2 _IOR('T', 42, struct termios2)
+#define TCSETS2 _IOW('T', 43, struct termios2)
+#define TCSETSW2 _IOW('T', 44, struct termios2)
+#define TCSETSF2 _IOW('T', 45, struct termios2)
+
#define TIOCSWINSZ _IOW('t', 103, struct winsize)
#define TIOCGWINSZ _IOR('t', 104, struct winsize)
#define TIOCSTART _IO('t', 110) /* start output, like ^Q */
diff --git a/arch/alpha/include/uapi/asm/termbits.h b/arch/alpha/include/uapi/asm/termbits.h
index de6c836..4575ba3 100644
--- a/arch/alpha/include/uapi/asm/termbits.h
+++ b/arch/alpha/include/uapi/asm/termbits.h
@@ -26,6 +26,19 @@ struct termios {
speed_t c_ospeed; /* output speed */
};
+/* Alpha has identical termios and termios2 */
+
+struct termios2 {
+ tcflag_t c_iflag; /* input mode flags */
+ tcflag_t c_oflag; /* output mode flags */
+ tcflag_t c_cflag; /* control mode flags */
+ tcflag_t c_lflag; /* local mode flags */
+ cc_t c_cc[NCCS]; /* control characters */
+ cc_t c_line; /* line discipline (== c_cc[19]) */
+ speed_t c_ispeed; /* input speed */
+ speed_t c_ospeed; /* output speed */
+};
+
/* Alpha has matching termios and ktermios */
struct ktermios {
@@ -152,6 +165,7 @@ struct ktermios {
#define B3000000 00034
#define B3500000 00035
#define B4000000 00036
+#define BOTHER 00037
#define CSIZE 00001400
#define CS5 00000000
@@ -169,6 +183,9 @@ struct ktermios {
#define CMSPAR 010000000000 /* mark or space (stick) parity */
#define CRTSCTS 020000000000 /* flow control */
+#define CIBAUD 07600000
+#define IBSHIFT 16
+
/* c_lflag bits */
#define ISIG 0x00000080
#define ICANON 0x00000100
diff --git a/arch/arm/boot/dts/imx53-ppd.dts b/arch/arm/boot/dts/imx53-ppd.dts
index b560ff8..5ff9a17 100644
--- a/arch/arm/boot/dts/imx53-ppd.dts
+++ b/arch/arm/boot/dts/imx53-ppd.dts
@@ -55,7 +55,7 @@
};
chosen {
- stdout-path = "&uart1:115200n8";
+ stdout-path = "serial0:115200n8";
};
memory@70000000 {
diff --git a/arch/arm/boot/dts/imx6sll.dtsi b/arch/arm/boot/dts/imx6sll.dtsi
index ed9a980..beefa1b 100644
--- a/arch/arm/boot/dts/imx6sll.dtsi
+++ b/arch/arm/boot/dts/imx6sll.dtsi
@@ -740,7 +740,7 @@
i2c1: i2c@21a0000 {
#address-cells = <1>;
#size-cells = <0>;
- compatible = "fs,imx6sll-i2c", "fsl,imx21-i2c";
+ compatible = "fsl,imx6sll-i2c", "fsl,imx21-i2c";
reg = <0x021a0000 0x4000>;
interrupts = <GIC_SPI 36 IRQ_TYPE_LEVEL_HIGH>;
clocks = <&clks IMX6SLL_CLK_I2C1>;
diff --git a/arch/arm/boot/dts/imx6sx-sdb.dtsi b/arch/arm/boot/dts/imx6sx-sdb.dtsi
index 53b3408..7d7d679 100644
--- a/arch/arm/boot/dts/imx6sx-sdb.dtsi
+++ b/arch/arm/boot/dts/imx6sx-sdb.dtsi
@@ -117,7 +117,9 @@
regulator-name = "enet_3v3";
regulator-min-microvolt = <3300000>;
regulator-max-microvolt = <3300000>;
- gpios = <&gpio2 6 GPIO_ACTIVE_LOW>;
+ gpio = <&gpio2 6 GPIO_ACTIVE_LOW>;
+ regulator-boot-on;
+ regulator-always-on;
};
reg_pcie_gpio: regulator-pcie-gpio {
@@ -180,6 +182,7 @@
phy-supply = <®_enet_3v3>;
phy-mode = "rgmii";
phy-handle = <ðphy1>;
+ phy-reset-gpios = <&gpio2 7 GPIO_ACTIVE_LOW>;
status = "okay";
mdio {
@@ -373,6 +376,8 @@
MX6SX_PAD_RGMII1_RD3__ENET1_RX_DATA_3 0x3081
MX6SX_PAD_RGMII1_RX_CTL__ENET1_RX_EN 0x3081
MX6SX_PAD_ENET2_RX_CLK__ENET2_REF_CLK_25M 0x91
+ /* phy reset */
+ MX6SX_PAD_ENET2_CRS__GPIO2_IO_7 0x10b0
>;
};
diff --git a/arch/arm/boot/dts/vf610m4-colibri.dts b/arch/arm/boot/dts/vf610m4-colibri.dts
index 41ec66a..ca62495 100644
--- a/arch/arm/boot/dts/vf610m4-colibri.dts
+++ b/arch/arm/boot/dts/vf610m4-colibri.dts
@@ -50,8 +50,8 @@
compatible = "fsl,vf610m4";
chosen {
- bootargs = "console=ttyLP2,115200 clk_ignore_unused init=/linuxrc rw";
- stdout-path = "&uart2";
+ bootargs = "clk_ignore_unused init=/linuxrc rw";
+ stdout-path = "serial2:115200";
};
memory@8c000000 {
diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig
index 1c76168..63af623 100644
--- a/arch/arm/configs/multi_v7_defconfig
+++ b/arch/arm/configs/multi_v7_defconfig
@@ -1,7 +1,6 @@
CONFIG_SYSVIPC=y
CONFIG_NO_HZ=y
CONFIG_HIGH_RES_TIMERS=y
-CONFIG_PREEMPT=y
CONFIG_CGROUPS=y
CONFIG_BLK_DEV_INITRD=y
CONFIG_EMBEDDED=y
diff --git a/arch/arm/include/asm/cputype.h b/arch/arm/include/asm/cputype.h
index 0d28924..775cac3 100644
--- a/arch/arm/include/asm/cputype.h
+++ b/arch/arm/include/asm/cputype.h
@@ -111,6 +111,7 @@
#include <linux/kernel.h>
extern unsigned int processor_id;
+struct proc_info_list *lookup_processor(u32 midr);
#ifdef CONFIG_CPU_CP15
#define read_cpuid(reg) \
diff --git a/arch/arm/include/asm/pgtable-2level.h b/arch/arm/include/asm/pgtable-2level.h
index 92fd2c8..12659ce 100644
--- a/arch/arm/include/asm/pgtable-2level.h
+++ b/arch/arm/include/asm/pgtable-2level.h
@@ -10,7 +10,7 @@
#ifndef _ASM_PGTABLE_2LEVEL_H
#define _ASM_PGTABLE_2LEVEL_H
-#define __PAGETABLE_PMD_FOLDED
+#define __PAGETABLE_PMD_FOLDED 1
/*
* Hardware-wise, we have a two level page table structure, where the first
diff --git a/arch/arm/include/asm/proc-fns.h b/arch/arm/include/asm/proc-fns.h
index e25f439..e1b6f28 100644
--- a/arch/arm/include/asm/proc-fns.h
+++ b/arch/arm/include/asm/proc-fns.h
@@ -23,7 +23,7 @@ struct mm_struct;
/*
* Don't change this structure - ASM code relies on it.
*/
-extern struct processor {
+struct processor {
/* MISC
* get data abort address/flags
*/
@@ -79,9 +79,13 @@ extern struct processor {
unsigned int suspend_size;
void (*do_suspend)(void *);
void (*do_resume)(void *);
-} processor;
+};
#ifndef MULTI_CPU
+static inline void init_proc_vtable(const struct processor *p)
+{
+}
+
extern void cpu_proc_init(void);
extern void cpu_proc_fin(void);
extern int cpu_do_idle(void);
@@ -98,17 +102,50 @@ extern void cpu_reset(unsigned long addr, bool hvc) __attribute__((noreturn));
extern void cpu_do_suspend(void *);
extern void cpu_do_resume(void *);
#else
-#define cpu_proc_init processor._proc_init
-#define cpu_proc_fin processor._proc_fin
-#define cpu_reset processor.reset
-#define cpu_do_idle processor._do_idle
-#define cpu_dcache_clean_area processor.dcache_clean_area
-#define cpu_set_pte_ext processor.set_pte_ext
-#define cpu_do_switch_mm processor.switch_mm
-/* These three are private to arch/arm/kernel/suspend.c */
-#define cpu_do_suspend processor.do_suspend
-#define cpu_do_resume processor.do_resume
+extern struct processor processor;
+#if defined(CONFIG_BIG_LITTLE) && defined(CONFIG_HARDEN_BRANCH_PREDICTOR)
+#include <linux/smp.h>
+/*
+ * This can't be a per-cpu variable because we need to access it before
+ * per-cpu has been initialised. We have a couple of functions that are
+ * called in a pre-emptible context, and so can't use smp_processor_id()
+ * there, hence PROC_TABLE(). We insist in init_proc_vtable() that the
+ * function pointers for these are identical across all CPUs.
+ */
+extern struct processor *cpu_vtable[];
+#define PROC_VTABLE(f) cpu_vtable[smp_processor_id()]->f
+#define PROC_TABLE(f) cpu_vtable[0]->f
+static inline void init_proc_vtable(const struct processor *p)
+{
+ unsigned int cpu = smp_processor_id();
+ *cpu_vtable[cpu] = *p;
+ WARN_ON_ONCE(cpu_vtable[cpu]->dcache_clean_area !=
+ cpu_vtable[0]->dcache_clean_area);
+ WARN_ON_ONCE(cpu_vtable[cpu]->set_pte_ext !=
+ cpu_vtable[0]->set_pte_ext);
+}
+#else
+#define PROC_VTABLE(f) processor.f
+#define PROC_TABLE(f) processor.f
+static inline void init_proc_vtable(const struct processor *p)
+{
+ processor = *p;
+}
+#endif
+
+#define cpu_proc_init PROC_VTABLE(_proc_init)
+#define cpu_check_bugs PROC_VTABLE(check_bugs)
+#define cpu_proc_fin PROC_VTABLE(_proc_fin)
+#define cpu_reset PROC_VTABLE(reset)
+#define cpu_do_idle PROC_VTABLE(_do_idle)
+#define cpu_dcache_clean_area PROC_TABLE(dcache_clean_area)
+#define cpu_set_pte_ext PROC_TABLE(set_pte_ext)
+#define cpu_do_switch_mm PROC_VTABLE(switch_mm)
+
+/* These two are private to arch/arm/kernel/suspend.c */
+#define cpu_do_suspend PROC_VTABLE(do_suspend)
+#define cpu_do_resume PROC_VTABLE(do_resume)
#endif
extern void cpu_resume(void);
diff --git a/arch/arm/kernel/bugs.c b/arch/arm/kernel/bugs.c
index 7be5113..d41d359 100644
--- a/arch/arm/kernel/bugs.c
+++ b/arch/arm/kernel/bugs.c
@@ -6,8 +6,8 @@
void check_other_bugs(void)
{
#ifdef MULTI_CPU
- if (processor.check_bugs)
- processor.check_bugs();
+ if (cpu_check_bugs)
+ cpu_check_bugs();
#endif
}
diff --git a/arch/arm/kernel/head-common.S b/arch/arm/kernel/head-common.S
index 6e0375e..997b023 100644
--- a/arch/arm/kernel/head-common.S
+++ b/arch/arm/kernel/head-common.S
@@ -145,6 +145,9 @@
#endif
.size __mmap_switched_data, . - __mmap_switched_data
+ __FINIT
+ .text
+
/*
* This provides a C-API version of __lookup_processor_type
*/
@@ -156,9 +159,6 @@
ldmfd sp!, {r4 - r6, r9, pc}
ENDPROC(lookup_processor_type)
- __FINIT
- .text
-
/*
* Read processor ID register (CP#15, CR0), and look up in the linker-built
* supported processor list. Note that we can't use the absolute addresses
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index ac7e088..375b13f 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -114,6 +114,11 @@ EXPORT_SYMBOL(elf_hwcap2);
#ifdef MULTI_CPU
struct processor processor __ro_after_init;
+#if defined(CONFIG_BIG_LITTLE) && defined(CONFIG_HARDEN_BRANCH_PREDICTOR)
+struct processor *cpu_vtable[NR_CPUS] = {
+ [0] = &processor,
+};
+#endif
#endif
#ifdef MULTI_TLB
struct cpu_tlb_fns cpu_tlb __ro_after_init;
@@ -666,28 +671,33 @@ static void __init smp_build_mpidr_hash(void)
}
#endif
+/*
+ * locate processor in the list of supported processor types. The linker
+ * builds this table for us from the entries in arch/arm/mm/proc-*.S
+ */
+struct proc_info_list *lookup_processor(u32 midr)
+{
+ struct proc_info_list *list = lookup_processor_type(midr);
+
+ if (!list) {
+ pr_err("CPU%u: configuration botched (ID %08x), CPU halted\n",
+ smp_processor_id(), midr);
+ while (1)
+ /* can't use cpu_relax() here as it may require MMU setup */;
+ }
+
+ return list;
+}
+
static void __init setup_processor(void)
{
- struct proc_info_list *list;
-
- /*
- * locate processor in the list of supported processor
- * types. The linker builds this table for us from the
- * entries in arch/arm/mm/proc-*.S
- */
- list = lookup_processor_type(read_cpuid_id());
- if (!list) {
- pr_err("CPU configuration botched (ID %08x), unable to continue.\n",
- read_cpuid_id());
- while (1);
- }
+ unsigned int midr = read_cpuid_id();
+ struct proc_info_list *list = lookup_processor(midr);
cpu_name = list->cpu_name;
__cpu_architecture = __get_cpu_architecture();
-#ifdef MULTI_CPU
- processor = *list->proc;
-#endif
+ init_proc_vtable(list->proc);
#ifdef MULTI_TLB
cpu_tlb = *list->tlb;
#endif
@@ -699,7 +709,7 @@ static void __init setup_processor(void)
#endif
pr_info("CPU: %s [%08x] revision %d (ARMv%s), cr=%08lx\n",
- cpu_name, read_cpuid_id(), read_cpuid_id() & 15,
+ list->cpu_name, midr, midr & 15,
proc_arch[cpu_architecture()], get_cr());
snprintf(init_utsname()->machine, __NEW_UTS_LEN + 1, "%s%c",
diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
index 0978282..12a6172 100644
--- a/arch/arm/kernel/smp.c
+++ b/arch/arm/kernel/smp.c
@@ -42,6 +42,7 @@
#include <asm/mmu_context.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
+#include <asm/procinfo.h>
#include <asm/processor.h>
#include <asm/sections.h>
#include <asm/tlbflush.h>
@@ -102,6 +103,30 @@ static unsigned long get_arch_pgd(pgd_t *pgd)
#endif
}
+#if defined(CONFIG_BIG_LITTLE) && defined(CONFIG_HARDEN_BRANCH_PREDICTOR)
+static int secondary_biglittle_prepare(unsigned int cpu)
+{
+ if (!cpu_vtable[cpu])
+ cpu_vtable[cpu] = kzalloc(sizeof(*cpu_vtable[cpu]), GFP_KERNEL);
+
+ return cpu_vtable[cpu] ? 0 : -ENOMEM;
+}
+
+static void secondary_biglittle_init(void)
+{
+ init_proc_vtable(lookup_processor(read_cpuid_id())->proc);
+}
+#else
+static int secondary_biglittle_prepare(unsigned int cpu)
+{
+ return 0;
+}
+
+static void secondary_biglittle_init(void)
+{
+}
+#endif
+
int __cpu_up(unsigned int cpu, struct task_struct *idle)
{
int ret;
@@ -109,6 +134,10 @@ int __cpu_up(unsigned int cpu, struct task_struct *idle)
if (!smp_ops.smp_boot_secondary)
return -ENOSYS;
+ ret = secondary_biglittle_prepare(cpu);
+ if (ret)
+ return ret;
+
/*
* We need to tell the secondary core where to find
* its stack and the page tables.
@@ -359,6 +388,8 @@ asmlinkage void secondary_start_kernel(void)
struct mm_struct *mm = &init_mm;
unsigned int cpu;
+ secondary_biglittle_init();
+
/*
* The identity mapping is uncached (strongly ordered), so
* switch away from it before attempting any exclusive accesses.
diff --git a/arch/arm/mach-omap2/display.c b/arch/arm/mach-omap2/display.c
index 9500b6e..f86b72d 100644
--- a/arch/arm/mach-omap2/display.c
+++ b/arch/arm/mach-omap2/display.c
@@ -209,11 +209,61 @@ static int __init omapdss_init_fbdev(void)
return 0;
}
-#else
-static inline int omapdss_init_fbdev(void)
+
+static const char * const omapdss_compat_names[] __initconst = {
+ "ti,omap2-dss",
+ "ti,omap3-dss",
+ "ti,omap4-dss",
+ "ti,omap5-dss",
+ "ti,dra7-dss",
+};
+
+static struct device_node * __init omapdss_find_dss_of_node(void)
{
- return 0;
+ struct device_node *node;
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(omapdss_compat_names); ++i) {
+ node = of_find_compatible_node(NULL, NULL,
+ omapdss_compat_names[i]);
+ if (node)
+ return node;
+ }
+
+ return NULL;
}
+
+static int __init omapdss_init_of(void)
+{
+ int r;
+ struct device_node *node;
+ struct platform_device *pdev;
+
+ /* only create dss helper devices if dss is enabled in the .dts */
+
+ node = omapdss_find_dss_of_node();
+ if (!node)
+ return 0;
+
+ if (!of_device_is_available(node))
+ return 0;
+
+ pdev = of_find_device_by_node(node);
+
+ if (!pdev) {
+ pr_err("Unable to find DSS platform device\n");
+ return -ENODEV;
+ }
+
+ r = of_platform_populate(node, NULL, NULL, &pdev->dev);
+ if (r) {
+ pr_err("Unable to populate DSS submodule devices\n");
+ return r;
+ }
+
+ return omapdss_init_fbdev();
+}
+omap_device_initcall(omapdss_init_of);
#endif /* CONFIG_FB_OMAP2 */
static void dispc_disable_outputs(void)
@@ -361,58 +411,3 @@ int omap_dss_reset(struct omap_hwmod *oh)
return r;
}
-
-static const char * const omapdss_compat_names[] __initconst = {
- "ti,omap2-dss",
- "ti,omap3-dss",
- "ti,omap4-dss",
- "ti,omap5-dss",
- "ti,dra7-dss",
-};
-
-static struct device_node * __init omapdss_find_dss_of_node(void)
-{
- struct device_node *node;
- int i;
-
- for (i = 0; i < ARRAY_SIZE(omapdss_compat_names); ++i) {
- node = of_find_compatible_node(NULL, NULL,
- omapdss_compat_names[i]);
- if (node)
- return node;
- }
-
- return NULL;
-}
-
-static int __init omapdss_init_of(void)
-{
- int r;
- struct device_node *node;
- struct platform_device *pdev;
-
- /* only create dss helper devices if dss is enabled in the .dts */
-
- node = omapdss_find_dss_of_node();
- if (!node)
- return 0;
-
- if (!of_device_is_available(node))
- return 0;
-
- pdev = of_find_device_by_node(node);
-
- if (!pdev) {
- pr_err("Unable to find DSS platform device\n");
- return -ENODEV;
- }
-
- r = of_platform_populate(node, NULL, NULL, &pdev->dev);
- if (r) {
- pr_err("Unable to populate DSS submodule devices\n");
- return r;
- }
-
- return omapdss_init_fbdev();
-}
-omap_device_initcall(omapdss_init_of);
diff --git a/arch/arm/mm/proc-v7-bugs.c b/arch/arm/mm/proc-v7-bugs.c
index 5544b82..9a07916 100644
--- a/arch/arm/mm/proc-v7-bugs.c
+++ b/arch/arm/mm/proc-v7-bugs.c
@@ -52,8 +52,6 @@ static void cpu_v7_spectre_init(void)
case ARM_CPU_PART_CORTEX_A17:
case ARM_CPU_PART_CORTEX_A73:
case ARM_CPU_PART_CORTEX_A75:
- if (processor.switch_mm != cpu_v7_bpiall_switch_mm)
- goto bl_error;
per_cpu(harden_branch_predictor_fn, cpu) =
harden_branch_predictor_bpiall;
spectre_v2_method = "BPIALL";
@@ -61,8 +59,6 @@ static void cpu_v7_spectre_init(void)
case ARM_CPU_PART_CORTEX_A15:
case ARM_CPU_PART_BRAHMA_B15:
- if (processor.switch_mm != cpu_v7_iciallu_switch_mm)
- goto bl_error;
per_cpu(harden_branch_predictor_fn, cpu) =
harden_branch_predictor_iciallu;
spectre_v2_method = "ICIALLU";
@@ -88,11 +84,9 @@ static void cpu_v7_spectre_init(void)
ARM_SMCCC_ARCH_WORKAROUND_1, &res);
if ((int)res.a0 != 0)
break;
- if (processor.switch_mm != cpu_v7_hvc_switch_mm && cpu)
- goto bl_error;
per_cpu(harden_branch_predictor_fn, cpu) =
call_hvc_arch_workaround_1;
- processor.switch_mm = cpu_v7_hvc_switch_mm;
+ cpu_do_switch_mm = cpu_v7_hvc_switch_mm;
spectre_v2_method = "hypervisor";
break;
@@ -101,11 +95,9 @@ static void cpu_v7_spectre_init(void)
ARM_SMCCC_ARCH_WORKAROUND_1, &res);
if ((int)res.a0 != 0)
break;
- if (processor.switch_mm != cpu_v7_smc_switch_mm && cpu)
- goto bl_error;
per_cpu(harden_branch_predictor_fn, cpu) =
call_smc_arch_workaround_1;
- processor.switch_mm = cpu_v7_smc_switch_mm;
+ cpu_do_switch_mm = cpu_v7_smc_switch_mm;
spectre_v2_method = "firmware";
break;
@@ -119,11 +111,6 @@ static void cpu_v7_spectre_init(void)
if (spectre_v2_method)
pr_info("CPU%u: Spectre v2: using %s workaround\n",
smp_processor_id(), spectre_v2_method);
- return;
-
-bl_error:
- pr_err("CPU%u: Spectre v2: incorrect context switching function, system vulnerable\n",
- cpu);
}
#else
static void cpu_v7_spectre_init(void)
diff --git a/arch/arm/mm/proc-v7.S b/arch/arm/mm/proc-v7.S
index 6fe52819..339eb17 100644
--- a/arch/arm/mm/proc-v7.S
+++ b/arch/arm/mm/proc-v7.S
@@ -112,7 +112,7 @@
hvc #0
ldmfd sp!, {r0 - r3}
b cpu_v7_switch_mm
-ENDPROC(cpu_v7_smc_switch_mm)
+ENDPROC(cpu_v7_hvc_switch_mm)
#endif
ENTRY(cpu_v7_iciallu_switch_mm)
mov r3, #0
diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c
index aff6e6e..ee7b079 100644
--- a/arch/arm/vfp/vfpmodule.c
+++ b/arch/arm/vfp/vfpmodule.c
@@ -573,7 +573,7 @@ int vfp_preserve_user_clear_hwstate(struct user_vfp *ufp,
*/
ufp_exc->fpexc = hwstate->fpexc;
ufp_exc->fpinst = hwstate->fpinst;
- ufp_exc->fpinst2 = ufp_exc->fpinst2;
+ ufp_exc->fpinst2 = hwstate->fpinst2;
/* Ensure that VFP is disabled. */
vfp_flush_hwstate(thread);
diff --git a/arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi b/arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi
index 8253a1a..fef7351 100644
--- a/arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi
+++ b/arch/arm64/boot/dts/altera/socfpga_stratix10.dtsi
@@ -139,6 +139,7 @@
clock-names = "stmmaceth";
tx-fifo-depth = <16384>;
rx-fifo-depth = <16384>;
+ snps,multicast-filter-bins = <256>;
status = "disabled";
};
@@ -154,6 +155,7 @@
clock-names = "stmmaceth";
tx-fifo-depth = <16384>;
rx-fifo-depth = <16384>;
+ snps,multicast-filter-bins = <256>;
status = "disabled";
};
@@ -169,6 +171,7 @@
clock-names = "stmmaceth";
tx-fifo-depth = <16384>;
rx-fifo-depth = <16384>;
+ snps,multicast-filter-bins = <256>;
status = "disabled";
};
diff --git a/arch/arm64/boot/dts/renesas/r8a7795.dtsi b/arch/arm64/boot/dts/renesas/r8a7795.dtsi
index b5f2273..a79c8d3 100644
--- a/arch/arm64/boot/dts/renesas/r8a7795.dtsi
+++ b/arch/arm64/boot/dts/renesas/r8a7795.dtsi
@@ -652,7 +652,7 @@
clock-names = "fck", "brg_int", "scif_clk";
dmas = <&dmac1 0x35>, <&dmac1 0x34>,
<&dmac2 0x35>, <&dmac2 0x34>;
- dma-names = "tx", "rx";
+ dma-names = "tx", "rx", "tx", "rx";
power-domains = <&sysc R8A7795_PD_ALWAYS_ON>;
resets = <&cpg 518>;
status = "disabled";
diff --git a/arch/arm64/boot/dts/renesas/r8a77980-condor.dts b/arch/arm64/boot/dts/renesas/r8a77980-condor.dts
index fe2e2c0..5a7012b 100644
--- a/arch/arm64/boot/dts/renesas/r8a77980-condor.dts
+++ b/arch/arm64/boot/dts/renesas/r8a77980-condor.dts
@@ -15,7 +15,7 @@
aliases {
serial0 = &scif0;
- ethernet0 = &avb;
+ ethernet0 = &gether;
};
chosen {
@@ -97,23 +97,6 @@
};
};
-&avb {
- pinctrl-0 = <&avb_pins>;
- pinctrl-names = "default";
-
- phy-mode = "rgmii-id";
- phy-handle = <&phy0>;
- renesas,no-ether-link;
- status = "okay";
-
- phy0: ethernet-phy@0 {
- rxc-skew-ps = <1500>;
- reg = <0>;
- interrupt-parent = <&gpio1>;
- interrupts = <17 IRQ_TYPE_LEVEL_LOW>;
- };
-};
-
&canfd {
pinctrl-0 = <&canfd0_pins>;
pinctrl-names = "default";
@@ -139,6 +122,23 @@
clock-frequency = <32768>;
};
+&gether {
+ pinctrl-0 = <&gether_pins>;
+ pinctrl-names = "default";
+
+ phy-mode = "rgmii-id";
+ phy-handle = <&phy0>;
+ renesas,no-ether-link;
+ status = "okay";
+
+ phy0: ethernet-phy@0 {
+ rxc-skew-ps = <1500>;
+ reg = <0>;
+ interrupt-parent = <&gpio4>;
+ interrupts = <23 IRQ_TYPE_LEVEL_LOW>;
+ };
+};
+
&i2c0 {
pinctrl-0 = <&i2c0_pins>;
pinctrl-names = "default";
@@ -236,16 +236,17 @@
};
&pfc {
- avb_pins: avb {
- groups = "avb_mdio", "avb_rgmii";
- function = "avb";
- };
-
canfd0_pins: canfd0 {
groups = "canfd0_data_a";
function = "canfd0";
};
+ gether_pins: gether {
+ groups = "gether_mdio_a", "gether_rgmii",
+ "gether_txcrefclk", "gether_txcrefclk_mega";
+ function = "gether";
+ };
+
i2c0_pins: i2c0 {
groups = "i2c0";
function = "i2c0";
diff --git a/arch/arm64/include/asm/processor.h b/arch/arm64/include/asm/processor.h
index 3e20917..6b0d4df 100644
--- a/arch/arm64/include/asm/processor.h
+++ b/arch/arm64/include/asm/processor.h
@@ -24,6 +24,14 @@
#define KERNEL_DS UL(-1)
#define USER_DS (TASK_SIZE_64 - 1)
+/*
+ * On arm64 systems, unaligned accesses by the CPU are cheap, and so there is
+ * no point in shifting all network buffers by 2 bytes just to make some IP
+ * header fields appear aligned in memory, potentially sacrificing some DMA
+ * performance on some platforms.
+ */
+#define NET_IP_ALIGN 0
+
#ifndef __ASSEMBLY__
#ifdef __KERNEL__
diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c
index 953e316..f4fc1e0 100644
--- a/arch/arm64/kernel/setup.c
+++ b/arch/arm64/kernel/setup.c
@@ -313,6 +313,7 @@ void __init setup_arch(char **cmdline_p)
arm64_memblock_init();
paging_init();
+ efi_apply_persistent_mem_reservations();
acpi_table_upgrade();
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index 9d9582c..9b432d9 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -483,8 +483,6 @@ void __init arm64_memblock_init(void)
high_memory = __va(memblock_end_of_DRAM() - 1) + 1;
dma_contiguous_reserve(arm64_dma_phys_limit);
-
- memblock_allow_resize();
}
void __init bootmem_init(void)
diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
index 394b8d55..d1d6601 100644
--- a/arch/arm64/mm/mmu.c
+++ b/arch/arm64/mm/mmu.c
@@ -659,6 +659,8 @@ void __init paging_init(void)
memblock_free(__pa_symbol(init_pg_dir),
__pa_symbol(init_pg_end) - __pa_symbol(init_pg_dir));
+
+ memblock_allow_resize();
}
/*
diff --git a/arch/m68k/include/asm/pgtable_mm.h b/arch/m68k/include/asm/pgtable_mm.h
index 6181e41..fe3ddd7 100644
--- a/arch/m68k/include/asm/pgtable_mm.h
+++ b/arch/m68k/include/asm/pgtable_mm.h
@@ -55,12 +55,12 @@
*/
#ifdef CONFIG_SUN3
#define PTRS_PER_PTE 16
-#define __PAGETABLE_PMD_FOLDED
+#define __PAGETABLE_PMD_FOLDED 1
#define PTRS_PER_PMD 1
#define PTRS_PER_PGD 2048
#elif defined(CONFIG_COLDFIRE)
#define PTRS_PER_PTE 512
-#define __PAGETABLE_PMD_FOLDED
+#define __PAGETABLE_PMD_FOLDED 1
#define PTRS_PER_PMD 1
#define PTRS_PER_PGD 1024
#else
diff --git a/arch/microblaze/include/asm/pgtable.h b/arch/microblaze/include/asm/pgtable.h
index f64ebb9..e14b662 100644
--- a/arch/microblaze/include/asm/pgtable.h
+++ b/arch/microblaze/include/asm/pgtable.h
@@ -63,7 +63,7 @@ extern int mem_init_done;
#include <asm-generic/4level-fixup.h>
-#define __PAGETABLE_PMD_FOLDED
+#define __PAGETABLE_PMD_FOLDED 1
#ifdef __KERNEL__
#ifndef __ASSEMBLY__
diff --git a/arch/mips/cavium-octeon/executive/cvmx-helper.c b/arch/mips/cavium-octeon/executive/cvmx-helper.c
index 75108ec..6c79e8a 100644
--- a/arch/mips/cavium-octeon/executive/cvmx-helper.c
+++ b/arch/mips/cavium-octeon/executive/cvmx-helper.c
@@ -67,7 +67,7 @@ void (*cvmx_override_pko_queue_priority) (int pko_port,
void (*cvmx_override_ipd_port_setup) (int ipd_port);
/* Port count per interface */
-static int interface_port_count[5];
+static int interface_port_count[9];
/**
* Return the number of interfaces the chip has. Each interface
diff --git a/arch/mips/mm/dma-noncoherent.c b/arch/mips/mm/dma-noncoherent.c
index e6c9485..cb38461 100644
--- a/arch/mips/mm/dma-noncoherent.c
+++ b/arch/mips/mm/dma-noncoherent.c
@@ -50,7 +50,7 @@ void *arch_dma_alloc(struct device *dev, size_t size,
void *ret;
ret = dma_direct_alloc_pages(dev, size, dma_handle, gfp, attrs);
- if (!ret && !(attrs & DMA_ATTR_NON_CONSISTENT)) {
+ if (ret && !(attrs & DMA_ATTR_NON_CONSISTENT)) {
dma_cache_wback_inv((unsigned long) ret, size);
ret = (void *)UNCAC_ADDR(ret);
}
diff --git a/arch/nds32/include/asm/pgtable.h b/arch/nds32/include/asm/pgtable.h
index d3e19a5..9f52db9 100644
--- a/arch/nds32/include/asm/pgtable.h
+++ b/arch/nds32/include/asm/pgtable.h
@@ -4,7 +4,7 @@
#ifndef _ASMNDS32_PGTABLE_H
#define _ASMNDS32_PGTABLE_H
-#define __PAGETABLE_PMD_FOLDED
+#define __PAGETABLE_PMD_FOLDED 1
#include <asm-generic/4level-fixup.h>
#include <asm-generic/sizes.h>
diff --git a/arch/parisc/include/asm/pgtable.h b/arch/parisc/include/asm/pgtable.h
index b941ac7d..c7bb74e 100644
--- a/arch/parisc/include/asm/pgtable.h
+++ b/arch/parisc/include/asm/pgtable.h
@@ -111,7 +111,7 @@ static inline void purge_tlb_entries(struct mm_struct *mm, unsigned long addr)
#if CONFIG_PGTABLE_LEVELS == 3
#define BITS_PER_PMD (PAGE_SHIFT + PMD_ORDER - BITS_PER_PMD_ENTRY)
#else
-#define __PAGETABLE_PMD_FOLDED
+#define __PAGETABLE_PMD_FOLDED 1
#define BITS_PER_PMD 0
#endif
#define PTRS_PER_PMD (1UL << BITS_PER_PMD)
diff --git a/arch/parisc/include/asm/spinlock.h b/arch/parisc/include/asm/spinlock.h
index 16aec9b..8a63515 100644
--- a/arch/parisc/include/asm/spinlock.h
+++ b/arch/parisc/include/asm/spinlock.h
@@ -37,8 +37,8 @@ static inline void arch_spin_unlock(arch_spinlock_t *x)
volatile unsigned int *a;
a = __ldcw_align(x);
- /* Release with ordered store. */
- __asm__ __volatile__("stw,ma %0,0(%1)" : : "r"(1), "r"(a) : "memory");
+ mb();
+ *a = 1;
}
static inline int arch_spin_trylock(arch_spinlock_t *x)
diff --git a/arch/parisc/kernel/syscall.S b/arch/parisc/kernel/syscall.S
index 9505c31..a9bc90d 100644
--- a/arch/parisc/kernel/syscall.S
+++ b/arch/parisc/kernel/syscall.S
@@ -640,7 +640,8 @@
sub,<> %r28, %r25, %r0
2: stw %r24, 0(%r26)
/* Free lock */
- stw,ma %r20, 0(%sr2,%r20)
+ sync
+ stw %r20, 0(%sr2,%r20)
#if ENABLE_LWS_DEBUG
/* Clear thread register indicator */
stw %r0, 4(%sr2,%r20)
@@ -654,7 +655,8 @@
3:
/* Error occurred on load or store */
/* Free lock */
- stw,ma %r20, 0(%sr2,%r20)
+ sync
+ stw %r20, 0(%sr2,%r20)
#if ENABLE_LWS_DEBUG
stw %r0, 4(%sr2,%r20)
#endif
@@ -855,7 +857,8 @@
cas2_end:
/* Free lock */
- stw,ma %r20, 0(%sr2,%r20)
+ sync
+ stw %r20, 0(%sr2,%r20)
/* Enable interrupts */
ssm PSW_SM_I, %r0
/* Return to userspace, set no error */
@@ -865,7 +868,8 @@
22:
/* Error occurred on load or store */
/* Free lock */
- stw,ma %r20, 0(%sr2,%r20)
+ sync
+ stw %r20, 0(%sr2,%r20)
ssm PSW_SM_I, %r0
ldo 1(%r0),%r28
b lws_exit
diff --git a/arch/powerpc/include/asm/io.h b/arch/powerpc/include/asm/io.h
index 3ef40b7..e746bec 100644
--- a/arch/powerpc/include/asm/io.h
+++ b/arch/powerpc/include/asm/io.h
@@ -268,19 +268,13 @@ extern void _memcpy_toio(volatile void __iomem *dest, const void *src,
* their hooks, a bitfield is reserved for use by the platform near the
* top of MMIO addresses (not PIO, those have to cope the hard way).
*
- * This bit field is 12 bits and is at the top of the IO virtual
- * addresses PCI_IO_INDIRECT_TOKEN_MASK.
+ * The highest address in the kernel virtual space are:
*
- * The kernel virtual space is thus:
+ * d0003fffffffffff # with Hash MMU
+ * c00fffffffffffff # with Radix MMU
*
- * 0xD000000000000000 : vmalloc
- * 0xD000080000000000 : PCI PHB IO space
- * 0xD000080080000000 : ioremap
- * 0xD0000fffffffffff : end of ioremap region
- *
- * Since the top 4 bits are reserved as the region ID, we use thus
- * the next 12 bits and keep 4 bits available for the future if the
- * virtual address space is ever to be extended.
+ * The top 4 bits are reserved as the region ID on hash, leaving us 8 bits
+ * that can be used for the field.
*
* The direct IO mapping operations will then mask off those bits
* before doing the actual access, though that only happen when
@@ -292,8 +286,8 @@ extern void _memcpy_toio(volatile void __iomem *dest, const void *src,
*/
#ifdef CONFIG_PPC_INDIRECT_MMIO
-#define PCI_IO_IND_TOKEN_MASK 0x0fff000000000000ul
-#define PCI_IO_IND_TOKEN_SHIFT 48
+#define PCI_IO_IND_TOKEN_SHIFT 52
+#define PCI_IO_IND_TOKEN_MASK (0xfful << PCI_IO_IND_TOKEN_SHIFT)
#define PCI_FIX_ADDR(addr) \
((PCI_IO_ADDR)(((unsigned long)(addr)) & ~PCI_IO_IND_TOKEN_MASK))
#define PCI_GET_ADDR_TOKEN(addr) \
diff --git a/arch/powerpc/include/asm/ppc-opcode.h b/arch/powerpc/include/asm/ppc-opcode.h
index 6093bc8..a6e9e31 100644
--- a/arch/powerpc/include/asm/ppc-opcode.h
+++ b/arch/powerpc/include/asm/ppc-opcode.h
@@ -493,6 +493,8 @@
__PPC_RS(t) | __PPC_RA0(a) | __PPC_RB(b))
#define PPC_SLBFEE_DOT(t, b) stringify_in_c(.long PPC_INST_SLBFEE | \
__PPC_RT(t) | __PPC_RB(b))
+#define __PPC_SLBFEE_DOT(t, b) stringify_in_c(.long PPC_INST_SLBFEE | \
+ ___PPC_RT(t) | ___PPC_RB(b))
#define PPC_ICBT(c,a,b) stringify_in_c(.long PPC_INST_ICBT | \
__PPC_CT(c) | __PPC_RA0(a) | __PPC_RB(b))
/* PASemi instructions */
diff --git a/arch/powerpc/include/asm/ptrace.h b/arch/powerpc/include/asm/ptrace.h
index f73886a1..0b8a735 100644
--- a/arch/powerpc/include/asm/ptrace.h
+++ b/arch/powerpc/include/asm/ptrace.h
@@ -54,6 +54,7 @@ struct pt_regs
#ifdef CONFIG_PPC64
unsigned long ppr;
+ unsigned long __pad; /* Maintain 16 byte interrupt stack alignment */
#endif
};
#endif
diff --git a/arch/powerpc/kernel/setup_64.c b/arch/powerpc/kernel/setup_64.c
index 2a51e4c..236c115 100644
--- a/arch/powerpc/kernel/setup_64.c
+++ b/arch/powerpc/kernel/setup_64.c
@@ -636,6 +636,8 @@ static void *__init alloc_stack(unsigned long limit, int cpu)
{
unsigned long pa;
+ BUILD_BUG_ON(STACK_INT_FRAME_SIZE % 16);
+
pa = memblock_alloc_base_nid(THREAD_SIZE, THREAD_SIZE, limit,
early_cpu_to_node(cpu), MEMBLOCK_NONE);
if (!pa) {
diff --git a/arch/powerpc/kvm/trace.h b/arch/powerpc/kvm/trace.h
index 491b0f7..ea1d7c8 100644
--- a/arch/powerpc/kvm/trace.h
+++ b/arch/powerpc/kvm/trace.h
@@ -6,8 +6,6 @@
#undef TRACE_SYSTEM
#define TRACE_SYSTEM kvm
-#define TRACE_INCLUDE_PATH .
-#define TRACE_INCLUDE_FILE trace
/*
* Tracepoint for guest mode entry.
@@ -120,4 +118,10 @@ TRACE_EVENT(kvm_check_requests,
#endif /* _TRACE_KVM_H */
/* This part must be outside protection */
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace
+
#include <trace/define_trace.h>
diff --git a/arch/powerpc/kvm/trace_booke.h b/arch/powerpc/kvm/trace_booke.h
index ac640e8..3837842 100644
--- a/arch/powerpc/kvm/trace_booke.h
+++ b/arch/powerpc/kvm/trace_booke.h
@@ -6,8 +6,6 @@
#undef TRACE_SYSTEM
#define TRACE_SYSTEM kvm_booke
-#define TRACE_INCLUDE_PATH .
-#define TRACE_INCLUDE_FILE trace_booke
#define kvm_trace_symbol_exit \
{0, "CRITICAL"}, \
@@ -218,4 +216,11 @@ TRACE_EVENT(kvm_booke_queue_irqprio,
#endif
/* This part must be outside protection */
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace_booke
+
#include <trace/define_trace.h>
diff --git a/arch/powerpc/kvm/trace_hv.h b/arch/powerpc/kvm/trace_hv.h
index bcfe8a9..8a1e3b0 100644
--- a/arch/powerpc/kvm/trace_hv.h
+++ b/arch/powerpc/kvm/trace_hv.h
@@ -9,8 +9,6 @@
#undef TRACE_SYSTEM
#define TRACE_SYSTEM kvm_hv
-#define TRACE_INCLUDE_PATH .
-#define TRACE_INCLUDE_FILE trace_hv
#define kvm_trace_symbol_hcall \
{H_REMOVE, "H_REMOVE"}, \
@@ -497,4 +495,11 @@ TRACE_EVENT(kvmppc_run_vcpu_exit,
#endif /* _TRACE_KVM_HV_H */
/* This part must be outside protection */
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace_hv
+
#include <trace/define_trace.h>
diff --git a/arch/powerpc/kvm/trace_pr.h b/arch/powerpc/kvm/trace_pr.h
index 2f9a882..46a46d3 100644
--- a/arch/powerpc/kvm/trace_pr.h
+++ b/arch/powerpc/kvm/trace_pr.h
@@ -8,8 +8,6 @@
#undef TRACE_SYSTEM
#define TRACE_SYSTEM kvm_pr
-#define TRACE_INCLUDE_PATH .
-#define TRACE_INCLUDE_FILE trace_pr
TRACE_EVENT(kvm_book3s_reenter,
TP_PROTO(int r, struct kvm_vcpu *vcpu),
@@ -257,4 +255,11 @@ TRACE_EVENT(kvm_exit,
#endif /* _TRACE_KVM_H */
/* This part must be outside protection */
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE trace_pr
+
#include <trace/define_trace.h>
diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index 3a048e9..ce28ae5 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -1178,7 +1178,7 @@ static long vphn_get_associativity(unsigned long cpu,
switch (rc) {
case H_FUNCTION:
- printk(KERN_INFO
+ printk_once(KERN_INFO
"VPHN is not supported. Disabling polling...\n");
stop_topology_update();
break;
diff --git a/arch/powerpc/mm/slb.c b/arch/powerpc/mm/slb.c
index c3fdf29..bc3914d 100644
--- a/arch/powerpc/mm/slb.c
+++ b/arch/powerpc/mm/slb.c
@@ -19,6 +19,7 @@
#include <asm/mmu.h>
#include <asm/mmu_context.h>
#include <asm/paca.h>
+#include <asm/ppc-opcode.h>
#include <asm/cputable.h>
#include <asm/cacheflush.h>
#include <asm/smp.h>
@@ -58,27 +59,19 @@ static inline unsigned long mk_vsid_data(unsigned long ea, int ssize,
return __mk_vsid_data(get_kernel_vsid(ea, ssize), ssize, flags);
}
-static void assert_slb_exists(unsigned long ea)
+static void assert_slb_presence(bool present, unsigned long ea)
{
#ifdef CONFIG_DEBUG_VM
unsigned long tmp;
WARN_ON_ONCE(mfmsr() & MSR_EE);
- asm volatile("slbfee. %0, %1" : "=r"(tmp) : "r"(ea) : "cr0");
- WARN_ON(tmp == 0);
-#endif
-}
+ if (!cpu_has_feature(CPU_FTR_ARCH_206))
+ return;
-static void assert_slb_notexists(unsigned long ea)
-{
-#ifdef CONFIG_DEBUG_VM
- unsigned long tmp;
+ asm volatile(__PPC_SLBFEE_DOT(%0, %1) : "=r"(tmp) : "r"(ea) : "cr0");
- WARN_ON_ONCE(mfmsr() & MSR_EE);
-
- asm volatile("slbfee. %0, %1" : "=r"(tmp) : "r"(ea) : "cr0");
- WARN_ON(tmp != 0);
+ WARN_ON(present == (tmp == 0));
#endif
}
@@ -114,7 +107,7 @@ static inline void create_shadowed_slbe(unsigned long ea, int ssize,
*/
slb_shadow_update(ea, ssize, flags, index);
- assert_slb_notexists(ea);
+ assert_slb_presence(false, ea);
asm volatile("slbmte %0,%1" :
: "r" (mk_vsid_data(ea, ssize, flags)),
"r" (mk_esid_data(ea, ssize, index))
@@ -137,7 +130,7 @@ void __slb_restore_bolted_realmode(void)
"r" (be64_to_cpu(p->save_area[index].esid)));
}
- assert_slb_exists(local_paca->kstack);
+ assert_slb_presence(true, local_paca->kstack);
}
/*
@@ -185,7 +178,7 @@ void slb_flush_and_restore_bolted(void)
:: "r" (be64_to_cpu(p->save_area[KSTACK_INDEX].vsid)),
"r" (be64_to_cpu(p->save_area[KSTACK_INDEX].esid))
: "memory");
- assert_slb_exists(get_paca()->kstack);
+ assert_slb_presence(true, get_paca()->kstack);
get_paca()->slb_cache_ptr = 0;
@@ -443,9 +436,9 @@ void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
ea = (unsigned long)
get_paca()->slb_cache[i] << SID_SHIFT;
/*
- * Could assert_slb_exists here, but hypervisor
- * or machine check could have come in and
- * removed the entry at this point.
+ * Could assert_slb_presence(true) here, but
+ * hypervisor or machine check could have come
+ * in and removed the entry at this point.
*/
slbie_data = ea;
@@ -676,7 +669,7 @@ static long slb_insert_entry(unsigned long ea, unsigned long context,
* User preloads should add isync afterwards in case the kernel
* accesses user memory before it returns to userspace with rfid.
*/
- assert_slb_notexists(ea);
+ assert_slb_presence(false, ea);
asm volatile("slbmte %0, %1" : : "r" (vsid_data), "r" (esid_data));
barrier();
@@ -715,7 +708,7 @@ static long slb_allocate_kernel(unsigned long ea, unsigned long id)
return -EFAULT;
if (ea < H_VMALLOC_END)
- flags = get_paca()->vmalloc_sllp;
+ flags = local_paca->vmalloc_sllp;
else
flags = SLB_VSID_KERNEL | mmu_psize_defs[mmu_io_psize].sllp;
} else {
diff --git a/arch/powerpc/platforms/powernv/npu-dma.c b/arch/powerpc/platforms/powernv/npu-dma.c
index 6f60e09..75b9352 100644
--- a/arch/powerpc/platforms/powernv/npu-dma.c
+++ b/arch/powerpc/platforms/powernv/npu-dma.c
@@ -102,63 +102,6 @@ struct pci_dev *pnv_pci_get_npu_dev(struct pci_dev *gpdev, int index)
}
EXPORT_SYMBOL(pnv_pci_get_npu_dev);
-#define NPU_DMA_OP_UNSUPPORTED() \
- dev_err_once(dev, "%s operation unsupported for NVLink devices\n", \
- __func__)
-
-static void *dma_npu_alloc(struct device *dev, size_t size,
- dma_addr_t *dma_handle, gfp_t flag,
- unsigned long attrs)
-{
- NPU_DMA_OP_UNSUPPORTED();
- return NULL;
-}
-
-static void dma_npu_free(struct device *dev, size_t size,
- void *vaddr, dma_addr_t dma_handle,
- unsigned long attrs)
-{
- NPU_DMA_OP_UNSUPPORTED();
-}
-
-static dma_addr_t dma_npu_map_page(struct device *dev, struct page *page,
- unsigned long offset, size_t size,
- enum dma_data_direction direction,
- unsigned long attrs)
-{
- NPU_DMA_OP_UNSUPPORTED();
- return 0;
-}
-
-static int dma_npu_map_sg(struct device *dev, struct scatterlist *sglist,
- int nelems, enum dma_data_direction direction,
- unsigned long attrs)
-{
- NPU_DMA_OP_UNSUPPORTED();
- return 0;
-}
-
-static int dma_npu_dma_supported(struct device *dev, u64 mask)
-{
- NPU_DMA_OP_UNSUPPORTED();
- return 0;
-}
-
-static u64 dma_npu_get_required_mask(struct device *dev)
-{
- NPU_DMA_OP_UNSUPPORTED();
- return 0;
-}
-
-static const struct dma_map_ops dma_npu_ops = {
- .map_page = dma_npu_map_page,
- .map_sg = dma_npu_map_sg,
- .alloc = dma_npu_alloc,
- .free = dma_npu_free,
- .dma_supported = dma_npu_dma_supported,
- .get_required_mask = dma_npu_get_required_mask,
-};
-
/*
* Returns the PE assoicated with the PCI device of the given
* NPU. Returns the linked pci device if pci_dev != NULL.
@@ -270,10 +213,11 @@ static void pnv_npu_dma_set_32(struct pnv_ioda_pe *npe)
rc = pnv_npu_set_window(npe, 0, gpe->table_group.tables[0]);
/*
- * We don't initialise npu_pe->tce32_table as we always use
- * dma_npu_ops which are nops.
+ * NVLink devices use the same TCE table configuration as
+ * their parent device so drivers shouldn't be doing DMA
+ * operations directly on these devices.
*/
- set_dma_ops(&npe->pdev->dev, &dma_npu_ops);
+ set_dma_ops(&npe->pdev->dev, NULL);
}
/*
diff --git a/arch/riscv/Makefile b/arch/riscv/Makefile
index d1014619..4af153a 100644
--- a/arch/riscv/Makefile
+++ b/arch/riscv/Makefile
@@ -77,4 +77,8 @@
libs-y += arch/riscv/lib/
+PHONY += vdso_install
+vdso_install:
+ $(Q)$(MAKE) $(build)=arch/riscv/kernel/vdso $@
+
all: vmlinux
diff --git a/arch/riscv/configs/defconfig b/arch/riscv/configs/defconfig
index 07fa9ea..ef4f15d 100644
--- a/arch/riscv/configs/defconfig
+++ b/arch/riscv/configs/defconfig
@@ -76,4 +76,5 @@
CONFIG_NFS_V4_2=y
CONFIG_ROOT_NFS=y
CONFIG_CRYPTO_USER_API_HASH=y
+CONFIG_PRINTK_TIME=y
# CONFIG_RCU_TRACE is not set
diff --git a/arch/riscv/include/asm/ptrace.h b/arch/riscv/include/asm/ptrace.h
index 2c5df94..bbe1862 100644
--- a/arch/riscv/include/asm/ptrace.h
+++ b/arch/riscv/include/asm/ptrace.h
@@ -56,8 +56,8 @@ struct pt_regs {
unsigned long sstatus;
unsigned long sbadaddr;
unsigned long scause;
- /* a0 value before the syscall */
- unsigned long orig_a0;
+ /* a0 value before the syscall */
+ unsigned long orig_a0;
};
#ifdef CONFIG_64BIT
diff --git a/arch/riscv/kernel/module.c b/arch/riscv/kernel/module.c
index 3303ed2..7dd3081 100644
--- a/arch/riscv/kernel/module.c
+++ b/arch/riscv/kernel/module.c
@@ -21,7 +21,7 @@ static int apply_r_riscv_32_rela(struct module *me, u32 *location, Elf_Addr v)
{
if (v != (u32)v) {
pr_err("%s: value %016llx out of range for 32-bit field\n",
- me->name, v);
+ me->name, (long long)v);
return -EINVAL;
}
*location = v;
@@ -102,7 +102,7 @@ static int apply_r_riscv_pcrel_hi20_rela(struct module *me, u32 *location,
if (offset != (s32)offset) {
pr_err(
"%s: target %016llx can not be addressed by the 32-bit offset from PC = %p\n",
- me->name, v, location);
+ me->name, (long long)v, location);
return -EINVAL;
}
@@ -144,7 +144,7 @@ static int apply_r_riscv_hi20_rela(struct module *me, u32 *location,
if (IS_ENABLED(CMODEL_MEDLOW)) {
pr_err(
"%s: target %016llx can not be addressed by the 32-bit offset from PC = %p\n",
- me->name, v, location);
+ me->name, (long long)v, location);
return -EINVAL;
}
@@ -188,7 +188,7 @@ static int apply_r_riscv_got_hi20_rela(struct module *me, u32 *location,
} else {
pr_err(
"%s: can not generate the GOT entry for symbol = %016llx from PC = %p\n",
- me->name, v, location);
+ me->name, (long long)v, location);
return -EINVAL;
}
@@ -212,7 +212,7 @@ static int apply_r_riscv_call_plt_rela(struct module *me, u32 *location,
} else {
pr_err(
"%s: target %016llx can not be addressed by the 32-bit offset from PC = %p\n",
- me->name, v, location);
+ me->name, (long long)v, location);
return -EINVAL;
}
}
@@ -234,7 +234,7 @@ static int apply_r_riscv_call_rela(struct module *me, u32 *location,
if (offset != fill_v) {
pr_err(
"%s: target %016llx can not be addressed by the 32-bit offset from PC = %p\n",
- me->name, v, location);
+ me->name, (long long)v, location);
return -EINVAL;
}
diff --git a/arch/riscv/lib/Makefile b/arch/riscv/lib/Makefile
index 5739bd0..4e2e600 100644
--- a/arch/riscv/lib/Makefile
+++ b/arch/riscv/lib/Makefile
@@ -3,6 +3,6 @@
lib-y += memset.o
lib-y += uaccess.o
-lib-(CONFIG_64BIT) += tishift.o
+lib-$(CONFIG_64BIT) += tishift.o
lib-$(CONFIG_32BIT) += udivdi3.o
diff --git a/arch/s390/Makefile b/arch/s390/Makefile
index 0b33577..e21053e 100644
--- a/arch/s390/Makefile
+++ b/arch/s390/Makefile
@@ -27,7 +27,7 @@
KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),-g)
KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO_DWARF4), $(call cc-option, -gdwarf-4,))
UTS_MACHINE := s390x
-STACK_SIZE := $(if $(CONFIG_KASAN),32768,16384)
+STACK_SIZE := $(if $(CONFIG_KASAN),65536,16384)
CHECKFLAGS += -D__s390__ -D__s390x__
export LD_BFD
diff --git a/arch/s390/boot/compressed/Makefile b/arch/s390/boot/compressed/Makefile
index 5930396..b1bdd15 100644
--- a/arch/s390/boot/compressed/Makefile
+++ b/arch/s390/boot/compressed/Makefile
@@ -22,10 +22,10 @@
OBJECTS := $(addprefix $(obj)/,$(obj-y))
LDFLAGS_vmlinux := --oformat $(LD_BFD) -e startup -T
-$(obj)/vmlinux: $(obj)/vmlinux.lds $(objtree)/arch/s390/boot/startup.a $(OBJECTS)
+$(obj)/vmlinux: $(obj)/vmlinux.lds $(objtree)/arch/s390/boot/startup.a $(OBJECTS) FORCE
$(call if_changed,ld)
-OBJCOPYFLAGS_info.bin := -O binary --only-section=.vmlinux.info
+OBJCOPYFLAGS_info.bin := -O binary --only-section=.vmlinux.info --set-section-flags .vmlinux.info=load
$(obj)/info.bin: vmlinux FORCE
$(call if_changed,objcopy)
@@ -46,17 +46,17 @@
suffix-$(CONFIG_KERNEL_LZO) := .lzo
suffix-$(CONFIG_KERNEL_XZ) := .xz
-$(obj)/vmlinux.bin.gz: $(vmlinux.bin.all-y)
+$(obj)/vmlinux.bin.gz: $(vmlinux.bin.all-y) FORCE
$(call if_changed,gzip)
-$(obj)/vmlinux.bin.bz2: $(vmlinux.bin.all-y)
+$(obj)/vmlinux.bin.bz2: $(vmlinux.bin.all-y) FORCE
$(call if_changed,bzip2)
-$(obj)/vmlinux.bin.lz4: $(vmlinux.bin.all-y)
+$(obj)/vmlinux.bin.lz4: $(vmlinux.bin.all-y) FORCE
$(call if_changed,lz4)
-$(obj)/vmlinux.bin.lzma: $(vmlinux.bin.all-y)
+$(obj)/vmlinux.bin.lzma: $(vmlinux.bin.all-y) FORCE
$(call if_changed,lzma)
-$(obj)/vmlinux.bin.lzo: $(vmlinux.bin.all-y)
+$(obj)/vmlinux.bin.lzo: $(vmlinux.bin.all-y) FORCE
$(call if_changed,lzo)
-$(obj)/vmlinux.bin.xz: $(vmlinux.bin.all-y)
+$(obj)/vmlinux.bin.xz: $(vmlinux.bin.all-y) FORCE
$(call if_changed,xzkern)
OBJCOPYFLAGS_piggy.o := -I binary -O elf64-s390 -B s390:64-bit --rename-section .data=.vmlinux.bin.compressed
diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig
index 259d169..c69cb04 100644
--- a/arch/s390/configs/debug_defconfig
+++ b/arch/s390/configs/debug_defconfig
@@ -64,6 +64,8 @@
CONFIG_PREEMPT=y
CONFIG_HZ_100=y
CONFIG_KEXEC_FILE=y
+CONFIG_EXPOLINE=y
+CONFIG_EXPOLINE_AUTO=y
CONFIG_MEMORY_HOTPLUG=y
CONFIG_MEMORY_HOTREMOVE=y
CONFIG_KSM=y
@@ -84,9 +86,11 @@
CONFIG_HOTPLUG_PCI=y
CONFIG_HOTPLUG_PCI_S390=y
CONFIG_CHSC_SCH=y
+CONFIG_VFIO_AP=m
CONFIG_CRASH_DUMP=y
CONFIG_BINFMT_MISC=m
CONFIG_HIBERNATION=y
+CONFIG_PM_DEBUG=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_PACKET_DIAG=m
@@ -161,8 +165,6 @@
CONFIG_NF_CT_NETLINK=m
CONFIG_NF_CT_NETLINK_TIMEOUT=m
CONFIG_NF_TABLES=m
-CONFIG_NFT_EXTHDR=m
-CONFIG_NFT_META=m
CONFIG_NFT_CT=m
CONFIG_NFT_COUNTER=m
CONFIG_NFT_LOG=m
@@ -365,6 +367,8 @@
CONFIG_NET_ACT_CSUM=m
CONFIG_DNS_RESOLVER=y
CONFIG_OPENVSWITCH=m
+CONFIG_VSOCKETS=m
+CONFIG_VIRTIO_VSOCKETS=m
CONFIG_NETLINK_DIAG=m
CONFIG_CGROUP_NET_PRIO=y
CONFIG_BPF_JIT=y
@@ -461,6 +465,7 @@
CONFIG_PPPOL2TP=m
CONFIG_PPP_ASYNC=m
CONFIG_PPP_SYNC_TTY=m
+CONFIG_ISM=m
CONFIG_INPUT_EVDEV=y
# CONFIG_INPUT_KEYBOARD is not set
# CONFIG_INPUT_MOUSE is not set
@@ -486,9 +491,12 @@
CONFIG_MLX5_INFINIBAND=m
CONFIG_VFIO=m
CONFIG_VFIO_PCI=m
+CONFIG_VFIO_MDEV=m
+CONFIG_VFIO_MDEV_DEVICE=m
CONFIG_VIRTIO_PCI=m
CONFIG_VIRTIO_BALLOON=m
CONFIG_VIRTIO_INPUT=y
+CONFIG_S390_AP_IOMMU=y
CONFIG_EXT4_FS=y
CONFIG_EXT4_FS_POSIX_ACL=y
CONFIG_EXT4_FS_SECURITY=y
@@ -615,7 +623,6 @@
CONFIG_RCU_TORTURE_TEST=m
CONFIG_RCU_CPU_STALL_TIMEOUT=300
CONFIG_NOTIFIER_ERROR_INJECTION=m
-CONFIG_PM_NOTIFIER_ERROR_INJECT=m
CONFIG_NETDEV_NOTIFIER_ERROR_INJECT=m
CONFIG_FAULT_INJECTION=y
CONFIG_FAILSLAB=y
@@ -727,3 +734,4 @@
CONFIG_KVM=m
CONFIG_KVM_S390_UCONTROL=y
CONFIG_VHOST_NET=m
+CONFIG_VHOST_VSOCK=m
diff --git a/arch/s390/configs/performance_defconfig b/arch/s390/configs/performance_defconfig
index 37fd60c..32f539d 100644
--- a/arch/s390/configs/performance_defconfig
+++ b/arch/s390/configs/performance_defconfig
@@ -65,6 +65,8 @@
CONFIG_NUMA=y
CONFIG_HZ_100=y
CONFIG_KEXEC_FILE=y
+CONFIG_EXPOLINE=y
+CONFIG_EXPOLINE_AUTO=y
CONFIG_MEMORY_HOTPLUG=y
CONFIG_MEMORY_HOTREMOVE=y
CONFIG_KSM=y
@@ -82,9 +84,11 @@
CONFIG_HOTPLUG_PCI=y
CONFIG_HOTPLUG_PCI_S390=y
CONFIG_CHSC_SCH=y
+CONFIG_VFIO_AP=m
CONFIG_CRASH_DUMP=y
CONFIG_BINFMT_MISC=m
CONFIG_HIBERNATION=y
+CONFIG_PM_DEBUG=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_PACKET_DIAG=m
@@ -159,8 +163,6 @@
CONFIG_NF_CT_NETLINK=m
CONFIG_NF_CT_NETLINK_TIMEOUT=m
CONFIG_NF_TABLES=m
-CONFIG_NFT_EXTHDR=m
-CONFIG_NFT_META=m
CONFIG_NFT_CT=m
CONFIG_NFT_COUNTER=m
CONFIG_NFT_LOG=m
@@ -362,6 +364,8 @@
CONFIG_NET_ACT_CSUM=m
CONFIG_DNS_RESOLVER=y
CONFIG_OPENVSWITCH=m
+CONFIG_VSOCKETS=m
+CONFIG_VIRTIO_VSOCKETS=m
CONFIG_NETLINK_DIAG=m
CONFIG_CGROUP_NET_PRIO=y
CONFIG_BPF_JIT=y
@@ -458,6 +462,7 @@
CONFIG_PPPOL2TP=m
CONFIG_PPP_ASYNC=m
CONFIG_PPP_SYNC_TTY=m
+CONFIG_ISM=m
CONFIG_INPUT_EVDEV=y
# CONFIG_INPUT_KEYBOARD is not set
# CONFIG_INPUT_MOUSE is not set
@@ -483,9 +488,12 @@
CONFIG_MLX5_INFINIBAND=m
CONFIG_VFIO=m
CONFIG_VFIO_PCI=m
+CONFIG_VFIO_MDEV=m
+CONFIG_VFIO_MDEV_DEVICE=m
CONFIG_VIRTIO_PCI=m
CONFIG_VIRTIO_BALLOON=m
CONFIG_VIRTIO_INPUT=y
+CONFIG_S390_AP_IOMMU=y
CONFIG_EXT4_FS=y
CONFIG_EXT4_FS_POSIX_ACL=y
CONFIG_EXT4_FS_SECURITY=y
@@ -666,3 +674,4 @@
CONFIG_KVM=m
CONFIG_KVM_S390_UCONTROL=y
CONFIG_VHOST_NET=m
+CONFIG_VHOST_VSOCK=m
diff --git a/arch/s390/defconfig b/arch/s390/defconfig
index 7cb6a52..4d58a92 100644
--- a/arch/s390/defconfig
+++ b/arch/s390/defconfig
@@ -26,14 +26,23 @@
CONFIG_CGROUP_PERF=y
CONFIG_NAMESPACES=y
CONFIG_USER_NS=y
+CONFIG_CHECKPOINT_RESTORE=y
CONFIG_BLK_DEV_INITRD=y
CONFIG_EXPERT=y
# CONFIG_SYSFS_SYSCALL is not set
-CONFIG_CHECKPOINT_RESTORE=y
CONFIG_BPF_SYSCALL=y
CONFIG_USERFAULTFD=y
# CONFIG_COMPAT_BRK is not set
CONFIG_PROFILING=y
+CONFIG_LIVEPATCH=y
+CONFIG_NR_CPUS=256
+CONFIG_NUMA=y
+CONFIG_HZ_100=y
+CONFIG_KEXEC_FILE=y
+CONFIG_CRASH_DUMP=y
+CONFIG_HIBERNATION=y
+CONFIG_PM_DEBUG=y
+CONFIG_CMM=m
CONFIG_OPROFILE=y
CONFIG_KPROBES=y
CONFIG_JUMP_LABEL=y
@@ -44,11 +53,7 @@
CONFIG_PARTITION_ADVANCED=y
CONFIG_IBM_PARTITION=y
CONFIG_DEFAULT_DEADLINE=y
-CONFIG_LIVEPATCH=y
-CONFIG_NR_CPUS=256
-CONFIG_NUMA=y
-CONFIG_HZ_100=y
-CONFIG_KEXEC_FILE=y
+CONFIG_BINFMT_MISC=m
CONFIG_MEMORY_HOTPLUG=y
CONFIG_MEMORY_HOTREMOVE=y
CONFIG_KSM=y
@@ -60,9 +65,6 @@
CONFIG_ZSMALLOC=m
CONFIG_ZSMALLOC_STAT=y
CONFIG_IDLE_PAGE_TRACKING=y
-CONFIG_CRASH_DUMP=y
-CONFIG_BINFMT_MISC=m
-CONFIG_HIBERNATION=y
CONFIG_NET=y
CONFIG_PACKET=y
CONFIG_UNIX=y
@@ -98,6 +100,7 @@
CONFIG_BLK_DEV_RAM=y
CONFIG_VIRTIO_BLK=y
CONFIG_SCSI=y
+# CONFIG_SCSI_MQ_DEFAULT is not set
CONFIG_BLK_DEV_SD=y
CONFIG_CHR_DEV_ST=y
CONFIG_BLK_DEV_SR=y
@@ -131,6 +134,7 @@
CONFIG_TUN=m
CONFIG_VIRTIO_NET=y
# CONFIG_NET_VENDOR_ALACRITECH is not set
+# CONFIG_NET_VENDOR_AURORA is not set
# CONFIG_NET_VENDOR_CORTINA is not set
# CONFIG_NET_VENDOR_SOLARFLARE is not set
# CONFIG_NET_VENDOR_SOCIONEXT is not set
@@ -157,33 +161,6 @@
CONFIG_TMPFS_POSIX_ACL=y
CONFIG_HUGETLBFS=y
# CONFIG_NETWORK_FILESYSTEMS is not set
-CONFIG_DEBUG_INFO=y
-CONFIG_DEBUG_INFO_DWARF4=y
-CONFIG_GDB_SCRIPTS=y
-CONFIG_UNUSED_SYMBOLS=y
-CONFIG_DEBUG_SECTION_MISMATCH=y
-CONFIG_DEBUG_FORCE_WEAK_PER_CPU=y
-CONFIG_MAGIC_SYSRQ=y
-CONFIG_DEBUG_PAGEALLOC=y
-CONFIG_DETECT_HUNG_TASK=y
-CONFIG_PANIC_ON_OOPS=y
-CONFIG_PROVE_LOCKING=y
-CONFIG_LOCK_STAT=y
-CONFIG_DEBUG_LOCKDEP=y
-CONFIG_DEBUG_ATOMIC_SLEEP=y
-CONFIG_DEBUG_LIST=y
-CONFIG_DEBUG_SG=y
-CONFIG_DEBUG_NOTIFIERS=y
-CONFIG_RCU_CPU_STALL_TIMEOUT=60
-CONFIG_LATENCYTOP=y
-CONFIG_SCHED_TRACER=y
-CONFIG_FTRACE_SYSCALLS=y
-CONFIG_TRACER_SNAPSHOT_PER_CPU_SWAP=y
-CONFIG_STACK_TRACER=y
-CONFIG_BLK_DEV_IO_TRACE=y
-CONFIG_FUNCTION_PROFILER=y
-# CONFIG_RUNTIME_TESTING_MENU is not set
-CONFIG_S390_PTDUMP=y
CONFIG_CRYPTO_CRYPTD=m
CONFIG_CRYPTO_AUTHENC=m
CONFIG_CRYPTO_TEST=m
@@ -193,6 +170,7 @@
CONFIG_CRYPTO_CFB=m
CONFIG_CRYPTO_CTS=m
CONFIG_CRYPTO_LRW=m
+CONFIG_CRYPTO_OFB=m
CONFIG_CRYPTO_PCBC=m
CONFIG_CRYPTO_XTS=m
CONFIG_CRYPTO_CMAC=m
@@ -231,7 +209,6 @@
CONFIG_CRYPTO_USER_API_SKCIPHER=m
CONFIG_CRYPTO_USER_API_RNG=m
CONFIG_ZCRYPT=m
-CONFIG_ZCRYPT_MULTIDEVNODES=y
CONFIG_PKEY=m
CONFIG_CRYPTO_PAES_S390=m
CONFIG_CRYPTO_SHA1_S390=m
@@ -247,4 +224,30 @@
# CONFIG_XZ_DEC_ARM is not set
# CONFIG_XZ_DEC_ARMTHUMB is not set
# CONFIG_XZ_DEC_SPARC is not set
-CONFIG_CMM=m
+CONFIG_DEBUG_INFO=y
+CONFIG_DEBUG_INFO_DWARF4=y
+CONFIG_GDB_SCRIPTS=y
+CONFIG_UNUSED_SYMBOLS=y
+CONFIG_DEBUG_SECTION_MISMATCH=y
+CONFIG_DEBUG_FORCE_WEAK_PER_CPU=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_PAGEALLOC=y
+CONFIG_DETECT_HUNG_TASK=y
+CONFIG_PANIC_ON_OOPS=y
+CONFIG_PROVE_LOCKING=y
+CONFIG_LOCK_STAT=y
+CONFIG_DEBUG_LOCKDEP=y
+CONFIG_DEBUG_ATOMIC_SLEEP=y
+CONFIG_DEBUG_LIST=y
+CONFIG_DEBUG_SG=y
+CONFIG_DEBUG_NOTIFIERS=y
+CONFIG_RCU_CPU_STALL_TIMEOUT=60
+CONFIG_LATENCYTOP=y
+CONFIG_SCHED_TRACER=y
+CONFIG_FTRACE_SYSCALLS=y
+CONFIG_TRACER_SNAPSHOT_PER_CPU_SWAP=y
+CONFIG_STACK_TRACER=y
+CONFIG_BLK_DEV_IO_TRACE=y
+CONFIG_FUNCTION_PROFILER=y
+# CONFIG_RUNTIME_TESTING_MENU is not set
+CONFIG_S390_PTDUMP=y
diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index dbd689d..ccbb53e 100644
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -46,8 +46,6 @@ static inline int init_new_context(struct task_struct *tsk,
mm->context.asce_limit = STACK_TOP_MAX;
mm->context.asce = __pa(mm->pgd) | _ASCE_TABLE_LENGTH |
_ASCE_USER_BITS | _ASCE_TYPE_REGION3;
- /* pgd_alloc() did not account this pud */
- mm_inc_nr_puds(mm);
break;
case -PAGE_SIZE:
/* forked 5-level task, set new asce with new_mm->pgd */
@@ -63,9 +61,6 @@ static inline int init_new_context(struct task_struct *tsk,
/* forked 2-level compat task, set new asce with new mm->pgd */
mm->context.asce = __pa(mm->pgd) | _ASCE_TABLE_LENGTH |
_ASCE_USER_BITS | _ASCE_TYPE_SEGMENT;
- /* pgd_alloc() did not account this pmd */
- mm_inc_nr_pmds(mm);
- mm_inc_nr_puds(mm);
}
crst_table_init((unsigned long *) mm->pgd, pgd_entry_type(mm));
return 0;
diff --git a/arch/s390/include/asm/pgalloc.h b/arch/s390/include/asm/pgalloc.h
index f0f9bcf..5ee7337 100644
--- a/arch/s390/include/asm/pgalloc.h
+++ b/arch/s390/include/asm/pgalloc.h
@@ -36,11 +36,11 @@ static inline void crst_table_init(unsigned long *crst, unsigned long entry)
static inline unsigned long pgd_entry_type(struct mm_struct *mm)
{
- if (mm->context.asce_limit <= _REGION3_SIZE)
+ if (mm_pmd_folded(mm))
return _SEGMENT_ENTRY_EMPTY;
- if (mm->context.asce_limit <= _REGION2_SIZE)
+ if (mm_pud_folded(mm))
return _REGION3_ENTRY_EMPTY;
- if (mm->context.asce_limit <= _REGION1_SIZE)
+ if (mm_p4d_folded(mm))
return _REGION2_ENTRY_EMPTY;
return _REGION1_ENTRY_EMPTY;
}
diff --git a/arch/s390/include/asm/pgtable.h b/arch/s390/include/asm/pgtable.h
index 411d435..0637324 100644
--- a/arch/s390/include/asm/pgtable.h
+++ b/arch/s390/include/asm/pgtable.h
@@ -493,6 +493,24 @@ static inline int is_module_addr(void *addr)
_REGION_ENTRY_PROTECT | \
_REGION_ENTRY_NOEXEC)
+static inline bool mm_p4d_folded(struct mm_struct *mm)
+{
+ return mm->context.asce_limit <= _REGION1_SIZE;
+}
+#define mm_p4d_folded(mm) mm_p4d_folded(mm)
+
+static inline bool mm_pud_folded(struct mm_struct *mm)
+{
+ return mm->context.asce_limit <= _REGION2_SIZE;
+}
+#define mm_pud_folded(mm) mm_pud_folded(mm)
+
+static inline bool mm_pmd_folded(struct mm_struct *mm)
+{
+ return mm->context.asce_limit <= _REGION3_SIZE;
+}
+#define mm_pmd_folded(mm) mm_pmd_folded(mm)
+
static inline int mm_has_pgste(struct mm_struct *mm)
{
#ifdef CONFIG_PGSTE
diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h
index 302795c..81038ab 100644
--- a/arch/s390/include/asm/processor.h
+++ b/arch/s390/include/asm/processor.h
@@ -236,7 +236,7 @@ static inline unsigned long current_stack_pointer(void)
return sp;
}
-static __no_sanitize_address_or_inline unsigned short stap(void)
+static __no_kasan_or_inline unsigned short stap(void)
{
unsigned short cpu_address;
@@ -330,7 +330,7 @@ static inline void __load_psw(psw_t psw)
* Set PSW mask to specified value, while leaving the
* PSW addr pointing to the next instruction.
*/
-static __no_sanitize_address_or_inline void __load_psw_mask(unsigned long mask)
+static __no_kasan_or_inline void __load_psw_mask(unsigned long mask)
{
unsigned long addr;
psw_t psw;
diff --git a/arch/s390/include/asm/thread_info.h b/arch/s390/include/asm/thread_info.h
index 27248f4..ce4e17c 100644
--- a/arch/s390/include/asm/thread_info.h
+++ b/arch/s390/include/asm/thread_info.h
@@ -14,7 +14,7 @@
* General size of kernel stacks
*/
#ifdef CONFIG_KASAN
-#define THREAD_SIZE_ORDER 3
+#define THREAD_SIZE_ORDER 4
#else
#define THREAD_SIZE_ORDER 2
#endif
diff --git a/arch/s390/include/asm/tlb.h b/arch/s390/include/asm/tlb.h
index 457b7ba..b31c779 100644
--- a/arch/s390/include/asm/tlb.h
+++ b/arch/s390/include/asm/tlb.h
@@ -136,7 +136,7 @@ static inline void pte_free_tlb(struct mmu_gather *tlb, pgtable_t pte,
static inline void pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd,
unsigned long address)
{
- if (tlb->mm->context.asce_limit <= _REGION3_SIZE)
+ if (mm_pmd_folded(tlb->mm))
return;
pgtable_pmd_page_dtor(virt_to_page(pmd));
tlb_remove_table(tlb, pmd);
@@ -152,7 +152,7 @@ static inline void pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd,
static inline void p4d_free_tlb(struct mmu_gather *tlb, p4d_t *p4d,
unsigned long address)
{
- if (tlb->mm->context.asce_limit <= _REGION1_SIZE)
+ if (mm_p4d_folded(tlb->mm))
return;
tlb_remove_table(tlb, p4d);
}
@@ -167,7 +167,7 @@ static inline void p4d_free_tlb(struct mmu_gather *tlb, p4d_t *p4d,
static inline void pud_free_tlb(struct mmu_gather *tlb, pud_t *pud,
unsigned long address)
{
- if (tlb->mm->context.asce_limit <= _REGION2_SIZE)
+ if (mm_pud_folded(tlb->mm))
return;
tlb_remove_table(tlb, pud);
}
diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index 724fba4..39191a0 100644
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -236,10 +236,10 @@
stmg %r6,%r15,__SF_GPRS(%r15) # store gprs of prev task
lghi %r4,__TASK_stack
lghi %r1,__TASK_thread
- lg %r5,0(%r4,%r3) # start of kernel stack of next
+ llill %r5,STACK_INIT
stg %r15,__THREAD_ksp(%r1,%r2) # store kernel stack of prev
- lgr %r15,%r5
- aghi %r15,STACK_INIT # end of kernel stack of next
+ lg %r15,0(%r4,%r3) # start of kernel stack of next
+ agr %r15,%r5 # end of kernel stack of next
stg %r3,__LC_CURRENT # store task struct of next
stg %r15,__LC_KERNEL_STACK # store end of kernel stack
lg %r15,__THREAD_ksp(%r1,%r3) # load kernel stack of next
diff --git a/arch/s390/kernel/perf_cpum_cf.c b/arch/s390/kernel/perf_cpum_cf.c
index cc085e2..74091fd 100644
--- a/arch/s390/kernel/perf_cpum_cf.c
+++ b/arch/s390/kernel/perf_cpum_cf.c
@@ -373,7 +373,7 @@ static int __hw_perf_event_init(struct perf_event *event)
return -ENOENT;
if (ev > PERF_CPUM_CF_MAX_CTR)
- return -EINVAL;
+ return -ENOENT;
/* Obtain the counter set to which the specified counter belongs */
set = get_counter_set(ev);
diff --git a/arch/s390/kernel/perf_cpum_sf.c b/arch/s390/kernel/perf_cpum_sf.c
index 7bf604f..bfabeb1 100644
--- a/arch/s390/kernel/perf_cpum_sf.c
+++ b/arch/s390/kernel/perf_cpum_sf.c
@@ -1842,10 +1842,30 @@ static void cpumsf_pmu_del(struct perf_event *event, int flags)
CPUMF_EVENT_ATTR(SF, SF_CYCLES_BASIC, PERF_EVENT_CPUM_SF);
CPUMF_EVENT_ATTR(SF, SF_CYCLES_BASIC_DIAG, PERF_EVENT_CPUM_SF_DIAG);
-static struct attribute *cpumsf_pmu_events_attr[] = {
- CPUMF_EVENT_PTR(SF, SF_CYCLES_BASIC),
- NULL,
- NULL,
+/* Attribute list for CPU_SF.
+ *
+ * The availablitiy depends on the CPU_MF sampling facility authorization
+ * for basic + diagnositic samples. This is determined at initialization
+ * time by the sampling facility device driver.
+ * If the authorization for basic samples is turned off, it should be
+ * also turned off for diagnostic sampling.
+ *
+ * During initialization of the device driver, check the authorization
+ * level for diagnostic sampling and installs the attribute
+ * file for diagnostic sampling if necessary.
+ *
+ * For now install a placeholder to reference all possible attributes:
+ * SF_CYCLES_BASIC and SF_CYCLES_BASIC_DIAG.
+ * Add another entry for the final NULL pointer.
+ */
+enum {
+ SF_CYCLES_BASIC_ATTR_IDX = 0,
+ SF_CYCLES_BASIC_DIAG_ATTR_IDX,
+ SF_CYCLES_ATTR_MAX
+};
+
+static struct attribute *cpumsf_pmu_events_attr[SF_CYCLES_ATTR_MAX + 1] = {
+ [SF_CYCLES_BASIC_ATTR_IDX] = CPUMF_EVENT_PTR(SF, SF_CYCLES_BASIC)
};
PMU_FORMAT_ATTR(event, "config:0-63");
@@ -2040,7 +2060,10 @@ static int __init init_cpum_sampling_pmu(void)
if (si.ad) {
sfb_set_limits(CPUM_SF_MIN_SDB, CPUM_SF_MAX_SDB);
- cpumsf_pmu_events_attr[1] =
+ /* Sampling of diagnostic data authorized,
+ * install event into attribute list of PMU device.
+ */
+ cpumsf_pmu_events_attr[SF_CYCLES_BASIC_DIAG_ATTR_IDX] =
CPUMF_EVENT_PTR(SF, SF_CYCLES_BASIC_DIAG);
}
diff --git a/arch/s390/kernel/vdso32/Makefile b/arch/s390/kernel/vdso32/Makefile
index eb8aebe..e76309f 100644
--- a/arch/s390/kernel/vdso32/Makefile
+++ b/arch/s390/kernel/vdso32/Makefile
@@ -37,7 +37,7 @@
$(obj)/vdso32_wrapper.o : $(obj)/vdso32.so
# link rule for the .so file, .lds has to be first
-$(obj)/vdso32.so.dbg: $(src)/vdso32.lds $(obj-vdso32)
+$(obj)/vdso32.so.dbg: $(src)/vdso32.lds $(obj-vdso32) FORCE
$(call if_changed,vdso32ld)
# strip rule for the .so file
@@ -46,12 +46,12 @@
$(call if_changed,objcopy)
# assembly rules for the .S files
-$(obj-vdso32): %.o: %.S
+$(obj-vdso32): %.o: %.S FORCE
$(call if_changed_dep,vdso32as)
# actual build commands
quiet_cmd_vdso32ld = VDSO32L $@
- cmd_vdso32ld = $(CC) $(c_flags) -Wl,-T $^ -o $@
+ cmd_vdso32ld = $(CC) $(c_flags) -Wl,-T $(filter %.lds %.o,$^) -o $@
quiet_cmd_vdso32as = VDSO32A $@
cmd_vdso32as = $(CC) $(a_flags) -c -o $@ $<
diff --git a/arch/s390/kernel/vdso64/Makefile b/arch/s390/kernel/vdso64/Makefile
index a22b2cf..f849ac6 100644
--- a/arch/s390/kernel/vdso64/Makefile
+++ b/arch/s390/kernel/vdso64/Makefile
@@ -37,7 +37,7 @@
$(obj)/vdso64_wrapper.o : $(obj)/vdso64.so
# link rule for the .so file, .lds has to be first
-$(obj)/vdso64.so.dbg: $(src)/vdso64.lds $(obj-vdso64)
+$(obj)/vdso64.so.dbg: $(src)/vdso64.lds $(obj-vdso64) FORCE
$(call if_changed,vdso64ld)
# strip rule for the .so file
@@ -46,12 +46,12 @@
$(call if_changed,objcopy)
# assembly rules for the .S files
-$(obj-vdso64): %.o: %.S
+$(obj-vdso64): %.o: %.S FORCE
$(call if_changed_dep,vdso64as)
# actual build commands
quiet_cmd_vdso64ld = VDSO64L $@
- cmd_vdso64ld = $(CC) $(c_flags) -Wl,-T $^ -o $@
+ cmd_vdso64ld = $(CC) $(c_flags) -Wl,-T $(filter %.lds %.o,$^) -o $@
quiet_cmd_vdso64as = VDSO64A $@
cmd_vdso64as = $(CC) $(a_flags) -c -o $@ $<
diff --git a/arch/s390/kernel/vmlinux.lds.S b/arch/s390/kernel/vmlinux.lds.S
index 21eb7407..8429ab0 100644
--- a/arch/s390/kernel/vmlinux.lds.S
+++ b/arch/s390/kernel/vmlinux.lds.S
@@ -154,14 +154,14 @@
* uncompressed image info used by the decompressor
* it should match struct vmlinux_info
*/
- .vmlinux.info 0 : {
+ .vmlinux.info 0 (INFO) : {
QUAD(_stext) /* default_lma */
QUAD(startup_continue) /* entry */
QUAD(__bss_start - _stext) /* image_size */
QUAD(__bss_stop - __bss_start) /* bss_size */
QUAD(__boot_data_start) /* bootdata_off */
QUAD(__boot_data_end - __boot_data_start) /* bootdata_size */
- }
+ } :NONE
/* Debugging sections. */
STABS_DEBUG
diff --git a/arch/s390/mm/pgalloc.c b/arch/s390/mm/pgalloc.c
index 76d89ee..814f265 100644
--- a/arch/s390/mm/pgalloc.c
+++ b/arch/s390/mm/pgalloc.c
@@ -101,6 +101,7 @@ int crst_table_upgrade(struct mm_struct *mm, unsigned long end)
mm->context.asce_limit = _REGION1_SIZE;
mm->context.asce = __pa(mm->pgd) | _ASCE_TABLE_LENGTH |
_ASCE_USER_BITS | _ASCE_TYPE_REGION2;
+ mm_inc_nr_puds(mm);
} else {
crst_table_init(table, _REGION1_ENTRY_EMPTY);
pgd_populate(mm, (pgd_t *) table, (p4d_t *) pgd);
diff --git a/arch/s390/numa/numa.c b/arch/s390/numa/numa.c
index ae0d9e8..d31bde0 100644
--- a/arch/s390/numa/numa.c
+++ b/arch/s390/numa/numa.c
@@ -53,6 +53,7 @@ int __node_distance(int a, int b)
{
return mode->distance ? mode->distance(a, b) : 0;
}
+EXPORT_SYMBOL(__node_distance);
int numa_debug_enabled;
diff --git a/arch/um/drivers/ubd_kern.c b/arch/um/drivers/ubd_kern.c
index 74c002d..28c4062 100644
--- a/arch/um/drivers/ubd_kern.c
+++ b/arch/um/drivers/ubd_kern.c
@@ -1305,6 +1305,7 @@ static int ubd_queue_one_vec(struct blk_mq_hw_ctx *hctx, struct request *req,
io_req->fds[0] = dev->cow.fd;
else
io_req->fds[0] = dev->fd;
+ io_req->error = 0;
if (req_op(req) == REQ_OP_FLUSH) {
io_req->op = UBD_FLUSH;
@@ -1313,9 +1314,7 @@ static int ubd_queue_one_vec(struct blk_mq_hw_ctx *hctx, struct request *req,
io_req->cow_offset = -1;
io_req->offset = off;
io_req->length = bvec->bv_len;
- io_req->error = 0;
io_req->sector_mask = 0;
-
io_req->op = rq_data_dir(req) == READ ? UBD_READ : UBD_WRITE;
io_req->offsets[0] = 0;
io_req->offsets[1] = dev->cow.data_offset;
@@ -1341,11 +1340,14 @@ static int ubd_queue_one_vec(struct blk_mq_hw_ctx *hctx, struct request *req,
static blk_status_t ubd_queue_rq(struct blk_mq_hw_ctx *hctx,
const struct blk_mq_queue_data *bd)
{
+ struct ubd *ubd_dev = hctx->queue->queuedata;
struct request *req = bd->rq;
int ret = 0;
blk_mq_start_request(req);
+ spin_lock_irq(&ubd_dev->lock);
+
if (req_op(req) == REQ_OP_FLUSH) {
ret = ubd_queue_one_vec(hctx, req, 0, NULL);
} else {
@@ -1361,9 +1363,11 @@ static blk_status_t ubd_queue_rq(struct blk_mq_hw_ctx *hctx,
}
}
out:
- if (ret < 0) {
+ spin_unlock_irq(&ubd_dev->lock);
+
+ if (ret < 0)
blk_mq_requeue_request(req, true);
- }
+
return BLK_STS_OK;
}
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index ba7e346..9d734f3 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -525,7 +525,6 @@
bool "ScaleMP vSMP"
select HYPERVISOR_GUEST
select PARAVIRT
- select PARAVIRT_XXL
depends on X86_64 && PCI
depends on X86_EXTENDED_PLATFORM
depends on SMP
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 5b562e4..88398fd 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -213,8 +213,6 @@
KBUILD_LDFLAGS += $(call ld-option, -z max-page-size=0x200000)
endif
-# Speed up the build
-KBUILD_CFLAGS += -pipe
# Workaround for a gcc prelease that unfortunately was shipped in a suse release
KBUILD_CFLAGS += -Wno-sign-compare
#
@@ -239,7 +237,7 @@
archmacros:
$(Q)$(MAKE) $(build)=arch/x86/kernel arch/x86/kernel/macros.s
-ASM_MACRO_FLAGS = -Wa,arch/x86/kernel/macros.s -Wa,-
+ASM_MACRO_FLAGS = -Wa,arch/x86/kernel/macros.s
export ASM_MACRO_FLAGS
KBUILD_CFLAGS += $(ASM_MACRO_FLAGS)
diff --git a/arch/x86/events/intel/uncore.h b/arch/x86/events/intel/uncore.h
index e17ab88..cb46d60 100644
--- a/arch/x86/events/intel/uncore.h
+++ b/arch/x86/events/intel/uncore.h
@@ -129,8 +129,15 @@ struct intel_uncore_box {
struct intel_uncore_extra_reg shared_regs[0];
};
-#define UNCORE_BOX_FLAG_INITIATED 0
-#define UNCORE_BOX_FLAG_CTL_OFFS8 1 /* event config registers are 8-byte apart */
+/* CFL uncore 8th cbox MSRs */
+#define CFL_UNC_CBO_7_PERFEVTSEL0 0xf70
+#define CFL_UNC_CBO_7_PER_CTR0 0xf76
+
+#define UNCORE_BOX_FLAG_INITIATED 0
+/* event config registers are 8-byte apart */
+#define UNCORE_BOX_FLAG_CTL_OFFS8 1
+/* CFL 8th CBOX has different MSR space */
+#define UNCORE_BOX_FLAG_CFL8_CBOX_MSR_OFFS 2
struct uncore_event_desc {
struct kobj_attribute attr;
@@ -297,17 +304,27 @@ unsigned int uncore_freerunning_counter(struct intel_uncore_box *box,
static inline
unsigned uncore_msr_event_ctl(struct intel_uncore_box *box, int idx)
{
- return box->pmu->type->event_ctl +
- (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx) +
- uncore_msr_box_offset(box);
+ if (test_bit(UNCORE_BOX_FLAG_CFL8_CBOX_MSR_OFFS, &box->flags)) {
+ return CFL_UNC_CBO_7_PERFEVTSEL0 +
+ (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx);
+ } else {
+ return box->pmu->type->event_ctl +
+ (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx) +
+ uncore_msr_box_offset(box);
+ }
}
static inline
unsigned uncore_msr_perf_ctr(struct intel_uncore_box *box, int idx)
{
- return box->pmu->type->perf_ctr +
- (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx) +
- uncore_msr_box_offset(box);
+ if (test_bit(UNCORE_BOX_FLAG_CFL8_CBOX_MSR_OFFS, &box->flags)) {
+ return CFL_UNC_CBO_7_PER_CTR0 +
+ (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx);
+ } else {
+ return box->pmu->type->perf_ctr +
+ (box->pmu->type->pair_ctr_ctl ? 2 * idx : idx) +
+ uncore_msr_box_offset(box);
+ }
}
static inline
diff --git a/arch/x86/events/intel/uncore_snb.c b/arch/x86/events/intel/uncore_snb.c
index 8527c3e..2593b0d 100644
--- a/arch/x86/events/intel/uncore_snb.c
+++ b/arch/x86/events/intel/uncore_snb.c
@@ -15,6 +15,25 @@
#define PCI_DEVICE_ID_INTEL_SKL_HQ_IMC 0x1910
#define PCI_DEVICE_ID_INTEL_SKL_SD_IMC 0x190f
#define PCI_DEVICE_ID_INTEL_SKL_SQ_IMC 0x191f
+#define PCI_DEVICE_ID_INTEL_KBL_Y_IMC 0x590c
+#define PCI_DEVICE_ID_INTEL_KBL_U_IMC 0x5904
+#define PCI_DEVICE_ID_INTEL_KBL_UQ_IMC 0x5914
+#define PCI_DEVICE_ID_INTEL_KBL_SD_IMC 0x590f
+#define PCI_DEVICE_ID_INTEL_KBL_SQ_IMC 0x591f
+#define PCI_DEVICE_ID_INTEL_CFL_2U_IMC 0x3ecc
+#define PCI_DEVICE_ID_INTEL_CFL_4U_IMC 0x3ed0
+#define PCI_DEVICE_ID_INTEL_CFL_4H_IMC 0x3e10
+#define PCI_DEVICE_ID_INTEL_CFL_6H_IMC 0x3ec4
+#define PCI_DEVICE_ID_INTEL_CFL_2S_D_IMC 0x3e0f
+#define PCI_DEVICE_ID_INTEL_CFL_4S_D_IMC 0x3e1f
+#define PCI_DEVICE_ID_INTEL_CFL_6S_D_IMC 0x3ec2
+#define PCI_DEVICE_ID_INTEL_CFL_8S_D_IMC 0x3e30
+#define PCI_DEVICE_ID_INTEL_CFL_4S_W_IMC 0x3e18
+#define PCI_DEVICE_ID_INTEL_CFL_6S_W_IMC 0x3ec6
+#define PCI_DEVICE_ID_INTEL_CFL_8S_W_IMC 0x3e31
+#define PCI_DEVICE_ID_INTEL_CFL_4S_S_IMC 0x3e33
+#define PCI_DEVICE_ID_INTEL_CFL_6S_S_IMC 0x3eca
+#define PCI_DEVICE_ID_INTEL_CFL_8S_S_IMC 0x3e32
/* SNB event control */
#define SNB_UNC_CTL_EV_SEL_MASK 0x000000ff
@@ -202,6 +221,10 @@ static void skl_uncore_msr_init_box(struct intel_uncore_box *box)
wrmsrl(SKL_UNC_PERF_GLOBAL_CTL,
SNB_UNC_GLOBAL_CTL_EN | SKL_UNC_GLOBAL_CTL_CORE_ALL);
}
+
+ /* The 8th CBOX has different MSR space */
+ if (box->pmu->pmu_idx == 7)
+ __set_bit(UNCORE_BOX_FLAG_CFL8_CBOX_MSR_OFFS, &box->flags);
}
static void skl_uncore_msr_enable_box(struct intel_uncore_box *box)
@@ -228,7 +251,7 @@ static struct intel_uncore_ops skl_uncore_msr_ops = {
static struct intel_uncore_type skl_uncore_cbox = {
.name = "cbox",
.num_counters = 4,
- .num_boxes = 5,
+ .num_boxes = 8,
.perf_ctr_bits = 44,
.fixed_ctr_bits = 48,
.perf_ctr = SNB_UNC_CBO_0_PER_CTR0,
@@ -569,7 +592,82 @@ static const struct pci_device_id skl_uncore_pci_ids[] = {
PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_SKL_SQ_IMC),
.driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
},
-
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_Y_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_U_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_UQ_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_SD_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_KBL_SQ_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_2U_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4U_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4H_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_6H_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_2S_D_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4S_D_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_6S_D_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_8S_D_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4S_W_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_6S_W_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_8S_W_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_4S_S_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_6S_S_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
+ { /* IMC */
+ PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_CFL_8S_S_IMC),
+ .driver_data = UNCORE_PCI_DEV_DATA(SNB_PCI_UNCORE_IMC, 0),
+ },
{ /* end: all zeroes */ },
};
@@ -618,6 +716,25 @@ static const struct imc_uncore_pci_dev desktop_imc_pci_ids[] = {
IMC_DEV(SKL_HQ_IMC, &skl_uncore_pci_driver), /* 6th Gen Core H Quad Core */
IMC_DEV(SKL_SD_IMC, &skl_uncore_pci_driver), /* 6th Gen Core S Dual Core */
IMC_DEV(SKL_SQ_IMC, &skl_uncore_pci_driver), /* 6th Gen Core S Quad Core */
+ IMC_DEV(KBL_Y_IMC, &skl_uncore_pci_driver), /* 7th Gen Core Y */
+ IMC_DEV(KBL_U_IMC, &skl_uncore_pci_driver), /* 7th Gen Core U */
+ IMC_DEV(KBL_UQ_IMC, &skl_uncore_pci_driver), /* 7th Gen Core U Quad Core */
+ IMC_DEV(KBL_SD_IMC, &skl_uncore_pci_driver), /* 7th Gen Core S Dual Core */
+ IMC_DEV(KBL_SQ_IMC, &skl_uncore_pci_driver), /* 7th Gen Core S Quad Core */
+ IMC_DEV(CFL_2U_IMC, &skl_uncore_pci_driver), /* 8th Gen Core U 2 Cores */
+ IMC_DEV(CFL_4U_IMC, &skl_uncore_pci_driver), /* 8th Gen Core U 4 Cores */
+ IMC_DEV(CFL_4H_IMC, &skl_uncore_pci_driver), /* 8th Gen Core H 4 Cores */
+ IMC_DEV(CFL_6H_IMC, &skl_uncore_pci_driver), /* 8th Gen Core H 6 Cores */
+ IMC_DEV(CFL_2S_D_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 2 Cores Desktop */
+ IMC_DEV(CFL_4S_D_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 4 Cores Desktop */
+ IMC_DEV(CFL_6S_D_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 6 Cores Desktop */
+ IMC_DEV(CFL_8S_D_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 8 Cores Desktop */
+ IMC_DEV(CFL_4S_W_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 4 Cores Work Station */
+ IMC_DEV(CFL_6S_W_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 6 Cores Work Station */
+ IMC_DEV(CFL_8S_W_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 8 Cores Work Station */
+ IMC_DEV(CFL_4S_S_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 4 Cores Server */
+ IMC_DEV(CFL_6S_S_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 6 Cores Server */
+ IMC_DEV(CFL_8S_S_IMC, &skl_uncore_pci_driver), /* 8th Gen Core S 8 Cores Server */
{ /* end marker */ }
};
diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h
index 4da9b1c..c1a812b 100644
--- a/arch/x86/include/asm/mce.h
+++ b/arch/x86/include/asm/mce.h
@@ -221,6 +221,8 @@ static inline void mce_hygon_feature_init(struct cpuinfo_x86 *c) { return mce_am
int mce_available(struct cpuinfo_x86 *c);
bool mce_is_memory_error(struct mce *m);
+bool mce_is_correctable(struct mce *m);
+int mce_usable_address(struct mce *m);
DECLARE_PER_CPU(unsigned, mce_exception_count);
DECLARE_PER_CPU(unsigned, mce_poll_count);
diff --git a/arch/x86/include/asm/mshyperv.h b/arch/x86/include/asm/mshyperv.h
index 0d6271c..1d0a777 100644
--- a/arch/x86/include/asm/mshyperv.h
+++ b/arch/x86/include/asm/mshyperv.h
@@ -232,7 +232,7 @@ static inline u64 hv_do_fast_hypercall16(u16 code, u64 input1, u64 input2)
: "cc");
}
#endif
- return hv_status;
+ return hv_status;
}
/*
diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h
index cd0cf1c..8f65728 100644
--- a/arch/x86/include/asm/page_64_types.h
+++ b/arch/x86/include/asm/page_64_types.h
@@ -33,12 +33,14 @@
/*
* Set __PAGE_OFFSET to the most negative possible address +
- * PGDIR_SIZE*16 (pgd slot 272). The gap is to allow a space for a
- * hypervisor to fit. Choosing 16 slots here is arbitrary, but it's
- * what Xen requires.
+ * PGDIR_SIZE*17 (pgd slot 273).
+ *
+ * The gap is to allow a space for LDT remap for PTI (1 pgd slot) and space for
+ * a hypervisor (16 slots). Choosing 16 slots for a hypervisor is arbitrary,
+ * but it's what Xen requires.
*/
-#define __PAGE_OFFSET_BASE_L5 _AC(0xff10000000000000, UL)
-#define __PAGE_OFFSET_BASE_L4 _AC(0xffff880000000000, UL)
+#define __PAGE_OFFSET_BASE_L5 _AC(0xff11000000000000, UL)
+#define __PAGE_OFFSET_BASE_L4 _AC(0xffff888000000000, UL)
#ifdef CONFIG_DYNAMIC_MEMORY_LAYOUT
#define __PAGE_OFFSET page_offset_base
diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h
index 04edd2d..84bd9bd 100644
--- a/arch/x86/include/asm/pgtable_64_types.h
+++ b/arch/x86/include/asm/pgtable_64_types.h
@@ -111,9 +111,7 @@ extern unsigned int ptrs_per_p4d;
*/
#define MAXMEM (1UL << MAX_PHYSMEM_BITS)
-#define LDT_PGD_ENTRY_L4 -3UL
-#define LDT_PGD_ENTRY_L5 -112UL
-#define LDT_PGD_ENTRY (pgtable_l5_enabled() ? LDT_PGD_ENTRY_L5 : LDT_PGD_ENTRY_L4)
+#define LDT_PGD_ENTRY -240UL
#define LDT_BASE_ADDR (LDT_PGD_ENTRY << PGDIR_SHIFT)
#define LDT_END_ADDR (LDT_BASE_ADDR + PGDIR_SIZE)
diff --git a/arch/x86/include/asm/qspinlock.h b/arch/x86/include/asm/qspinlock.h
index 87623c6..bd5ac6c 100644
--- a/arch/x86/include/asm/qspinlock.h
+++ b/arch/x86/include/asm/qspinlock.h
@@ -13,12 +13,15 @@
#define queued_fetch_set_pending_acquire queued_fetch_set_pending_acquire
static __always_inline u32 queued_fetch_set_pending_acquire(struct qspinlock *lock)
{
- u32 val = 0;
+ u32 val;
- if (GEN_BINARY_RMWcc(LOCK_PREFIX "btsl", lock->val.counter, c,
- "I", _Q_PENDING_OFFSET))
- val |= _Q_PENDING_VAL;
-
+ /*
+ * We can't use GEN_BINARY_RMWcc() inside an if() stmt because asm goto
+ * and CONFIG_PROFILE_ALL_BRANCHES=y results in a label inside a
+ * statement expression, which GCC doesn't like.
+ */
+ val = GEN_BINARY_RMWcc(LOCK_PREFIX "btsl", lock->val.counter, c,
+ "I", _Q_PENDING_OFFSET) * _Q_PENDING_VAL;
val |= atomic_read(&lock->val) & ~_Q_PENDING_MASK;
return val;
diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h
index 123e669..790ce08 100644
--- a/arch/x86/include/asm/xen/page.h
+++ b/arch/x86/include/asm/xen/page.h
@@ -9,7 +9,7 @@
#include <linux/mm.h>
#include <linux/device.h>
-#include <linux/uaccess.h>
+#include <asm/extable.h>
#include <asm/page.h>
#include <asm/pgtable.h>
@@ -93,12 +93,39 @@ clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops,
*/
static inline int xen_safe_write_ulong(unsigned long *addr, unsigned long val)
{
- return __put_user(val, (unsigned long __user *)addr);
+ int ret = 0;
+
+ asm volatile("1: mov %[val], %[ptr]\n"
+ "2:\n"
+ ".section .fixup, \"ax\"\n"
+ "3: sub $1, %[ret]\n"
+ " jmp 2b\n"
+ ".previous\n"
+ _ASM_EXTABLE(1b, 3b)
+ : [ret] "+r" (ret), [ptr] "=m" (*addr)
+ : [val] "r" (val));
+
+ return ret;
}
-static inline int xen_safe_read_ulong(unsigned long *addr, unsigned long *val)
+static inline int xen_safe_read_ulong(const unsigned long *addr,
+ unsigned long *val)
{
- return __get_user(*val, (unsigned long __user *)addr);
+ int ret = 0;
+ unsigned long rval = ~0ul;
+
+ asm volatile("1: mov %[ptr], %[rval]\n"
+ "2:\n"
+ ".section .fixup, \"ax\"\n"
+ "3: sub $1, %[ret]\n"
+ " jmp 2b\n"
+ ".previous\n"
+ _ASM_EXTABLE(1b, 3b)
+ : [ret] "+r" (ret), [rval] "+r" (rval)
+ : [ptr] "m" (*addr));
+ *val = rval;
+
+ return ret;
}
#ifdef CONFIG_XEN_PV
diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 8c66d2f..36d2696 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -485,7 +485,7 @@ static void mce_report_event(struct pt_regs *regs)
* be somewhat complicated (e.g. segment offset would require an instruction
* parser). So only support physical addresses up to page granuality for now.
*/
-static int mce_usable_address(struct mce *m)
+int mce_usable_address(struct mce *m)
{
if (!(m->status & MCI_STATUS_ADDRV))
return 0;
@@ -505,6 +505,7 @@ static int mce_usable_address(struct mce *m)
return 1;
}
+EXPORT_SYMBOL_GPL(mce_usable_address);
bool mce_is_memory_error(struct mce *m)
{
@@ -534,7 +535,7 @@ bool mce_is_memory_error(struct mce *m)
}
EXPORT_SYMBOL_GPL(mce_is_memory_error);
-static bool mce_is_correctable(struct mce *m)
+bool mce_is_correctable(struct mce *m)
{
if (m->cpuvendor == X86_VENDOR_AMD && m->status & MCI_STATUS_DEFERRED)
return false;
@@ -547,6 +548,7 @@ static bool mce_is_correctable(struct mce *m)
return true;
}
+EXPORT_SYMBOL_GPL(mce_is_correctable);
static bool cec_add_mce(struct mce *m)
{
diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
index 1c72f38..e81a2db 100644
--- a/arch/x86/kernel/cpu/mshyperv.c
+++ b/arch/x86/kernel/cpu/mshyperv.c
@@ -20,6 +20,7 @@
#include <linux/interrupt.h>
#include <linux/irq.h>
#include <linux/kexec.h>
+#include <linux/i8253.h>
#include <asm/processor.h>
#include <asm/hypervisor.h>
#include <asm/hyperv-tlfs.h>
@@ -295,6 +296,16 @@ static void __init ms_hyperv_init_platform(void)
if (efi_enabled(EFI_BOOT))
x86_platform.get_nmi_reason = hv_get_nmi_reason;
+ /*
+ * Hyper-V VMs have a PIT emulation quirk such that zeroing the
+ * counter register during PIT shutdown restarts the PIT. So it
+ * continues to interrupt @18.2 HZ. Setting i8253_clear_counter
+ * to false tells pit_shutdown() not to zero the counter so that
+ * the PIT really is shutdown. Generation 2 VMs don't have a PIT,
+ * and setting this value has no effect.
+ */
+ i8253_clear_counter_on_shutdown = false;
+
#if IS_ENABLED(CONFIG_HYPERV)
/*
* Setup the hook to get control post apic initialization.
diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c
index d9ab49b..0eda91f 100644
--- a/arch/x86/kernel/cpu/vmware.c
+++ b/arch/x86/kernel/cpu/vmware.c
@@ -77,7 +77,7 @@ static __init int setup_vmw_sched_clock(char *s)
}
early_param("no-vmw-sched-clock", setup_vmw_sched_clock);
-static unsigned long long vmware_sched_clock(void)
+static unsigned long long notrace vmware_sched_clock(void)
{
unsigned long long ns;
diff --git a/arch/x86/kernel/ldt.c b/arch/x86/kernel/ldt.c
index ab18e08..6135ae8 100644
--- a/arch/x86/kernel/ldt.c
+++ b/arch/x86/kernel/ldt.c
@@ -199,14 +199,6 @@ static void sanity_check_ldt_mapping(struct mm_struct *mm)
/*
* If PTI is enabled, this maps the LDT into the kernelmode and
* usermode tables for the given mm.
- *
- * There is no corresponding unmap function. Even if the LDT is freed, we
- * leave the PTEs around until the slot is reused or the mm is destroyed.
- * This is harmless: the LDT is always in ordinary memory, and no one will
- * access the freed slot.
- *
- * If we wanted to unmap freed LDTs, we'd also need to do a flush to make
- * it useful, and the flush would slow down modify_ldt().
*/
static int
map_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt, int slot)
@@ -214,8 +206,7 @@ map_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt, int slot)
unsigned long va;
bool is_vmalloc;
spinlock_t *ptl;
- pgd_t *pgd;
- int i;
+ int i, nr_pages;
if (!static_cpu_has(X86_FEATURE_PTI))
return 0;
@@ -229,16 +220,11 @@ map_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt, int slot)
/* Check if the current mappings are sane */
sanity_check_ldt_mapping(mm);
- /*
- * Did we already have the top level entry allocated? We can't
- * use pgd_none() for this because it doens't do anything on
- * 4-level page table kernels.
- */
- pgd = pgd_offset(mm, LDT_BASE_ADDR);
-
is_vmalloc = is_vmalloc_addr(ldt->entries);
- for (i = 0; i * PAGE_SIZE < ldt->nr_entries * LDT_ENTRY_SIZE; i++) {
+ nr_pages = DIV_ROUND_UP(ldt->nr_entries * LDT_ENTRY_SIZE, PAGE_SIZE);
+
+ for (i = 0; i < nr_pages; i++) {
unsigned long offset = i << PAGE_SHIFT;
const void *src = (char *)ldt->entries + offset;
unsigned long pfn;
@@ -272,13 +258,39 @@ map_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt, int slot)
/* Propagate LDT mapping to the user page-table */
map_ldt_struct_to_user(mm);
- va = (unsigned long)ldt_slot_va(slot);
- flush_tlb_mm_range(mm, va, va + LDT_SLOT_STRIDE, PAGE_SHIFT, false);
-
ldt->slot = slot;
return 0;
}
+static void unmap_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt)
+{
+ unsigned long va;
+ int i, nr_pages;
+
+ if (!ldt)
+ return;
+
+ /* LDT map/unmap is only required for PTI */
+ if (!static_cpu_has(X86_FEATURE_PTI))
+ return;
+
+ nr_pages = DIV_ROUND_UP(ldt->nr_entries * LDT_ENTRY_SIZE, PAGE_SIZE);
+
+ for (i = 0; i < nr_pages; i++) {
+ unsigned long offset = i << PAGE_SHIFT;
+ spinlock_t *ptl;
+ pte_t *ptep;
+
+ va = (unsigned long)ldt_slot_va(ldt->slot) + offset;
+ ptep = get_locked_pte(mm, va, &ptl);
+ pte_clear(mm, va, ptep);
+ pte_unmap_unlock(ptep, ptl);
+ }
+
+ va = (unsigned long)ldt_slot_va(ldt->slot);
+ flush_tlb_mm_range(mm, va, va + nr_pages * PAGE_SIZE, PAGE_SHIFT, false);
+}
+
#else /* !CONFIG_PAGE_TABLE_ISOLATION */
static int
@@ -286,6 +298,10 @@ map_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt, int slot)
{
return 0;
}
+
+static void unmap_ldt_struct(struct mm_struct *mm, struct ldt_struct *ldt)
+{
+}
#endif /* CONFIG_PAGE_TABLE_ISOLATION */
static void free_ldt_pgtables(struct mm_struct *mm)
@@ -524,6 +540,7 @@ static int write_ldt(void __user *ptr, unsigned long bytecount, int oldmode)
}
install_ldt(mm, new_ldt);
+ unmap_ldt_struct(mm, old_ldt);
free_ldt_struct(old_ldt);
error = 0;
diff --git a/arch/x86/kernel/vsmp_64.c b/arch/x86/kernel/vsmp_64.c
index 1eae5af..891a75d 100644
--- a/arch/x86/kernel/vsmp_64.c
+++ b/arch/x86/kernel/vsmp_64.c
@@ -26,65 +26,8 @@
#define TOPOLOGY_REGISTER_OFFSET 0x10
-#if defined CONFIG_PCI && defined CONFIG_PARAVIRT_XXL
-/*
- * Interrupt control on vSMPowered systems:
- * ~AC is a shadow of IF. If IF is 'on' AC should be 'off'
- * and vice versa.
- */
-
-asmlinkage __visible unsigned long vsmp_save_fl(void)
-{
- unsigned long flags = native_save_fl();
-
- if (!(flags & X86_EFLAGS_IF) || (flags & X86_EFLAGS_AC))
- flags &= ~X86_EFLAGS_IF;
- return flags;
-}
-PV_CALLEE_SAVE_REGS_THUNK(vsmp_save_fl);
-
-__visible void vsmp_restore_fl(unsigned long flags)
-{
- if (flags & X86_EFLAGS_IF)
- flags &= ~X86_EFLAGS_AC;
- else
- flags |= X86_EFLAGS_AC;
- native_restore_fl(flags);
-}
-PV_CALLEE_SAVE_REGS_THUNK(vsmp_restore_fl);
-
-asmlinkage __visible void vsmp_irq_disable(void)
-{
- unsigned long flags = native_save_fl();
-
- native_restore_fl((flags & ~X86_EFLAGS_IF) | X86_EFLAGS_AC);
-}
-PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_disable);
-
-asmlinkage __visible void vsmp_irq_enable(void)
-{
- unsigned long flags = native_save_fl();
-
- native_restore_fl((flags | X86_EFLAGS_IF) & (~X86_EFLAGS_AC));
-}
-PV_CALLEE_SAVE_REGS_THUNK(vsmp_irq_enable);
-
-static unsigned __init vsmp_patch(u8 type, void *ibuf,
- unsigned long addr, unsigned len)
-{
- switch (type) {
- case PARAVIRT_PATCH(irq.irq_enable):
- case PARAVIRT_PATCH(irq.irq_disable):
- case PARAVIRT_PATCH(irq.save_fl):
- case PARAVIRT_PATCH(irq.restore_fl):
- return paravirt_patch_default(type, ibuf, addr, len);
- default:
- return native_patch(type, ibuf, addr, len);
- }
-
-}
-
-static void __init set_vsmp_pv_ops(void)
+#ifdef CONFIG_PCI
+static void __init set_vsmp_ctl(void)
{
void __iomem *address;
unsigned int cap, ctl, cfg;
@@ -109,28 +52,12 @@ static void __init set_vsmp_pv_ops(void)
}
#endif
- if (cap & ctl & (1 << 4)) {
- /* Setup irq ops and turn on vSMP IRQ fastpath handling */
- pv_ops.irq.irq_disable = PV_CALLEE_SAVE(vsmp_irq_disable);
- pv_ops.irq.irq_enable = PV_CALLEE_SAVE(vsmp_irq_enable);
- pv_ops.irq.save_fl = PV_CALLEE_SAVE(vsmp_save_fl);
- pv_ops.irq.restore_fl = PV_CALLEE_SAVE(vsmp_restore_fl);
- pv_ops.init.patch = vsmp_patch;
- ctl &= ~(1 << 4);
- }
writel(ctl, address + 4);
ctl = readl(address + 4);
pr_info("vSMP CTL: control set to:0x%08x\n", ctl);
early_iounmap(address, 8);
}
-#else
-static void __init set_vsmp_pv_ops(void)
-{
-}
-#endif
-
-#ifdef CONFIG_PCI
static int is_vsmp = -1;
static void __init detect_vsmp_box(void)
@@ -164,11 +91,14 @@ static int is_vsmp_box(void)
{
return 0;
}
+static void __init set_vsmp_ctl(void)
+{
+}
#endif
static void __init vsmp_cap_cpus(void)
{
-#if !defined(CONFIG_X86_VSMP) && defined(CONFIG_SMP)
+#if !defined(CONFIG_X86_VSMP) && defined(CONFIG_SMP) && defined(CONFIG_PCI)
void __iomem *address;
unsigned int cfg, topology, node_shift, maxcpus;
@@ -221,6 +151,6 @@ void __init vsmp_init(void)
vsmp_cap_cpus();
- set_vsmp_pv_ops();
+ set_vsmp_ctl();
return;
}
diff --git a/arch/x86/xen/mmu_pv.c b/arch/x86/xen/mmu_pv.c
index 0d7b3ae..a5d7ed1 100644
--- a/arch/x86/xen/mmu_pv.c
+++ b/arch/x86/xen/mmu_pv.c
@@ -1905,7 +1905,7 @@ void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
init_top_pgt[0] = __pgd(0);
/* Pre-constructed entries are in pfn, so convert to mfn */
- /* L4[272] -> level3_ident_pgt */
+ /* L4[273] -> level3_ident_pgt */
/* L4[511] -> level3_kernel_pgt */
convert_pfn_mfn(init_top_pgt);
@@ -1925,8 +1925,8 @@ void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
addr[0] = (unsigned long)pgd;
addr[1] = (unsigned long)l3;
addr[2] = (unsigned long)l2;
- /* Graft it onto L4[272][0]. Note that we creating an aliasing problem:
- * Both L4[272][0] and L4[511][510] have entries that point to the same
+ /* Graft it onto L4[273][0]. Note that we creating an aliasing problem:
+ * Both L4[273][0] and L4[511][510] have entries that point to the same
* L2 (PMD) tables. Meaning that if you modify it in __va space
* it will be also modified in the __ka space! (But if you just
* modify the PMD table to point to other PTE's or none, then you
diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c
index b067317..055e37e 100644
--- a/arch/x86/xen/p2m.c
+++ b/arch/x86/xen/p2m.c
@@ -656,8 +656,7 @@ bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn)
/*
* The interface requires atomic updates on p2m elements.
- * xen_safe_write_ulong() is using __put_user which does an atomic
- * store via asm().
+ * xen_safe_write_ulong() is using an atomic store via asm().
*/
if (likely(!xen_safe_write_ulong(xen_p2m_addr + pfn, mfn)))
return true;
diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
index 441c882..1c8a881 100644
--- a/arch/x86/xen/spinlock.c
+++ b/arch/x86/xen/spinlock.c
@@ -9,6 +9,7 @@
#include <linux/log2.h>
#include <linux/gfp.h>
#include <linux/slab.h>
+#include <linux/atomic.h>
#include <asm/paravirt.h>
#include <asm/qspinlock.h>
@@ -21,6 +22,7 @@
static DEFINE_PER_CPU(int, lock_kicker_irq) = -1;
static DEFINE_PER_CPU(char *, irq_name);
+static DEFINE_PER_CPU(atomic_t, xen_qlock_wait_nest);
static bool xen_pvspin = true;
static void xen_qlock_kick(int cpu)
@@ -39,25 +41,25 @@ static void xen_qlock_kick(int cpu)
*/
static void xen_qlock_wait(u8 *byte, u8 val)
{
- unsigned long flags;
int irq = __this_cpu_read(lock_kicker_irq);
+ atomic_t *nest_cnt = this_cpu_ptr(&xen_qlock_wait_nest);
/* If kicker interrupts not initialized yet, just spin */
if (irq == -1 || in_nmi())
return;
- /* Guard against reentry. */
- local_irq_save(flags);
+ /* Detect reentry. */
+ atomic_inc(nest_cnt);
- /* If irq pending already clear it. */
- if (xen_test_irq_pending(irq)) {
+ /* If irq pending already and no nested call clear it. */
+ if (atomic_read(nest_cnt) == 1 && xen_test_irq_pending(irq)) {
xen_clear_irq_pending(irq);
} else if (READ_ONCE(*byte) == val) {
/* Block until irq becomes pending (or a spurious wakeup) */
xen_poll_irq(irq);
}
- local_irq_restore(flags);
+ atomic_dec(nest_cnt);
}
static irqreturn_t dummy_handler(int irq, void *dev_id)
diff --git a/arch/xtensa/include/asm/processor.h b/arch/xtensa/include/asm/processor.h
index be9bfd9..34a2301 100644
--- a/arch/xtensa/include/asm/processor.h
+++ b/arch/xtensa/include/asm/processor.h
@@ -23,7 +23,11 @@
# error Linux requires the Xtensa Windowed Registers Option.
#endif
-#define ARCH_SLAB_MINALIGN XCHAL_DATA_WIDTH
+/* Xtensa ABI requires stack alignment to be at least 16 */
+
+#define STACK_ALIGN (XCHAL_DATA_WIDTH > 16 ? XCHAL_DATA_WIDTH : 16)
+
+#define ARCH_SLAB_MINALIGN STACK_ALIGN
/*
* User space process size: 1 GB.
diff --git a/arch/xtensa/kernel/head.S b/arch/xtensa/kernel/head.S
index 2f76118..9053a56 100644
--- a/arch/xtensa/kernel/head.S
+++ b/arch/xtensa/kernel/head.S
@@ -88,9 +88,12 @@
initialize_mmu
#if defined(CONFIG_MMU) && XCHAL_HAVE_PTP_MMU && XCHAL_HAVE_SPANNING_WAY
rsr a2, excsave1
- movi a3, 0x08000000
+ movi a3, XCHAL_KSEG_PADDR
+ bltu a2, a3, 1f
+ sub a2, a2, a3
+ movi a3, XCHAL_KSEG_SIZE
bgeu a2, a3, 1f
- movi a3, 0xd0000000
+ movi a3, XCHAL_KSEG_CACHED_VADDR
add a2, a2, a3
wsr a2, excsave1
1:
diff --git a/block/bio.c b/block/bio.c
index d5368a4..4f4d988 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -605,6 +605,7 @@ void __bio_clone_fast(struct bio *bio, struct bio *bio_src)
if (bio_flagged(bio_src, BIO_THROTTLED))
bio_set_flag(bio, BIO_THROTTLED);
bio->bi_opf = bio_src->bi_opf;
+ bio->bi_ioprio = bio_src->bi_ioprio;
bio->bi_write_hint = bio_src->bi_write_hint;
bio->bi_iter = bio_src->bi_iter;
bio->bi_io_vec = bio_src->bi_io_vec;
@@ -1260,6 +1261,7 @@ struct bio *bio_copy_user_iov(struct request_queue *q,
if (ret)
goto cleanup;
} else {
+ zero_fill_bio(bio);
iov_iter_advance(iter, bio->bi_iter.bi_size);
}
diff --git a/block/blk-core.c b/block/blk-core.c
index ce12515f..deb5693 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -798,9 +798,8 @@ void blk_cleanup_queue(struct request_queue *q)
* dispatch may still be in-progress since we dispatch requests
* from more than one contexts.
*
- * No need to quiesce queue if it isn't initialized yet since
- * blk_freeze_queue() should be enough for cases of passthrough
- * request.
+ * We rely on driver to deal with the race in case that queue
+ * initialization isn't done.
*/
if (q->mq_ops && blk_queue_init_done(q))
blk_mq_quiesce_queue(q);
diff --git a/block/blk-lib.c b/block/blk-lib.c
index 76f867e..5f2c429 100644
--- a/block/blk-lib.c
+++ b/block/blk-lib.c
@@ -51,16 +51,14 @@ int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
if ((sector | nr_sects) & bs_mask)
return -EINVAL;
+ if (!nr_sects)
+ return -EINVAL;
+
while (nr_sects) {
- unsigned int req_sects = nr_sects;
- sector_t end_sect;
+ sector_t req_sects = min_t(sector_t, nr_sects,
+ bio_allowed_max_sectors(q));
- if (!req_sects)
- goto fail;
- if (req_sects > UINT_MAX >> 9)
- req_sects = UINT_MAX >> 9;
-
- end_sect = sector + req_sects;
+ WARN_ON_ONCE((req_sects << 9) > UINT_MAX);
bio = blk_next_bio(bio, 0, gfp_mask);
bio->bi_iter.bi_sector = sector;
@@ -68,8 +66,8 @@ int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
bio_set_op_attrs(bio, op, 0);
bio->bi_iter.bi_size = req_sects << 9;
+ sector += req_sects;
nr_sects -= req_sects;
- sector = end_sect;
/*
* We can loop for a long time in here, if someone does
@@ -82,14 +80,6 @@ int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
*biop = bio;
return 0;
-
-fail:
- if (bio) {
- submit_bio_wait(bio);
- bio_put(bio);
- }
- *biop = NULL;
- return -EOPNOTSUPP;
}
EXPORT_SYMBOL(__blkdev_issue_discard);
@@ -161,7 +151,7 @@ static int __blkdev_issue_write_same(struct block_device *bdev, sector_t sector,
return -EOPNOTSUPP;
/* Ensure that max_write_same_sectors doesn't overflow bi_size */
- max_write_same_sectors = UINT_MAX >> 9;
+ max_write_same_sectors = bio_allowed_max_sectors(q);
while (nr_sects) {
bio = blk_next_bio(bio, 1, gfp_mask);
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 6b5ad27..e7696c4 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -46,7 +46,7 @@ static inline bool bio_will_gap(struct request_queue *q,
bio_get_first_bvec(prev_rq->bio, &pb);
else
bio_get_first_bvec(prev, &pb);
- if (pb.bv_offset)
+ if (pb.bv_offset & queue_virt_boundary(q))
return true;
/*
@@ -90,7 +90,8 @@ static struct bio *blk_bio_discard_split(struct request_queue *q,
/* Zero-sector (unknown) and one-sector granularities are the same. */
granularity = max(q->limits.discard_granularity >> 9, 1U);
- max_discard_sectors = min(q->limits.max_discard_sectors, UINT_MAX >> 9);
+ max_discard_sectors = min(q->limits.max_discard_sectors,
+ bio_allowed_max_sectors(q));
max_discard_sectors -= max_discard_sectors % granularity;
if (unlikely(!max_discard_sectors)) {
diff --git a/block/blk.h b/block/blk.h
index a1841b8..0089fef 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -169,7 +169,7 @@ static inline bool biovec_phys_mergeable(struct request_queue *q,
static inline bool __bvec_gap_to_prev(struct request_queue *q,
struct bio_vec *bprv, unsigned int offset)
{
- return offset ||
+ return (offset & queue_virt_boundary(q)) ||
((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q));
}
@@ -396,6 +396,16 @@ static inline unsigned long blk_rq_deadline(struct request *rq)
}
/*
+ * The max size one bio can handle is UINT_MAX becasue bvec_iter.bi_size
+ * is defined as 'unsigned int', meantime it has to aligned to with logical
+ * block size which is the minimum accepted unit by hardware.
+ */
+static inline unsigned int bio_allowed_max_sectors(struct request_queue *q)
+{
+ return round_down(UINT_MAX, queue_logical_block_size(q)) >> 9;
+}
+
+/*
* Internal io_context interface
*/
void get_io_context(struct io_context *ioc);
diff --git a/block/bounce.c b/block/bounce.c
index 36869af..559c55b 100644
--- a/block/bounce.c
+++ b/block/bounce.c
@@ -248,6 +248,7 @@ static struct bio *bounce_clone_bio(struct bio *bio_src, gfp_t gfp_mask,
return NULL;
bio->bi_disk = bio_src->bi_disk;
bio->bi_opf = bio_src->bi_opf;
+ bio->bi_ioprio = bio_src->bi_ioprio;
bio->bi_write_hint = bio_src->bi_write_hint;
bio->bi_iter.bi_sector = bio_src->bi_iter.bi_sector;
bio->bi_iter.bi_size = bio_src->bi_iter.bi_size;
diff --git a/crypto/crypto_user_base.c b/crypto/crypto_user_base.c
index e41f6cc..784748d 100644
--- a/crypto/crypto_user_base.c
+++ b/crypto/crypto_user_base.c
@@ -84,7 +84,7 @@ static int crypto_report_cipher(struct sk_buff *skb, struct crypto_alg *alg)
{
struct crypto_report_cipher rcipher;
- strlcpy(rcipher.type, "cipher", sizeof(rcipher.type));
+ strncpy(rcipher.type, "cipher", sizeof(rcipher.type));
rcipher.blocksize = alg->cra_blocksize;
rcipher.min_keysize = alg->cra_cipher.cia_min_keysize;
@@ -103,7 +103,7 @@ static int crypto_report_comp(struct sk_buff *skb, struct crypto_alg *alg)
{
struct crypto_report_comp rcomp;
- strlcpy(rcomp.type, "compression", sizeof(rcomp.type));
+ strncpy(rcomp.type, "compression", sizeof(rcomp.type));
if (nla_put(skb, CRYPTOCFGA_REPORT_COMPRESS,
sizeof(struct crypto_report_comp), &rcomp))
goto nla_put_failure;
@@ -117,7 +117,7 @@ static int crypto_report_acomp(struct sk_buff *skb, struct crypto_alg *alg)
{
struct crypto_report_acomp racomp;
- strlcpy(racomp.type, "acomp", sizeof(racomp.type));
+ strncpy(racomp.type, "acomp", sizeof(racomp.type));
if (nla_put(skb, CRYPTOCFGA_REPORT_ACOMP,
sizeof(struct crypto_report_acomp), &racomp))
@@ -132,7 +132,7 @@ static int crypto_report_akcipher(struct sk_buff *skb, struct crypto_alg *alg)
{
struct crypto_report_akcipher rakcipher;
- strlcpy(rakcipher.type, "akcipher", sizeof(rakcipher.type));
+ strncpy(rakcipher.type, "akcipher", sizeof(rakcipher.type));
if (nla_put(skb, CRYPTOCFGA_REPORT_AKCIPHER,
sizeof(struct crypto_report_akcipher), &rakcipher))
@@ -147,7 +147,7 @@ static int crypto_report_kpp(struct sk_buff *skb, struct crypto_alg *alg)
{
struct crypto_report_kpp rkpp;
- strlcpy(rkpp.type, "kpp", sizeof(rkpp.type));
+ strncpy(rkpp.type, "kpp", sizeof(rkpp.type));
if (nla_put(skb, CRYPTOCFGA_REPORT_KPP,
sizeof(struct crypto_report_kpp), &rkpp))
@@ -161,10 +161,10 @@ static int crypto_report_kpp(struct sk_buff *skb, struct crypto_alg *alg)
static int crypto_report_one(struct crypto_alg *alg,
struct crypto_user_alg *ualg, struct sk_buff *skb)
{
- strlcpy(ualg->cru_name, alg->cra_name, sizeof(ualg->cru_name));
- strlcpy(ualg->cru_driver_name, alg->cra_driver_name,
+ strncpy(ualg->cru_name, alg->cra_name, sizeof(ualg->cru_name));
+ strncpy(ualg->cru_driver_name, alg->cra_driver_name,
sizeof(ualg->cru_driver_name));
- strlcpy(ualg->cru_module_name, module_name(alg->cra_module),
+ strncpy(ualg->cru_module_name, module_name(alg->cra_module),
sizeof(ualg->cru_module_name));
ualg->cru_type = 0;
@@ -177,7 +177,7 @@ static int crypto_report_one(struct crypto_alg *alg,
if (alg->cra_flags & CRYPTO_ALG_LARVAL) {
struct crypto_report_larval rl;
- strlcpy(rl.type, "larval", sizeof(rl.type));
+ strncpy(rl.type, "larval", sizeof(rl.type));
if (nla_put(skb, CRYPTOCFGA_REPORT_LARVAL,
sizeof(struct crypto_report_larval), &rl))
goto nla_put_failure;
diff --git a/crypto/crypto_user_stat.c b/crypto/crypto_user_stat.c
index 021ad06..1dfaa0c 100644
--- a/crypto/crypto_user_stat.c
+++ b/crypto/crypto_user_stat.c
@@ -37,6 +37,8 @@ static int crypto_report_aead(struct sk_buff *skb, struct crypto_alg *alg)
u64 v64;
u32 v32;
+ memset(&raead, 0, sizeof(raead));
+
strncpy(raead.type, "aead", sizeof(raead.type));
v32 = atomic_read(&alg->encrypt_cnt);
@@ -65,6 +67,8 @@ static int crypto_report_cipher(struct sk_buff *skb, struct crypto_alg *alg)
u64 v64;
u32 v32;
+ memset(&rcipher, 0, sizeof(rcipher));
+
strlcpy(rcipher.type, "cipher", sizeof(rcipher.type));
v32 = atomic_read(&alg->encrypt_cnt);
@@ -93,6 +97,8 @@ static int crypto_report_comp(struct sk_buff *skb, struct crypto_alg *alg)
u64 v64;
u32 v32;
+ memset(&rcomp, 0, sizeof(rcomp));
+
strlcpy(rcomp.type, "compression", sizeof(rcomp.type));
v32 = atomic_read(&alg->compress_cnt);
rcomp.stat_compress_cnt = v32;
@@ -120,6 +126,8 @@ static int crypto_report_acomp(struct sk_buff *skb, struct crypto_alg *alg)
u64 v64;
u32 v32;
+ memset(&racomp, 0, sizeof(racomp));
+
strlcpy(racomp.type, "acomp", sizeof(racomp.type));
v32 = atomic_read(&alg->compress_cnt);
racomp.stat_compress_cnt = v32;
@@ -147,6 +155,8 @@ static int crypto_report_akcipher(struct sk_buff *skb, struct crypto_alg *alg)
u64 v64;
u32 v32;
+ memset(&rakcipher, 0, sizeof(rakcipher));
+
strncpy(rakcipher.type, "akcipher", sizeof(rakcipher.type));
v32 = atomic_read(&alg->encrypt_cnt);
rakcipher.stat_encrypt_cnt = v32;
@@ -177,6 +187,8 @@ static int crypto_report_kpp(struct sk_buff *skb, struct crypto_alg *alg)
struct crypto_stat rkpp;
u32 v;
+ memset(&rkpp, 0, sizeof(rkpp));
+
strlcpy(rkpp.type, "kpp", sizeof(rkpp.type));
v = atomic_read(&alg->setsecret_cnt);
@@ -203,6 +215,8 @@ static int crypto_report_ahash(struct sk_buff *skb, struct crypto_alg *alg)
u64 v64;
u32 v32;
+ memset(&rhash, 0, sizeof(rhash));
+
strncpy(rhash.type, "ahash", sizeof(rhash.type));
v32 = atomic_read(&alg->hash_cnt);
@@ -227,6 +241,8 @@ static int crypto_report_shash(struct sk_buff *skb, struct crypto_alg *alg)
u64 v64;
u32 v32;
+ memset(&rhash, 0, sizeof(rhash));
+
strncpy(rhash.type, "shash", sizeof(rhash.type));
v32 = atomic_read(&alg->hash_cnt);
@@ -251,6 +267,8 @@ static int crypto_report_rng(struct sk_buff *skb, struct crypto_alg *alg)
u64 v64;
u32 v32;
+ memset(&rrng, 0, sizeof(rrng));
+
strncpy(rrng.type, "rng", sizeof(rrng.type));
v32 = atomic_read(&alg->generate_cnt);
@@ -275,6 +293,8 @@ static int crypto_reportstat_one(struct crypto_alg *alg,
struct crypto_user_alg *ualg,
struct sk_buff *skb)
{
+ memset(ualg, 0, sizeof(*ualg));
+
strlcpy(ualg->cru_name, alg->cra_name, sizeof(ualg->cru_name));
strlcpy(ualg->cru_driver_name, alg->cra_driver_name,
sizeof(ualg->cru_driver_name));
@@ -291,6 +311,7 @@ static int crypto_reportstat_one(struct crypto_alg *alg,
if (alg->cra_flags & CRYPTO_ALG_LARVAL) {
struct crypto_stat rl;
+ memset(&rl, 0, sizeof(rl));
strlcpy(rl.type, "larval", sizeof(rl.type));
if (nla_put(skb, CRYPTOCFGA_STAT_LARVAL,
sizeof(struct crypto_stat), &rl))
diff --git a/crypto/simd.c b/crypto/simd.c
index ea7240b..78e8d03 100644
--- a/crypto/simd.c
+++ b/crypto/simd.c
@@ -124,8 +124,9 @@ static int simd_skcipher_init(struct crypto_skcipher *tfm)
ctx->cryptd_tfm = cryptd_tfm;
- reqsize = sizeof(struct skcipher_request);
- reqsize += crypto_skcipher_reqsize(&cryptd_tfm->base);
+ reqsize = crypto_skcipher_reqsize(cryptd_skcipher_child(cryptd_tfm));
+ reqsize = max(reqsize, crypto_skcipher_reqsize(&cryptd_tfm->base));
+ reqsize += sizeof(struct skcipher_request);
crypto_skcipher_set_reqsize(tfm, reqsize);
diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig
index 8f3a444..7cea769c 100644
--- a/drivers/acpi/Kconfig
+++ b/drivers/acpi/Kconfig
@@ -512,7 +512,7 @@
config XPOWER_PMIC_OPREGION
bool "ACPI operation region support for XPower AXP288 PMIC"
- depends on MFD_AXP20X_I2C && IOSF_MBI
+ depends on MFD_AXP20X_I2C && IOSF_MBI=y
help
This config adds ACPI operation region support for XPower AXP288 PMIC.
diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index f8c638f..14d9f5b 100644
--- a/drivers/acpi/nfit/core.c
+++ b/drivers/acpi/nfit/core.c
@@ -2928,9 +2928,9 @@ static int acpi_nfit_query_poison(struct acpi_nfit_desc *acpi_desc)
return rc;
if (ars_status_process_records(acpi_desc))
- return -ENOMEM;
+ dev_err(acpi_desc->dev, "Failed to process ARS records\n");
- return 0;
+ return rc;
}
static int ars_register(struct acpi_nfit_desc *acpi_desc,
@@ -3341,8 +3341,6 @@ static int acpi_nfit_clear_to_send(struct nvdimm_bus_descriptor *nd_desc,
struct nvdimm *nvdimm, unsigned int cmd)
{
struct acpi_nfit_desc *acpi_desc = to_acpi_nfit_desc(nd_desc);
- struct nfit_spa *nfit_spa;
- int rc = 0;
if (nvdimm)
return 0;
@@ -3355,17 +3353,10 @@ static int acpi_nfit_clear_to_send(struct nvdimm_bus_descriptor *nd_desc,
* just needs guarantees that any ARS it initiates are not
* interrupted by any intervening start requests from userspace.
*/
- mutex_lock(&acpi_desc->init_mutex);
- list_for_each_entry(nfit_spa, &acpi_desc->spas, list)
- if (acpi_desc->scrub_spa
- || test_bit(ARS_REQ_SHORT, &nfit_spa->ars_state)
- || test_bit(ARS_REQ_LONG, &nfit_spa->ars_state)) {
- rc = -EBUSY;
- break;
- }
- mutex_unlock(&acpi_desc->init_mutex);
+ if (work_busy(&acpi_desc->dwork.work))
+ return -EBUSY;
- return rc;
+ return 0;
}
int acpi_nfit_ars_rescan(struct acpi_nfit_desc *acpi_desc,
diff --git a/drivers/acpi/nfit/mce.c b/drivers/acpi/nfit/mce.c
index e9626bf..d6c1b10 100644
--- a/drivers/acpi/nfit/mce.c
+++ b/drivers/acpi/nfit/mce.c
@@ -25,8 +25,12 @@ static int nfit_handle_mce(struct notifier_block *nb, unsigned long val,
struct acpi_nfit_desc *acpi_desc;
struct nfit_spa *nfit_spa;
- /* We only care about memory errors */
- if (!mce_is_memory_error(mce))
+ /* We only care about uncorrectable memory errors */
+ if (!mce_is_memory_error(mce) || mce_is_correctable(mce))
+ return NOTIFY_DONE;
+
+ /* Verify the address reported in the MCE is valid. */
+ if (!mce_usable_address(mce))
return NOTIFY_DONE;
/*
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index 6e59464..a7f5202 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4553,7 +4553,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
/* These specific Samsung models/firmware-revs do not handle LPM well */
{ "SAMSUNG MZMPC128HBFU-000MV", "CXM14M1Q", ATA_HORKAGE_NOLPM, },
{ "SAMSUNG SSD PM830 mSATA *", "CXM13D1Q", ATA_HORKAGE_NOLPM, },
- { "SAMSUNG MZ7TD256HAFV-000L9", "DXT02L5Q", ATA_HORKAGE_NOLPM, },
+ { "SAMSUNG MZ7TD256HAFV-000L9", NULL, ATA_HORKAGE_NOLPM, },
/* devices that don't properly handle queued TRIM commands */
{ "Micron_M500IT_*", "MU01", ATA_HORKAGE_NO_NCQ_TRIM |
diff --git a/drivers/ata/sata_rcar.c b/drivers/ata/sata_rcar.c
index 10ecb232..4b1ff5b 100644
--- a/drivers/ata/sata_rcar.c
+++ b/drivers/ata/sata_rcar.c
@@ -1,14 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0+
/*
* Renesas R-Car SATA driver
*
* Author: Vladimir Barinov <source@cogentembedded.com>
* Copyright (C) 2013-2015 Cogent Embedded, Inc.
* Copyright (C) 2013-2015 Renesas Solutions Corp.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the
- * Free Software Foundation; either version 2 of the License, or (at your
- * option) any later version.
*/
#include <linux/kernel.h>
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index a8cfa01..fb23578 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -4148,10 +4148,11 @@ static int __floppy_read_block_0(struct block_device *bdev, int drive)
bio.bi_end_io = floppy_rb0_cb;
bio_set_op_attrs(&bio, REQ_OP_READ, 0);
+ init_completion(&cbdata.complete);
+
submit_bio(&bio);
process_fd_request();
- init_completion(&cbdata.complete);
wait_for_completion(&cbdata.complete);
__free_page(page);
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index 56452ca..0ed4b20 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -1919,6 +1919,7 @@ static int negotiate_mq(struct blkfront_info *info)
GFP_KERNEL);
if (!info->rinfo) {
xenbus_dev_fatal(info->xbdev, -ENOMEM, "allocating ring_info structure");
+ info->nr_rings = 0;
return -ENOMEM;
}
diff --git a/drivers/clk/clk-fixed-factor.c b/drivers/clk/clk-fixed-factor.c
index ef0ca94..ff83e89 100644
--- a/drivers/clk/clk-fixed-factor.c
+++ b/drivers/clk/clk-fixed-factor.c
@@ -210,6 +210,7 @@ static int of_fixed_factor_clk_remove(struct platform_device *pdev)
{
struct clk *clk = platform_get_drvdata(pdev);
+ of_clk_del_provider(pdev->dev.of_node);
clk_unregister_fixed_factor(clk);
return 0;
diff --git a/drivers/clk/meson/axg.c b/drivers/clk/meson/axg.c
index c981159..792735d 100644
--- a/drivers/clk/meson/axg.c
+++ b/drivers/clk/meson/axg.c
@@ -325,6 +325,7 @@ static struct clk_regmap axg_fclk_div2 = {
.ops = &clk_regmap_gate_ops,
.parent_names = (const char *[]){ "fclk_div2_div" },
.num_parents = 1,
+ .flags = CLK_IS_CRITICAL,
},
};
@@ -349,6 +350,18 @@ static struct clk_regmap axg_fclk_div3 = {
.ops = &clk_regmap_gate_ops,
.parent_names = (const char *[]){ "fclk_div3_div" },
.num_parents = 1,
+ /*
+ * FIXME:
+ * This clock, as fdiv2, is used by the SCPI FW and is required
+ * by the platform to operate correctly.
+ * Until the following condition are met, we need this clock to
+ * be marked as critical:
+ * a) The SCPI generic driver claims and enable all the clocks
+ * it needs
+ * b) CCF has a clock hand-off mechanism to make the sure the
+ * clock stays on until the proper driver comes along
+ */
+ .flags = CLK_IS_CRITICAL,
},
};
diff --git a/drivers/clk/meson/gxbb.c b/drivers/clk/meson/gxbb.c
index 9309cfa..4ada966 100644
--- a/drivers/clk/meson/gxbb.c
+++ b/drivers/clk/meson/gxbb.c
@@ -506,6 +506,18 @@ static struct clk_regmap gxbb_fclk_div3 = {
.ops = &clk_regmap_gate_ops,
.parent_names = (const char *[]){ "fclk_div3_div" },
.num_parents = 1,
+ /*
+ * FIXME:
+ * This clock, as fdiv2, is used by the SCPI FW and is required
+ * by the platform to operate correctly.
+ * Until the following condition are met, we need this clock to
+ * be marked as critical:
+ * a) The SCPI generic driver claims and enable all the clocks
+ * it needs
+ * b) CCF has a clock hand-off mechanism to make the sure the
+ * clock stays on until the proper driver comes along
+ */
+ .flags = CLK_IS_CRITICAL,
},
};
diff --git a/drivers/clk/qcom/gcc-qcs404.c b/drivers/clk/qcom/gcc-qcs404.c
index e4ca6a4..ef1b267 100644
--- a/drivers/clk/qcom/gcc-qcs404.c
+++ b/drivers/clk/qcom/gcc-qcs404.c
@@ -265,7 +265,7 @@ static struct clk_fixed_factor cxo = {
.div = 1,
.hw.init = &(struct clk_init_data){
.name = "cxo",
- .parent_names = (const char *[]){ "xo_board" },
+ .parent_names = (const char *[]){ "xo-board" },
.num_parents = 1,
.ops = &clk_fixed_factor_ops,
},
diff --git a/drivers/clocksource/i8253.c b/drivers/clocksource/i8253.c
index 9c38895..d4350bb 100644
--- a/drivers/clocksource/i8253.c
+++ b/drivers/clocksource/i8253.c
@@ -20,6 +20,13 @@
DEFINE_RAW_SPINLOCK(i8253_lock);
EXPORT_SYMBOL(i8253_lock);
+/*
+ * Handle PIT quirk in pit_shutdown() where zeroing the counter register
+ * restarts the PIT, negating the shutdown. On platforms with the quirk,
+ * platform specific code can set this to false.
+ */
+bool i8253_clear_counter_on_shutdown __ro_after_init = true;
+
#ifdef CONFIG_CLKSRC_I8253
/*
* Since the PIT overflows every tick, its not very useful
@@ -109,8 +116,11 @@ static int pit_shutdown(struct clock_event_device *evt)
raw_spin_lock(&i8253_lock);
outb_p(0x30, PIT_MODE);
- outb_p(0, PIT_CH0);
- outb_p(0, PIT_CH0);
+
+ if (i8253_clear_counter_on_shutdown) {
+ outb_p(0, PIT_CH0);
+ outb_p(0, PIT_CH0);
+ }
raw_spin_unlock(&i8253_lock);
return 0;
diff --git a/drivers/cpufreq/imx6q-cpufreq.c b/drivers/cpufreq/imx6q-cpufreq.c
index 8cfee0a..d8c3595 100644
--- a/drivers/cpufreq/imx6q-cpufreq.c
+++ b/drivers/cpufreq/imx6q-cpufreq.c
@@ -160,8 +160,13 @@ static int imx6q_set_target(struct cpufreq_policy *policy, unsigned int index)
/* Ensure the arm clock divider is what we expect */
ret = clk_set_rate(clks[ARM].clk, new_freq * 1000);
if (ret) {
+ int ret1;
+
dev_err(cpu_dev, "failed to set clock rate: %d\n", ret);
- regulator_set_voltage_tol(arm_reg, volt_old, 0);
+ ret1 = regulator_set_voltage_tol(arm_reg, volt_old, 0);
+ if (ret1)
+ dev_warn(cpu_dev,
+ "failed to restore vddarm voltage: %d\n", ret1);
return ret;
}
diff --git a/drivers/cpuidle/cpuidle-arm.c b/drivers/cpuidle/cpuidle-arm.c
index 073557f..3a407a3 100644
--- a/drivers/cpuidle/cpuidle-arm.c
+++ b/drivers/cpuidle/cpuidle-arm.c
@@ -82,7 +82,6 @@ static int __init arm_idle_init_cpu(int cpu)
{
int ret;
struct cpuidle_driver *drv;
- struct cpuidle_device *dev;
drv = kmemdup(&arm_idle_driver, sizeof(*drv), GFP_KERNEL);
if (!drv)
@@ -103,13 +102,6 @@ static int __init arm_idle_init_cpu(int cpu)
goto out_kfree_drv;
}
- ret = cpuidle_register_driver(drv);
- if (ret) {
- if (ret != -EBUSY)
- pr_err("Failed to register cpuidle driver\n");
- goto out_kfree_drv;
- }
-
/*
* Call arch CPU operations in order to initialize
* idle states suspend back-end specific data
@@ -117,37 +109,21 @@ static int __init arm_idle_init_cpu(int cpu)
ret = arm_cpuidle_init(cpu);
/*
- * Skip the cpuidle device initialization if the reported
+ * Allow the initialization to continue for other CPUs, if the reported
* failure is a HW misconfiguration/breakage (-ENXIO).
*/
- if (ret == -ENXIO)
- return 0;
-
if (ret) {
pr_err("CPU %d failed to init idle CPU ops\n", cpu);
- goto out_unregister_drv;
+ ret = ret == -ENXIO ? 0 : ret;
+ goto out_kfree_drv;
}
- dev = kzalloc(sizeof(*dev), GFP_KERNEL);
- if (!dev) {
- ret = -ENOMEM;
- goto out_unregister_drv;
- }
- dev->cpu = cpu;
-
- ret = cpuidle_register_device(dev);
- if (ret) {
- pr_err("Failed to register cpuidle device for CPU %d\n",
- cpu);
- goto out_kfree_dev;
- }
+ ret = cpuidle_register(drv, NULL);
+ if (ret)
+ goto out_kfree_drv;
return 0;
-out_kfree_dev:
- kfree(dev);
-out_unregister_drv:
- cpuidle_unregister_driver(drv);
out_kfree_drv:
kfree(drv);
return ret;
@@ -178,9 +154,7 @@ static int __init arm_idle_init(void)
while (--cpu >= 0) {
dev = per_cpu(cpuidle_devices, cpu);
drv = cpuidle_get_cpu_driver(dev);
- cpuidle_unregister_device(dev);
- cpuidle_unregister_driver(drv);
- kfree(dev);
+ cpuidle_unregister(drv);
kfree(drv);
}
diff --git a/drivers/crypto/hisilicon/sec/sec_algs.c b/drivers/crypto/hisilicon/sec/sec_algs.c
index f7d6d69..cdc4f9a 100644
--- a/drivers/crypto/hisilicon/sec/sec_algs.c
+++ b/drivers/crypto/hisilicon/sec/sec_algs.c
@@ -732,6 +732,7 @@ static int sec_alg_skcipher_crypto(struct skcipher_request *skreq,
int *splits_in_nents;
int *splits_out_nents = NULL;
struct sec_request_el *el, *temp;
+ bool split = skreq->src != skreq->dst;
mutex_init(&sec_req->lock);
sec_req->req_base = &skreq->base;
@@ -750,7 +751,7 @@ static int sec_alg_skcipher_crypto(struct skcipher_request *skreq,
if (ret)
goto err_free_split_sizes;
- if (skreq->src != skreq->dst) {
+ if (split) {
sec_req->len_out = sg_nents(skreq->dst);
ret = sec_map_and_split_sg(skreq->dst, split_sizes, steps,
&splits_out, &splits_out_nents,
@@ -785,8 +786,9 @@ static int sec_alg_skcipher_crypto(struct skcipher_request *skreq,
split_sizes[i],
skreq->src != skreq->dst,
splits_in[i], splits_in_nents[i],
- splits_out[i],
- splits_out_nents[i], info);
+ split ? splits_out[i] : NULL,
+ split ? splits_out_nents[i] : 0,
+ info);
if (IS_ERR(el)) {
ret = PTR_ERR(el);
goto err_free_elements;
@@ -806,13 +808,6 @@ static int sec_alg_skcipher_crypto(struct skcipher_request *skreq,
* more refined but this is unlikely to happen so no need.
*/
- /* Cleanup - all elements in pointer arrays have been coppied */
- kfree(splits_in_nents);
- kfree(splits_in);
- kfree(splits_out_nents);
- kfree(splits_out);
- kfree(split_sizes);
-
/* Grab a big lock for a long time to avoid concurrency issues */
mutex_lock(&queue->queuelock);
@@ -827,13 +822,13 @@ static int sec_alg_skcipher_crypto(struct skcipher_request *skreq,
(!queue->havesoftqueue ||
kfifo_avail(&queue->softqueue) > steps)) ||
!list_empty(&ctx->backlog)) {
+ ret = -EBUSY;
if ((skreq->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG)) {
list_add_tail(&sec_req->backlog_head, &ctx->backlog);
mutex_unlock(&queue->queuelock);
- return -EBUSY;
+ goto out;
}
- ret = -EBUSY;
mutex_unlock(&queue->queuelock);
goto err_free_elements;
}
@@ -842,7 +837,15 @@ static int sec_alg_skcipher_crypto(struct skcipher_request *skreq,
if (ret)
goto err_free_elements;
- return -EINPROGRESS;
+ ret = -EINPROGRESS;
+out:
+ /* Cleanup - all elements in pointer arrays have been copied */
+ kfree(splits_in_nents);
+ kfree(splits_in);
+ kfree(splits_out_nents);
+ kfree(splits_out);
+ kfree(split_sizes);
+ return ret;
err_free_elements:
list_for_each_entry_safe(el, temp, &sec_req->elements, head) {
@@ -854,7 +857,7 @@ static int sec_alg_skcipher_crypto(struct skcipher_request *skreq,
crypto_skcipher_ivsize(atfm),
DMA_BIDIRECTIONAL);
err_unmap_out_sg:
- if (skreq->src != skreq->dst)
+ if (split)
sec_unmap_sg_on_err(skreq->dst, steps, splits_out,
splits_out_nents, sec_req->len_out,
info->dev);
diff --git a/drivers/firmware/efi/arm-init.c b/drivers/firmware/efi/arm-init.c
index 388a929..1a6a77d 100644
--- a/drivers/firmware/efi/arm-init.c
+++ b/drivers/firmware/efi/arm-init.c
@@ -265,6 +265,10 @@ void __init efi_init(void)
(params.mmap & ~PAGE_MASK)));
init_screen_info();
+
+ /* ARM does not permit early mappings to persist across paging_init() */
+ if (IS_ENABLED(CONFIG_ARM))
+ efi_memmap_unmap();
}
static int __init register_gop_device(void)
diff --git a/drivers/firmware/efi/arm-runtime.c b/drivers/firmware/efi/arm-runtime.c
index 922cfb8..a00934d 100644
--- a/drivers/firmware/efi/arm-runtime.c
+++ b/drivers/firmware/efi/arm-runtime.c
@@ -110,7 +110,7 @@ static int __init arm_enable_runtime_services(void)
{
u64 mapsize;
- if (!efi_enabled(EFI_BOOT) || !efi_enabled(EFI_MEMMAP)) {
+ if (!efi_enabled(EFI_BOOT)) {
pr_info("EFI services will not be available.\n");
return 0;
}
diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c
index 249eb70..fad7c62 100644
--- a/drivers/firmware/efi/efi.c
+++ b/drivers/firmware/efi/efi.c
@@ -592,7 +592,11 @@ int __init efi_config_parse_tables(void *config_tables, int count, int sz,
early_memunmap(tbl, sizeof(*tbl));
}
+ return 0;
+}
+int __init efi_apply_persistent_mem_reservations(void)
+{
if (efi.mem_reserve != EFI_INVALID_TABLE_ADDR) {
unsigned long prsv = efi.mem_reserve;
@@ -963,37 +967,44 @@ bool efi_is_table_address(unsigned long phys_addr)
}
static DEFINE_SPINLOCK(efi_mem_reserve_persistent_lock);
+static struct linux_efi_memreserve *efi_memreserve_root __ro_after_init;
int efi_mem_reserve_persistent(phys_addr_t addr, u64 size)
{
- struct linux_efi_memreserve *rsv, *parent;
+ struct linux_efi_memreserve *rsv;
- if (efi.mem_reserve == EFI_INVALID_TABLE_ADDR)
+ if (!efi_memreserve_root)
return -ENODEV;
- rsv = kmalloc(sizeof(*rsv), GFP_KERNEL);
+ rsv = kmalloc(sizeof(*rsv), GFP_ATOMIC);
if (!rsv)
return -ENOMEM;
- parent = memremap(efi.mem_reserve, sizeof(*rsv), MEMREMAP_WB);
- if (!parent) {
- kfree(rsv);
- return -ENOMEM;
- }
-
rsv->base = addr;
rsv->size = size;
spin_lock(&efi_mem_reserve_persistent_lock);
- rsv->next = parent->next;
- parent->next = __pa(rsv);
+ rsv->next = efi_memreserve_root->next;
+ efi_memreserve_root->next = __pa(rsv);
spin_unlock(&efi_mem_reserve_persistent_lock);
- memunmap(parent);
-
return 0;
}
+static int __init efi_memreserve_root_init(void)
+{
+ if (efi.mem_reserve == EFI_INVALID_TABLE_ADDR)
+ return -ENODEV;
+
+ efi_memreserve_root = memremap(efi.mem_reserve,
+ sizeof(*efi_memreserve_root),
+ MEMREMAP_WB);
+ if (!efi_memreserve_root)
+ return -ENOMEM;
+ return 0;
+}
+early_initcall(efi_memreserve_root_init);
+
#ifdef CONFIG_KEXEC
static int update_efi_random_seed(struct notifier_block *nb,
unsigned long code, void *unused)
diff --git a/drivers/firmware/efi/libstub/arm-stub.c b/drivers/firmware/efi/libstub/arm-stub.c
index 30ac0c9..3d36142 100644
--- a/drivers/firmware/efi/libstub/arm-stub.c
+++ b/drivers/firmware/efi/libstub/arm-stub.c
@@ -75,6 +75,9 @@ void install_memreserve_table(efi_system_table_t *sys_table_arg)
efi_guid_t memreserve_table_guid = LINUX_EFI_MEMRESERVE_TABLE_GUID;
efi_status_t status;
+ if (IS_ENABLED(CONFIG_ARM))
+ return;
+
status = efi_call_early(allocate_pool, EFI_LOADER_DATA, sizeof(*rsv),
(void **)&rsv);
if (status != EFI_SUCCESS) {
diff --git a/drivers/firmware/efi/libstub/fdt.c b/drivers/firmware/efi/libstub/fdt.c
index 8830fa6..0c0d231 100644
--- a/drivers/firmware/efi/libstub/fdt.c
+++ b/drivers/firmware/efi/libstub/fdt.c
@@ -158,6 +158,10 @@ static efi_status_t update_fdt(efi_system_table_t *sys_table, void *orig_fdt,
return efi_status;
}
}
+
+ /* shrink the FDT back to its minimum size */
+ fdt_pack(fdt);
+
return EFI_SUCCESS;
fdt_set_fail:
diff --git a/drivers/firmware/efi/memmap.c b/drivers/firmware/efi/memmap.c
index fa2904f..38b686c 100644
--- a/drivers/firmware/efi/memmap.c
+++ b/drivers/firmware/efi/memmap.c
@@ -118,6 +118,9 @@ int __init efi_memmap_init_early(struct efi_memory_map_data *data)
void __init efi_memmap_unmap(void)
{
+ if (!efi_enabled(EFI_MEMMAP))
+ return;
+
if (!efi.memmap.late) {
unsigned long size;
diff --git a/drivers/firmware/efi/runtime-wrappers.c b/drivers/firmware/efi/runtime-wrappers.c
index a19d845b..8903b9c 100644
--- a/drivers/firmware/efi/runtime-wrappers.c
+++ b/drivers/firmware/efi/runtime-wrappers.c
@@ -67,7 +67,7 @@ struct efi_runtime_work efi_rts_work;
} \
\
init_completion(&efi_rts_work.efi_rts_comp); \
- INIT_WORK_ONSTACK(&efi_rts_work.work, efi_call_rts); \
+ INIT_WORK(&efi_rts_work.work, efi_call_rts); \
efi_rts_work.arg1 = _arg1; \
efi_rts_work.arg2 = _arg2; \
efi_rts_work.arg3 = _arg3; \
diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile
index 576ba98..7f3be35 100644
--- a/drivers/gpu/drm/Makefile
+++ b/drivers/gpu/drm/Makefile
@@ -11,7 +11,7 @@
drm_sysfs.o drm_hashtab.o drm_mm.o \
drm_crtc.o drm_fourcc.o drm_modes.o drm_edid.o \
drm_info.o drm_encoder_slave.o \
- drm_trace_points.o drm_global.o drm_prime.o \
+ drm_trace_points.o drm_prime.o \
drm_rect.o drm_vma_manager.o drm_flip_work.o \
drm_modeset_lock.o drm_atomic.o drm_bridge.o \
drm_framebuffer.o drm_connector.o drm_blend.o \
diff --git a/drivers/gpu/drm/amd/amdgpu/Makefile b/drivers/gpu/drm/amd/amdgpu/Makefile
index 138cb78..f76bcb9 100644
--- a/drivers/gpu/drm/amd/amdgpu/Makefile
+++ b/drivers/gpu/drm/amd/amdgpu/Makefile
@@ -53,7 +53,7 @@
amdgpu_ucode.o amdgpu_bo_list.o amdgpu_ctx.o amdgpu_sync.o \
amdgpu_gtt_mgr.o amdgpu_vram_mgr.o amdgpu_virt.o amdgpu_atomfirmware.o \
amdgpu_vf_error.o amdgpu_sched.o amdgpu_debugfs.o amdgpu_ids.o \
- amdgpu_gmc.o amdgpu_xgmi.o
+ amdgpu_gmc.o amdgpu_xgmi.o amdgpu_csa.o
# add asic specific block
amdgpu-$(CONFIG_DRM_AMDGPU_CIK)+= cik.o cik_ih.o kv_smc.o kv_dpm.o \
@@ -105,6 +105,7 @@
# add GFX block
amdgpu-y += \
amdgpu_gfx.o \
+ amdgpu_rlc.o \
gfx_v8_0.o \
gfx_v9_0.o
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index d0102cf..42f882c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -75,6 +75,7 @@
#include "amdgpu_sdma.h"
#include "amdgpu_dm.h"
#include "amdgpu_virt.h"
+#include "amdgpu_csa.h"
#include "amdgpu_gart.h"
#include "amdgpu_debugfs.h"
#include "amdgpu_job.h"
@@ -151,6 +152,7 @@ extern int amdgpu_compute_multipipe;
extern int amdgpu_gpu_recovery;
extern int amdgpu_emu_mode;
extern uint amdgpu_smu_memory_pool_size;
+extern uint amdgpu_dc_feature_mask;
extern struct amdgpu_mgpu_info mgpu_info;
#ifdef CONFIG_DRM_AMDGPU_SI
@@ -432,7 +434,7 @@ typedef enum _AMDGPU_DOORBELL64_ASSIGNMENT
* default non-graphics QWORD index is 0xe0 - 0xFF inclusive
*/
- /* sDMA engines reserved from 0xe0 -oxef */
+ /* sDMA engines reserved from 0xe0 -0xef */
AMDGPU_DOORBELL64_sDMA_ENGINE0 = 0xE0,
AMDGPU_DOORBELL64_sDMA_HI_PRI_ENGINE0 = 0xE1,
AMDGPU_DOORBELL64_sDMA_ENGINE1 = 0xE8,
@@ -830,7 +832,6 @@ struct amdgpu_device {
bool need_dma32;
bool need_swiotlb;
bool accel_working;
- struct work_struct reset_work;
struct notifier_block acpi_nb;
struct amdgpu_i2c_chan *i2c_bus[AMDGPU_MAX_I2C_BUS];
struct amdgpu_debugfs debugfs[AMDGPU_DEBUGFS_MAX_COMPONENTS];
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index c31a884..60f9a87 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -144,7 +144,7 @@ void amdgpu_amdkfd_device_init(struct amdgpu_device *adev)
KGD_MAX_QUEUES);
/* remove the KIQ bit as well */
- if (adev->gfx.kiq.ring.ready)
+ if (adev->gfx.kiq.ring.sched.ready)
clear_bit(amdgpu_gfx_queue_to_bit(adev,
adev->gfx.kiq.ring.me - 1,
adev->gfx.kiq.ring.pipe,
@@ -268,9 +268,9 @@ void amdgpu_amdkfd_gpu_reset(struct kgd_dev *kgd)
amdgpu_device_gpu_recover(adev, NULL);
}
-int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
- void **mem_obj, uint64_t *gpu_addr,
- void **cpu_ptr, bool mqd_gfx9)
+int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
+ void **mem_obj, uint64_t *gpu_addr,
+ void **cpu_ptr, bool mqd_gfx9)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
struct amdgpu_bo *bo = NULL;
@@ -340,7 +340,7 @@ int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
return r;
}
-void free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
+void amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
{
struct amdgpu_bo *bo = (struct amdgpu_bo *) mem_obj;
@@ -351,8 +351,8 @@ void free_gtt_mem(struct kgd_dev *kgd, void *mem_obj)
amdgpu_bo_unref(&(bo));
}
-void get_local_mem_info(struct kgd_dev *kgd,
- struct kfd_local_mem_info *mem_info)
+void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd,
+ struct kfd_local_mem_info *mem_info)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
uint64_t address_mask = adev->dev->dma_mask ? ~*adev->dev->dma_mask :
@@ -383,7 +383,7 @@ void get_local_mem_info(struct kgd_dev *kgd,
mem_info->mem_clk_max = 100;
}
-uint64_t get_gpu_clock_counter(struct kgd_dev *kgd)
+uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct kgd_dev *kgd)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
@@ -392,7 +392,7 @@ uint64_t get_gpu_clock_counter(struct kgd_dev *kgd)
return 0;
}
-uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
+uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
@@ -405,7 +405,7 @@ uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd)
return 100;
}
-void get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info)
+void amdgpu_amdkfd_get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info)
{
struct amdgpu_device *adev = (struct amdgpu_device *)kgd;
struct amdgpu_cu_info acu_info = adev->gfx.cu_info;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index 8e0d4f7..bcf587b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -134,16 +134,16 @@ int amdgpu_amdkfd_post_reset(struct amdgpu_device *adev);
void amdgpu_amdkfd_gpu_reset(struct kgd_dev *kgd);
/* Shared API */
-int alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
- void **mem_obj, uint64_t *gpu_addr,
- void **cpu_ptr, bool mqd_gfx9);
-void free_gtt_mem(struct kgd_dev *kgd, void *mem_obj);
-void get_local_mem_info(struct kgd_dev *kgd,
- struct kfd_local_mem_info *mem_info);
-uint64_t get_gpu_clock_counter(struct kgd_dev *kgd);
+int amdgpu_amdkfd_alloc_gtt_mem(struct kgd_dev *kgd, size_t size,
+ void **mem_obj, uint64_t *gpu_addr,
+ void **cpu_ptr, bool mqd_gfx9);
+void amdgpu_amdkfd_free_gtt_mem(struct kgd_dev *kgd, void *mem_obj);
+void amdgpu_amdkfd_get_local_mem_info(struct kgd_dev *kgd,
+ struct kfd_local_mem_info *mem_info);
+uint64_t amdgpu_amdkfd_get_gpu_clock_counter(struct kgd_dev *kgd);
-uint32_t get_max_engine_clock_in_mhz(struct kgd_dev *kgd);
-void get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info);
+uint32_t amdgpu_amdkfd_get_max_engine_clock_in_mhz(struct kgd_dev *kgd);
+void amdgpu_amdkfd_get_cu_info(struct kgd_dev *kgd, struct kfd_cu_info *cu_info);
uint64_t amdgpu_amdkfd_get_vram_usage(struct kgd_dev *kgd);
uint64_t amdgpu_amdkfd_get_hive_id(struct kgd_dev *kgd);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
index 244d983..72a357d 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v7.c
@@ -173,13 +173,6 @@ static int get_tile_config(struct kgd_dev *kgd,
}
static const struct kfd2kgd_calls kfd2kgd = {
- .init_gtt_mem_allocation = alloc_gtt_mem,
- .free_gtt_mem = free_gtt_mem,
- .get_local_mem_info = get_local_mem_info,
- .get_gpu_clock_counter = get_gpu_clock_counter,
- .get_max_engine_clock_in_mhz = get_max_engine_clock_in_mhz,
- .alloc_pasid = amdgpu_pasid_alloc,
- .free_pasid = amdgpu_pasid_free,
.program_sh_mem_settings = kgd_program_sh_mem_settings,
.set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
.init_interrupts = kgd_init_interrupts,
@@ -200,28 +193,10 @@ static const struct kfd2kgd_calls kfd2kgd = {
.get_fw_version = get_fw_version,
.set_scratch_backing_va = set_scratch_backing_va,
.get_tile_config = get_tile_config,
- .get_cu_info = get_cu_info,
- .get_vram_usage = amdgpu_amdkfd_get_vram_usage,
- .create_process_vm = amdgpu_amdkfd_gpuvm_create_process_vm,
- .acquire_process_vm = amdgpu_amdkfd_gpuvm_acquire_process_vm,
- .destroy_process_vm = amdgpu_amdkfd_gpuvm_destroy_process_vm,
- .release_process_vm = amdgpu_amdkfd_gpuvm_release_process_vm,
- .get_process_page_dir = amdgpu_amdkfd_gpuvm_get_process_page_dir,
.set_vm_context_page_table_base = set_vm_context_page_table_base,
- .alloc_memory_of_gpu = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu,
- .free_memory_of_gpu = amdgpu_amdkfd_gpuvm_free_memory_of_gpu,
- .map_memory_to_gpu = amdgpu_amdkfd_gpuvm_map_memory_to_gpu,
- .unmap_memory_to_gpu = amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu,
- .sync_memory = amdgpu_amdkfd_gpuvm_sync_memory,
- .map_gtt_bo_to_kernel = amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel,
- .restore_process_bos = amdgpu_amdkfd_gpuvm_restore_process_bos,
.invalidate_tlbs = invalidate_tlbs,
.invalidate_tlbs_vmid = invalidate_tlbs_vmid,
- .submit_ib = amdgpu_amdkfd_submit_ib,
- .get_vm_fault_info = amdgpu_amdkfd_gpuvm_get_vm_fault_info,
.read_vmid_from_vmfault_reg = read_vmid_from_vmfault_reg,
- .gpu_recover = amdgpu_amdkfd_gpu_reset,
- .set_compute_idle = amdgpu_amdkfd_set_compute_idle
};
struct kfd2kgd_calls *amdgpu_amdkfd_gfx_7_get_functions(void)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
index 9f14991..0e2a56b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v8.c
@@ -128,13 +128,6 @@ static int get_tile_config(struct kgd_dev *kgd,
}
static const struct kfd2kgd_calls kfd2kgd = {
- .init_gtt_mem_allocation = alloc_gtt_mem,
- .free_gtt_mem = free_gtt_mem,
- .get_local_mem_info = get_local_mem_info,
- .get_gpu_clock_counter = get_gpu_clock_counter,
- .get_max_engine_clock_in_mhz = get_max_engine_clock_in_mhz,
- .alloc_pasid = amdgpu_pasid_alloc,
- .free_pasid = amdgpu_pasid_free,
.program_sh_mem_settings = kgd_program_sh_mem_settings,
.set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
.init_interrupts = kgd_init_interrupts,
@@ -157,27 +150,9 @@ static const struct kfd2kgd_calls kfd2kgd = {
.get_fw_version = get_fw_version,
.set_scratch_backing_va = set_scratch_backing_va,
.get_tile_config = get_tile_config,
- .get_cu_info = get_cu_info,
- .get_vram_usage = amdgpu_amdkfd_get_vram_usage,
- .create_process_vm = amdgpu_amdkfd_gpuvm_create_process_vm,
- .acquire_process_vm = amdgpu_amdkfd_gpuvm_acquire_process_vm,
- .destroy_process_vm = amdgpu_amdkfd_gpuvm_destroy_process_vm,
- .release_process_vm = amdgpu_amdkfd_gpuvm_release_process_vm,
- .get_process_page_dir = amdgpu_amdkfd_gpuvm_get_process_page_dir,
.set_vm_context_page_table_base = set_vm_context_page_table_base,
- .alloc_memory_of_gpu = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu,
- .free_memory_of_gpu = amdgpu_amdkfd_gpuvm_free_memory_of_gpu,
- .map_memory_to_gpu = amdgpu_amdkfd_gpuvm_map_memory_to_gpu,
- .unmap_memory_to_gpu = amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu,
- .sync_memory = amdgpu_amdkfd_gpuvm_sync_memory,
- .map_gtt_bo_to_kernel = amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel,
- .restore_process_bos = amdgpu_amdkfd_gpuvm_restore_process_bos,
.invalidate_tlbs = invalidate_tlbs,
.invalidate_tlbs_vmid = invalidate_tlbs_vmid,
- .submit_ib = amdgpu_amdkfd_submit_ib,
- .get_vm_fault_info = amdgpu_amdkfd_gpuvm_get_vm_fault_info,
- .gpu_recover = amdgpu_amdkfd_gpu_reset,
- .set_compute_idle = amdgpu_amdkfd_set_compute_idle
};
struct kfd2kgd_calls *amdgpu_amdkfd_gfx_8_0_get_functions(void)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
index 42cb4c4..03b604c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c
@@ -46,38 +46,9 @@
#include "v9_structs.h"
#include "soc15.h"
#include "soc15d.h"
+#include "mmhub_v1_0.h"
+#include "gfxhub_v1_0.h"
-/* HACK: MMHUB and GC both have VM-related register with the same
- * names but different offsets. Define the MMHUB register we need here
- * with a prefix. A proper solution would be to move the functions
- * programming these registers into gfx_v9_0.c and mmhub_v1_0.c
- * respectively.
- */
-#define mmMMHUB_VM_INVALIDATE_ENG16_REQ 0x06f3
-#define mmMMHUB_VM_INVALIDATE_ENG16_REQ_BASE_IDX 0
-
-#define mmMMHUB_VM_INVALIDATE_ENG16_ACK 0x0705
-#define mmMMHUB_VM_INVALIDATE_ENG16_ACK_BASE_IDX 0
-
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32 0x072b
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32_BASE_IDX 0
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32 0x072c
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32_BASE_IDX 0
-
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32 0x074b
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32_BASE_IDX 0
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32 0x074c
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32_BASE_IDX 0
-
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32 0x076b
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32_BASE_IDX 0
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32 0x076c
-#define mmMMHUB_VM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32_BASE_IDX 0
-
-#define mmMMHUB_VM_INVALIDATE_ENG16_ADDR_RANGE_LO32 0x0727
-#define mmMMHUB_VM_INVALIDATE_ENG16_ADDR_RANGE_LO32_BASE_IDX 0
-#define mmMMHUB_VM_INVALIDATE_ENG16_ADDR_RANGE_HI32 0x0728
-#define mmMMHUB_VM_INVALIDATE_ENG16_ADDR_RANGE_HI32_BASE_IDX 0
#define V9_PIPE_PER_MEC (4)
#define V9_QUEUES_PER_PIPE_MEC (8)
@@ -167,13 +138,6 @@ static int amdgpu_amdkfd_get_tile_config(struct kgd_dev *kgd,
}
static const struct kfd2kgd_calls kfd2kgd = {
- .init_gtt_mem_allocation = alloc_gtt_mem,
- .free_gtt_mem = free_gtt_mem,
- .get_local_mem_info = get_local_mem_info,
- .get_gpu_clock_counter = get_gpu_clock_counter,
- .get_max_engine_clock_in_mhz = get_max_engine_clock_in_mhz,
- .alloc_pasid = amdgpu_pasid_alloc,
- .free_pasid = amdgpu_pasid_free,
.program_sh_mem_settings = kgd_program_sh_mem_settings,
.set_pasid_vmid_mapping = kgd_set_pasid_vmid_mapping,
.init_interrupts = kgd_init_interrupts,
@@ -196,26 +160,9 @@ static const struct kfd2kgd_calls kfd2kgd = {
.get_fw_version = get_fw_version,
.set_scratch_backing_va = set_scratch_backing_va,
.get_tile_config = amdgpu_amdkfd_get_tile_config,
- .get_cu_info = get_cu_info,
- .get_vram_usage = amdgpu_amdkfd_get_vram_usage,
- .create_process_vm = amdgpu_amdkfd_gpuvm_create_process_vm,
- .acquire_process_vm = amdgpu_amdkfd_gpuvm_acquire_process_vm,
- .destroy_process_vm = amdgpu_amdkfd_gpuvm_destroy_process_vm,
- .release_process_vm = amdgpu_amdkfd_gpuvm_release_process_vm,
- .get_process_page_dir = amdgpu_amdkfd_gpuvm_get_process_page_dir,
.set_vm_context_page_table_base = set_vm_context_page_table_base,
- .alloc_memory_of_gpu = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu,
- .free_memory_of_gpu = amdgpu_amdkfd_gpuvm_free_memory_of_gpu,
- .map_memory_to_gpu = amdgpu_amdkfd_gpuvm_map_memory_to_gpu,
- .unmap_memory_to_gpu = amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu,
- .sync_memory = amdgpu_amdkfd_gpuvm_sync_memory,
- .map_gtt_bo_to_kernel = amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel,
- .restore_process_bos = amdgpu_amdkfd_gpuvm_restore_process_bos,
.invalidate_tlbs = invalidate_tlbs,
.invalidate_tlbs_vmid = invalidate_tlbs_vmid,
- .submit_ib = amdgpu_amdkfd_submit_ib,
- .gpu_recover = amdgpu_amdkfd_gpu_reset,
- .set_compute_idle = amdgpu_amdkfd_set_compute_idle,
.get_hive_id = amdgpu_amdkfd_get_hive_id,
};
@@ -785,15 +732,6 @@ static uint16_t get_atc_vmid_pasid_mapping_pasid(struct kgd_dev *kgd,
static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid)
{
struct amdgpu_device *adev = (struct amdgpu_device *) kgd;
- uint32_t req = (1 << vmid) |
- (0 << VM_INVALIDATE_ENG16_REQ__FLUSH_TYPE__SHIFT) | /* legacy */
- VM_INVALIDATE_ENG16_REQ__INVALIDATE_L2_PTES_MASK |
- VM_INVALIDATE_ENG16_REQ__INVALIDATE_L2_PDE0_MASK |
- VM_INVALIDATE_ENG16_REQ__INVALIDATE_L2_PDE1_MASK |
- VM_INVALIDATE_ENG16_REQ__INVALIDATE_L2_PDE2_MASK |
- VM_INVALIDATE_ENG16_REQ__INVALIDATE_L1_PTES_MASK;
-
- mutex_lock(&adev->srbm_mutex);
/* Use legacy mode tlb invalidation.
*
@@ -810,34 +748,7 @@ static void write_vmid_invalidate_request(struct kgd_dev *kgd, uint8_t vmid)
* TODO 2: support range-based invalidation, requires kfg2kgd
* interface change
*/
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_INVALIDATE_ENG16_ADDR_RANGE_LO32),
- 0xffffffff);
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_INVALIDATE_ENG16_ADDR_RANGE_HI32),
- 0x0000001f);
-
- WREG32(SOC15_REG_OFFSET(MMHUB, 0,
- mmMMHUB_VM_INVALIDATE_ENG16_ADDR_RANGE_LO32),
- 0xffffffff);
- WREG32(SOC15_REG_OFFSET(MMHUB, 0,
- mmMMHUB_VM_INVALIDATE_ENG16_ADDR_RANGE_HI32),
- 0x0000001f);
-
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_INVALIDATE_ENG16_REQ), req);
-
- WREG32(SOC15_REG_OFFSET(MMHUB, 0, mmMMHUB_VM_INVALIDATE_ENG16_REQ),
- req);
-
- while (!(RREG32(SOC15_REG_OFFSET(GC, 0, mmVM_INVALIDATE_ENG16_ACK)) &
- (1 << vmid)))
- cpu_relax();
-
- while (!(RREG32(SOC15_REG_OFFSET(MMHUB, 0,
- mmMMHUB_VM_INVALIDATE_ENG16_ACK)) &
- (1 << vmid)))
- cpu_relax();
-
- mutex_unlock(&adev->srbm_mutex);
-
+ amdgpu_gmc_flush_gpu_tlb(adev, vmid, 0);
}
static int invalidate_tlbs_with_kiq(struct amdgpu_device *adev, uint16_t pasid)
@@ -876,7 +787,7 @@ static int invalidate_tlbs(struct kgd_dev *kgd, uint16_t pasid)
if (adev->in_gpu_reset)
return -EIO;
- if (ring->ready)
+ if (ring->sched.ready)
return invalidate_tlbs_with_kiq(adev, pasid);
for (vmid = 0; vmid < 16; vmid++) {
@@ -1016,7 +927,6 @@ static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
uint64_t page_table_base)
{
struct amdgpu_device *adev = get_amdgpu_device(kgd);
- uint64_t base = page_table_base | AMDGPU_PTE_VALID;
if (!amdgpu_amdkfd_is_kfd_vmid(adev, vmid)) {
pr_err("trying to set page table base for wrong VMID %u\n",
@@ -1028,25 +938,7 @@ static void set_vm_context_page_table_base(struct kgd_dev *kgd, uint32_t vmid,
* now, all processes share the same address space size, like
* on GFX8 and older.
*/
- WREG32(SOC15_REG_OFFSET(MMHUB, 0, mmMMHUB_VM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32) + (vmid*2), 0);
- WREG32(SOC15_REG_OFFSET(MMHUB, 0, mmMMHUB_VM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32) + (vmid*2), 0);
+ mmhub_v1_0_setup_vm_pt_regs(adev, vmid, page_table_base);
- WREG32(SOC15_REG_OFFSET(MMHUB, 0, mmMMHUB_VM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32) + (vmid*2),
- lower_32_bits(adev->vm_manager.max_pfn - 1));
- WREG32(SOC15_REG_OFFSET(MMHUB, 0, mmMMHUB_VM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32) + (vmid*2),
- upper_32_bits(adev->vm_manager.max_pfn - 1));
-
- WREG32(SOC15_REG_OFFSET(MMHUB, 0, mmMMHUB_VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32) + (vmid*2), lower_32_bits(base));
- WREG32(SOC15_REG_OFFSET(MMHUB, 0, mmMMHUB_VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32) + (vmid*2), upper_32_bits(base));
-
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32) + (vmid*2), 0);
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_START_ADDR_HI32) + (vmid*2), 0);
-
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_END_ADDR_LO32) + (vmid*2),
- lower_32_bits(adev->vm_manager.max_pfn - 1));
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_END_ADDR_HI32) + (vmid*2),
- upper_32_bits(adev->vm_manager.max_pfn - 1));
-
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32) + (vmid*2), lower_32_bits(base));
- WREG32(SOC15_REG_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32) + (vmid*2), upper_32_bits(base));
+ gfxhub_v1_0_setup_vm_pt_regs(adev, vmid, page_table_base);
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
index 8816c69..ceadeea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
@@ -330,7 +330,9 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
case CHIP_TOPAZ:
if (((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0x81)) ||
((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0x83)) ||
- ((adev->pdev->device == 0x6907) && (adev->pdev->revision == 0x87))) {
+ ((adev->pdev->device == 0x6907) && (adev->pdev->revision == 0x87)) ||
+ ((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0xD1)) ||
+ ((adev->pdev->device == 0x6900) && (adev->pdev->revision == 0xD3))) {
info->is_kicker = true;
strcpy(fw_name, "amdgpu/topaz_k_smc.bin");
} else
@@ -351,7 +353,6 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
if (type == CGS_UCODE_ID_SMU) {
if (((adev->pdev->device == 0x67ef) &&
((adev->pdev->revision == 0xe0) ||
- (adev->pdev->revision == 0xe2) ||
(adev->pdev->revision == 0xe5))) ||
((adev->pdev->device == 0x67ff) &&
((adev->pdev->revision == 0xcf) ||
@@ -359,8 +360,13 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
(adev->pdev->revision == 0xff)))) {
info->is_kicker = true;
strcpy(fw_name, "amdgpu/polaris11_k_smc.bin");
- } else
+ } else if ((adev->pdev->device == 0x67ef) &&
+ (adev->pdev->revision == 0xe2)) {
+ info->is_kicker = true;
+ strcpy(fw_name, "amdgpu/polaris11_k2_smc.bin");
+ } else {
strcpy(fw_name, "amdgpu/polaris11_smc.bin");
+ }
} else if (type == CGS_UCODE_ID_SMU_SK) {
strcpy(fw_name, "amdgpu/polaris11_smc_sk.bin");
}
@@ -378,14 +384,31 @@ static int amdgpu_cgs_get_firmware_info(struct cgs_device *cgs_device,
(adev->pdev->revision == 0xef))) {
info->is_kicker = true;
strcpy(fw_name, "amdgpu/polaris10_k_smc.bin");
- } else
+ } else if ((adev->pdev->device == 0x67df) &&
+ ((adev->pdev->revision == 0xe1) ||
+ (adev->pdev->revision == 0xf7))) {
+ info->is_kicker = true;
+ strcpy(fw_name, "amdgpu/polaris10_k2_smc.bin");
+ } else {
strcpy(fw_name, "amdgpu/polaris10_smc.bin");
+ }
} else if (type == CGS_UCODE_ID_SMU_SK) {
strcpy(fw_name, "amdgpu/polaris10_smc_sk.bin");
}
break;
case CHIP_POLARIS12:
- strcpy(fw_name, "amdgpu/polaris12_smc.bin");
+ if (((adev->pdev->device == 0x6987) &&
+ ((adev->pdev->revision == 0xc0) ||
+ (adev->pdev->revision == 0xc3))) ||
+ ((adev->pdev->device == 0x6981) &&
+ ((adev->pdev->revision == 0x00) ||
+ (adev->pdev->revision == 0x01) ||
+ (adev->pdev->revision == 0x10)))) {
+ info->is_kicker = true;
+ strcpy(fw_name, "amdgpu/polaris12_k_smc.bin");
+ } else {
+ strcpy(fw_name, "amdgpu/polaris12_smc.bin");
+ }
break;
case CHIP_VEGAM:
strcpy(fw_name, "amdgpu/vegam_smc.bin");
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 35bc8fc3b..024dfbd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -1260,8 +1260,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
return 0;
error_abort:
- dma_fence_put(&job->base.s_fence->finished);
- job->base.s_fence = NULL;
+ drm_sched_job_cleanup(&job->base);
amdgpu_mn_unlock(p->mn);
error_unlock:
@@ -1285,7 +1284,7 @@ int amdgpu_cs_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
r = amdgpu_cs_parser_init(&parser, data);
if (r) {
- DRM_ERROR("Failed to initialize parser !\n");
+ DRM_ERROR("Failed to initialize parser %d!\n", r);
goto out;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
new file mode 100644
index 0000000..0c590dd
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.c
@@ -0,0 +1,117 @@
+/*
+ * Copyright 2016 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+
+ * * Author: Monk.liu@amd.com
+ */
+
+#include "amdgpu.h"
+
+uint64_t amdgpu_csa_vaddr(struct amdgpu_device *adev)
+{
+ uint64_t addr = adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT;
+
+ addr -= AMDGPU_VA_RESERVED_SIZE;
+ addr = amdgpu_gmc_sign_extend(addr);
+
+ return addr;
+}
+
+int amdgpu_allocate_static_csa(struct amdgpu_device *adev, struct amdgpu_bo **bo,
+ u32 domain, uint32_t size)
+{
+ int r;
+ void *ptr;
+
+ r = amdgpu_bo_create_kernel(adev, size, PAGE_SIZE,
+ domain, bo,
+ NULL, &ptr);
+ if (!bo)
+ return -ENOMEM;
+
+ memset(ptr, 0, size);
+ return 0;
+}
+
+void amdgpu_free_static_csa(struct amdgpu_bo **bo)
+{
+ amdgpu_bo_free_kernel(bo, NULL, NULL);
+}
+
+/*
+ * amdgpu_map_static_csa should be called during amdgpu_vm_init
+ * it maps virtual address amdgpu_csa_vaddr() to this VM, and each command
+ * submission of GFX should use this virtual address within META_DATA init
+ * package to support SRIOV gfx preemption.
+ */
+int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+ struct amdgpu_bo *bo, struct amdgpu_bo_va **bo_va,
+ uint64_t csa_addr, uint32_t size)
+{
+ struct ww_acquire_ctx ticket;
+ struct list_head list;
+ struct amdgpu_bo_list_entry pd;
+ struct ttm_validate_buffer csa_tv;
+ int r;
+
+ INIT_LIST_HEAD(&list);
+ INIT_LIST_HEAD(&csa_tv.head);
+ csa_tv.bo = &bo->tbo;
+ csa_tv.shared = true;
+
+ list_add(&csa_tv.head, &list);
+ amdgpu_vm_get_pd_bo(vm, &list, &pd);
+
+ r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL);
+ if (r) {
+ DRM_ERROR("failed to reserve CSA,PD BOs: err=%d\n", r);
+ return r;
+ }
+
+ *bo_va = amdgpu_vm_bo_add(adev, vm, bo);
+ if (!*bo_va) {
+ ttm_eu_backoff_reservation(&ticket, &list);
+ DRM_ERROR("failed to create bo_va for static CSA\n");
+ return -ENOMEM;
+ }
+
+ r = amdgpu_vm_alloc_pts(adev, (*bo_va)->base.vm, csa_addr,
+ size);
+ if (r) {
+ DRM_ERROR("failed to allocate pts for static CSA, err=%d\n", r);
+ amdgpu_vm_bo_rmv(adev, *bo_va);
+ ttm_eu_backoff_reservation(&ticket, &list);
+ return r;
+ }
+
+ r = amdgpu_vm_bo_map(adev, *bo_va, csa_addr, 0, size,
+ AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
+ AMDGPU_PTE_EXECUTABLE);
+
+ if (r) {
+ DRM_ERROR("failed to do bo_map on static CSA, err=%d\n", r);
+ amdgpu_vm_bo_rmv(adev, *bo_va);
+ ttm_eu_backoff_reservation(&ticket, &list);
+ return r;
+ }
+
+ ttm_eu_backoff_reservation(&ticket, &list);
+ return 0;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h
new file mode 100644
index 0000000..524b443
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_csa.h
@@ -0,0 +1,39 @@
+/*
+ * Copyright 2016 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Author: Monk.liu@amd.com
+ */
+
+#ifndef AMDGPU_CSA_MANAGER_H
+#define AMDGPU_CSA_MANAGER_H
+
+#define AMDGPU_CSA_SIZE (128 * 1024)
+
+uint32_t amdgpu_get_total_csa_size(struct amdgpu_device *adev);
+uint64_t amdgpu_csa_vaddr(struct amdgpu_device *adev);
+int amdgpu_allocate_static_csa(struct amdgpu_device *adev, struct amdgpu_bo **bo,
+ u32 domain, uint32_t size);
+int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
+ struct amdgpu_bo *bo, struct amdgpu_bo_va **bo_va,
+ uint64_t csa_addr, uint32_t size);
+void amdgpu_free_static_csa(struct amdgpu_bo **bo);
+
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index 30bc345..590588a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1656,7 +1656,9 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
/* right after GMC hw init, we create CSA */
if (amdgpu_sriov_vf(adev)) {
- r = amdgpu_allocate_static_csa(adev);
+ r = amdgpu_allocate_static_csa(adev, &adev->virt.csa_obj,
+ AMDGPU_GEM_DOMAIN_VRAM,
+ AMDGPU_CSA_SIZE);
if (r) {
DRM_ERROR("allocate CSA failed %d\n", r);
return r;
@@ -1681,7 +1683,8 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
if (r)
return r;
- amdgpu_xgmi_add_device(adev);
+ if (adev->gmc.xgmi.num_physical_nodes > 1)
+ amdgpu_xgmi_add_device(adev);
amdgpu_amdkfd_device_init(adev);
if (amdgpu_sriov_vf(adev))
@@ -1890,7 +1893,7 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev)
if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_GMC) {
amdgpu_ucode_free_bo(adev);
- amdgpu_free_static_csa(adev);
+ amdgpu_free_static_csa(&adev->virt.csa_obj);
amdgpu_device_wb_fini(adev);
amdgpu_device_vram_scratch_fini(adev);
}
@@ -3295,13 +3298,35 @@ bool amdgpu_device_should_recover_gpu(struct amdgpu_device *adev)
return false;
}
- if (amdgpu_gpu_recovery == 0 || (amdgpu_gpu_recovery == -1 &&
- !amdgpu_sriov_vf(adev))) {
- DRM_INFO("GPU recovery disabled.\n");
- return false;
+ if (amdgpu_gpu_recovery == 0)
+ goto disabled;
+
+ if (amdgpu_sriov_vf(adev))
+ return true;
+
+ if (amdgpu_gpu_recovery == -1) {
+ switch (adev->asic_type) {
+ case CHIP_TOPAZ:
+ case CHIP_TONGA:
+ case CHIP_FIJI:
+ case CHIP_POLARIS10:
+ case CHIP_POLARIS11:
+ case CHIP_POLARIS12:
+ case CHIP_VEGAM:
+ case CHIP_VEGA20:
+ case CHIP_VEGA10:
+ case CHIP_VEGA12:
+ break;
+ default:
+ goto disabled;
+ }
}
return true;
+
+disabled:
+ DRM_INFO("GPU recovery disabled.\n");
+ return false;
}
/**
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index 943dbf3..8de55f7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -127,6 +127,9 @@ int amdgpu_compute_multipipe = -1;
int amdgpu_gpu_recovery = -1; /* auto */
int amdgpu_emu_mode = 0;
uint amdgpu_smu_memory_pool_size = 0;
+/* FBC (bit 0) disabled by default*/
+uint amdgpu_dc_feature_mask = 0;
+
struct amdgpu_mgpu_info mgpu_info = {
.mutex = __MUTEX_INITIALIZER(mgpu_info.mutex),
};
@@ -631,6 +634,14 @@ module_param(halt_if_hws_hang, int, 0644);
MODULE_PARM_DESC(halt_if_hws_hang, "Halt if HWS hang is detected (0 = off (default), 1 = on)");
#endif
+/**
+ * DOC: dcfeaturemask (uint)
+ * Override display features enabled. See enum DC_FEATURE_MASK in drivers/gpu/drm/amd/include/amd_shared.h.
+ * The default is the current set of stable display features.
+ */
+MODULE_PARM_DESC(dcfeaturemask, "all stable DC features enabled (default))");
+module_param_named(dcfeaturemask, amdgpu_dc_feature_mask, uint, 0444);
+
static const struct pci_device_id pciidlist[] = {
#ifdef CONFIG_DRM_AMDGPU_SI
{0x1002, 0x6780, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_TAHITI},
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 5448cf2..ee47c11 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -398,9 +398,9 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
ring->fence_drv.irq_type = irq_type;
ring->fence_drv.initialized = true;
- dev_dbg(adev->dev, "fence driver on ring %d use gpu addr 0x%016llx, "
- "cpu addr 0x%p\n", ring->idx,
- ring->fence_drv.gpu_addr, ring->fence_drv.cpu_addr);
+ DRM_DEV_DEBUG(adev->dev, "fence driver on ring %s use gpu addr "
+ "0x%016llx, cpu addr 0x%p\n", ring->name,
+ ring->fence_drv.gpu_addr, ring->fence_drv.cpu_addr);
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
index 11fea28..6d11e17 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c
@@ -248,7 +248,7 @@ int amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset,
}
mb();
amdgpu_asic_flush_hdp(adev, NULL);
- amdgpu_gmc_flush_gpu_tlb(adev, 0);
+ amdgpu_gmc_flush_gpu_tlb(adev, 0, 0);
return 0;
}
@@ -259,6 +259,8 @@ int amdgpu_gart_unbind(struct amdgpu_device *adev, uint64_t offset,
* @offset: offset into the GPU's gart aperture
* @pages: number of pages to bind
* @dma_addr: DMA addresses of pages
+ * @flags: page table entry flags
+ * @dst: CPU address of the gart table
*
* Map the dma_addresses into GART entries (all asics).
* Returns 0 for success, -EINVAL for failure.
@@ -331,7 +333,7 @@ int amdgpu_gart_bind(struct amdgpu_device *adev, uint64_t offset,
mb();
amdgpu_asic_flush_hdp(adev, NULL);
- amdgpu_gmc_flush_gpu_tlb(adev, 0);
+ amdgpu_gmc_flush_gpu_tlb(adev, 0, 0);
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h
index 9ff6288..afa2e28 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.h
@@ -41,6 +41,7 @@ struct amdgpu_bo;
struct amdgpu_gart {
struct amdgpu_bo *bo;
+ /* CPU kmapped address of gart table */
void *ptr;
unsigned num_gpu_pages;
unsigned num_cpu_pages;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
index 1a656b8..6a70c0b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.c
@@ -25,6 +25,7 @@
#include <drm/drmP.h>
#include "amdgpu.h"
#include "amdgpu_gfx.h"
+#include "amdgpu_rlc.h"
/* delay 0.1 second to enable gfx off feature */
#define GFX_OFF_DELAY_ENABLE msecs_to_jiffies(100)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
index b61b5c1..f790e15 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gfx.h
@@ -29,6 +29,7 @@
*/
#include "clearstate_defs.h"
#include "amdgpu_ring.h"
+#include "amdgpu_rlc.h"
/* GFX current status */
#define AMDGPU_GFX_NORMAL_MODE 0x00000000L
@@ -37,59 +38,6 @@
#define AMDGPU_GFX_CG_DISABLED_MODE 0x00000004L
#define AMDGPU_GFX_LBPW_DISABLED_MODE 0x00000008L
-
-struct amdgpu_rlc_funcs {
- void (*enter_safe_mode)(struct amdgpu_device *adev);
- void (*exit_safe_mode)(struct amdgpu_device *adev);
-};
-
-struct amdgpu_rlc {
- /* for power gating */
- struct amdgpu_bo *save_restore_obj;
- uint64_t save_restore_gpu_addr;
- volatile uint32_t *sr_ptr;
- const u32 *reg_list;
- u32 reg_list_size;
- /* for clear state */
- struct amdgpu_bo *clear_state_obj;
- uint64_t clear_state_gpu_addr;
- volatile uint32_t *cs_ptr;
- const struct cs_section_def *cs_data;
- u32 clear_state_size;
- /* for cp tables */
- struct amdgpu_bo *cp_table_obj;
- uint64_t cp_table_gpu_addr;
- volatile uint32_t *cp_table_ptr;
- u32 cp_table_size;
-
- /* safe mode for updating CG/PG state */
- bool in_safe_mode;
- const struct amdgpu_rlc_funcs *funcs;
-
- /* for firmware data */
- u32 save_and_restore_offset;
- u32 clear_state_descriptor_offset;
- u32 avail_scratch_ram_locations;
- u32 reg_restore_list_size;
- u32 reg_list_format_start;
- u32 reg_list_format_separate_start;
- u32 starting_offsets_start;
- u32 reg_list_format_size_bytes;
- u32 reg_list_size_bytes;
- u32 reg_list_format_direct_reg_list_length;
- u32 save_restore_list_cntl_size_bytes;
- u32 save_restore_list_gpm_size_bytes;
- u32 save_restore_list_srm_size_bytes;
-
- u32 *register_list_format;
- u32 *register_restore;
- u8 *save_restore_list_cntl;
- u8 *save_restore_list_gpm;
- u8 *save_restore_list_srm;
-
- bool is_rlc_v2_1;
-};
-
#define AMDGPU_MAX_COMPUTE_QUEUES KGD_MAX_QUEUES
struct amdgpu_mec {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
index 6fa7ef4..8c57924 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h
@@ -64,7 +64,7 @@ struct amdgpu_vmhub {
struct amdgpu_gmc_funcs {
/* flush the vm tlb via mmio */
void (*flush_gpu_tlb)(struct amdgpu_device *adev,
- uint32_t vmid);
+ uint32_t vmid, uint32_t flush_type);
/* flush the vm tlb via ring */
uint64_t (*emit_flush_gpu_tlb)(struct amdgpu_ring *ring, unsigned vmid,
uint64_t pd_addr);
@@ -89,7 +89,7 @@ struct amdgpu_gmc_funcs {
struct amdgpu_xgmi {
/* from psp */
- u64 device_id;
+ u64 node_id;
u64 hive_id;
/* fixed per family */
u64 node_segment_size;
@@ -151,7 +151,7 @@ struct amdgpu_gmc {
struct amdgpu_xgmi xgmi;
};
-#define amdgpu_gmc_flush_gpu_tlb(adev, vmid) (adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid))
+#define amdgpu_gmc_flush_gpu_tlb(adev, vmid, type) (adev)->gmc.gmc_funcs->flush_gpu_tlb((adev), (vmid), (type))
#define amdgpu_gmc_emit_flush_gpu_tlb(r, vmid, addr) (r)->adev->gmc.gmc_funcs->emit_flush_gpu_tlb((r), (vmid), (addr))
#define amdgpu_gmc_emit_pasid_mapping(r, vmid, pasid) (r)->adev->gmc.gmc_funcs->emit_pasid_mapping((r), (vmid), (pasid))
#define amdgpu_gmc_set_pte_pde(adev, pt, idx, addr, flags) (adev)->gmc.gmc_funcs->set_pte_pde((adev), (pt), (idx), (addr), (flags))
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index b8963b7..c48207b3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -146,7 +146,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
fence_ctx = 0;
}
- if (!ring->ready) {
+ if (!ring->sched.ready) {
dev_err(adev->dev, "couldn't schedule ib on ring <%s>\n", ring->name);
return -EINVAL;
}
@@ -221,8 +221,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
!amdgpu_sriov_vf(adev)) /* for SRIOV preemption, Preamble CE ib must be inserted anyway */
continue;
- amdgpu_ring_emit_ib(ring, ib, job ? job->vmid : 0,
- need_ctx_switch);
+ amdgpu_ring_emit_ib(ring, job, ib, need_ctx_switch);
need_ctx_switch = false;
}
@@ -347,19 +346,14 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev)
tmo_gfx = 8 * AMDGPU_IB_TEST_TIMEOUT;
}
- for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+ for (i = 0; i < adev->num_rings; ++i) {
struct amdgpu_ring *ring = adev->rings[i];
long tmo;
- if (!ring || !ring->ready)
- continue;
-
- /* skip IB tests for KIQ in general for the below reasons:
- * 1. We never submit IBs to the KIQ
- * 2. KIQ doesn't use the EOP interrupts,
- * we use some other CP interrupt.
+ /* KIQ rings don't have an IB test because we never submit IBs
+ * to them and they have no interrupt support.
*/
- if (ring->funcs->type == AMDGPU_RING_TYPE_KIQ)
+ if (!ring->sched.ready || !ring->funcs->test_ib)
continue;
/* MM engine need more time */
@@ -374,20 +368,23 @@ int amdgpu_ib_ring_tests(struct amdgpu_device *adev)
tmo = tmo_gfx;
r = amdgpu_ring_test_ib(ring, tmo);
- if (r) {
- ring->ready = false;
+ if (!r) {
+ DRM_DEV_DEBUG(adev->dev, "ib test on %s succeeded\n",
+ ring->name);
+ continue;
+ }
- if (ring == &adev->gfx.gfx_ring[0]) {
- /* oh, oh, that's really bad */
- DRM_ERROR("amdgpu: failed testing IB on GFX ring (%d).\n", r);
- adev->accel_working = false;
- return r;
+ ring->sched.ready = false;
+ DRM_DEV_ERROR(adev->dev, "IB test failed on %s (%d).\n",
+ ring->name, r);
- } else {
- /* still not good, but we can live with it */
- DRM_ERROR("amdgpu: failed testing IB on ring %d (%d).\n", i, r);
- ret = r;
- }
+ if (ring == &adev->gfx.gfx_ring[0]) {
+ /* oh, oh, that's really bad */
+ adev->accel_working = false;
+ return r;
+
+ } else {
+ ret = r;
}
}
return ret;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
index 52c17f6..6b6524f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c
@@ -94,23 +94,6 @@ static void amdgpu_hotplug_work_func(struct work_struct *work)
}
/**
- * amdgpu_irq_reset_work_func - execute GPU reset
- *
- * @work: work struct pointer
- *
- * Execute scheduled GPU reset (Cayman+).
- * This function is called when the IRQ handler thinks we need a GPU reset.
- */
-static void amdgpu_irq_reset_work_func(struct work_struct *work)
-{
- struct amdgpu_device *adev = container_of(work, struct amdgpu_device,
- reset_work);
-
- if (!amdgpu_sriov_vf(adev) && amdgpu_device_should_recover_gpu(adev))
- amdgpu_device_gpu_recover(adev, NULL);
-}
-
-/**
* amdgpu_irq_disable_all - disable *all* interrupts
*
* @adev: amdgpu device pointer
@@ -262,15 +245,12 @@ int amdgpu_irq_init(struct amdgpu_device *adev)
amdgpu_hotplug_work_func);
}
- INIT_WORK(&adev->reset_work, amdgpu_irq_reset_work_func);
-
adev->irq.installed = true;
r = drm_irq_install(adev->ddev, adev->ddev->pdev->irq);
if (r) {
adev->irq.installed = false;
if (!amdgpu_device_has_dc_support(adev))
flush_work(&adev->hotplug_work);
- cancel_work_sync(&adev->reset_work);
return r;
}
adev->ddev->max_vblank_count = 0x00ffffff;
@@ -299,7 +279,6 @@ void amdgpu_irq_fini(struct amdgpu_device *adev)
pci_disable_msi(adev->pdev);
if (!amdgpu_device_has_dc_support(adev))
flush_work(&adev->hotplug_work);
- cancel_work_sync(&adev->reset_work);
}
for (i = 0; i < AMDGPU_IRQ_CLIENTID_MAX; ++i) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index 755f733..e0af44f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -112,6 +112,8 @@ static void amdgpu_job_free_cb(struct drm_sched_job *s_job)
struct amdgpu_ring *ring = to_amdgpu_ring(s_job->sched);
struct amdgpu_job *job = to_amdgpu_job(s_job);
+ drm_sched_job_cleanup(s_job);
+
amdgpu_ring_priority_put(ring, s_job->s_priority);
dma_fence_put(job->fence);
amdgpu_sync_free(&job->sync);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
index 57cfe78..e1b46a6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.h
@@ -33,6 +33,8 @@
#define to_amdgpu_job(sched_job) \
container_of((sched_job), struct amdgpu_job, base)
+#define AMDGPU_JOB_GET_VMID(job) ((job) ? (job)->vmid : 0)
+
struct amdgpu_fence;
struct amdgpu_job {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 81732a8..9b3164c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -336,7 +336,7 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
case AMDGPU_HW_IP_GFX:
type = AMD_IP_BLOCK_TYPE_GFX;
for (i = 0; i < adev->gfx.num_gfx_rings; i++)
- if (adev->gfx.gfx_ring[i].ready)
+ if (adev->gfx.gfx_ring[i].sched.ready)
++num_rings;
ib_start_alignment = 32;
ib_size_alignment = 32;
@@ -344,7 +344,7 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
case AMDGPU_HW_IP_COMPUTE:
type = AMD_IP_BLOCK_TYPE_GFX;
for (i = 0; i < adev->gfx.num_compute_rings; i++)
- if (adev->gfx.compute_ring[i].ready)
+ if (adev->gfx.compute_ring[i].sched.ready)
++num_rings;
ib_start_alignment = 32;
ib_size_alignment = 32;
@@ -352,7 +352,7 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
case AMDGPU_HW_IP_DMA:
type = AMD_IP_BLOCK_TYPE_SDMA;
for (i = 0; i < adev->sdma.num_instances; i++)
- if (adev->sdma.instance[i].ring.ready)
+ if (adev->sdma.instance[i].ring.sched.ready)
++num_rings;
ib_start_alignment = 256;
ib_size_alignment = 4;
@@ -363,7 +363,7 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
if (adev->uvd.harvest_config & (1 << i))
continue;
- if (adev->uvd.inst[i].ring.ready)
+ if (adev->uvd.inst[i].ring.sched.ready)
++num_rings;
}
ib_start_alignment = 64;
@@ -372,7 +372,7 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
case AMDGPU_HW_IP_VCE:
type = AMD_IP_BLOCK_TYPE_VCE;
for (i = 0; i < adev->vce.num_rings; i++)
- if (adev->vce.ring[i].ready)
+ if (adev->vce.ring[i].sched.ready)
++num_rings;
ib_start_alignment = 4;
ib_size_alignment = 1;
@@ -384,7 +384,7 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
continue;
for (j = 0; j < adev->uvd.num_enc_rings; j++)
- if (adev->uvd.inst[i].ring_enc[j].ready)
+ if (adev->uvd.inst[i].ring_enc[j].sched.ready)
++num_rings;
}
ib_start_alignment = 64;
@@ -392,7 +392,7 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
break;
case AMDGPU_HW_IP_VCN_DEC:
type = AMD_IP_BLOCK_TYPE_VCN;
- if (adev->vcn.ring_dec.ready)
+ if (adev->vcn.ring_dec.sched.ready)
++num_rings;
ib_start_alignment = 16;
ib_size_alignment = 16;
@@ -400,14 +400,14 @@ static int amdgpu_hw_ip_info(struct amdgpu_device *adev,
case AMDGPU_HW_IP_VCN_ENC:
type = AMD_IP_BLOCK_TYPE_VCN;
for (i = 0; i < adev->vcn.num_enc_rings; i++)
- if (adev->vcn.ring_enc[i].ready)
+ if (adev->vcn.ring_enc[i].sched.ready)
++num_rings;
ib_start_alignment = 64;
ib_size_alignment = 1;
break;
case AMDGPU_HW_IP_VCN_JPEG:
type = AMD_IP_BLOCK_TYPE_VCN;
- if (adev->vcn.ring_jpeg.ready)
+ if (adev->vcn.ring_jpeg.sched.ready)
++num_rings;
ib_start_alignment = 16;
ib_size_alignment = 16;
@@ -978,7 +978,10 @@ int amdgpu_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv)
}
if (amdgpu_sriov_vf(adev)) {
- r = amdgpu_map_static_csa(adev, &fpriv->vm, &fpriv->csa_va);
+ uint64_t csa_addr = amdgpu_csa_vaddr(adev) & AMDGPU_GMC_HOLE_MASK;
+
+ r = amdgpu_map_static_csa(adev, &fpriv->vm, adev->virt.csa_obj,
+ &fpriv->csa_va, csa_addr, AMDGPU_CSA_SIZE);
if (r)
goto error_vm;
}
@@ -1048,8 +1051,8 @@ void amdgpu_driver_postclose_kms(struct drm_device *dev,
pasid = fpriv->vm.pasid;
pd = amdgpu_bo_ref(fpriv->vm.root.base.bo);
- amdgpu_vm_fini(adev, &fpriv->vm);
amdgpu_ctx_mgr_fini(&fpriv->ctx_mgr);
+ amdgpu_vm_fini(adev, &fpriv->vm);
if (pasid)
amdgpu_pasid_free_delayed(pd->tbo.resv, pasid);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
index b9e9e8b..11723d8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mode.h
@@ -57,7 +57,6 @@ struct amdgpu_hpd;
#define to_amdgpu_connector(x) container_of(x, struct amdgpu_connector, base)
#define to_amdgpu_encoder(x) container_of(x, struct amdgpu_encoder, base)
#define to_amdgpu_framebuffer(x) container_of(x, struct amdgpu_framebuffer, base)
-#define to_amdgpu_plane(x) container_of(x, struct amdgpu_plane, base)
#define to_dm_plane_state(x) container_of(x, struct dm_plane_state, base);
@@ -325,7 +324,7 @@ struct amdgpu_mode_info {
struct card_info *atom_card_info;
bool mode_config_initialized;
struct amdgpu_crtc *crtcs[AMDGPU_MAX_CRTCS];
- struct amdgpu_plane *planes[AMDGPU_MAX_PLANES];
+ struct drm_plane *planes[AMDGPU_MAX_PLANES];
struct amdgpu_afmt *afmt[AMDGPU_MAX_AFMT_BLOCKS];
/* DVI-I properties */
struct drm_property *coherent_mode_property;
@@ -434,11 +433,6 @@ struct amdgpu_crtc {
struct drm_pending_vblank_event *event;
};
-struct amdgpu_plane {
- struct drm_plane base;
- enum drm_plane_type plane_type;
-};
-
struct amdgpu_encoder_atom_dig {
bool linkb;
/* atom dig */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
index 59cc678..7235cd0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
@@ -2129,7 +2129,7 @@ void amdgpu_pm_compute_clocks(struct amdgpu_device *adev)
for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
struct amdgpu_ring *ring = adev->rings[i];
- if (ring && ring->ready)
+ if (ring && ring->sched.ready)
amdgpu_fence_wait_empty(ring);
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
index 25d2f3e..e05dc66 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c
@@ -90,6 +90,8 @@ static int psp_sw_fini(void *handle)
adev->psp.sos_fw = NULL;
release_firmware(adev->psp.asd_fw);
adev->psp.asd_fw = NULL;
+ release_firmware(adev->psp.ta_fw);
+ adev->psp.ta_fw = NULL;
return 0;
}
@@ -118,22 +120,26 @@ int psp_wait_for(struct psp_context *psp, uint32_t reg_index,
static int
psp_cmd_submit_buf(struct psp_context *psp,
struct amdgpu_firmware_info *ucode,
- struct psp_gfx_cmd_resp *cmd, uint64_t fence_mc_addr,
- int index)
+ struct psp_gfx_cmd_resp *cmd, uint64_t fence_mc_addr)
{
int ret;
+ int index;
memset(psp->cmd_buf_mem, 0, PSP_CMD_BUFFER_SIZE);
memcpy(psp->cmd_buf_mem, cmd, sizeof(struct psp_gfx_cmd_resp));
+ index = atomic_inc_return(&psp->fence_value);
ret = psp_cmd_submit(psp, ucode, psp->cmd_buf_mc_addr,
fence_mc_addr, index);
-
- while (*((unsigned int *)psp->fence_buf) != index) {
- msleep(1);
+ if (ret) {
+ atomic_dec(&psp->fence_value);
+ return ret;
}
+ while (*((unsigned int *)psp->fence_buf) != index)
+ msleep(1);
+
/* the status field must be 0 after FW is loaded */
if (ucode && psp->cmd_buf_mem->resp.status) {
DRM_ERROR("failed loading with status (%d) and ucode id (%d)\n",
@@ -191,7 +197,7 @@ static int psp_tmr_load(struct psp_context *psp)
PSP_TMR_SIZE, psp->tmr_mc_addr);
ret = psp_cmd_submit_buf(psp, NULL, cmd,
- psp->fence_buf_mc_addr, 1);
+ psp->fence_buf_mc_addr);
if (ret)
goto failed;
@@ -258,13 +264,194 @@ static int psp_asd_load(struct psp_context *psp)
psp->asd_ucode_size, PSP_ASD_SHARED_MEM_SIZE);
ret = psp_cmd_submit_buf(psp, NULL, cmd,
- psp->fence_buf_mc_addr, 2);
+ psp->fence_buf_mc_addr);
kfree(cmd);
return ret;
}
+static void psp_prep_xgmi_ta_load_cmd_buf(struct psp_gfx_cmd_resp *cmd,
+ uint64_t xgmi_ta_mc, uint64_t xgmi_mc_shared,
+ uint32_t xgmi_ta_size, uint32_t shared_size)
+{
+ cmd->cmd_id = GFX_CMD_ID_LOAD_TA;
+ cmd->cmd.cmd_load_ta.app_phy_addr_lo = lower_32_bits(xgmi_ta_mc);
+ cmd->cmd.cmd_load_ta.app_phy_addr_hi = upper_32_bits(xgmi_ta_mc);
+ cmd->cmd.cmd_load_ta.app_len = xgmi_ta_size;
+
+ cmd->cmd.cmd_load_ta.cmd_buf_phy_addr_lo = lower_32_bits(xgmi_mc_shared);
+ cmd->cmd.cmd_load_ta.cmd_buf_phy_addr_hi = upper_32_bits(xgmi_mc_shared);
+ cmd->cmd.cmd_load_ta.cmd_buf_len = shared_size;
+}
+
+static int psp_xgmi_init_shared_buf(struct psp_context *psp)
+{
+ int ret;
+
+ /*
+ * Allocate 16k memory aligned to 4k from Frame Buffer (local
+ * physical) for xgmi ta <-> Driver
+ */
+ ret = amdgpu_bo_create_kernel(psp->adev, PSP_XGMI_SHARED_MEM_SIZE,
+ PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+ &psp->xgmi_context.xgmi_shared_bo,
+ &psp->xgmi_context.xgmi_shared_mc_addr,
+ &psp->xgmi_context.xgmi_shared_buf);
+
+ return ret;
+}
+
+static int psp_xgmi_load(struct psp_context *psp)
+{
+ int ret;
+ struct psp_gfx_cmd_resp *cmd;
+
+ /*
+ * TODO: bypass the loading in sriov for now
+ */
+ if (amdgpu_sriov_vf(psp->adev))
+ return 0;
+
+ cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+ if (!cmd)
+ return -ENOMEM;
+
+ memset(psp->fw_pri_buf, 0, PSP_1_MEG);
+ memcpy(psp->fw_pri_buf, psp->ta_xgmi_start_addr, psp->ta_xgmi_ucode_size);
+
+ psp_prep_xgmi_ta_load_cmd_buf(cmd, psp->fw_pri_mc_addr,
+ psp->xgmi_context.xgmi_shared_mc_addr,
+ psp->ta_xgmi_ucode_size, PSP_XGMI_SHARED_MEM_SIZE);
+
+ ret = psp_cmd_submit_buf(psp, NULL, cmd,
+ psp->fence_buf_mc_addr);
+
+ if (!ret) {
+ psp->xgmi_context.initialized = 1;
+ psp->xgmi_context.session_id = cmd->resp.session_id;
+ }
+
+ kfree(cmd);
+
+ return ret;
+}
+
+static void psp_prep_xgmi_ta_unload_cmd_buf(struct psp_gfx_cmd_resp *cmd,
+ uint32_t xgmi_session_id)
+{
+ cmd->cmd_id = GFX_CMD_ID_UNLOAD_TA;
+ cmd->cmd.cmd_unload_ta.session_id = xgmi_session_id;
+}
+
+static int psp_xgmi_unload(struct psp_context *psp)
+{
+ int ret;
+ struct psp_gfx_cmd_resp *cmd;
+
+ /*
+ * TODO: bypass the unloading in sriov for now
+ */
+ if (amdgpu_sriov_vf(psp->adev))
+ return 0;
+
+ cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+ if (!cmd)
+ return -ENOMEM;
+
+ psp_prep_xgmi_ta_unload_cmd_buf(cmd, psp->xgmi_context.session_id);
+
+ ret = psp_cmd_submit_buf(psp, NULL, cmd,
+ psp->fence_buf_mc_addr);
+
+ kfree(cmd);
+
+ return ret;
+}
+
+static void psp_prep_xgmi_ta_invoke_cmd_buf(struct psp_gfx_cmd_resp *cmd,
+ uint32_t ta_cmd_id,
+ uint32_t xgmi_session_id)
+{
+ cmd->cmd_id = GFX_CMD_ID_INVOKE_CMD;
+ cmd->cmd.cmd_invoke_cmd.session_id = xgmi_session_id;
+ cmd->cmd.cmd_invoke_cmd.ta_cmd_id = ta_cmd_id;
+ /* Note: cmd_invoke_cmd.buf is not used for now */
+}
+
+int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id)
+{
+ int ret;
+ struct psp_gfx_cmd_resp *cmd;
+
+ /*
+ * TODO: bypass the loading in sriov for now
+ */
+ if (amdgpu_sriov_vf(psp->adev))
+ return 0;
+
+ cmd = kzalloc(sizeof(struct psp_gfx_cmd_resp), GFP_KERNEL);
+ if (!cmd)
+ return -ENOMEM;
+
+ psp_prep_xgmi_ta_invoke_cmd_buf(cmd, ta_cmd_id,
+ psp->xgmi_context.session_id);
+
+ ret = psp_cmd_submit_buf(psp, NULL, cmd,
+ psp->fence_buf_mc_addr);
+
+ kfree(cmd);
+
+ return ret;
+}
+
+static int psp_xgmi_terminate(struct psp_context *psp)
+{
+ int ret;
+
+ if (!psp->xgmi_context.initialized)
+ return 0;
+
+ ret = psp_xgmi_unload(psp);
+ if (ret)
+ return ret;
+
+ psp->xgmi_context.initialized = 0;
+
+ /* free xgmi shared memory */
+ amdgpu_bo_free_kernel(&psp->xgmi_context.xgmi_shared_bo,
+ &psp->xgmi_context.xgmi_shared_mc_addr,
+ &psp->xgmi_context.xgmi_shared_buf);
+
+ return 0;
+}
+
+static int psp_xgmi_initialize(struct psp_context *psp)
+{
+ struct ta_xgmi_shared_memory *xgmi_cmd;
+ int ret;
+
+ if (!psp->xgmi_context.initialized) {
+ ret = psp_xgmi_init_shared_buf(psp);
+ if (ret)
+ return ret;
+ }
+
+ /* Load XGMI TA */
+ ret = psp_xgmi_load(psp);
+ if (ret)
+ return ret;
+
+ /* Initialize XGMI session */
+ xgmi_cmd = (struct ta_xgmi_shared_memory *)(psp->xgmi_context.xgmi_shared_buf);
+ memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory));
+ xgmi_cmd->cmd_id = TA_COMMAND_XGMI__INITIALIZE;
+
+ ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id);
+
+ return ret;
+}
+
static int psp_hw_start(struct psp_context *psp)
{
struct amdgpu_device *adev = psp->adev;
@@ -292,6 +479,15 @@ static int psp_hw_start(struct psp_context *psp)
if (ret)
return ret;
+ if (adev->gmc.xgmi.num_physical_nodes > 1) {
+ ret = psp_xgmi_initialize(psp);
+ /* Warning the XGMI seesion initialize failure
+ * Instead of stop driver initialization
+ */
+ if (ret)
+ dev_err(psp->adev->dev,
+ "XGMI: Failed to initialize XGMI session\n");
+ }
return 0;
}
@@ -321,7 +517,7 @@ static int psp_np_fw_load(struct psp_context *psp)
return ret;
ret = psp_cmd_submit_buf(psp, ucode, psp->cmd,
- psp->fence_buf_mc_addr, i + 3);
+ psp->fence_buf_mc_addr);
if (ret)
return ret;
@@ -452,6 +648,10 @@ static int psp_hw_fini(void *handle)
if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
return 0;
+ if (adev->gmc.xgmi.num_physical_nodes > 1 &&
+ psp->xgmi_context.initialized == 1)
+ psp_xgmi_terminate(psp);
+
psp_ring_destroy(psp, PSP_RING_TYPE__KM);
amdgpu_bo_free_kernel(&psp->tmr_bo, &psp->tmr_mc_addr, &psp->tmr_buf);
@@ -479,6 +679,15 @@ static int psp_suspend(void *handle)
if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP)
return 0;
+ if (adev->gmc.xgmi.num_physical_nodes > 1 &&
+ psp->xgmi_context.initialized == 1) {
+ ret = psp_xgmi_terminate(psp);
+ if (ret) {
+ DRM_ERROR("Failed to terminate xgmi ta\n");
+ return ret;
+ }
+ }
+
ret = psp_ring_stop(psp, PSP_RING_TYPE__KM);
if (ret) {
DRM_ERROR("PSP ring stop failed\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
index 8b8720e..9ec5d1a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.h
@@ -27,14 +27,17 @@
#include "amdgpu.h"
#include "psp_gfx_if.h"
+#include "ta_xgmi_if.h"
#define PSP_FENCE_BUFFER_SIZE 0x1000
#define PSP_CMD_BUFFER_SIZE 0x1000
-#define PSP_ASD_SHARED_MEM_SIZE 0x4000
+#define PSP_ASD_SHARED_MEM_SIZE 0x4000
+#define PSP_XGMI_SHARED_MEM_SIZE 0x4000
#define PSP_1_MEG 0x100000
#define PSP_TMR_SIZE 0x400000
struct psp_context;
+struct psp_xgmi_node_info;
struct psp_xgmi_topology_info;
enum psp_ring_type
@@ -80,12 +83,20 @@ struct psp_funcs
enum AMDGPU_UCODE_ID ucode_type);
bool (*smu_reload_quirk)(struct psp_context *psp);
int (*mode1_reset)(struct psp_context *psp);
- uint64_t (*xgmi_get_device_id)(struct psp_context *psp);
+ uint64_t (*xgmi_get_node_id)(struct psp_context *psp);
uint64_t (*xgmi_get_hive_id)(struct psp_context *psp);
int (*xgmi_get_topology_info)(struct psp_context *psp, int number_devices,
- struct psp_xgmi_topology_info *topology);
+ struct psp_xgmi_topology_info *topology);
int (*xgmi_set_topology_info)(struct psp_context *psp, int number_devices,
- struct psp_xgmi_topology_info *topology);
+ struct psp_xgmi_topology_info *topology);
+};
+
+struct psp_xgmi_context {
+ uint8_t initialized;
+ uint32_t session_id;
+ struct amdgpu_bo *xgmi_shared_bo;
+ uint64_t xgmi_shared_mc_addr;
+ void *xgmi_shared_buf;
};
struct psp_context
@@ -96,7 +107,7 @@ struct psp_context
const struct psp_funcs *funcs;
- /* fence buffer */
+ /* firmware buffer */
struct amdgpu_bo *fw_pri_bo;
uint64_t fw_pri_mc_addr;
void *fw_pri_buf;
@@ -134,6 +145,16 @@ struct psp_context
struct amdgpu_bo *cmd_buf_bo;
uint64_t cmd_buf_mc_addr;
struct psp_gfx_cmd_resp *cmd_buf_mem;
+
+ /* fence value associated with cmd buffer */
+ atomic_t fence_value;
+
+ /* xgmi ta firmware and buffer */
+ const struct firmware *ta_fw;
+ uint32_t ta_xgmi_ucode_version;
+ uint32_t ta_xgmi_ucode_size;
+ uint8_t *ta_xgmi_start_addr;
+ struct psp_xgmi_context xgmi_context;
};
struct amdgpu_psp_funcs {
@@ -141,21 +162,17 @@ struct amdgpu_psp_funcs {
enum AMDGPU_UCODE_ID);
};
+#define AMDGPU_XGMI_MAX_CONNECTED_NODES 64
+struct psp_xgmi_node_info {
+ uint64_t node_id;
+ uint8_t num_hops;
+ uint8_t is_sharing_enabled;
+ enum ta_xgmi_assigned_sdma_engine sdma_engine;
+};
+
struct psp_xgmi_topology_info {
- /* Generated by PSP to identify the GPU instance within xgmi connection */
- uint64_t device_id;
- /*
- * If all bits set to 0 , driver indicates it wants to retrieve the xgmi
- * connection vector topology, but not access enable the connections
- * if some or all bits are set to 1, driver indicates it want to retrieve the
- * current xgmi topology and access enable the link to GPU[i] associated
- * with the bit position in the vector.
- * On return,: bits indicated which xgmi links are present/active depending
- * on the value passed in. The relative bit offset for the relative GPU index
- * within the hive is always marked active.
- */
- uint32_t connection_mask;
- uint32_t reserved; /* must be 0 */
+ uint32_t num_nodes;
+ struct psp_xgmi_node_info nodes[AMDGPU_XGMI_MAX_CONNECTED_NODES];
};
#define psp_prep_cmd_buf(ucode, type) (psp)->funcs->prep_cmd_buf((ucode), (type))
@@ -177,8 +194,8 @@ struct psp_xgmi_topology_info {
((psp)->funcs->smu_reload_quirk ? (psp)->funcs->smu_reload_quirk((psp)) : false)
#define psp_mode1_reset(psp) \
((psp)->funcs->mode1_reset ? (psp)->funcs->mode1_reset((psp)) : false)
-#define psp_xgmi_get_device_id(psp) \
- ((psp)->funcs->xgmi_get_device_id ? (psp)->funcs->xgmi_get_device_id((psp)) : 0)
+#define psp_xgmi_get_node_id(psp) \
+ ((psp)->funcs->xgmi_get_node_id ? (psp)->funcs->xgmi_get_node_id((psp)) : 0)
#define psp_xgmi_get_hive_id(psp) \
((psp)->funcs->xgmi_get_hive_id ? (psp)->funcs->xgmi_get_hive_id((psp)) : 0)
#define psp_xgmi_get_topology_info(psp, num_device, topology) \
@@ -199,6 +216,8 @@ extern int psp_wait_for(struct psp_context *psp, uint32_t reg_index,
extern const struct amdgpu_ip_block_version psp_v10_0_ip_block;
int psp_gpu_reset(struct amdgpu_device *adev);
+int psp_xgmi_invoke(struct psp_context *psp, uint32_t ta_cmd_id);
+
extern const struct amdgpu_ip_block_version psp_v11_0_ip_block;
#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index b70e85e..5b75bdc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -338,7 +338,7 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
*/
void amdgpu_ring_fini(struct amdgpu_ring *ring)
{
- ring->ready = false;
+ ring->sched.ready = false;
/* Not to finish a ring which is not initialized */
if (!(ring->adev) || !(ring->adev->rings[ring->idx]))
@@ -500,3 +500,29 @@ static void amdgpu_debugfs_ring_fini(struct amdgpu_ring *ring)
debugfs_remove(ring->ent);
#endif
}
+
+/**
+ * amdgpu_ring_test_helper - tests ring and set sched readiness status
+ *
+ * @ring: ring to try the recovery on
+ *
+ * Tests ring and set sched readiness status
+ *
+ * Returns 0 on success, error on failure.
+ */
+int amdgpu_ring_test_helper(struct amdgpu_ring *ring)
+{
+ struct amdgpu_device *adev = ring->adev;
+ int r;
+
+ r = amdgpu_ring_test_ring(ring);
+ if (r)
+ DRM_DEV_ERROR(adev->dev, "ring %s test failed (%d)\n",
+ ring->name, r);
+ else
+ DRM_DEV_DEBUG(adev->dev, "ring test on %s succeeded\n",
+ ring->name);
+
+ ring->sched.ready = !r;
+ return r;
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
index 4caa301..0beb01f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.h
@@ -129,8 +129,9 @@ struct amdgpu_ring_funcs {
unsigned emit_ib_size;
/* command emit functions */
void (*emit_ib)(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch);
+ bool ctx_switch);
void (*emit_fence)(struct amdgpu_ring *ring, uint64_t addr,
uint64_t seq, unsigned flags);
void (*emit_pipeline_sync)(struct amdgpu_ring *ring);
@@ -189,7 +190,6 @@ struct amdgpu_ring {
uint64_t gpu_addr;
uint64_t ptr_mask;
uint32_t buf_mask;
- bool ready;
u32 idx;
u32 me;
u32 pipe;
@@ -229,7 +229,7 @@ struct amdgpu_ring {
#define amdgpu_ring_get_rptr(r) (r)->funcs->get_rptr((r))
#define amdgpu_ring_get_wptr(r) (r)->funcs->get_wptr((r))
#define amdgpu_ring_set_wptr(r) (r)->funcs->set_wptr((r))
-#define amdgpu_ring_emit_ib(r, ib, vmid, c) (r)->funcs->emit_ib((r), (ib), (vmid), (c))
+#define amdgpu_ring_emit_ib(r, job, ib, c) ((r)->funcs->emit_ib((r), (job), (ib), (c)))
#define amdgpu_ring_emit_pipeline_sync(r) (r)->funcs->emit_pipeline_sync((r))
#define amdgpu_ring_emit_vm_flush(r, vmid, addr) (r)->funcs->emit_vm_flush((r), (vmid), (addr))
#define amdgpu_ring_emit_fence(r, addr, seq, flags) (r)->funcs->emit_fence((r), (addr), (seq), (flags))
@@ -313,4 +313,6 @@ static inline void amdgpu_ring_write_multiple(struct amdgpu_ring *ring,
ring->count_dw -= count_dw;
}
+int amdgpu_ring_test_helper(struct amdgpu_ring *ring);
+
#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
new file mode 100644
index 0000000..c8793e6
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.c
@@ -0,0 +1,282 @@
+/*
+ * Copyright 2014 Advanced Micro Devices, Inc.
+ * Copyright 2008 Red Hat Inc.
+ * Copyright 2009 Jerome Glisse.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+#include <linux/firmware.h>
+#include "amdgpu.h"
+#include "amdgpu_gfx.h"
+#include "amdgpu_rlc.h"
+
+/**
+ * amdgpu_gfx_rlc_enter_safe_mode - Set RLC into safe mode
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Set RLC enter into safe mode if RLC is enabled and haven't in safe mode.
+ */
+void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev)
+{
+ if (adev->gfx.rlc.in_safe_mode)
+ return;
+
+ /* if RLC is not enabled, do nothing */
+ if (!adev->gfx.rlc.funcs->is_rlc_enabled(adev))
+ return;
+
+ if (adev->cg_flags &
+ (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG |
+ AMD_CG_SUPPORT_GFX_3D_CGCG)) {
+ adev->gfx.rlc.funcs->set_safe_mode(adev);
+ adev->gfx.rlc.in_safe_mode = true;
+ }
+}
+
+/**
+ * amdgpu_gfx_rlc_exit_safe_mode - Set RLC out of safe mode
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Set RLC exit safe mode if RLC is enabled and have entered into safe mode.
+ */
+void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev)
+{
+ if (!(adev->gfx.rlc.in_safe_mode))
+ return;
+
+ /* if RLC is not enabled, do nothing */
+ if (!adev->gfx.rlc.funcs->is_rlc_enabled(adev))
+ return;
+
+ if (adev->cg_flags &
+ (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG |
+ AMD_CG_SUPPORT_GFX_3D_CGCG)) {
+ adev->gfx.rlc.funcs->unset_safe_mode(adev);
+ adev->gfx.rlc.in_safe_mode = false;
+ }
+}
+
+/**
+ * amdgpu_gfx_rlc_init_sr - Init save restore block
+ *
+ * @adev: amdgpu_device pointer
+ * @dws: the size of save restore block
+ *
+ * Allocate and setup value to save restore block of rlc.
+ * Returns 0 on succeess or negative error code if allocate failed.
+ */
+int amdgpu_gfx_rlc_init_sr(struct amdgpu_device *adev, u32 dws)
+{
+ const u32 *src_ptr;
+ volatile u32 *dst_ptr;
+ u32 i;
+ int r;
+
+ /* allocate save restore block */
+ r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
+ AMDGPU_GEM_DOMAIN_VRAM,
+ &adev->gfx.rlc.save_restore_obj,
+ &adev->gfx.rlc.save_restore_gpu_addr,
+ (void **)&adev->gfx.rlc.sr_ptr);
+ if (r) {
+ dev_warn(adev->dev, "(%d) create RLC sr bo failed\n", r);
+ amdgpu_gfx_rlc_fini(adev);
+ return r;
+ }
+
+ /* write the sr buffer */
+ src_ptr = adev->gfx.rlc.reg_list;
+ dst_ptr = adev->gfx.rlc.sr_ptr;
+ for (i = 0; i < adev->gfx.rlc.reg_list_size; i++)
+ dst_ptr[i] = cpu_to_le32(src_ptr[i]);
+ amdgpu_bo_kunmap(adev->gfx.rlc.save_restore_obj);
+ amdgpu_bo_unreserve(adev->gfx.rlc.save_restore_obj);
+
+ return 0;
+}
+
+/**
+ * amdgpu_gfx_rlc_init_csb - Init clear state block
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Allocate and setup value to clear state block of rlc.
+ * Returns 0 on succeess or negative error code if allocate failed.
+ */
+int amdgpu_gfx_rlc_init_csb(struct amdgpu_device *adev)
+{
+ volatile u32 *dst_ptr;
+ u32 dws;
+ int r;
+
+ /* allocate clear state block */
+ adev->gfx.rlc.clear_state_size = dws = adev->gfx.rlc.funcs->get_csb_size(adev);
+ r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
+ AMDGPU_GEM_DOMAIN_VRAM,
+ &adev->gfx.rlc.clear_state_obj,
+ &adev->gfx.rlc.clear_state_gpu_addr,
+ (void **)&adev->gfx.rlc.cs_ptr);
+ if (r) {
+ dev_err(adev->dev, "(%d) failed to create rlc csb bo\n", r);
+ amdgpu_gfx_rlc_fini(adev);
+ return r;
+ }
+
+ /* set up the cs buffer */
+ dst_ptr = adev->gfx.rlc.cs_ptr;
+ adev->gfx.rlc.funcs->get_csb_buffer(adev, dst_ptr);
+ amdgpu_bo_kunmap(adev->gfx.rlc.clear_state_obj);
+ amdgpu_bo_unpin(adev->gfx.rlc.clear_state_obj);
+ amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
+
+ return 0;
+}
+
+/**
+ * amdgpu_gfx_rlc_init_cpt - Init cp table
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Allocate and setup value to cp table of rlc.
+ * Returns 0 on succeess or negative error code if allocate failed.
+ */
+int amdgpu_gfx_rlc_init_cpt(struct amdgpu_device *adev)
+{
+ int r;
+
+ r = amdgpu_bo_create_reserved(adev, adev->gfx.rlc.cp_table_size,
+ PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
+ &adev->gfx.rlc.cp_table_obj,
+ &adev->gfx.rlc.cp_table_gpu_addr,
+ (void **)&adev->gfx.rlc.cp_table_ptr);
+ if (r) {
+ dev_err(adev->dev, "(%d) failed to create cp table bo\n", r);
+ amdgpu_gfx_rlc_fini(adev);
+ return r;
+ }
+
+ /* set up the cp table */
+ amdgpu_gfx_rlc_setup_cp_table(adev);
+ amdgpu_bo_kunmap(adev->gfx.rlc.cp_table_obj);
+ amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
+
+ return 0;
+}
+
+/**
+ * amdgpu_gfx_rlc_setup_cp_table - setup cp the buffer of cp table
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Write cp firmware data into cp table.
+ */
+void amdgpu_gfx_rlc_setup_cp_table(struct amdgpu_device *adev)
+{
+ const __le32 *fw_data;
+ volatile u32 *dst_ptr;
+ int me, i, max_me;
+ u32 bo_offset = 0;
+ u32 table_offset, table_size;
+
+ max_me = adev->gfx.rlc.funcs->get_cp_table_num(adev);
+
+ /* write the cp table buffer */
+ dst_ptr = adev->gfx.rlc.cp_table_ptr;
+ for (me = 0; me < max_me; me++) {
+ if (me == 0) {
+ const struct gfx_firmware_header_v1_0 *hdr =
+ (const struct gfx_firmware_header_v1_0 *)adev->gfx.ce_fw->data;
+ fw_data = (const __le32 *)
+ (adev->gfx.ce_fw->data +
+ le32_to_cpu(hdr->header.ucode_array_offset_bytes));
+ table_offset = le32_to_cpu(hdr->jt_offset);
+ table_size = le32_to_cpu(hdr->jt_size);
+ } else if (me == 1) {
+ const struct gfx_firmware_header_v1_0 *hdr =
+ (const struct gfx_firmware_header_v1_0 *)adev->gfx.pfp_fw->data;
+ fw_data = (const __le32 *)
+ (adev->gfx.pfp_fw->data +
+ le32_to_cpu(hdr->header.ucode_array_offset_bytes));
+ table_offset = le32_to_cpu(hdr->jt_offset);
+ table_size = le32_to_cpu(hdr->jt_size);
+ } else if (me == 2) {
+ const struct gfx_firmware_header_v1_0 *hdr =
+ (const struct gfx_firmware_header_v1_0 *)adev->gfx.me_fw->data;
+ fw_data = (const __le32 *)
+ (adev->gfx.me_fw->data +
+ le32_to_cpu(hdr->header.ucode_array_offset_bytes));
+ table_offset = le32_to_cpu(hdr->jt_offset);
+ table_size = le32_to_cpu(hdr->jt_size);
+ } else if (me == 3) {
+ const struct gfx_firmware_header_v1_0 *hdr =
+ (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
+ fw_data = (const __le32 *)
+ (adev->gfx.mec_fw->data +
+ le32_to_cpu(hdr->header.ucode_array_offset_bytes));
+ table_offset = le32_to_cpu(hdr->jt_offset);
+ table_size = le32_to_cpu(hdr->jt_size);
+ } else if (me == 4) {
+ const struct gfx_firmware_header_v1_0 *hdr =
+ (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec2_fw->data;
+ fw_data = (const __le32 *)
+ (adev->gfx.mec2_fw->data +
+ le32_to_cpu(hdr->header.ucode_array_offset_bytes));
+ table_offset = le32_to_cpu(hdr->jt_offset);
+ table_size = le32_to_cpu(hdr->jt_size);
+ }
+
+ for (i = 0; i < table_size; i ++) {
+ dst_ptr[bo_offset + i] =
+ cpu_to_le32(le32_to_cpu(fw_data[table_offset + i]));
+ }
+
+ bo_offset += table_size;
+ }
+}
+
+/**
+ * amdgpu_gfx_rlc_fini - Free BO which used for RLC
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Free three BO which is used for rlc_save_restore_block, rlc_clear_state_block
+ * and rlc_jump_table_block.
+ */
+void amdgpu_gfx_rlc_fini(struct amdgpu_device *adev)
+{
+ /* save restore block */
+ if (adev->gfx.rlc.save_restore_obj) {
+ amdgpu_bo_free_kernel(&adev->gfx.rlc.save_restore_obj,
+ &adev->gfx.rlc.save_restore_gpu_addr,
+ (void **)&adev->gfx.rlc.sr_ptr);
+ }
+
+ /* clear state block */
+ amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj,
+ &adev->gfx.rlc.clear_state_gpu_addr,
+ (void **)&adev->gfx.rlc.cs_ptr);
+
+ /* jump table block */
+ amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj,
+ &adev->gfx.rlc.cp_table_gpu_addr,
+ (void **)&adev->gfx.rlc.cp_table_ptr);
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
new file mode 100644
index 0000000..49a8ab5
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_rlc.h
@@ -0,0 +1,98 @@
+/*
+ * Copyright 2014 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __AMDGPU_RLC_H__
+#define __AMDGPU_RLC_H__
+
+#include "clearstate_defs.h"
+
+struct amdgpu_rlc_funcs {
+ bool (*is_rlc_enabled)(struct amdgpu_device *adev);
+ void (*set_safe_mode)(struct amdgpu_device *adev);
+ void (*unset_safe_mode)(struct amdgpu_device *adev);
+ int (*init)(struct amdgpu_device *adev);
+ u32 (*get_csb_size)(struct amdgpu_device *adev);
+ void (*get_csb_buffer)(struct amdgpu_device *adev, volatile u32 *buffer);
+ int (*get_cp_table_num)(struct amdgpu_device *adev);
+ int (*resume)(struct amdgpu_device *adev);
+ void (*stop)(struct amdgpu_device *adev);
+ void (*reset)(struct amdgpu_device *adev);
+ void (*start)(struct amdgpu_device *adev);
+};
+
+struct amdgpu_rlc {
+ /* for power gating */
+ struct amdgpu_bo *save_restore_obj;
+ uint64_t save_restore_gpu_addr;
+ volatile uint32_t *sr_ptr;
+ const u32 *reg_list;
+ u32 reg_list_size;
+ /* for clear state */
+ struct amdgpu_bo *clear_state_obj;
+ uint64_t clear_state_gpu_addr;
+ volatile uint32_t *cs_ptr;
+ const struct cs_section_def *cs_data;
+ u32 clear_state_size;
+ /* for cp tables */
+ struct amdgpu_bo *cp_table_obj;
+ uint64_t cp_table_gpu_addr;
+ volatile uint32_t *cp_table_ptr;
+ u32 cp_table_size;
+
+ /* safe mode for updating CG/PG state */
+ bool in_safe_mode;
+ const struct amdgpu_rlc_funcs *funcs;
+
+ /* for firmware data */
+ u32 save_and_restore_offset;
+ u32 clear_state_descriptor_offset;
+ u32 avail_scratch_ram_locations;
+ u32 reg_restore_list_size;
+ u32 reg_list_format_start;
+ u32 reg_list_format_separate_start;
+ u32 starting_offsets_start;
+ u32 reg_list_format_size_bytes;
+ u32 reg_list_size_bytes;
+ u32 reg_list_format_direct_reg_list_length;
+ u32 save_restore_list_cntl_size_bytes;
+ u32 save_restore_list_gpm_size_bytes;
+ u32 save_restore_list_srm_size_bytes;
+
+ u32 *register_list_format;
+ u32 *register_restore;
+ u8 *save_restore_list_cntl;
+ u8 *save_restore_list_gpm;
+ u8 *save_restore_list_srm;
+
+ bool is_rlc_v2_1;
+};
+
+void amdgpu_gfx_rlc_enter_safe_mode(struct amdgpu_device *adev);
+void amdgpu_gfx_rlc_exit_safe_mode(struct amdgpu_device *adev);
+int amdgpu_gfx_rlc_init_sr(struct amdgpu_device *adev, u32 dws);
+int amdgpu_gfx_rlc_init_csb(struct amdgpu_device *adev);
+int amdgpu_gfx_rlc_init_cpt(struct amdgpu_device *adev);
+void amdgpu_gfx_rlc_setup_cp_table(struct amdgpu_device *adev);
+void amdgpu_gfx_rlc_fini(struct amdgpu_device *adev);
+
+#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
index bc9244b..115bb0c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.c
@@ -28,17 +28,31 @@
* GPU SDMA IP block helpers function.
*/
-struct amdgpu_sdma_instance * amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
+struct amdgpu_sdma_instance *amdgpu_sdma_get_instance_from_ring(struct amdgpu_ring *ring)
{
struct amdgpu_device *adev = ring->adev;
int i;
for (i = 0; i < adev->sdma.num_instances; i++)
- if (&adev->sdma.instance[i].ring == ring)
- break;
+ if (ring == &adev->sdma.instance[i].ring ||
+ ring == &adev->sdma.instance[i].page)
+ return &adev->sdma.instance[i];
- if (i < AMDGPU_MAX_SDMA_INSTANCES)
- return &adev->sdma.instance[i];
- else
- return NULL;
+ return NULL;
+}
+
+int amdgpu_sdma_get_index_from_ring(struct amdgpu_ring *ring, uint32_t *index)
+{
+ struct amdgpu_device *adev = ring->adev;
+ int i;
+
+ for (i = 0; i < adev->sdma.num_instances; i++) {
+ if (ring == &adev->sdma.instance[i].ring ||
+ ring == &adev->sdma.instance[i].page) {
+ *index = i;
+ return 0;
+ }
+ }
+
+ return -EINVAL;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
index 500113e..16b1a6a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sdma.h
@@ -41,6 +41,7 @@ struct amdgpu_sdma_instance {
uint32_t feature_version;
struct amdgpu_ring ring;
+ struct amdgpu_ring page;
bool burst_nop;
};
@@ -50,6 +51,7 @@ struct amdgpu_sdma {
struct amdgpu_irq_src illegal_inst_irq;
int num_instances;
uint32_t srbm_soft_reset;
+ bool has_page_queue;
};
/*
@@ -92,6 +94,7 @@ struct amdgpu_buffer_funcs {
#define amdgpu_emit_fill_buffer(adev, ib, s, d, b) (adev)->mman.buffer_funcs->emit_fill_buffer((ib), (s), (d), (b))
struct amdgpu_sdma_instance *
-amdgpu_get_sdma_instance(struct amdgpu_ring *ring);
+amdgpu_sdma_get_instance_from_ring(struct amdgpu_ring *ring);
+int amdgpu_sdma_get_index_from_ring(struct amdgpu_ring *ring, uint32_t *index);
#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index e9bf70e..626abca 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -218,6 +218,7 @@ TRACE_EVENT(amdgpu_vm_grab_id,
TP_ARGS(vm, ring, job),
TP_STRUCT__entry(
__field(u32, pasid)
+ __string(ring, ring->name)
__field(u32, ring)
__field(u32, vmid)
__field(u32, vm_hub)
@@ -227,14 +228,14 @@ TRACE_EVENT(amdgpu_vm_grab_id,
TP_fast_assign(
__entry->pasid = vm->pasid;
- __entry->ring = ring->idx;
+ __assign_str(ring, ring->name)
__entry->vmid = job->vmid;
__entry->vm_hub = ring->funcs->vmhub,
__entry->pd_addr = job->vm_pd_addr;
__entry->needs_flush = job->vm_needs_flush;
),
- TP_printk("pasid=%d, ring=%u, id=%u, hub=%u, pd_addr=%010Lx needs_flush=%u",
- __entry->pasid, __entry->ring, __entry->vmid,
+ TP_printk("pasid=%d, ring=%s, id=%u, hub=%u, pd_addr=%010Lx needs_flush=%u",
+ __entry->pasid, __get_str(ring), __entry->vmid,
__entry->vm_hub, __entry->pd_addr, __entry->needs_flush)
);
@@ -366,20 +367,20 @@ TRACE_EVENT(amdgpu_vm_flush,
uint64_t pd_addr),
TP_ARGS(ring, vmid, pd_addr),
TP_STRUCT__entry(
- __field(u32, ring)
+ __string(ring, ring->name)
__field(u32, vmid)
__field(u32, vm_hub)
__field(u64, pd_addr)
),
TP_fast_assign(
- __entry->ring = ring->idx;
+ __assign_str(ring, ring->name)
__entry->vmid = vmid;
__entry->vm_hub = ring->funcs->vmhub;
__entry->pd_addr = pd_addr;
),
- TP_printk("ring=%u, id=%u, hub=%u, pd_addr=%010Lx",
- __entry->ring, __entry->vmid,
+ TP_printk("ring=%s, id=%u, hub=%u, pd_addr=%010Lx",
+ __get_str(ring), __entry->vmid,
__entry->vm_hub,__entry->pd_addr)
);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index a44fc12..c91ec31 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -61,100 +61,6 @@ static int amdgpu_map_buffer(struct ttm_buffer_object *bo,
static int amdgpu_ttm_debugfs_init(struct amdgpu_device *adev);
static void amdgpu_ttm_debugfs_fini(struct amdgpu_device *adev);
-/*
- * Global memory.
- */
-
-/**
- * amdgpu_ttm_mem_global_init - Initialize and acquire reference to
- * memory object
- *
- * @ref: Object for initialization.
- *
- * This is called by drm_global_item_ref() when an object is being
- * initialized.
- */
-static int amdgpu_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-/**
- * amdgpu_ttm_mem_global_release - Drop reference to a memory object
- *
- * @ref: Object being removed
- *
- * This is called by drm_global_item_unref() when an object is being
- * released.
- */
-static void amdgpu_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-/**
- * amdgpu_ttm_global_init - Initialize global TTM memory reference structures.
- *
- * @adev: AMDGPU device for which the global structures need to be registered.
- *
- * This is called as part of the AMDGPU ttm init from amdgpu_ttm_init()
- * during bring up.
- */
-static int amdgpu_ttm_global_init(struct amdgpu_device *adev)
-{
- struct drm_global_reference *global_ref;
- int r;
-
- /* ensure reference is false in case init fails */
- adev->mman.mem_global_referenced = false;
-
- global_ref = &adev->mman.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &amdgpu_ttm_mem_global_init;
- global_ref->release = &amdgpu_ttm_mem_global_release;
- r = drm_global_item_ref(global_ref);
- if (r) {
- DRM_ERROR("Failed setting up TTM memory accounting "
- "subsystem.\n");
- goto error_mem;
- }
-
- adev->mman.bo_global_ref.mem_glob =
- adev->mman.mem_global_ref.object;
- global_ref = &adev->mman.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- r = drm_global_item_ref(global_ref);
- if (r) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- goto error_bo;
- }
-
- mutex_init(&adev->mman.gtt_window_lock);
-
- adev->mman.mem_global_referenced = true;
-
- return 0;
-
-error_bo:
- drm_global_item_unref(&adev->mman.mem_global_ref);
-error_mem:
- return r;
-}
-
-static void amdgpu_ttm_global_fini(struct amdgpu_device *adev)
-{
- if (adev->mman.mem_global_referenced) {
- mutex_destroy(&adev->mman.gtt_window_lock);
- drm_global_item_unref(&adev->mman.bo_global_ref.ref);
- drm_global_item_unref(&adev->mman.mem_global_ref);
- adev->mman.mem_global_referenced = false;
- }
-}
-
static int amdgpu_invalidate_caches(struct ttm_bo_device *bdev, uint32_t flags)
{
return 0;
@@ -1758,14 +1664,10 @@ int amdgpu_ttm_init(struct amdgpu_device *adev)
int r;
u64 vis_vram_limit;
- /* initialize global references for vram/gtt */
- r = amdgpu_ttm_global_init(adev);
- if (r) {
- return r;
- }
+ mutex_init(&adev->mman.gtt_window_lock);
+
/* No others user of address space so set it to 0 */
r = ttm_bo_device_init(&adev->mman.bdev,
- adev->mman.bo_global_ref.ref.object,
&amdgpu_bo_driver,
adev->ddev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET,
@@ -1922,7 +1824,6 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
ttm_bo_clean_mm(&adev->mman.bdev, AMDGPU_PL_GWS);
ttm_bo_clean_mm(&adev->mman.bdev, AMDGPU_PL_OA);
ttm_bo_device_release(&adev->mman.bdev);
- amdgpu_ttm_global_fini(adev);
adev->mman.initialized = false;
DRM_INFO("amdgpu: ttm finalized\n");
}
@@ -2069,7 +1970,7 @@ int amdgpu_copy_buffer(struct amdgpu_ring *ring, uint64_t src_offset,
unsigned i;
int r;
- if (direct_submit && !ring->ready) {
+ if (direct_submit && !ring->sched.ready) {
DRM_ERROR("Trying to move memory with ring turned off.\n");
return -EINVAL;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
index fe8f276..b5b2d10 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h
@@ -39,8 +39,6 @@
#define AMDGPU_GTT_NUM_TRANSFER_WINDOWS 2
struct amdgpu_mman {
- struct ttm_bo_global_ref bo_global_ref;
- struct drm_global_reference mem_global_ref;
struct ttm_bo_device bdev;
bool mem_global_referenced;
bool initialized;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
index aa6641b..7ac25a1 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ucode.h
@@ -58,6 +58,17 @@ struct psp_firmware_header_v1_0 {
};
/* version_major=1, version_minor=0 */
+struct ta_firmware_header_v1_0 {
+ struct common_firmware_header header;
+ uint32_t ta_xgmi_ucode_version;
+ uint32_t ta_xgmi_offset_bytes;
+ uint32_t ta_xgmi_size_bytes;
+ uint32_t ta_ras_ucode_version;
+ uint32_t ta_ras_offset_bytes;
+ uint32_t ta_ras_size_bytes;
+};
+
+/* version_major=1, version_minor=0 */
struct gfx_firmware_header_v1_0 {
struct common_firmware_header header;
uint32_t ucode_feature_version;
@@ -170,6 +181,7 @@ union amdgpu_firmware_header {
struct mc_firmware_header_v1_0 mc;
struct smc_firmware_header_v1_0 smc;
struct psp_firmware_header_v1_0 psp;
+ struct ta_firmware_header_v1_0 ta;
struct gfx_firmware_header_v1_0 gfx;
struct rlc_firmware_header_v1_0 rlc;
struct rlc_firmware_header_v2_0 rlc_v2_0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index e5a6db6..69896f4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -1243,30 +1243,20 @@ int amdgpu_uvd_ring_test_ib(struct amdgpu_ring *ring, long timeout)
{
struct dma_fence *fence;
long r;
- uint32_t ip_instance = ring->me;
r = amdgpu_uvd_get_create_msg(ring, 1, NULL);
- if (r) {
- DRM_ERROR("amdgpu: (%d)failed to get create msg (%ld).\n", ip_instance, r);
+ if (r)
goto error;
- }
r = amdgpu_uvd_get_destroy_msg(ring, 1, true, &fence);
- if (r) {
- DRM_ERROR("amdgpu: (%d)failed to get destroy ib (%ld).\n", ip_instance, r);
+ if (r)
goto error;
- }
r = dma_fence_wait_timeout(fence, false, timeout);
- if (r == 0) {
- DRM_ERROR("amdgpu: (%d)IB test timed out.\n", ip_instance);
+ if (r == 0)
r = -ETIMEDOUT;
- } else if (r < 0) {
- DRM_ERROR("amdgpu: (%d)fence wait failed (%ld).\n", ip_instance, r);
- } else {
- DRM_DEBUG("ib test on (%d)ring %d succeeded\n", ip_instance, ring->idx);
+ else if (r > 0)
r = 0;
- }
dma_fence_put(fence);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index 5f3f540..98a1b2c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -1032,8 +1032,10 @@ int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, uint32_t ib_idx)
* @ib: the IB to execute
*
*/
-void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
amdgpu_ring_write(ring, VCE_CMD_IB);
amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
@@ -1079,11 +1081,9 @@ int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring)
return 0;
r = amdgpu_ring_alloc(ring, 16);
- if (r) {
- DRM_ERROR("amdgpu: vce failed to lock ring %d (%d).\n",
- ring->idx, r);
+ if (r)
return r;
- }
+
amdgpu_ring_write(ring, VCE_CMD_END);
amdgpu_ring_commit(ring);
@@ -1093,14 +1093,8 @@ int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed\n",
- ring->idx);
+ if (i >= timeout)
r = -ETIMEDOUT;
- }
return r;
}
@@ -1121,27 +1115,19 @@ int amdgpu_vce_ring_test_ib(struct amdgpu_ring *ring, long timeout)
return 0;
r = amdgpu_vce_get_create_msg(ring, 1, NULL);
- if (r) {
- DRM_ERROR("amdgpu: failed to get create msg (%ld).\n", r);
+ if (r)
goto error;
- }
r = amdgpu_vce_get_destroy_msg(ring, 1, true, &fence);
- if (r) {
- DRM_ERROR("amdgpu: failed to get destroy ib (%ld).\n", r);
+ if (r)
goto error;
- }
r = dma_fence_wait_timeout(fence, false, timeout);
- if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out.\n");
+ if (r == 0)
r = -ETIMEDOUT;
- } else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
- } else {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ else if (r > 0)
r = 0;
- }
+
error:
dma_fence_put(fence);
return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
index a1f209e..5029365 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.h
@@ -65,8 +65,8 @@ int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp);
int amdgpu_vce_ring_parse_cs(struct amdgpu_cs_parser *p, uint32_t ib_idx);
int amdgpu_vce_ring_parse_cs_vm(struct amdgpu_cs_parser *p, uint32_t ib_idx);
-void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch);
+void amdgpu_vce_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job,
+ struct amdgpu_ib *ib, bool ctx_switch);
void amdgpu_vce_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64 seq,
unsigned flags);
int amdgpu_vce_ring_test_ring(struct amdgpu_ring *ring);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
index 27da13d..e2e42e3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
@@ -425,11 +425,9 @@ int amdgpu_vcn_dec_ring_test_ring(struct amdgpu_ring *ring)
WREG32(SOC15_REG_OFFSET(UVD, 0, mmUVD_SCRATCH9), 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n",
- ring->idx, r);
+ if (r)
return r;
- }
+
amdgpu_ring_write(ring,
PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_SCRATCH9), 0));
amdgpu_ring_write(ring, 0xDEADBEEF);
@@ -441,14 +439,9 @@ int amdgpu_vcn_dec_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
return r;
}
@@ -570,30 +563,20 @@ int amdgpu_vcn_dec_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_vcn_dec_get_create_msg(ring, 1, NULL);
- if (r) {
- DRM_ERROR("amdgpu: failed to get create msg (%ld).\n", r);
+ if (r)
goto error;
- }
r = amdgpu_vcn_dec_get_destroy_msg(ring, 1, &fence);
- if (r) {
- DRM_ERROR("amdgpu: failed to get destroy ib (%ld).\n", r);
+ if (r)
goto error;
- }
r = dma_fence_wait_timeout(fence, false, timeout);
- if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out.\n");
+ if (r == 0)
r = -ETIMEDOUT;
- } else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
- } else {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ else if (r > 0)
r = 0;
- }
dma_fence_put(fence);
-
error:
return r;
}
@@ -606,11 +589,9 @@ int amdgpu_vcn_enc_ring_test_ring(struct amdgpu_ring *ring)
int r;
r = amdgpu_ring_alloc(ring, 16);
- if (r) {
- DRM_ERROR("amdgpu: vcn enc failed to lock ring %d (%d).\n",
- ring->idx, r);
+ if (r)
return r;
- }
+
amdgpu_ring_write(ring, VCN_ENC_CMD_END);
amdgpu_ring_commit(ring);
@@ -620,14 +601,8 @@ int amdgpu_vcn_enc_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed\n",
- ring->idx);
+ if (i >= adev->usec_timeout)
r = -ETIMEDOUT;
- }
return r;
}
@@ -742,27 +717,19 @@ int amdgpu_vcn_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_vcn_enc_get_create_msg(ring, 1, NULL);
- if (r) {
- DRM_ERROR("amdgpu: failed to get create msg (%ld).\n", r);
+ if (r)
goto error;
- }
r = amdgpu_vcn_enc_get_destroy_msg(ring, 1, &fence);
- if (r) {
- DRM_ERROR("amdgpu: failed to get destroy ib (%ld).\n", r);
+ if (r)
goto error;
- }
r = dma_fence_wait_timeout(fence, false, timeout);
- if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out.\n");
+ if (r == 0)
r = -ETIMEDOUT;
- } else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
- } else {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ else if (r > 0)
r = 0;
- }
+
error:
dma_fence_put(fence);
return r;
@@ -778,11 +745,8 @@ int amdgpu_vcn_jpeg_ring_test_ring(struct amdgpu_ring *ring)
WREG32(SOC15_REG_OFFSET(UVD, 0, mmUVD_SCRATCH9), 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n",
- ring->idx, r);
+ if (r)
return r;
- }
amdgpu_ring_write(ring,
PACKETJ(SOC15_REG_OFFSET(UVD, 0, mmUVD_SCRATCH9), 0, 0, 0));
@@ -796,14 +760,8 @@ int amdgpu_vcn_jpeg_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
return r;
}
@@ -856,21 +814,18 @@ int amdgpu_vcn_jpeg_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r = 0;
r = amdgpu_vcn_jpeg_set_reg(ring, 1, &fence);
- if (r) {
- DRM_ERROR("amdgpu: failed to set jpeg register (%ld).\n", r);
+ if (r)
goto error;
- }
r = dma_fence_wait_timeout(fence, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out.\n");
r = -ETIMEDOUT;
goto error;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto error;
- } else
+ } else {
r = 0;
+ }
for (i = 0; i < adev->usec_timeout; i++) {
tmp = RREG32(SOC15_REG_OFFSET(UVD, 0, mmUVD_SCRATCH9));
@@ -879,15 +834,10 @@ int amdgpu_vcn_jpeg_ring_test_ib(struct amdgpu_ring *ring, long timeout)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout)
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
- else {
- DRM_ERROR("ib test failed (0x%08X)\n", tmp);
- r = -EINVAL;
- }
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
dma_fence_put(fence);
-
error:
return r;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index f2f358a..cfee747 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -23,16 +23,6 @@
#include "amdgpu.h"
-uint64_t amdgpu_csa_vaddr(struct amdgpu_device *adev)
-{
- uint64_t addr = adev->vm_manager.max_pfn << AMDGPU_GPU_PAGE_SHIFT;
-
- addr -= AMDGPU_VA_RESERVED_SIZE;
- addr = amdgpu_gmc_sign_extend(addr);
-
- return addr;
-}
-
bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev)
{
/* By now all MMIO pages except mailbox are blocked */
@@ -41,88 +31,6 @@ bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev)
return RREG32_NO_KIQ(0xc040) == 0xffffffff;
}
-int amdgpu_allocate_static_csa(struct amdgpu_device *adev)
-{
- int r;
- void *ptr;
-
- r = amdgpu_bo_create_kernel(adev, AMDGPU_CSA_SIZE, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_VRAM, &adev->virt.csa_obj,
- &adev->virt.csa_vmid0_addr, &ptr);
- if (r)
- return r;
-
- memset(ptr, 0, AMDGPU_CSA_SIZE);
- return 0;
-}
-
-void amdgpu_free_static_csa(struct amdgpu_device *adev) {
- amdgpu_bo_free_kernel(&adev->virt.csa_obj,
- &adev->virt.csa_vmid0_addr,
- NULL);
-}
-
-/*
- * amdgpu_map_static_csa should be called during amdgpu_vm_init
- * it maps virtual address amdgpu_csa_vaddr() to this VM, and each command
- * submission of GFX should use this virtual address within META_DATA init
- * package to support SRIOV gfx preemption.
- */
-int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
- struct amdgpu_bo_va **bo_va)
-{
- uint64_t csa_addr = amdgpu_csa_vaddr(adev) & AMDGPU_GMC_HOLE_MASK;
- struct ww_acquire_ctx ticket;
- struct list_head list;
- struct amdgpu_bo_list_entry pd;
- struct ttm_validate_buffer csa_tv;
- int r;
-
- INIT_LIST_HEAD(&list);
- INIT_LIST_HEAD(&csa_tv.head);
- csa_tv.bo = &adev->virt.csa_obj->tbo;
- csa_tv.shared = true;
-
- list_add(&csa_tv.head, &list);
- amdgpu_vm_get_pd_bo(vm, &list, &pd);
-
- r = ttm_eu_reserve_buffers(&ticket, &list, true, NULL);
- if (r) {
- DRM_ERROR("failed to reserve CSA,PD BOs: err=%d\n", r);
- return r;
- }
-
- *bo_va = amdgpu_vm_bo_add(adev, vm, adev->virt.csa_obj);
- if (!*bo_va) {
- ttm_eu_backoff_reservation(&ticket, &list);
- DRM_ERROR("failed to create bo_va for static CSA\n");
- return -ENOMEM;
- }
-
- r = amdgpu_vm_alloc_pts(adev, (*bo_va)->base.vm, csa_addr,
- AMDGPU_CSA_SIZE);
- if (r) {
- DRM_ERROR("failed to allocate pts for static CSA, err=%d\n", r);
- amdgpu_vm_bo_rmv(adev, *bo_va);
- ttm_eu_backoff_reservation(&ticket, &list);
- return r;
- }
-
- r = amdgpu_vm_bo_map(adev, *bo_va, csa_addr, 0, AMDGPU_CSA_SIZE,
- AMDGPU_PTE_READABLE | AMDGPU_PTE_WRITEABLE |
- AMDGPU_PTE_EXECUTABLE);
-
- if (r) {
- DRM_ERROR("failed to do bo_map on static CSA, err=%d\n", r);
- amdgpu_vm_bo_rmv(adev, *bo_va);
- ttm_eu_backoff_reservation(&ticket, &list);
- return r;
- }
-
- ttm_eu_backoff_reservation(&ticket, &list);
- return 0;
-}
-
void amdgpu_virt_init_setting(struct amdgpu_device *adev)
{
/* enable virtual display */
@@ -162,9 +70,7 @@ uint32_t amdgpu_virt_kiq_rreg(struct amdgpu_device *adev, uint32_t reg)
if (r < 1 && (adev->in_gpu_reset || in_interrupt()))
goto failed_kiq_read;
- if (in_interrupt())
- might_sleep();
-
+ might_sleep();
while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) {
msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
@@ -210,9 +116,7 @@ void amdgpu_virt_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v)
if (r < 1 && (adev->in_gpu_reset || in_interrupt()))
goto failed_kiq_write;
- if (in_interrupt())
- might_sleep();
-
+ might_sleep();
while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) {
msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
@@ -228,6 +132,46 @@ void amdgpu_virt_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v)
pr_err("failed to write reg:%x\n", reg);
}
+void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev,
+ uint32_t reg0, uint32_t reg1,
+ uint32_t ref, uint32_t mask)
+{
+ struct amdgpu_kiq *kiq = &adev->gfx.kiq;
+ struct amdgpu_ring *ring = &kiq->ring;
+ signed long r, cnt = 0;
+ unsigned long flags;
+ uint32_t seq;
+
+ spin_lock_irqsave(&kiq->ring_lock, flags);
+ amdgpu_ring_alloc(ring, 32);
+ amdgpu_ring_emit_reg_write_reg_wait(ring, reg0, reg1,
+ ref, mask);
+ amdgpu_fence_emit_polling(ring, &seq);
+ amdgpu_ring_commit(ring);
+ spin_unlock_irqrestore(&kiq->ring_lock, flags);
+
+ r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
+
+ /* don't wait anymore for IRQ context */
+ if (r < 1 && in_interrupt())
+ goto failed_kiq;
+
+ might_sleep();
+ while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) {
+
+ msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
+ r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
+ }
+
+ if (cnt > MAX_KIQ_REG_TRY)
+ goto failed_kiq;
+
+ return;
+
+failed_kiq:
+ pr_err("failed to write reg %x wait reg %x\n", reg0, reg1);
+}
+
/**
* amdgpu_virt_request_full_gpu() - request full gpu access
* @amdgpu: amdgpu device.
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
index 880ac11..0728fbc 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.h
@@ -238,7 +238,6 @@ typedef struct amdgim_vf2pf_info_v2 amdgim_vf2pf_info ;
struct amdgpu_virt {
uint32_t caps;
struct amdgpu_bo *csa_obj;
- uint64_t csa_vmid0_addr;
bool chained_ib_support;
uint32_t reg_val_offs;
struct amdgpu_irq_src ack_irq;
@@ -251,8 +250,6 @@ struct amdgpu_virt {
uint32_t gim_feature;
};
-#define AMDGPU_CSA_SIZE (8 * 1024)
-
#define amdgpu_sriov_enabled(adev) \
((adev)->virt.caps & AMDGPU_SRIOV_CAPS_ENABLE_IOV)
@@ -277,17 +274,13 @@ static inline bool is_virtual_machine(void)
#endif
}
-struct amdgpu_vm;
-
-uint64_t amdgpu_csa_vaddr(struct amdgpu_device *adev);
bool amdgpu_virt_mmio_blocked(struct amdgpu_device *adev);
-int amdgpu_allocate_static_csa(struct amdgpu_device *adev);
-int amdgpu_map_static_csa(struct amdgpu_device *adev, struct amdgpu_vm *vm,
- struct amdgpu_bo_va **bo_va);
-void amdgpu_free_static_csa(struct amdgpu_device *adev);
void amdgpu_virt_init_setting(struct amdgpu_device *adev);
uint32_t amdgpu_virt_kiq_rreg(struct amdgpu_device *adev, uint32_t reg);
void amdgpu_virt_kiq_wreg(struct amdgpu_device *adev, uint32_t reg, uint32_t v);
+void amdgpu_virt_kiq_reg_write_reg_wait(struct amdgpu_device *adev,
+ uint32_t reg0, uint32_t rreg1,
+ uint32_t ref, uint32_t mask);
int amdgpu_virt_request_full_gpu(struct amdgpu_device *adev, bool init);
int amdgpu_virt_release_full_gpu(struct amdgpu_device *adev, bool init);
int amdgpu_virt_reset_gpu(struct amdgpu_device *adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index d6c4797..58a2363 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -1632,13 +1632,6 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
continue;
}
- /* First check if the entry is already handled */
- if (cursor.pfn < frag_start) {
- cursor.entry->huge = true;
- amdgpu_vm_pt_next(adev, &cursor);
- continue;
- }
-
/* If it isn't already handled it can't be a huge page */
if (cursor.entry->huge) {
/* Add the entry to the relocated list to update it. */
@@ -1701,8 +1694,17 @@ static int amdgpu_vm_update_ptes(struct amdgpu_pte_update_params *params,
}
} while (frag_start < entry_end);
- if (frag >= shift)
+ if (amdgpu_vm_pt_descendant(adev, &cursor)) {
+ /* Mark all child entries as huge */
+ while (cursor.pfn < frag_start) {
+ cursor.entry->huge = true;
+ amdgpu_vm_pt_next(adev, &cursor);
+ }
+
+ } else if (frag >= shift) {
+ /* or just move on to the next on the same level. */
amdgpu_vm_pt_next(adev, &cursor);
+ }
}
return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
index 897afbb..909216a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c
@@ -63,7 +63,7 @@ static struct amdgpu_hive_info *amdgpu_get_xgmi_hive(struct amdgpu_device *adev)
int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
{
- struct psp_xgmi_topology_info tmp_topology[AMDGPU_MAX_XGMI_DEVICE_PER_HIVE];
+ struct psp_xgmi_topology_info *tmp_topology;
struct amdgpu_hive_info *hive;
struct amdgpu_xgmi *entry;
struct amdgpu_device *tmp_adev;
@@ -73,10 +73,12 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
if ((adev->asic_type < CHIP_VEGA20) ||
(adev->flags & AMD_IS_APU) )
return 0;
- adev->gmc.xgmi.device_id = psp_xgmi_get_device_id(&adev->psp);
+ adev->gmc.xgmi.node_id = psp_xgmi_get_node_id(&adev->psp);
adev->gmc.xgmi.hive_id = psp_xgmi_get_hive_id(&adev->psp);
- memset(&tmp_topology[0], 0, sizeof(tmp_topology));
+ tmp_topology = kzalloc(sizeof(struct psp_xgmi_topology_info), GFP_KERNEL);
+ if (!tmp_topology)
+ return -ENOMEM;
mutex_lock(&xgmi_mutex);
hive = amdgpu_get_xgmi_hive(adev);
if (!hive)
@@ -84,23 +86,28 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
list_add_tail(&adev->gmc.xgmi.head, &hive->device_list);
list_for_each_entry(entry, &hive->device_list, head)
- tmp_topology[count++].device_id = entry->device_id;
+ tmp_topology->nodes[count++].node_id = entry->node_id;
- ret = psp_xgmi_get_topology_info(&adev->psp, count, tmp_topology);
- if (ret) {
- dev_err(adev->dev,
- "XGMI: Get topology failure on device %llx, hive %llx, ret %d",
- adev->gmc.xgmi.device_id,
- adev->gmc.xgmi.hive_id, ret);
- goto exit;
+ /* Each psp need to get the latest topology */
+ list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
+ ret = psp_xgmi_get_topology_info(&tmp_adev->psp, count, tmp_topology);
+ if (ret) {
+ dev_err(tmp_adev->dev,
+ "XGMI: Get topology failure on device %llx, hive %llx, ret %d",
+ tmp_adev->gmc.xgmi.node_id,
+ tmp_adev->gmc.xgmi.hive_id, ret);
+ /* To do : continue with some node failed or disable the whole hive */
+ break;
+ }
}
+
/* Each psp need to set the latest topology */
list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) {
ret = psp_xgmi_set_topology_info(&tmp_adev->psp, count, tmp_topology);
if (ret) {
dev_err(tmp_adev->dev,
"XGMI: Set topology failure on device %llx, hive %llx, ret %d",
- tmp_adev->gmc.xgmi.device_id,
+ tmp_adev->gmc.xgmi.node_id,
tmp_adev->gmc.xgmi.hive_id, ret);
/* To do : continue with some node failed or disable the whole hive */
break;
@@ -113,7 +120,6 @@ int amdgpu_xgmi_add_device(struct amdgpu_device *adev)
exit:
mutex_unlock(&xgmi_mutex);
+ kfree(tmp_topology);
return ret;
}
-
-
diff --git a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
index 79220a9..86e14c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
@@ -743,19 +743,19 @@ static int ci_enable_didt(struct amdgpu_device *adev, bool enable)
if (pi->caps_sq_ramping || pi->caps_db_ramping ||
pi->caps_td_ramping || pi->caps_tcp_ramping) {
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
if (enable) {
ret = ci_program_pt_config_registers(adev, didt_config_ci);
if (ret) {
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return ret;
}
}
ci_do_enable_didt(adev, enable);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index b918c88..4579519 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -198,7 +198,7 @@ static void cik_sdma_ring_set_wptr(struct amdgpu_ring *ring)
static void cik_sdma_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
{
- struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
+ struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
int i;
for (i = 0; i < count; i++)
@@ -218,9 +218,11 @@ static void cik_sdma_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
* Schedule an IB in the DMA ring (CIK).
*/
static void cik_sdma_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
u32 extra_bits = vmid & 0xf;
/* IB packet must end on a 8 DW boundary */
@@ -316,8 +318,8 @@ static void cik_sdma_gfx_stop(struct amdgpu_device *adev)
WREG32(mmSDMA0_GFX_RB_CNTL + sdma_offsets[i], rb_cntl);
WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], 0);
}
- sdma0->ready = false;
- sdma1->ready = false;
+ sdma0->sched.ready = false;
+ sdma1->sched.ready = false;
}
/**
@@ -494,18 +496,16 @@ static int cik_sdma_gfx_resume(struct amdgpu_device *adev)
/* enable DMA IBs */
WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
- ring->ready = true;
+ ring->sched.ready = true;
}
cik_sdma_enable(adev, true);
for (i = 0; i < adev->sdma.num_instances; i++) {
ring = &adev->sdma.instance[i].ring;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
return r;
- }
if (adev->mman.buffer_funcs_ring == ring)
amdgpu_ttm_set_buffer_funcs_status(adev, true);
@@ -618,21 +618,17 @@ static int cik_sdma_ring_test_ring(struct amdgpu_ring *ring)
u64 gpu_addr;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
r = amdgpu_ring_alloc(ring, 5);
- if (r) {
- DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
- amdgpu_device_wb_free(adev, index);
- return r;
- }
+ if (r)
+ goto error_free_wb;
+
amdgpu_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_WRITE, SDMA_WRITE_SUB_OPCODE_LINEAR, 0));
amdgpu_ring_write(ring, lower_32_bits(gpu_addr));
amdgpu_ring_write(ring, upper_32_bits(gpu_addr));
@@ -647,15 +643,11 @@ static int cik_sdma_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n", ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
- amdgpu_device_wb_free(adev, index);
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+error_free_wb:
+ amdgpu_device_wb_free(adev, index);
return r;
}
@@ -678,20 +670,16 @@ static int cik_sdma_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 256, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err0;
- }
ib.ptr[0] = SDMA_PACKET(SDMA_OPCODE_WRITE,
SDMA_WRITE_SUB_OPCODE_LINEAR, 0);
@@ -706,21 +694,16 @@ static int cik_sdma_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out\n");
r = -ETIMEDOUT;
goto err1;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto err1;
}
tmp = le32_to_cpu(adev->wb.wb[index]);
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ if (tmp == 0xDEADBEEF)
r = 0;
- } else {
- DRM_ERROR("amdgpu: ib test failed (0x%08X)\n", tmp);
+ else
r = -EINVAL;
- }
err1:
amdgpu_ib_free(adev, &ib, NULL);
@@ -822,7 +805,7 @@ static void cik_sdma_vm_set_pte_pde(struct amdgpu_ib *ib, uint64_t pe,
*/
static void cik_sdma_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib)
{
- struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
+ struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
u32 pad_count;
int i;
@@ -1214,8 +1197,11 @@ static int cik_sdma_process_illegal_inst_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
+ u8 instance_id;
+
DRM_ERROR("Illegal instruction in SDMA command stream\n");
- schedule_work(&adev->reset_work);
+ instance_id = (entry->ring_id & 0x3) >> 0;
+ drm_sched_fault(&adev->sdma.instance[instance_id].ring.sched);
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
index d76eb27..1dc3013 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v6_0.c
@@ -1775,18 +1775,15 @@ static int gfx_v6_0_ring_test_ring(struct amdgpu_ring *ring)
int r;
r = amdgpu_gfx_scratch_get(adev, &scratch);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to get scratch reg (%d).\n", r);
+ if (r)
return r;
- }
+
WREG32(scratch, 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n", ring->idx, r);
- amdgpu_gfx_scratch_free(adev, scratch);
- return r;
- }
+ if (r)
+ goto error_free_scratch;
+
amdgpu_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1));
amdgpu_ring_write(ring, (scratch - PACKET3_SET_CONFIG_REG_START));
amdgpu_ring_write(ring, 0xDEADBEEF);
@@ -1798,13 +1795,11 @@ static int gfx_v6_0_ring_test_ring(struct amdgpu_ring *ring)
break;
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n", ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (scratch(0x%04X)=0x%08X)\n",
- ring->idx, scratch, tmp);
- r = -EINVAL;
- }
+
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
+error_free_scratch:
amdgpu_gfx_scratch_free(adev, scratch);
return r;
}
@@ -1845,9 +1840,11 @@ static void gfx_v6_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr,
}
static void gfx_v6_0_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
u32 header, control = 0;
/* insert SWITCH_BUFFER packet before first IB in the ring frame */
@@ -1892,17 +1889,15 @@ static int gfx_v6_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_gfx_scratch_get(adev, &scratch);
- if (r) {
- DRM_ERROR("amdgpu: failed to get scratch reg (%ld).\n", r);
+ if (r)
return r;
- }
+
WREG32(scratch, 0xCAFEDEAD);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 256, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err1;
- }
+
ib.ptr[0] = PACKET3(PACKET3_SET_CONFIG_REG, 1);
ib.ptr[1] = ((scratch - PACKET3_SET_CONFIG_REG_START));
ib.ptr[2] = 0xDEADBEEF;
@@ -1914,22 +1909,16 @@ static int gfx_v6_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out\n");
r = -ETIMEDOUT;
goto err2;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto err2;
}
tmp = RREG32(scratch);
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ if (tmp == 0xDEADBEEF)
r = 0;
- } else {
- DRM_ERROR("amdgpu: ib test failed (scratch(0x%04X)=0x%08X)\n",
- scratch, tmp);
+ else
r = -EINVAL;
- }
err2:
amdgpu_ib_free(adev, &ib, NULL);
@@ -1950,9 +1939,9 @@ static void gfx_v6_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable)
CP_ME_CNTL__CE_HALT_MASK));
WREG32(mmSCRATCH_UMSK, 0);
for (i = 0; i < adev->gfx.num_gfx_rings; i++)
- adev->gfx.gfx_ring[i].ready = false;
+ adev->gfx.gfx_ring[i].sched.ready = false;
for (i = 0; i < adev->gfx.num_compute_rings; i++)
- adev->gfx.compute_ring[i].ready = false;
+ adev->gfx.compute_ring[i].sched.ready = false;
}
udelay(50);
}
@@ -2124,12 +2113,9 @@ static int gfx_v6_0_cp_gfx_resume(struct amdgpu_device *adev)
/* start the rings */
gfx_v6_0_cp_gfx_start(adev);
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
return r;
- }
return 0;
}
@@ -2227,14 +2213,11 @@ static int gfx_v6_0_cp_compute_resume(struct amdgpu_device *adev)
WREG32(mmCP_RB2_CNTL, tmp);
WREG32(mmCP_RB2_BASE, ring->gpu_addr >> 8);
- adev->gfx.compute_ring[0].ready = false;
- adev->gfx.compute_ring[1].ready = false;
for (i = 0; i < 2; i++) {
- r = amdgpu_ring_test_ring(&adev->gfx.compute_ring[i]);
+ r = amdgpu_ring_test_helper(&adev->gfx.compute_ring[i]);
if (r)
return r;
- adev->gfx.compute_ring[i].ready = true;
}
return 0;
@@ -2368,18 +2351,11 @@ static void gfx_v6_0_ring_emit_wreg(struct amdgpu_ring *ring,
amdgpu_ring_write(ring, val);
}
-static void gfx_v6_0_rlc_fini(struct amdgpu_device *adev)
-{
- amdgpu_bo_free_kernel(&adev->gfx.rlc.save_restore_obj, NULL, NULL);
- amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, NULL, NULL);
- amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, NULL, NULL);
-}
-
static int gfx_v6_0_rlc_init(struct amdgpu_device *adev)
{
const u32 *src_ptr;
volatile u32 *dst_ptr;
- u32 dws, i;
+ u32 dws;
u64 reg_list_mc_addr;
const struct cs_section_def *cs_data;
int r;
@@ -2394,26 +2370,10 @@ static int gfx_v6_0_rlc_init(struct amdgpu_device *adev)
cs_data = adev->gfx.rlc.cs_data;
if (src_ptr) {
- /* save restore block */
- r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_VRAM,
- &adev->gfx.rlc.save_restore_obj,
- &adev->gfx.rlc.save_restore_gpu_addr,
- (void **)&adev->gfx.rlc.sr_ptr);
- if (r) {
- dev_warn(adev->dev, "(%d) create RLC sr bo failed\n",
- r);
- gfx_v6_0_rlc_fini(adev);
+ /* init save restore block */
+ r = amdgpu_gfx_rlc_init_sr(adev, dws);
+ if (r)
return r;
- }
-
- /* write the sr buffer */
- dst_ptr = adev->gfx.rlc.sr_ptr;
- for (i = 0; i < adev->gfx.rlc.reg_list_size; i++)
- dst_ptr[i] = cpu_to_le32(src_ptr[i]);
-
- amdgpu_bo_kunmap(adev->gfx.rlc.save_restore_obj);
- amdgpu_bo_unreserve(adev->gfx.rlc.save_restore_obj);
}
if (cs_data) {
@@ -2428,7 +2388,7 @@ static int gfx_v6_0_rlc_init(struct amdgpu_device *adev)
(void **)&adev->gfx.rlc.cs_ptr);
if (r) {
dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
- gfx_v6_0_rlc_fini(adev);
+ amdgpu_gfx_rlc_fini(adev);
return r;
}
@@ -2549,8 +2509,8 @@ static int gfx_v6_0_rlc_resume(struct amdgpu_device *adev)
if (!adev->gfx.rlc_fw)
return -EINVAL;
- gfx_v6_0_rlc_stop(adev);
- gfx_v6_0_rlc_reset(adev);
+ adev->gfx.rlc.funcs->stop(adev);
+ adev->gfx.rlc.funcs->reset(adev);
gfx_v6_0_init_pg(adev);
gfx_v6_0_init_cg(adev);
@@ -2578,7 +2538,7 @@ static int gfx_v6_0_rlc_resume(struct amdgpu_device *adev)
WREG32(mmRLC_UCODE_ADDR, 0);
gfx_v6_0_enable_lbpw(adev, gfx_v6_0_lbpw_supported(adev));
- gfx_v6_0_rlc_start(adev);
+ adev->gfx.rlc.funcs->start(adev);
return 0;
}
@@ -3075,6 +3035,14 @@ static const struct amdgpu_gfx_funcs gfx_v6_0_gfx_funcs = {
.select_me_pipe_q = &gfx_v6_0_select_me_pipe_q
};
+static const struct amdgpu_rlc_funcs gfx_v6_0_rlc_funcs = {
+ .init = gfx_v6_0_rlc_init,
+ .resume = gfx_v6_0_rlc_resume,
+ .stop = gfx_v6_0_rlc_stop,
+ .reset = gfx_v6_0_rlc_reset,
+ .start = gfx_v6_0_rlc_start
+};
+
static int gfx_v6_0_early_init(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -3082,6 +3050,7 @@ static int gfx_v6_0_early_init(void *handle)
adev->gfx.num_gfx_rings = GFX6_NUM_GFX_RINGS;
adev->gfx.num_compute_rings = GFX6_NUM_COMPUTE_RINGS;
adev->gfx.funcs = &gfx_v6_0_gfx_funcs;
+ adev->gfx.rlc.funcs = &gfx_v6_0_rlc_funcs;
gfx_v6_0_set_ring_funcs(adev);
gfx_v6_0_set_irq_funcs(adev);
@@ -3114,7 +3083,7 @@ static int gfx_v6_0_sw_init(void *handle)
return r;
}
- r = gfx_v6_0_rlc_init(adev);
+ r = adev->gfx.rlc.funcs->init(adev);
if (r) {
DRM_ERROR("Failed to init rlc BOs!\n");
return r;
@@ -3165,7 +3134,7 @@ static int gfx_v6_0_sw_fini(void *handle)
for (i = 0; i < adev->gfx.num_compute_rings; i++)
amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
- gfx_v6_0_rlc_fini(adev);
+ amdgpu_gfx_rlc_fini(adev);
return 0;
}
@@ -3177,7 +3146,7 @@ static int gfx_v6_0_hw_init(void *handle)
gfx_v6_0_constants_init(adev);
- r = gfx_v6_0_rlc_resume(adev);
+ r = adev->gfx.rlc.funcs->resume(adev);
if (r)
return r;
@@ -3195,7 +3164,7 @@ static int gfx_v6_0_hw_fini(void *handle)
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
gfx_v6_0_cp_enable(adev, false);
- gfx_v6_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
gfx_v6_0_fini_pg(adev);
return 0;
@@ -3393,12 +3362,31 @@ static int gfx_v6_0_eop_irq(struct amdgpu_device *adev,
return 0;
}
+static void gfx_v6_0_fault(struct amdgpu_device *adev,
+ struct amdgpu_iv_entry *entry)
+{
+ struct amdgpu_ring *ring;
+
+ switch (entry->ring_id) {
+ case 0:
+ ring = &adev->gfx.gfx_ring[0];
+ break;
+ case 1:
+ case 2:
+ ring = &adev->gfx.compute_ring[entry->ring_id - 1];
+ break;
+ default:
+ return;
+ }
+ drm_sched_fault(&ring->sched);
+}
+
static int gfx_v6_0_priv_reg_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
DRM_ERROR("Illegal register access in command stream\n");
- schedule_work(&adev->reset_work);
+ gfx_v6_0_fault(adev, entry);
return 0;
}
@@ -3407,7 +3395,7 @@ static int gfx_v6_0_priv_inst_irq(struct amdgpu_device *adev,
struct amdgpu_iv_entry *entry)
{
DRM_ERROR("Illegal instruction in command stream\n");
- schedule_work(&adev->reset_work);
+ gfx_v6_0_fault(adev, entry);
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 0e72bc0..f467b9b 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -882,7 +882,6 @@ static const u32 kalindi_rlc_save_restore_register_list[] =
static u32 gfx_v7_0_get_csb_size(struct amdgpu_device *adev);
static void gfx_v7_0_get_csb_buffer(struct amdgpu_device *adev, volatile u32 *buffer);
-static void gfx_v7_0_init_cp_pg_table(struct amdgpu_device *adev);
static void gfx_v7_0_init_pg(struct amdgpu_device *adev);
static void gfx_v7_0_get_cu_info(struct amdgpu_device *adev);
@@ -2064,17 +2063,14 @@ static int gfx_v7_0_ring_test_ring(struct amdgpu_ring *ring)
int r;
r = amdgpu_gfx_scratch_get(adev, &scratch);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to get scratch reg (%d).\n", r);
+ if (r)
return r;
- }
+
WREG32(scratch, 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n", ring->idx, r);
- amdgpu_gfx_scratch_free(adev, scratch);
- return r;
- }
+ if (r)
+ goto error_free_scratch;
+
amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1));
amdgpu_ring_write(ring, (scratch - PACKET3_SET_UCONFIG_REG_START));
amdgpu_ring_write(ring, 0xDEADBEEF);
@@ -2086,13 +2082,10 @@ static int gfx_v7_0_ring_test_ring(struct amdgpu_ring *ring)
break;
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n", ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (scratch(0x%04X)=0x%08X)\n",
- ring->idx, scratch, tmp);
- r = -EINVAL;
- }
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
+error_free_scratch:
amdgpu_gfx_scratch_free(adev, scratch);
return r;
}
@@ -2233,9 +2226,11 @@ static void gfx_v7_0_ring_emit_fence_compute(struct amdgpu_ring *ring,
* on the gfx ring for execution by the GPU.
*/
static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
u32 header, control = 0;
/* insert SWITCH_BUFFER packet before first IB in the ring frame */
@@ -2262,9 +2257,11 @@ static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
}
static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
@@ -2316,17 +2313,15 @@ static int gfx_v7_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_gfx_scratch_get(adev, &scratch);
- if (r) {
- DRM_ERROR("amdgpu: failed to get scratch reg (%ld).\n", r);
+ if (r)
return r;
- }
+
WREG32(scratch, 0xCAFEDEAD);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 256, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err1;
- }
+
ib.ptr[0] = PACKET3(PACKET3_SET_UCONFIG_REG, 1);
ib.ptr[1] = ((scratch - PACKET3_SET_UCONFIG_REG_START));
ib.ptr[2] = 0xDEADBEEF;
@@ -2338,22 +2333,16 @@ static int gfx_v7_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out\n");
r = -ETIMEDOUT;
goto err2;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto err2;
}
tmp = RREG32(scratch);
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ if (tmp == 0xDEADBEEF)
r = 0;
- } else {
- DRM_ERROR("amdgpu: ib test failed (scratch(0x%04X)=0x%08X)\n",
- scratch, tmp);
+ else
r = -EINVAL;
- }
err2:
amdgpu_ib_free(adev, &ib, NULL);
@@ -2403,7 +2392,7 @@ static void gfx_v7_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable)
} else {
WREG32(mmCP_ME_CNTL, (CP_ME_CNTL__ME_HALT_MASK | CP_ME_CNTL__PFP_HALT_MASK | CP_ME_CNTL__CE_HALT_MASK));
for (i = 0; i < adev->gfx.num_gfx_rings; i++)
- adev->gfx.gfx_ring[i].ready = false;
+ adev->gfx.gfx_ring[i].sched.ready = false;
}
udelay(50);
}
@@ -2613,12 +2602,9 @@ static int gfx_v7_0_cp_gfx_resume(struct amdgpu_device *adev)
/* start the ring */
gfx_v7_0_cp_gfx_start(adev);
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
return r;
- }
return 0;
}
@@ -2675,7 +2661,7 @@ static void gfx_v7_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
} else {
WREG32(mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK));
for (i = 0; i < adev->gfx.num_compute_rings; i++)
- adev->gfx.compute_ring[i].ready = false;
+ adev->gfx.compute_ring[i].sched.ready = false;
}
udelay(50);
}
@@ -2781,7 +2767,7 @@ static int gfx_v7_0_mec_init(struct amdgpu_device *adev)
* GFX7_MEC_HPD_SIZE * 2;
r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_GTT,
+ AMDGPU_GEM_DOMAIN_VRAM,
&adev->gfx.mec.hpd_eop_obj,
&adev->gfx.mec.hpd_eop_gpu_addr,
(void **)&hpd);
@@ -3106,10 +3092,7 @@ static int gfx_v7_0_cp_compute_resume(struct amdgpu_device *adev)
for (i = 0; i < adev->gfx.num_compute_rings; i++) {
ring = &adev->gfx.compute_ring[i];
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r)
- ring->ready = false;
+ amdgpu_ring_test_helper(ring);
}
return 0;
@@ -3268,18 +3251,10 @@ static void gfx_v7_0_ring_emit_wreg(struct amdgpu_ring *ring,
* The RLC is a multi-purpose microengine that handles a
* variety of functions.
*/
-static void gfx_v7_0_rlc_fini(struct amdgpu_device *adev)
-{
- amdgpu_bo_free_kernel(&adev->gfx.rlc.save_restore_obj, NULL, NULL);
- amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, NULL, NULL);
- amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, NULL, NULL);
-}
-
static int gfx_v7_0_rlc_init(struct amdgpu_device *adev)
{
const u32 *src_ptr;
- volatile u32 *dst_ptr;
- u32 dws, i;
+ u32 dws;
const struct cs_section_def *cs_data;
int r;
@@ -3306,66 +3281,23 @@ static int gfx_v7_0_rlc_init(struct amdgpu_device *adev)
cs_data = adev->gfx.rlc.cs_data;
if (src_ptr) {
- /* save restore block */
- r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_VRAM,
- &adev->gfx.rlc.save_restore_obj,
- &adev->gfx.rlc.save_restore_gpu_addr,
- (void **)&adev->gfx.rlc.sr_ptr);
- if (r) {
- dev_warn(adev->dev, "(%d) create, pin or map of RLC sr bo failed\n", r);
- gfx_v7_0_rlc_fini(adev);
+ /* init save restore block */
+ r = amdgpu_gfx_rlc_init_sr(adev, dws);
+ if (r)
return r;
- }
-
- /* write the sr buffer */
- dst_ptr = adev->gfx.rlc.sr_ptr;
- for (i = 0; i < adev->gfx.rlc.reg_list_size; i++)
- dst_ptr[i] = cpu_to_le32(src_ptr[i]);
- amdgpu_bo_kunmap(adev->gfx.rlc.save_restore_obj);
- amdgpu_bo_unreserve(adev->gfx.rlc.save_restore_obj);
}
if (cs_data) {
- /* clear state block */
- adev->gfx.rlc.clear_state_size = dws = gfx_v7_0_get_csb_size(adev);
-
- r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_VRAM,
- &adev->gfx.rlc.clear_state_obj,
- &adev->gfx.rlc.clear_state_gpu_addr,
- (void **)&adev->gfx.rlc.cs_ptr);
- if (r) {
- dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
- gfx_v7_0_rlc_fini(adev);
+ /* init clear state block */
+ r = amdgpu_gfx_rlc_init_csb(adev);
+ if (r)
return r;
- }
-
- /* set up the cs buffer */
- dst_ptr = adev->gfx.rlc.cs_ptr;
- gfx_v7_0_get_csb_buffer(adev, dst_ptr);
- amdgpu_bo_kunmap(adev->gfx.rlc.clear_state_obj);
- amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
}
if (adev->gfx.rlc.cp_table_size) {
-
- r = amdgpu_bo_create_reserved(adev, adev->gfx.rlc.cp_table_size,
- PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
- &adev->gfx.rlc.cp_table_obj,
- &adev->gfx.rlc.cp_table_gpu_addr,
- (void **)&adev->gfx.rlc.cp_table_ptr);
- if (r) {
- dev_warn(adev->dev, "(%d) create RLC cp table bo failed\n", r);
- gfx_v7_0_rlc_fini(adev);
+ r = amdgpu_gfx_rlc_init_cpt(adev);
+ if (r)
return r;
- }
-
- gfx_v7_0_init_cp_pg_table(adev);
-
- amdgpu_bo_kunmap(adev->gfx.rlc.cp_table_obj);
- amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
-
}
return 0;
@@ -3446,7 +3378,12 @@ static u32 gfx_v7_0_halt_rlc(struct amdgpu_device *adev)
return orig;
}
-static void gfx_v7_0_enter_rlc_safe_mode(struct amdgpu_device *adev)
+static bool gfx_v7_0_is_rlc_enabled(struct amdgpu_device *adev)
+{
+ return true;
+}
+
+static void gfx_v7_0_set_safe_mode(struct amdgpu_device *adev)
{
u32 tmp, i, mask;
@@ -3468,7 +3405,7 @@ static void gfx_v7_0_enter_rlc_safe_mode(struct amdgpu_device *adev)
}
}
-static void gfx_v7_0_exit_rlc_safe_mode(struct amdgpu_device *adev)
+static void gfx_v7_0_unset_safe_mode(struct amdgpu_device *adev)
{
u32 tmp;
@@ -3545,13 +3482,13 @@ static int gfx_v7_0_rlc_resume(struct amdgpu_device *adev)
adev->gfx.rlc_feature_version = le32_to_cpu(
hdr->ucode_feature_version);
- gfx_v7_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
/* disable CG */
tmp = RREG32(mmRLC_CGCG_CGLS_CTRL) & 0xfffffffc;
WREG32(mmRLC_CGCG_CGLS_CTRL, tmp);
- gfx_v7_0_rlc_reset(adev);
+ adev->gfx.rlc.funcs->reset(adev);
gfx_v7_0_init_pg(adev);
@@ -3582,7 +3519,7 @@ static int gfx_v7_0_rlc_resume(struct amdgpu_device *adev)
if (adev->asic_type == CHIP_BONAIRE)
WREG32(mmRLC_DRIVER_CPDMA_STATUS, 0);
- gfx_v7_0_rlc_start(adev);
+ adev->gfx.rlc.funcs->start(adev);
return 0;
}
@@ -3784,72 +3721,12 @@ static void gfx_v7_0_enable_gds_pg(struct amdgpu_device *adev, bool enable)
WREG32(mmRLC_PG_CNTL, data);
}
-static void gfx_v7_0_init_cp_pg_table(struct amdgpu_device *adev)
+static int gfx_v7_0_cp_pg_table_num(struct amdgpu_device *adev)
{
- const __le32 *fw_data;
- volatile u32 *dst_ptr;
- int me, i, max_me = 4;
- u32 bo_offset = 0;
- u32 table_offset, table_size;
-
if (adev->asic_type == CHIP_KAVERI)
- max_me = 5;
-
- if (adev->gfx.rlc.cp_table_ptr == NULL)
- return;
-
- /* write the cp table buffer */
- dst_ptr = adev->gfx.rlc.cp_table_ptr;
- for (me = 0; me < max_me; me++) {
- if (me == 0) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.ce_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.ce_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 1) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.pfp_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.pfp_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 2) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.me_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.me_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 3) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.mec_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec2_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.mec2_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- }
-
- for (i = 0; i < table_size; i ++) {
- dst_ptr[bo_offset + i] =
- cpu_to_le32(le32_to_cpu(fw_data[table_offset + i]));
- }
-
- bo_offset += table_size;
- }
+ return 5;
+ else
+ return 4;
}
static void gfx_v7_0_enable_gfx_cgpg(struct amdgpu_device *adev,
@@ -4288,8 +4165,17 @@ static const struct amdgpu_gfx_funcs gfx_v7_0_gfx_funcs = {
};
static const struct amdgpu_rlc_funcs gfx_v7_0_rlc_funcs = {
- .enter_safe_mode = gfx_v7_0_enter_rlc_safe_mode,
- .exit_safe_mode = gfx_v7_0_exit_rlc_safe_mode
+ .is_rlc_enabled = gfx_v7_0_is_rlc_enabled,
+ .set_safe_mode = gfx_v7_0_set_safe_mode,
+ .unset_safe_mode = gfx_v7_0_unset_safe_mode,
+ .init = gfx_v7_0_rlc_init,
+ .get_csb_size = gfx_v7_0_get_csb_size,
+ .get_csb_buffer = gfx_v7_0_get_csb_buffer,
+ .get_cp_table_num = gfx_v7_0_cp_pg_table_num,
+ .resume = gfx_v7_0_rlc_resume,
+ .stop = gfx_v7_0_rlc_stop,
+ .reset = gfx_v7_0_rlc_reset,
+ .start = gfx_v7_0_rlc_start
};
static int gfx_v7_0_early_init(void *handle)
@@ -4540,7 +4426,7 @@ static int gfx_v7_0_sw_init(void *handle)
return r;
}
- r = gfx_v7_0_rlc_init(adev);
+ r = adev->gfx.rlc.funcs->init(adev);
if (r) {
DRM_ERROR("Failed to init rlc BOs!\n");
return r;
@@ -4604,7 +4490,7 @@ static int gfx_v7_0_sw_fini(void *handle)
amdgpu_ring_fini(&adev->gfx.compute_ring[i]);
gfx_v7_0_cp_compute_fini(adev);
- gfx_v7_0_rlc_fini(adev);
+ amdgpu_gfx_rlc_fini(adev);
gfx_v7_0_mec_fini(adev);
amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj,
&adev->gfx.rlc.clear_state_gpu_addr,
@@ -4627,7 +4513,7 @@ static int gfx_v7_0_hw_init(void *handle)
gfx_v7_0_constants_init(adev);
/* init rlc */
- r = gfx_v7_0_rlc_resume(adev);
+ r = adev->gfx.rlc.funcs->resume(adev);
if (r)
return r;
@@ -4645,7 +4531,7 @@ static int gfx_v7_0_hw_fini(void *handle)
amdgpu_irq_put(adev, &adev->gfx.priv_reg_irq, 0);
amdgpu_irq_put(adev, &adev->gfx.priv_inst_irq, 0);
gfx_v7_0_cp_enable(adev, false);
- gfx_v7_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
gfx_v7_0_fini_pg(adev);
return 0;
@@ -4730,7 +4616,7 @@ static int gfx_v7_0_soft_reset(void *handle)
gfx_v7_0_update_cg(adev, false);
/* stop the rlc */
- gfx_v7_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
/* Disable GFX parsing/prefetching */
WREG32(mmCP_ME_CNTL, CP_ME_CNTL__ME_HALT_MASK | CP_ME_CNTL__PFP_HALT_MASK | CP_ME_CNTL__CE_HALT_MASK);
@@ -4959,12 +4845,36 @@ static int gfx_v7_0_eop_irq(struct amdgpu_device *adev,
return 0;
}
+static void gfx_v7_0_fault(struct amdgpu_device *adev,
+ struct amdgpu_iv_entry *entry)
+{
+ struct amdgpu_ring *ring;
+ u8 me_id, pipe_id;
+ int i;
+
+ me_id = (entry->ring_id & 0x0c) >> 2;
+ pipe_id = (entry->ring_id & 0x03) >> 0;
+ switch (me_id) {
+ case 0:
+ drm_sched_fault(&adev->gfx.gfx_ring[0].sched);
+ break;
+ case 1:
+ case 2:
+ for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+ ring = &adev->gfx.compute_ring[i];
+ if ((ring->me == me_id) && (ring->pipe == pipe_id))
+ drm_sched_fault(&ring->sched);
+ }
+ break;
+ }
+}
+
static int gfx_v7_0_priv_reg_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
DRM_ERROR("Illegal register access in command stream\n");
- schedule_work(&adev->reset_work);
+ gfx_v7_0_fault(adev, entry);
return 0;
}
@@ -4974,7 +4884,7 @@ static int gfx_v7_0_priv_inst_irq(struct amdgpu_device *adev,
{
DRM_ERROR("Illegal instruction in command stream\n");
// XXX soft reset the gfx block only
- schedule_work(&adev->reset_work);
+ gfx_v7_0_fault(adev, entry);
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 617b0c8..cb066a8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -54,7 +54,7 @@
#include "ivsrcid/ivsrcid_vislands30.h"
#define GFX8_NUM_GFX_RINGS 1
-#define GFX8_MEC_HPD_SIZE 2048
+#define GFX8_MEC_HPD_SIZE 4096
#define TOPAZ_GB_ADDR_CONFIG_GOLDEN 0x22010001
#define CARRIZO_GB_ADDR_CONFIG_GOLDEN 0x22010001
@@ -839,18 +839,14 @@ static int gfx_v8_0_ring_test_ring(struct amdgpu_ring *ring)
int r;
r = amdgpu_gfx_scratch_get(adev, &scratch);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to get scratch reg (%d).\n", r);
+ if (r)
return r;
- }
+
WREG32(scratch, 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n",
- ring->idx, r);
- amdgpu_gfx_scratch_free(adev, scratch);
- return r;
- }
+ if (r)
+ goto error_free_scratch;
+
amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1));
amdgpu_ring_write(ring, (scratch - PACKET3_SET_UCONFIG_REG_START));
amdgpu_ring_write(ring, 0xDEADBEEF);
@@ -862,14 +858,11 @@ static int gfx_v8_0_ring_test_ring(struct amdgpu_ring *ring)
break;
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (scratch(0x%04X)=0x%08X)\n",
- ring->idx, scratch, tmp);
- r = -EINVAL;
- }
+
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
+error_free_scratch:
amdgpu_gfx_scratch_free(adev, scratch);
return r;
}
@@ -886,19 +879,16 @@ static int gfx_v8_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 16, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err1;
- }
+
ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3);
ib.ptr[1] = WRITE_DATA_DST_SEL(5) | WR_CONFIRM;
ib.ptr[2] = lower_32_bits(gpu_addr);
@@ -912,22 +902,17 @@ static int gfx_v8_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out.\n");
r = -ETIMEDOUT;
goto err2;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto err2;
}
tmp = adev->wb.wb[index];
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ if (tmp == 0xDEADBEEF)
r = 0;
- } else {
- DRM_ERROR("ib test on ring %d failed\n", ring->idx);
+ else
r = -EINVAL;
- }
err2:
amdgpu_ib_free(adev, &ib, NULL);
@@ -1298,81 +1283,16 @@ static void gfx_v8_0_get_csb_buffer(struct amdgpu_device *adev,
buffer[count++] = cpu_to_le32(0);
}
-static void cz_init_cp_jump_table(struct amdgpu_device *adev)
+static int gfx_v8_0_cp_jump_table_num(struct amdgpu_device *adev)
{
- const __le32 *fw_data;
- volatile u32 *dst_ptr;
- int me, i, max_me = 4;
- u32 bo_offset = 0;
- u32 table_offset, table_size;
-
if (adev->asic_type == CHIP_CARRIZO)
- max_me = 5;
-
- /* write the cp table buffer */
- dst_ptr = adev->gfx.rlc.cp_table_ptr;
- for (me = 0; me < max_me; me++) {
- if (me == 0) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.ce_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.ce_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 1) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.pfp_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.pfp_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 2) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.me_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.me_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 3) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.mec_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 4) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec2_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.mec2_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- }
-
- for (i = 0; i < table_size; i ++) {
- dst_ptr[bo_offset + i] =
- cpu_to_le32(le32_to_cpu(fw_data[table_offset + i]));
- }
-
- bo_offset += table_size;
- }
-}
-
-static void gfx_v8_0_rlc_fini(struct amdgpu_device *adev)
-{
- amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj, NULL, NULL);
- amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj, NULL, NULL);
+ return 5;
+ else
+ return 4;
}
static int gfx_v8_0_rlc_init(struct amdgpu_device *adev)
{
- volatile u32 *dst_ptr;
- u32 dws;
const struct cs_section_def *cs_data;
int r;
@@ -1381,44 +1301,18 @@ static int gfx_v8_0_rlc_init(struct amdgpu_device *adev)
cs_data = adev->gfx.rlc.cs_data;
if (cs_data) {
- /* clear state block */
- adev->gfx.rlc.clear_state_size = dws = gfx_v8_0_get_csb_size(adev);
-
- r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_VRAM,
- &adev->gfx.rlc.clear_state_obj,
- &adev->gfx.rlc.clear_state_gpu_addr,
- (void **)&adev->gfx.rlc.cs_ptr);
- if (r) {
- dev_warn(adev->dev, "(%d) create RLC c bo failed\n", r);
- gfx_v8_0_rlc_fini(adev);
+ /* init clear state block */
+ r = amdgpu_gfx_rlc_init_csb(adev);
+ if (r)
return r;
- }
-
- /* set up the cs buffer */
- dst_ptr = adev->gfx.rlc.cs_ptr;
- gfx_v8_0_get_csb_buffer(adev, dst_ptr);
- amdgpu_bo_kunmap(adev->gfx.rlc.clear_state_obj);
- amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
}
if ((adev->asic_type == CHIP_CARRIZO) ||
(adev->asic_type == CHIP_STONEY)) {
adev->gfx.rlc.cp_table_size = ALIGN(96 * 5 * 4, 2048) + (64 * 1024); /* JT + GDS */
- r = amdgpu_bo_create_reserved(adev, adev->gfx.rlc.cp_table_size,
- PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
- &adev->gfx.rlc.cp_table_obj,
- &adev->gfx.rlc.cp_table_gpu_addr,
- (void **)&adev->gfx.rlc.cp_table_ptr);
- if (r) {
- dev_warn(adev->dev, "(%d) create RLC cp table bo failed\n", r);
+ r = amdgpu_gfx_rlc_init_cpt(adev);
+ if (r)
return r;
- }
-
- cz_init_cp_jump_table(adev);
-
- amdgpu_bo_kunmap(adev->gfx.rlc.cp_table_obj);
- amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
}
return 0;
@@ -1443,7 +1337,7 @@ static int gfx_v8_0_mec_init(struct amdgpu_device *adev)
mec_hpd_size = adev->gfx.num_compute_rings * GFX8_MEC_HPD_SIZE;
r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_GTT,
+ AMDGPU_GEM_DOMAIN_VRAM,
&adev->gfx.mec.hpd_eop_obj,
&adev->gfx.mec.hpd_eop_gpu_addr,
(void **)&hpd);
@@ -1629,7 +1523,7 @@ static int gfx_v8_0_do_edc_gpr_workarounds(struct amdgpu_device *adev)
return 0;
/* bail if the compute ring is not ready */
- if (!ring->ready)
+ if (!ring->sched.ready)
return 0;
tmp = RREG32(mmGB_EDC_MODE);
@@ -2088,7 +1982,7 @@ static int gfx_v8_0_sw_init(void *handle)
return r;
}
- r = gfx_v8_0_rlc_init(adev);
+ r = adev->gfx.rlc.funcs->init(adev);
if (r) {
DRM_ERROR("Failed to init rlc BOs!\n");
return r;
@@ -2181,7 +2075,7 @@ static int gfx_v8_0_sw_fini(void *handle)
amdgpu_gfx_kiq_fini(adev);
gfx_v8_0_mec_fini(adev);
- gfx_v8_0_rlc_fini(adev);
+ amdgpu_gfx_rlc_fini(adev);
amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj,
&adev->gfx.rlc.clear_state_gpu_addr,
(void **)&adev->gfx.rlc.cs_ptr);
@@ -4175,10 +4069,10 @@ static void gfx_v8_0_rlc_start(struct amdgpu_device *adev)
static int gfx_v8_0_rlc_resume(struct amdgpu_device *adev)
{
- gfx_v8_0_rlc_stop(adev);
- gfx_v8_0_rlc_reset(adev);
+ adev->gfx.rlc.funcs->stop(adev);
+ adev->gfx.rlc.funcs->reset(adev);
gfx_v8_0_init_pg(adev);
- gfx_v8_0_rlc_start(adev);
+ adev->gfx.rlc.funcs->start(adev);
return 0;
}
@@ -4197,7 +4091,7 @@ static void gfx_v8_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable)
tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, PFP_HALT, 1);
tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_HALT, 1);
for (i = 0; i < adev->gfx.num_gfx_rings; i++)
- adev->gfx.gfx_ring[i].ready = false;
+ adev->gfx.gfx_ring[i].sched.ready = false;
}
WREG32(mmCP_ME_CNTL, tmp);
udelay(50);
@@ -4379,10 +4273,8 @@ static int gfx_v8_0_cp_gfx_resume(struct amdgpu_device *adev)
/* start the ring */
amdgpu_ring_clear_ring(ring);
gfx_v8_0_cp_gfx_start(adev);
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r)
- ring->ready = false;
+ ring->sched.ready = true;
+ r = amdgpu_ring_test_helper(ring);
return r;
}
@@ -4396,8 +4288,8 @@ static void gfx_v8_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
} else {
WREG32(mmCP_MEC_CNTL, (CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK));
for (i = 0; i < adev->gfx.num_compute_rings; i++)
- adev->gfx.compute_ring[i].ready = false;
- adev->gfx.kiq.ring.ready = false;
+ adev->gfx.compute_ring[i].sched.ready = false;
+ adev->gfx.kiq.ring.sched.ready = false;
}
udelay(50);
}
@@ -4473,11 +4365,9 @@ static int gfx_v8_0_kiq_kcq_enable(struct amdgpu_device *adev)
amdgpu_ring_write(kiq_ring, upper_32_bits(wptr_addr));
}
- r = amdgpu_ring_test_ring(kiq_ring);
- if (r) {
+ r = amdgpu_ring_test_helper(kiq_ring);
+ if (r)
DRM_ERROR("KCQ enable failed\n");
- kiq_ring->ready = false;
- }
return r;
}
@@ -4781,7 +4671,7 @@ static int gfx_v8_0_kiq_resume(struct amdgpu_device *adev)
amdgpu_bo_kunmap(ring->mqd_obj);
ring->mqd_ptr = NULL;
amdgpu_bo_unreserve(ring->mqd_obj);
- ring->ready = true;
+ ring->sched.ready = true;
return 0;
}
@@ -4820,10 +4710,7 @@ static int gfx_v8_0_kcq_resume(struct amdgpu_device *adev)
*/
for (i = adev->gfx.num_compute_rings - 1; i >= 0; i--) {
ring = &adev->gfx.compute_ring[i];
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r)
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
}
done:
@@ -4867,7 +4754,7 @@ static int gfx_v8_0_hw_init(void *handle)
gfx_v8_0_init_golden_registers(adev);
gfx_v8_0_constants_init(adev);
- r = gfx_v8_0_rlc_resume(adev);
+ r = adev->gfx.rlc.funcs->resume(adev);
if (r)
return r;
@@ -4899,7 +4786,7 @@ static int gfx_v8_0_kcq_disable(struct amdgpu_device *adev)
amdgpu_ring_write(kiq_ring, 0);
amdgpu_ring_write(kiq_ring, 0);
}
- r = amdgpu_ring_test_ring(kiq_ring);
+ r = amdgpu_ring_test_helper(kiq_ring);
if (r)
DRM_ERROR("KCQ disable failed\n");
@@ -4973,16 +4860,16 @@ static int gfx_v8_0_hw_fini(void *handle)
pr_debug("For SRIOV client, shouldn't do anything.\n");
return 0;
}
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
if (!gfx_v8_0_wait_for_idle(adev))
gfx_v8_0_cp_enable(adev, false);
else
pr_err("cp is busy, skip halt cp\n");
if (!gfx_v8_0_wait_for_rlc_idle(adev))
- gfx_v8_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
else
pr_err("rlc is busy, skip halt rlc\n");
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return 0;
}
@@ -5071,7 +4958,7 @@ static int gfx_v8_0_pre_soft_reset(void *handle)
srbm_soft_reset = adev->gfx.srbm_soft_reset;
/* stop the rlc */
- gfx_v8_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
if (REG_GET_FIELD(grbm_soft_reset, GRBM_SOFT_RESET, SOFT_RESET_CP) ||
REG_GET_FIELD(grbm_soft_reset, GRBM_SOFT_RESET, SOFT_RESET_GFX))
@@ -5197,7 +5084,7 @@ static int gfx_v8_0_post_soft_reset(void *handle)
REG_GET_FIELD(grbm_soft_reset, GRBM_SOFT_RESET, SOFT_RESET_GFX))
gfx_v8_0_cp_gfx_resume(adev);
- gfx_v8_0_rlc_start(adev);
+ adev->gfx.rlc.funcs->start(adev);
return 0;
}
@@ -5445,7 +5332,7 @@ static int gfx_v8_0_set_powergating_state(void *handle,
AMD_PG_SUPPORT_RLC_SMU_HS |
AMD_PG_SUPPORT_CP |
AMD_PG_SUPPORT_GFX_DMG))
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
switch (adev->asic_type) {
case CHIP_CARRIZO:
case CHIP_STONEY:
@@ -5499,7 +5386,7 @@ static int gfx_v8_0_set_powergating_state(void *handle,
AMD_PG_SUPPORT_RLC_SMU_HS |
AMD_PG_SUPPORT_CP |
AMD_PG_SUPPORT_GFX_DMG))
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return 0;
}
@@ -5593,57 +5480,53 @@ static void gfx_v8_0_send_serdes_cmd(struct amdgpu_device *adev,
#define RLC_GPR_REG2__MESSAGE__SHIFT 0x00000001
#define RLC_GPR_REG2__MESSAGE_MASK 0x0000001e
-static void iceland_enter_rlc_safe_mode(struct amdgpu_device *adev)
+static bool gfx_v8_0_is_rlc_enabled(struct amdgpu_device *adev)
{
- u32 data;
+ uint32_t rlc_setting;
+
+ rlc_setting = RREG32(mmRLC_CNTL);
+ if (!(rlc_setting & RLC_CNTL__RLC_ENABLE_F32_MASK))
+ return false;
+
+ return true;
+}
+
+static void gfx_v8_0_set_safe_mode(struct amdgpu_device *adev)
+{
+ uint32_t data;
unsigned i;
-
data = RREG32(mmRLC_CNTL);
- if (!(data & RLC_CNTL__RLC_ENABLE_F32_MASK))
- return;
+ data |= RLC_SAFE_MODE__CMD_MASK;
+ data &= ~RLC_SAFE_MODE__MESSAGE_MASK;
+ data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT);
+ WREG32(mmRLC_SAFE_MODE, data);
- if (adev->cg_flags & (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG)) {
- data |= RLC_SAFE_MODE__CMD_MASK;
- data &= ~RLC_SAFE_MODE__MESSAGE_MASK;
- data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT);
- WREG32(mmRLC_SAFE_MODE, data);
-
- for (i = 0; i < adev->usec_timeout; i++) {
- if ((RREG32(mmRLC_GPM_STAT) &
- (RLC_GPM_STAT__GFX_CLOCK_STATUS_MASK |
- RLC_GPM_STAT__GFX_POWER_STATUS_MASK)) ==
- (RLC_GPM_STAT__GFX_CLOCK_STATUS_MASK |
- RLC_GPM_STAT__GFX_POWER_STATUS_MASK))
- break;
- udelay(1);
- }
-
- for (i = 0; i < adev->usec_timeout; i++) {
- if (!REG_GET_FIELD(RREG32(mmRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
- break;
- udelay(1);
- }
- adev->gfx.rlc.in_safe_mode = true;
+ /* wait for RLC_SAFE_MODE */
+ for (i = 0; i < adev->usec_timeout; i++) {
+ if ((RREG32(mmRLC_GPM_STAT) &
+ (RLC_GPM_STAT__GFX_CLOCK_STATUS_MASK |
+ RLC_GPM_STAT__GFX_POWER_STATUS_MASK)) ==
+ (RLC_GPM_STAT__GFX_CLOCK_STATUS_MASK |
+ RLC_GPM_STAT__GFX_POWER_STATUS_MASK))
+ break;
+ udelay(1);
+ }
+ for (i = 0; i < adev->usec_timeout; i++) {
+ if (!REG_GET_FIELD(RREG32(mmRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
+ break;
+ udelay(1);
}
}
-static void iceland_exit_rlc_safe_mode(struct amdgpu_device *adev)
+static void gfx_v8_0_unset_safe_mode(struct amdgpu_device *adev)
{
- u32 data = 0;
+ uint32_t data;
unsigned i;
data = RREG32(mmRLC_CNTL);
- if (!(data & RLC_CNTL__RLC_ENABLE_F32_MASK))
- return;
-
- if (adev->cg_flags & (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG)) {
- if (adev->gfx.rlc.in_safe_mode) {
- data |= RLC_SAFE_MODE__CMD_MASK;
- data &= ~RLC_SAFE_MODE__MESSAGE_MASK;
- WREG32(mmRLC_SAFE_MODE, data);
- adev->gfx.rlc.in_safe_mode = false;
- }
- }
+ data |= RLC_SAFE_MODE__CMD_MASK;
+ data &= ~RLC_SAFE_MODE__MESSAGE_MASK;
+ WREG32(mmRLC_SAFE_MODE, data);
for (i = 0; i < adev->usec_timeout; i++) {
if (!REG_GET_FIELD(RREG32(mmRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
@@ -5653,8 +5536,17 @@ static void iceland_exit_rlc_safe_mode(struct amdgpu_device *adev)
}
static const struct amdgpu_rlc_funcs iceland_rlc_funcs = {
- .enter_safe_mode = iceland_enter_rlc_safe_mode,
- .exit_safe_mode = iceland_exit_rlc_safe_mode
+ .is_rlc_enabled = gfx_v8_0_is_rlc_enabled,
+ .set_safe_mode = gfx_v8_0_set_safe_mode,
+ .unset_safe_mode = gfx_v8_0_unset_safe_mode,
+ .init = gfx_v8_0_rlc_init,
+ .get_csb_size = gfx_v8_0_get_csb_size,
+ .get_csb_buffer = gfx_v8_0_get_csb_buffer,
+ .get_cp_table_num = gfx_v8_0_cp_jump_table_num,
+ .resume = gfx_v8_0_rlc_resume,
+ .stop = gfx_v8_0_rlc_stop,
+ .reset = gfx_v8_0_rlc_reset,
+ .start = gfx_v8_0_rlc_start
};
static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev,
@@ -5662,7 +5554,7 @@ static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev
{
uint32_t temp, data;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
/* It is disabled by HW by default */
if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
@@ -5758,7 +5650,7 @@ static void gfx_v8_0_update_medium_grain_clock_gating(struct amdgpu_device *adev
gfx_v8_0_wait_for_rlc_serdes(adev);
}
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
@@ -5768,7 +5660,7 @@ static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev
temp = data = RREG32(mmRLC_CGCG_CGLS_CTRL);
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
temp1 = data1 = RREG32(mmRLC_CGTT_MGCG_OVERRIDE);
@@ -5851,7 +5743,7 @@ static void gfx_v8_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev
gfx_v8_0_wait_for_rlc_serdes(adev);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
static int gfx_v8_0_update_gfx_clock_gating(struct amdgpu_device *adev,
bool enable)
@@ -6131,9 +6023,11 @@ static void gfx_v8_0_ring_emit_vgt_flush(struct amdgpu_ring *ring)
}
static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
u32 header, control = 0;
if (ib->flags & AMDGPU_IB_FLAG_CE)
@@ -6161,9 +6055,11 @@ static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
}
static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
@@ -6738,12 +6634,39 @@ static int gfx_v8_0_eop_irq(struct amdgpu_device *adev,
return 0;
}
+static void gfx_v8_0_fault(struct amdgpu_device *adev,
+ struct amdgpu_iv_entry *entry)
+{
+ u8 me_id, pipe_id, queue_id;
+ struct amdgpu_ring *ring;
+ int i;
+
+ me_id = (entry->ring_id & 0x0c) >> 2;
+ pipe_id = (entry->ring_id & 0x03) >> 0;
+ queue_id = (entry->ring_id & 0x70) >> 4;
+
+ switch (me_id) {
+ case 0:
+ drm_sched_fault(&adev->gfx.gfx_ring[0].sched);
+ break;
+ case 1:
+ case 2:
+ for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+ ring = &adev->gfx.compute_ring[i];
+ if (ring->me == me_id && ring->pipe == pipe_id &&
+ ring->queue == queue_id)
+ drm_sched_fault(&ring->sched);
+ }
+ break;
+ }
+}
+
static int gfx_v8_0_priv_reg_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
DRM_ERROR("Illegal register access in command stream\n");
- schedule_work(&adev->reset_work);
+ gfx_v8_0_fault(adev, entry);
return 0;
}
@@ -6752,7 +6675,7 @@ static int gfx_v8_0_priv_inst_irq(struct amdgpu_device *adev,
struct amdgpu_iv_entry *entry)
{
DRM_ERROR("Illegal instruction in command stream\n");
- schedule_work(&adev->reset_work);
+ gfx_v8_0_fault(adev, entry);
return 0;
}
@@ -6976,10 +6899,8 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_kiq = {
17 + /* gfx_v8_0_ring_emit_vm_flush */
7 + 7 + 7, /* gfx_v8_0_ring_emit_fence_kiq x3 for user fence, vm fence */
.emit_ib_size = 4, /* gfx_v8_0_ring_emit_ib_compute */
- .emit_ib = gfx_v8_0_ring_emit_ib_compute,
.emit_fence = gfx_v8_0_ring_emit_fence_kiq,
.test_ring = gfx_v8_0_ring_test_ring,
- .test_ib = gfx_v8_0_ring_test_ib,
.insert_nop = amdgpu_ring_insert_nop,
.pad_ib = amdgpu_ring_generic_pad_ib,
.emit_rreg = gfx_v8_0_ring_emit_rreg,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
index 6d7baf5..c27caa1 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -41,7 +41,7 @@
#include "ivsrcid/gfx/irqsrcs_gfx_9_0.h"
#define GFX9_NUM_GFX_RINGS 1
-#define GFX9_MEC_HPD_SIZE 2048
+#define GFX9_MEC_HPD_SIZE 4096
#define RLCG_UCODE_LOADING_START_ADDRESS 0x00002000L
#define RLC_SAVE_RESTORE_ADDR_STARTING_OFFSET 0x00000000L
@@ -396,18 +396,14 @@ static int gfx_v9_0_ring_test_ring(struct amdgpu_ring *ring)
int r;
r = amdgpu_gfx_scratch_get(adev, &scratch);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to get scratch reg (%d).\n", r);
+ if (r)
return r;
- }
+
WREG32(scratch, 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n",
- ring->idx, r);
- amdgpu_gfx_scratch_free(adev, scratch);
- return r;
- }
+ if (r)
+ goto error_free_scratch;
+
amdgpu_ring_write(ring, PACKET3(PACKET3_SET_UCONFIG_REG, 1));
amdgpu_ring_write(ring, (scratch - PACKET3_SET_UCONFIG_REG_START));
amdgpu_ring_write(ring, 0xDEADBEEF);
@@ -419,14 +415,11 @@ static int gfx_v9_0_ring_test_ring(struct amdgpu_ring *ring)
break;
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (scratch(0x%04X)=0x%08X)\n",
- ring->idx, scratch, tmp);
- r = -EINVAL;
- }
+
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
+error_free_scratch:
amdgpu_gfx_scratch_free(adev, scratch);
return r;
}
@@ -443,19 +436,16 @@ static int gfx_v9_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
adev->wb.wb[index] = cpu_to_le32(0xCAFEDEAD);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 16, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err1;
- }
+
ib.ptr[0] = PACKET3(PACKET3_WRITE_DATA, 3);
ib.ptr[1] = WRITE_DATA_DST_SEL(5) | WR_CONFIRM;
ib.ptr[2] = lower_32_bits(gpu_addr);
@@ -469,22 +459,17 @@ static int gfx_v9_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out.\n");
- r = -ETIMEDOUT;
- goto err2;
+ r = -ETIMEDOUT;
+ goto err2;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
- goto err2;
+ goto err2;
}
tmp = adev->wb.wb[index];
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
- r = 0;
- } else {
- DRM_ERROR("ib test on ring %d failed\n", ring->idx);
- r = -EINVAL;
- }
+ if (tmp == 0xDEADBEEF)
+ r = 0;
+ else
+ r = -EINVAL;
err2:
amdgpu_ib_free(adev, &ib, NULL);
@@ -1065,85 +1050,13 @@ static void gfx_v9_0_enable_lbpw(struct amdgpu_device *adev, bool enable)
WREG32_FIELD15(GC, 0, RLC_LB_CNTL, LOAD_BALANCE_ENABLE, enable ? 1 : 0);
}
-static void rv_init_cp_jump_table(struct amdgpu_device *adev)
+static int gfx_v9_0_cp_jump_table_num(struct amdgpu_device *adev)
{
- const __le32 *fw_data;
- volatile u32 *dst_ptr;
- int me, i, max_me = 5;
- u32 bo_offset = 0;
- u32 table_offset, table_size;
-
- /* write the cp table buffer */
- dst_ptr = adev->gfx.rlc.cp_table_ptr;
- for (me = 0; me < max_me; me++) {
- if (me == 0) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.ce_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.ce_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 1) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.pfp_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.pfp_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 2) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.me_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.me_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 3) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.mec_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- } else if (me == 4) {
- const struct gfx_firmware_header_v1_0 *hdr =
- (const struct gfx_firmware_header_v1_0 *)adev->gfx.mec2_fw->data;
- fw_data = (const __le32 *)
- (adev->gfx.mec2_fw->data +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- table_offset = le32_to_cpu(hdr->jt_offset);
- table_size = le32_to_cpu(hdr->jt_size);
- }
-
- for (i = 0; i < table_size; i ++) {
- dst_ptr[bo_offset + i] =
- cpu_to_le32(le32_to_cpu(fw_data[table_offset + i]));
- }
-
- bo_offset += table_size;
- }
-}
-
-static void gfx_v9_0_rlc_fini(struct amdgpu_device *adev)
-{
- /* clear state block */
- amdgpu_bo_free_kernel(&adev->gfx.rlc.clear_state_obj,
- &adev->gfx.rlc.clear_state_gpu_addr,
- (void **)&adev->gfx.rlc.cs_ptr);
-
- /* jump table block */
- amdgpu_bo_free_kernel(&adev->gfx.rlc.cp_table_obj,
- &adev->gfx.rlc.cp_table_gpu_addr,
- (void **)&adev->gfx.rlc.cp_table_ptr);
+ return 5;
}
static int gfx_v9_0_rlc_init(struct amdgpu_device *adev)
{
- volatile u32 *dst_ptr;
- u32 dws;
const struct cs_section_def *cs_data;
int r;
@@ -1152,45 +1065,18 @@ static int gfx_v9_0_rlc_init(struct amdgpu_device *adev)
cs_data = adev->gfx.rlc.cs_data;
if (cs_data) {
- /* clear state block */
- adev->gfx.rlc.clear_state_size = dws = gfx_v9_0_get_csb_size(adev);
- r = amdgpu_bo_create_reserved(adev, dws * 4, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_VRAM,
- &adev->gfx.rlc.clear_state_obj,
- &adev->gfx.rlc.clear_state_gpu_addr,
- (void **)&adev->gfx.rlc.cs_ptr);
- if (r) {
- dev_err(adev->dev, "(%d) failed to create rlc csb bo\n",
- r);
- gfx_v9_0_rlc_fini(adev);
+ /* init clear state block */
+ r = amdgpu_gfx_rlc_init_csb(adev);
+ if (r)
return r;
- }
- /* set up the cs buffer */
- dst_ptr = adev->gfx.rlc.cs_ptr;
- gfx_v9_0_get_csb_buffer(adev, dst_ptr);
- amdgpu_bo_kunmap(adev->gfx.rlc.clear_state_obj);
- amdgpu_bo_unpin(adev->gfx.rlc.clear_state_obj);
- amdgpu_bo_unreserve(adev->gfx.rlc.clear_state_obj);
}
if (adev->asic_type == CHIP_RAVEN) {
/* TODO: double check the cp_table_size for RV */
adev->gfx.rlc.cp_table_size = ALIGN(96 * 5 * 4, 2048) + (64 * 1024); /* JT + GDS */
- r = amdgpu_bo_create_reserved(adev, adev->gfx.rlc.cp_table_size,
- PAGE_SIZE, AMDGPU_GEM_DOMAIN_VRAM,
- &adev->gfx.rlc.cp_table_obj,
- &adev->gfx.rlc.cp_table_gpu_addr,
- (void **)&adev->gfx.rlc.cp_table_ptr);
- if (r) {
- dev_err(adev->dev,
- "(%d) failed to create cp table bo\n", r);
- gfx_v9_0_rlc_fini(adev);
+ r = amdgpu_gfx_rlc_init_cpt(adev);
+ if (r)
return r;
- }
-
- rv_init_cp_jump_table(adev);
- amdgpu_bo_kunmap(adev->gfx.rlc.cp_table_obj);
- amdgpu_bo_unreserve(adev->gfx.rlc.cp_table_obj);
}
switch (adev->asic_type) {
@@ -1264,7 +1150,7 @@ static int gfx_v9_0_mec_init(struct amdgpu_device *adev)
mec_hpd_size = adev->gfx.num_compute_rings * GFX9_MEC_HPD_SIZE;
r = amdgpu_bo_create_reserved(adev, mec_hpd_size, PAGE_SIZE,
- AMDGPU_GEM_DOMAIN_GTT,
+ AMDGPU_GEM_DOMAIN_VRAM,
&adev->gfx.mec.hpd_eop_obj,
&adev->gfx.mec.hpd_eop_gpu_addr,
(void **)&hpd);
@@ -1635,8 +1521,8 @@ static int gfx_v9_0_ngg_en(struct amdgpu_device *adev)
/* Clear GDS reserved memory */
r = amdgpu_ring_alloc(ring, 17);
if (r) {
- DRM_ERROR("amdgpu: NGG failed to lock ring %d (%d).\n",
- ring->idx, r);
+ DRM_ERROR("amdgpu: NGG failed to lock ring %s (%d).\n",
+ ring->name, r);
return r;
}
@@ -1748,7 +1634,7 @@ static int gfx_v9_0_sw_init(void *handle)
return r;
}
- r = gfx_v9_0_rlc_init(adev);
+ r = adev->gfx.rlc.funcs->init(adev);
if (r) {
DRM_ERROR("Failed to init rlc BOs!\n");
return r;
@@ -2498,12 +2384,12 @@ static int gfx_v9_0_rlc_resume(struct amdgpu_device *adev)
return 0;
}
- gfx_v9_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
/* disable CG */
WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL, 0);
- gfx_v9_0_rlc_reset(adev);
+ adev->gfx.rlc.funcs->reset(adev);
gfx_v9_0_init_pg(adev);
@@ -2514,15 +2400,24 @@ static int gfx_v9_0_rlc_resume(struct amdgpu_device *adev)
return r;
}
- if (adev->asic_type == CHIP_RAVEN ||
- adev->asic_type == CHIP_VEGA20) {
- if (amdgpu_lbpw != 0)
+ switch (adev->asic_type) {
+ case CHIP_RAVEN:
+ if (amdgpu_lbpw == 0)
+ gfx_v9_0_enable_lbpw(adev, false);
+ else
+ gfx_v9_0_enable_lbpw(adev, true);
+ break;
+ case CHIP_VEGA20:
+ if (amdgpu_lbpw > 0)
gfx_v9_0_enable_lbpw(adev, true);
else
gfx_v9_0_enable_lbpw(adev, false);
+ break;
+ default:
+ break;
}
- gfx_v9_0_rlc_start(adev);
+ adev->gfx.rlc.funcs->start(adev);
return 0;
}
@@ -2537,7 +2432,7 @@ static void gfx_v9_0_cp_gfx_enable(struct amdgpu_device *adev, bool enable)
tmp = REG_SET_FIELD(tmp, CP_ME_CNTL, CE_HALT, enable ? 0 : 1);
if (!enable) {
for (i = 0; i < adev->gfx.num_gfx_rings; i++)
- adev->gfx.gfx_ring[i].ready = false;
+ adev->gfx.gfx_ring[i].sched.ready = false;
}
WREG32_SOC15(GC, 0, mmCP_ME_CNTL, tmp);
udelay(50);
@@ -2727,7 +2622,7 @@ static int gfx_v9_0_cp_gfx_resume(struct amdgpu_device *adev)
/* start the ring */
gfx_v9_0_cp_gfx_start(adev);
- ring->ready = true;
+ ring->sched.ready = true;
return 0;
}
@@ -2742,8 +2637,8 @@ static void gfx_v9_0_cp_compute_enable(struct amdgpu_device *adev, bool enable)
WREG32_SOC15(GC, 0, mmCP_MEC_CNTL,
(CP_MEC_CNTL__MEC_ME1_HALT_MASK | CP_MEC_CNTL__MEC_ME2_HALT_MASK));
for (i = 0; i < adev->gfx.num_compute_rings; i++)
- adev->gfx.compute_ring[i].ready = false;
- adev->gfx.kiq.ring.ready = false;
+ adev->gfx.compute_ring[i].sched.ready = false;
+ adev->gfx.kiq.ring.sched.ready = false;
}
udelay(50);
}
@@ -2866,11 +2761,9 @@ static int gfx_v9_0_kiq_kcq_enable(struct amdgpu_device *adev)
amdgpu_ring_write(kiq_ring, upper_32_bits(wptr_addr));
}
- r = amdgpu_ring_test_ring(kiq_ring);
- if (r) {
+ r = amdgpu_ring_test_helper(kiq_ring);
+ if (r)
DRM_ERROR("KCQ enable failed\n");
- kiq_ring->ready = false;
- }
return r;
}
@@ -3249,7 +3142,7 @@ static int gfx_v9_0_kiq_resume(struct amdgpu_device *adev)
amdgpu_bo_kunmap(ring->mqd_obj);
ring->mqd_ptr = NULL;
amdgpu_bo_unreserve(ring->mqd_obj);
- ring->ready = true;
+ ring->sched.ready = true;
return 0;
}
@@ -3314,19 +3207,13 @@ static int gfx_v9_0_cp_resume(struct amdgpu_device *adev)
return r;
ring = &adev->gfx.gfx_ring[0];
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
return r;
- }
for (i = 0; i < adev->gfx.num_compute_rings; i++) {
ring = &adev->gfx.compute_ring[i];
-
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r)
- ring->ready = false;
+ amdgpu_ring_test_helper(ring);
}
gfx_v9_0_enable_gui_idle_interrupt(adev, true);
@@ -3353,7 +3240,7 @@ static int gfx_v9_0_hw_init(void *handle)
if (r)
return r;
- r = gfx_v9_0_rlc_resume(adev);
+ r = adev->gfx.rlc.funcs->resume(adev);
if (r)
return r;
@@ -3391,7 +3278,7 @@ static int gfx_v9_0_kcq_disable(struct amdgpu_device *adev)
amdgpu_ring_write(kiq_ring, 0);
amdgpu_ring_write(kiq_ring, 0);
}
- r = amdgpu_ring_test_ring(kiq_ring);
+ r = amdgpu_ring_test_helper(kiq_ring);
if (r)
DRM_ERROR("KCQ disable failed\n");
@@ -3433,7 +3320,7 @@ static int gfx_v9_0_hw_fini(void *handle)
}
gfx_v9_0_cp_enable(adev, false);
- gfx_v9_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
gfx_v9_0_csb_vram_unpin(adev);
@@ -3508,7 +3395,7 @@ static int gfx_v9_0_soft_reset(void *handle)
if (grbm_soft_reset) {
/* stop the rlc */
- gfx_v9_0_rlc_stop(adev);
+ adev->gfx.rlc.funcs->stop(adev);
/* Disable GFX parsing/prefetching */
gfx_v9_0_cp_gfx_enable(adev, false);
@@ -3607,64 +3494,47 @@ static int gfx_v9_0_late_init(void *handle)
return 0;
}
-static void gfx_v9_0_enter_rlc_safe_mode(struct amdgpu_device *adev)
+static bool gfx_v9_0_is_rlc_enabled(struct amdgpu_device *adev)
{
- uint32_t rlc_setting, data;
- unsigned i;
-
- if (adev->gfx.rlc.in_safe_mode)
- return;
+ uint32_t rlc_setting;
/* if RLC is not enabled, do nothing */
rlc_setting = RREG32_SOC15(GC, 0, mmRLC_CNTL);
if (!(rlc_setting & RLC_CNTL__RLC_ENABLE_F32_MASK))
- return;
+ return false;
- if (adev->cg_flags &
- (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG |
- AMD_CG_SUPPORT_GFX_3D_CGCG)) {
- data = RLC_SAFE_MODE__CMD_MASK;
- data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT);
- WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE, data);
+ return true;
+}
- /* wait for RLC_SAFE_MODE */
- for (i = 0; i < adev->usec_timeout; i++) {
- if (!REG_GET_FIELD(RREG32_SOC15(GC, 0, mmRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
- break;
- udelay(1);
- }
- adev->gfx.rlc.in_safe_mode = true;
+static void gfx_v9_0_set_safe_mode(struct amdgpu_device *adev)
+{
+ uint32_t data;
+ unsigned i;
+
+ data = RLC_SAFE_MODE__CMD_MASK;
+ data |= (1 << RLC_SAFE_MODE__MESSAGE__SHIFT);
+ WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE, data);
+
+ /* wait for RLC_SAFE_MODE */
+ for (i = 0; i < adev->usec_timeout; i++) {
+ if (!REG_GET_FIELD(RREG32_SOC15(GC, 0, mmRLC_SAFE_MODE), RLC_SAFE_MODE, CMD))
+ break;
+ udelay(1);
}
}
-static void gfx_v9_0_exit_rlc_safe_mode(struct amdgpu_device *adev)
+static void gfx_v9_0_unset_safe_mode(struct amdgpu_device *adev)
{
- uint32_t rlc_setting, data;
+ uint32_t data;
- if (!adev->gfx.rlc.in_safe_mode)
- return;
-
- /* if RLC is not enabled, do nothing */
- rlc_setting = RREG32_SOC15(GC, 0, mmRLC_CNTL);
- if (!(rlc_setting & RLC_CNTL__RLC_ENABLE_F32_MASK))
- return;
-
- if (adev->cg_flags &
- (AMD_CG_SUPPORT_GFX_CGCG | AMD_CG_SUPPORT_GFX_MGCG)) {
- /*
- * Try to exit safe mode only if it is already in safe
- * mode.
- */
- data = RLC_SAFE_MODE__CMD_MASK;
- WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE, data);
- adev->gfx.rlc.in_safe_mode = false;
- }
+ data = RLC_SAFE_MODE__CMD_MASK;
+ WREG32_SOC15(GC, 0, mmRLC_SAFE_MODE, data);
}
static void gfx_v9_0_update_gfx_cg_power_gating(struct amdgpu_device *adev,
bool enable)
{
- gfx_v9_0_enter_rlc_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
if ((adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) && enable) {
gfx_v9_0_enable_gfx_cg_power_gating(adev, true);
@@ -3675,7 +3545,7 @@ static void gfx_v9_0_update_gfx_cg_power_gating(struct amdgpu_device *adev,
gfx_v9_0_enable_gfx_pipeline_powergating(adev, false);
}
- gfx_v9_0_exit_rlc_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
static void gfx_v9_0_update_gfx_mg_power_gating(struct amdgpu_device *adev,
@@ -3773,7 +3643,7 @@ static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev,
{
uint32_t data, def;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
/* Enable 3D CGCG/CGLS */
if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_3D_CGCG)) {
@@ -3813,7 +3683,7 @@ static void gfx_v9_0_update_3d_clock_gating(struct amdgpu_device *adev,
WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL_3D, data);
}
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev,
@@ -3821,7 +3691,7 @@ static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev
{
uint32_t def, data;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
def = data = RREG32_SOC15(GC, 0, mmRLC_CGTT_MGCG_OVERRIDE);
@@ -3861,7 +3731,7 @@ static void gfx_v9_0_update_coarse_grain_clock_gating(struct amdgpu_device *adev
WREG32_SOC15(GC, 0, mmRLC_CGCG_CGLS_CTRL, data);
}
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
static int gfx_v9_0_update_gfx_clock_gating(struct amdgpu_device *adev,
@@ -3890,8 +3760,17 @@ static int gfx_v9_0_update_gfx_clock_gating(struct amdgpu_device *adev,
}
static const struct amdgpu_rlc_funcs gfx_v9_0_rlc_funcs = {
- .enter_safe_mode = gfx_v9_0_enter_rlc_safe_mode,
- .exit_safe_mode = gfx_v9_0_exit_rlc_safe_mode
+ .is_rlc_enabled = gfx_v9_0_is_rlc_enabled,
+ .set_safe_mode = gfx_v9_0_set_safe_mode,
+ .unset_safe_mode = gfx_v9_0_unset_safe_mode,
+ .init = gfx_v9_0_rlc_init,
+ .get_csb_size = gfx_v9_0_get_csb_size,
+ .get_csb_buffer = gfx_v9_0_get_csb_buffer,
+ .get_cp_table_num = gfx_v9_0_cp_jump_table_num,
+ .resume = gfx_v9_0_rlc_resume,
+ .stop = gfx_v9_0_rlc_stop,
+ .reset = gfx_v9_0_rlc_reset,
+ .start = gfx_v9_0_rlc_start
};
static int gfx_v9_0_set_powergating_state(void *handle,
@@ -4072,9 +3951,11 @@ static void gfx_v9_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
}
static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
u32 header, control = 0;
if (ib->flags & AMDGPU_IB_FLAG_CE)
@@ -4103,20 +3984,22 @@ static void gfx_v9_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
}
static void gfx_v9_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
- u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+ u32 control = INDIRECT_BUFFER_VALID | ib->length_dw | (vmid << 24);
- amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
+ amdgpu_ring_write(ring, PACKET3(PACKET3_INDIRECT_BUFFER, 2));
BUG_ON(ib->gpu_addr & 0x3); /* Dword align */
- amdgpu_ring_write(ring,
+ amdgpu_ring_write(ring,
#ifdef __BIG_ENDIAN
- (2 << 0) |
+ (2 << 0) |
#endif
- lower_32_bits(ib->gpu_addr));
- amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
- amdgpu_ring_write(ring, control);
+ lower_32_bits(ib->gpu_addr));
+ amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr));
+ amdgpu_ring_write(ring, control);
}
static void gfx_v9_0_ring_emit_fence(struct amdgpu_ring *ring, u64 addr,
@@ -4695,12 +4578,39 @@ static int gfx_v9_0_eop_irq(struct amdgpu_device *adev,
return 0;
}
+static void gfx_v9_0_fault(struct amdgpu_device *adev,
+ struct amdgpu_iv_entry *entry)
+{
+ u8 me_id, pipe_id, queue_id;
+ struct amdgpu_ring *ring;
+ int i;
+
+ me_id = (entry->ring_id & 0x0c) >> 2;
+ pipe_id = (entry->ring_id & 0x03) >> 0;
+ queue_id = (entry->ring_id & 0x70) >> 4;
+
+ switch (me_id) {
+ case 0:
+ drm_sched_fault(&adev->gfx.gfx_ring[0].sched);
+ break;
+ case 1:
+ case 2:
+ for (i = 0; i < adev->gfx.num_compute_rings; i++) {
+ ring = &adev->gfx.compute_ring[i];
+ if (ring->me == me_id && ring->pipe == pipe_id &&
+ ring->queue == queue_id)
+ drm_sched_fault(&ring->sched);
+ }
+ break;
+ }
+}
+
static int gfx_v9_0_priv_reg_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
DRM_ERROR("Illegal register access in command stream\n");
- schedule_work(&adev->reset_work);
+ gfx_v9_0_fault(adev, entry);
return 0;
}
@@ -4709,7 +4619,7 @@ static int gfx_v9_0_priv_inst_irq(struct amdgpu_device *adev,
struct amdgpu_iv_entry *entry)
{
DRM_ERROR("Illegal instruction in command stream\n");
- schedule_work(&adev->reset_work);
+ gfx_v9_0_fault(adev, entry);
return 0;
}
@@ -4836,10 +4746,8 @@ static const struct amdgpu_ring_funcs gfx_v9_0_ring_funcs_kiq = {
2 + /* gfx_v9_0_ring_emit_vm_flush */
8 + 8 + 8, /* gfx_v9_0_ring_emit_fence_kiq x3 for user fence, vm fence */
.emit_ib_size = 4, /* gfx_v9_0_ring_emit_ib_compute */
- .emit_ib = gfx_v9_0_ring_emit_ib_compute,
.emit_fence = gfx_v9_0_ring_emit_fence_kiq,
.test_ring = gfx_v9_0_ring_test_ring,
- .test_ib = gfx_v9_0_ring_test_ib,
.insert_nop = amdgpu_ring_insert_nop,
.pad_ib = amdgpu_ring_generic_pad_ib,
.emit_rreg = gfx_v9_0_ring_emit_rreg,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
index ceb7847..f5edddf 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c
@@ -35,20 +35,25 @@ u64 gfxhub_v1_0_get_mc_fb_offset(struct amdgpu_device *adev)
return (u64)RREG32_SOC15(GC, 0, mmMC_VM_FB_OFFSET) << 24;
}
-static void gfxhub_v1_0_init_gart_pt_regs(struct amdgpu_device *adev)
+void gfxhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+ uint64_t page_table_base)
{
- uint64_t value = amdgpu_gmc_pd_addr(adev->gart.bo);
+ /* two registers distance between mmVM_CONTEXT0_* to mmVM_CONTEXT1_* */
+ int offset = mmVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32
+ - mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
- WREG32_SOC15(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
- lower_32_bits(value));
+ WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+ offset * vmid, lower_32_bits(page_table_base));
- WREG32_SOC15(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
- upper_32_bits(value));
+ WREG32_SOC15_OFFSET(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+ offset * vmid, upper_32_bits(page_table_base));
}
static void gfxhub_v1_0_init_gart_aperture_regs(struct amdgpu_device *adev)
{
- gfxhub_v1_0_init_gart_pt_regs(adev);
+ uint64_t pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+
+ gfxhub_v1_0_setup_vm_pt_regs(adev, 0, pt_base);
WREG32_SOC15(GC, 0, mmVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
(u32)(adev->gmc.gart_start >> 12));
@@ -72,7 +77,7 @@ static void gfxhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev)
/* Program the system aperture low logical page number. */
WREG32_SOC15(GC, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR,
- min(adev->gmc.vram_start, adev->gmc.agp_start) >> 18);
+ min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
if (adev->asic_type == CHIP_RAVEN && adev->rev_id >= 0x8)
/*
@@ -82,11 +87,11 @@ static void gfxhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev)
* to get rid of the VM fault and hardware hang.
*/
WREG32_SOC15(GC, 0, mmMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
- max((adev->gmc.vram_end >> 18) + 0x1,
+ max((adev->gmc.fb_end >> 18) + 0x1,
adev->gmc.agp_end >> 18));
else
WREG32_SOC15(GC, 0, mmMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
- max(adev->gmc.vram_end, adev->gmc.agp_end) >> 18);
+ max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
/* Set default page address. */
value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start
diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
index 206e29c..92d3a70 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.h
@@ -30,5 +30,7 @@ void gfxhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev,
bool value);
void gfxhub_v1_0_init(struct amdgpu_device *adev);
u64 gfxhub_v1_0_get_mc_fb_offset(struct amdgpu_device *adev);
+void gfxhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+ uint64_t page_table_base);
#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
index e1c2b4e..2821d1d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c
@@ -358,7 +358,8 @@ static int gmc_v6_0_mc_init(struct amdgpu_device *adev)
return 0;
}
-static void gmc_v6_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid)
+static void gmc_v6_0_flush_gpu_tlb(struct amdgpu_device *adev,
+ uint32_t vmid, uint32_t flush_type)
{
WREG32(mmVM_INVALIDATE_REQUEST, 1 << vmid);
}
@@ -580,7 +581,7 @@ static int gmc_v6_0_gart_enable(struct amdgpu_device *adev)
else
gmc_v6_0_set_fault_enable_default(adev, true);
- gmc_v6_0_flush_gpu_tlb(adev, 0);
+ gmc_v6_0_flush_gpu_tlb(adev, 0, 0);
dev_info(adev->dev, "PCIE GART of %uM enabled (table at 0x%016llX).\n",
(unsigned)(adev->gmc.gart_size >> 20),
(unsigned long long)table_addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 910c4ce..761dcfb 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -430,7 +430,8 @@ static int gmc_v7_0_mc_init(struct amdgpu_device *adev)
*
* Flush the TLB for the requested page table (CIK).
*/
-static void gmc_v7_0_flush_gpu_tlb(struct amdgpu_device *adev, uint32_t vmid)
+static void gmc_v7_0_flush_gpu_tlb(struct amdgpu_device *adev,
+ uint32_t vmid, uint32_t flush_type)
{
/* bits 0-15 are the VM contexts0-15 */
WREG32(mmVM_INVALIDATE_REQUEST, 1 << vmid);
@@ -698,7 +699,7 @@ static int gmc_v7_0_gart_enable(struct amdgpu_device *adev)
WREG32(mmCHUB_CONTROL, tmp);
}
- gmc_v7_0_flush_gpu_tlb(adev, 0);
+ gmc_v7_0_flush_gpu_tlb(adev, 0, 0);
DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
(unsigned)(adev->gmc.gart_size >> 20),
(unsigned long long)table_addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 1d3265c..531aaf3 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -611,7 +611,7 @@ static int gmc_v8_0_mc_init(struct amdgpu_device *adev)
* Flush the TLB for the requested page table (CIK).
*/
static void gmc_v8_0_flush_gpu_tlb(struct amdgpu_device *adev,
- uint32_t vmid)
+ uint32_t vmid, uint32_t flush_type)
{
/* bits 0-15 are the VM contexts0-15 */
WREG32(mmVM_INVALIDATE_REQUEST, 1 << vmid);
@@ -920,7 +920,7 @@ static int gmc_v8_0_gart_enable(struct amdgpu_device *adev)
else
gmc_v8_0_set_fault_enable_default(adev, true);
- gmc_v8_0_flush_gpu_tlb(adev, 0);
+ gmc_v8_0_flush_gpu_tlb(adev, 0, 0);
DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
(unsigned)(adev->gmc.gart_size >> 20),
(unsigned long long)table_addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index f35d7a5..811231e 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -293,14 +293,14 @@ static void gmc_v9_0_set_irq_funcs(struct amdgpu_device *adev)
adev->gmc.vm_fault.funcs = &gmc_v9_0_irq_funcs;
}
-static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid)
+static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid,
+ uint32_t flush_type)
{
u32 req = 0;
- /* invalidate using legacy mode on vmid*/
req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ,
PER_VMID_INVALIDATE_REQ, 1 << vmid);
- req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, 0);
+ req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, FLUSH_TYPE, flush_type);
req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PTES, 1);
req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE0, 1);
req = REG_SET_FIELD(req, VM_INVALIDATE_ENG0_REQ, INVALIDATE_L2_PDE1, 1);
@@ -312,48 +312,6 @@ static uint32_t gmc_v9_0_get_invalidate_req(unsigned int vmid)
return req;
}
-static signed long amdgpu_kiq_reg_write_reg_wait(struct amdgpu_device *adev,
- uint32_t reg0, uint32_t reg1,
- uint32_t ref, uint32_t mask)
-{
- signed long r, cnt = 0;
- unsigned long flags;
- uint32_t seq;
- struct amdgpu_kiq *kiq = &adev->gfx.kiq;
- struct amdgpu_ring *ring = &kiq->ring;
-
- spin_lock_irqsave(&kiq->ring_lock, flags);
-
- amdgpu_ring_alloc(ring, 32);
- amdgpu_ring_emit_reg_write_reg_wait(ring, reg0, reg1,
- ref, mask);
- amdgpu_fence_emit_polling(ring, &seq);
- amdgpu_ring_commit(ring);
- spin_unlock_irqrestore(&kiq->ring_lock, flags);
-
- r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
-
- /* don't wait anymore for IRQ context */
- if (r < 1 && in_interrupt())
- goto failed_kiq;
-
- might_sleep();
-
- while (r < 1 && cnt++ < MAX_KIQ_REG_TRY) {
- msleep(MAX_KIQ_REG_BAILOUT_INTERVAL);
- r = amdgpu_fence_wait_polling(ring, seq, MAX_KIQ_REG_WAIT);
- }
-
- if (cnt > MAX_KIQ_REG_TRY)
- goto failed_kiq;
-
- return 0;
-
-failed_kiq:
- pr_err("failed to invalidate tlb with kiq\n");
- return r;
-}
-
/*
* GART
* VMID 0 is the physical GPU addresses as used by the kernel.
@@ -362,64 +320,47 @@ static signed long amdgpu_kiq_reg_write_reg_wait(struct amdgpu_device *adev,
*/
/**
- * gmc_v9_0_flush_gpu_tlb - gart tlb flush callback
+ * gmc_v9_0_flush_gpu_tlb - tlb flush with certain type
*
* @adev: amdgpu_device pointer
* @vmid: vm instance to flush
+ * @flush_type: the flush type
*
- * Flush the TLB for the requested page table.
+ * Flush the TLB for the requested page table using certain type.
*/
static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev,
- uint32_t vmid)
+ uint32_t vmid, uint32_t flush_type)
{
- /* Use register 17 for GART */
const unsigned eng = 17;
unsigned i, j;
- int r;
for (i = 0; i < AMDGPU_MAX_VMHUBS; ++i) {
struct amdgpu_vmhub *hub = &adev->vmhub[i];
- u32 tmp = gmc_v9_0_get_invalidate_req(vmid);
+ u32 tmp = gmc_v9_0_get_invalidate_req(vmid, flush_type);
- if (adev->gfx.kiq.ring.ready &&
- (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev)) &&
- !adev->in_gpu_reset) {
- r = amdgpu_kiq_reg_write_reg_wait(adev, hub->vm_inv_eng0_req + eng,
- hub->vm_inv_eng0_ack + eng, tmp, 1 << vmid);
- if (!r)
- continue;
+ if (i == AMDGPU_GFXHUB && !adev->in_gpu_reset &&
+ adev->gfx.kiq.ring.sched.ready &&
+ (amdgpu_sriov_runtime(adev) || !amdgpu_sriov_vf(adev))) {
+ uint32_t req = hub->vm_inv_eng0_req + eng;
+ uint32_t ack = hub->vm_inv_eng0_ack + eng;
+
+ amdgpu_virt_kiq_reg_write_reg_wait(adev, req, ack, tmp,
+ 1 << vmid);
+ continue;
}
spin_lock(&adev->gmc.invalidate_lock);
-
WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, tmp);
-
- /* Busy wait for ACK.*/
- for (j = 0; j < 100; j++) {
- tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng);
- tmp &= 1 << vmid;
- if (tmp)
- break;
- cpu_relax();
- }
- if (j < 100) {
- spin_unlock(&adev->gmc.invalidate_lock);
- continue;
- }
-
- /* Wait for ACK with a delay.*/
for (j = 0; j < adev->usec_timeout; j++) {
tmp = RREG32_NO_KIQ(hub->vm_inv_eng0_ack + eng);
- tmp &= 1 << vmid;
- if (tmp)
+ if (tmp & (1 << vmid))
break;
udelay(1);
}
- if (j < adev->usec_timeout) {
- spin_unlock(&adev->gmc.invalidate_lock);
- continue;
- }
spin_unlock(&adev->gmc.invalidate_lock);
+ if (j < adev->usec_timeout)
+ continue;
+
DRM_ERROR("Timeout waiting for VM flush ACK!\n");
}
}
@@ -429,7 +370,7 @@ static uint64_t gmc_v9_0_emit_flush_gpu_tlb(struct amdgpu_ring *ring,
{
struct amdgpu_device *adev = ring->adev;
struct amdgpu_vmhub *hub = &adev->vmhub[ring->funcs->vmhub];
- uint32_t req = gmc_v9_0_get_invalidate_req(vmid);
+ uint32_t req = gmc_v9_0_get_invalidate_req(vmid, 0);
unsigned eng = ring->vm_inv_eng;
amdgpu_ring_emit_wreg(ring, hub->ctx0_ptb_addr_lo32 + (2 * vmid),
@@ -739,9 +680,8 @@ static int gmc_v9_0_late_init(void *handle)
unsigned vmhub = ring->funcs->vmhub;
ring->vm_inv_eng = vm_inv_eng[vmhub]++;
- dev_info(adev->dev, "ring %u(%s) uses VM inv eng %u on hub %u\n",
- ring->idx, ring->name, ring->vm_inv_eng,
- ring->funcs->vmhub);
+ dev_info(adev->dev, "ring %s uses VM inv eng %u on hub %u\n",
+ ring->name, ring->vm_inv_eng, ring->funcs->vmhub);
}
/* Engine 16 is used for KFD and 17 for GART flushes */
@@ -1122,7 +1062,7 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev)
gfxhub_v1_0_set_fault_enable_default(adev, value);
mmhub_v1_0_set_fault_enable_default(adev, value);
- gmc_v9_0_flush_gpu_tlb(adev, 0);
+ gmc_v9_0_flush_gpu_tlb(adev, 0, 0);
DRM_INFO("PCIE GART of %uM enabled (table at 0x%016llX).\n",
(unsigned)(adev->gmc.gart_size >> 20),
diff --git a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
index d0e478f..0c9a2c0 100644
--- a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
@@ -508,19 +508,19 @@ static int kv_enable_didt(struct amdgpu_device *adev, bool enable)
pi->caps_db_ramping ||
pi->caps_td_ramping ||
pi->caps_tcp_ramping) {
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
if (enable) {
ret = kv_program_pt_config_registers(adev, didt_config_kv);
if (ret) {
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return ret;
}
}
kv_do_enable_didt(adev, enable);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
index fd23ba1..d0d966d 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
@@ -52,20 +52,25 @@ u64 mmhub_v1_0_get_fb_location(struct amdgpu_device *adev)
return base;
}
-static void mmhub_v1_0_init_gart_pt_regs(struct amdgpu_device *adev)
+void mmhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+ uint64_t page_table_base)
{
- uint64_t value = amdgpu_gmc_pd_addr(adev->gart.bo);
+ /* two registers distance between mmVM_CONTEXT0_* to mmVM_CONTEXT1_* */
+ int offset = mmVM_CONTEXT1_PAGE_TABLE_BASE_ADDR_LO32
+ - mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32;
- WREG32_SOC15(MMHUB, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
- lower_32_bits(value));
+ WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_LO32,
+ offset * vmid, lower_32_bits(page_table_base));
- WREG32_SOC15(MMHUB, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
- upper_32_bits(value));
+ WREG32_SOC15_OFFSET(MMHUB, 0, mmVM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32,
+ offset * vmid, upper_32_bits(page_table_base));
}
static void mmhub_v1_0_init_gart_aperture_regs(struct amdgpu_device *adev)
{
- mmhub_v1_0_init_gart_pt_regs(adev);
+ uint64_t pt_base = amdgpu_gmc_pd_addr(adev->gart.bo);
+
+ mmhub_v1_0_setup_vm_pt_regs(adev, 0, pt_base);
WREG32_SOC15(MMHUB, 0, mmVM_CONTEXT0_PAGE_TABLE_START_ADDR_LO32,
(u32)(adev->gmc.gart_start >> 12));
@@ -90,7 +95,7 @@ static void mmhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev)
/* Program the system aperture low logical page number. */
WREG32_SOC15(MMHUB, 0, mmMC_VM_SYSTEM_APERTURE_LOW_ADDR,
- min(adev->gmc.vram_start, adev->gmc.agp_start) >> 18);
+ min(adev->gmc.fb_start, adev->gmc.agp_start) >> 18);
if (adev->asic_type == CHIP_RAVEN && adev->rev_id >= 0x8)
/*
@@ -100,11 +105,11 @@ static void mmhub_v1_0_init_system_aperture_regs(struct amdgpu_device *adev)
* to get rid of the VM fault and hardware hang.
*/
WREG32_SOC15(MMHUB, 0, mmMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
- max((adev->gmc.vram_end >> 18) + 0x1,
+ max((adev->gmc.fb_end >> 18) + 0x1,
adev->gmc.agp_end >> 18));
else
WREG32_SOC15(MMHUB, 0, mmMC_VM_SYSTEM_APERTURE_HIGH_ADDR,
- max(adev->gmc.vram_end, adev->gmc.agp_end) >> 18);
+ max(adev->gmc.fb_end, adev->gmc.agp_end) >> 18);
/* Set default page address. */
value = adev->vram_scratch.gpu_addr - adev->gmc.vram_start +
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
index bef3d0c..0de0fdf 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.h
@@ -34,5 +34,7 @@ int mmhub_v1_0_set_clockgating(struct amdgpu_device *adev,
void mmhub_v1_0_get_clockgating(struct amdgpu_device *adev, u32 *flags);
void mmhub_v1_0_update_power_gating(struct amdgpu_device *adev,
bool enable);
+void mmhub_v1_0_setup_vm_pt_regs(struct amdgpu_device *adev, uint32_t vmid,
+ uint64_t page_table_base);
#endif
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
index 3f3fac2..e5dd052 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v11_0.c
@@ -34,6 +34,7 @@
#include "nbio/nbio_7_4_offset.h"
MODULE_FIRMWARE("amdgpu/vega20_sos.bin");
+MODULE_FIRMWARE("amdgpu/vega20_ta.bin");
/* address block */
#define smnMP1_FIRMWARE_FLAGS 0x3010024
@@ -98,7 +99,8 @@ static int psp_v11_0_init_microcode(struct psp_context *psp)
const char *chip_name;
char fw_name[30];
int err = 0;
- const struct psp_firmware_header_v1_0 *hdr;
+ const struct psp_firmware_header_v1_0 *sos_hdr;
+ const struct ta_firmware_header_v1_0 *ta_hdr;
DRM_DEBUG("\n");
@@ -119,16 +121,32 @@ static int psp_v11_0_init_microcode(struct psp_context *psp)
if (err)
goto out;
- hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.sos_fw->data;
- adev->psp.sos_fw_version = le32_to_cpu(hdr->header.ucode_version);
- adev->psp.sos_feature_version = le32_to_cpu(hdr->ucode_feature_version);
- adev->psp.sos_bin_size = le32_to_cpu(hdr->sos_size_bytes);
- adev->psp.sys_bin_size = le32_to_cpu(hdr->header.ucode_size_bytes) -
- le32_to_cpu(hdr->sos_size_bytes);
- adev->psp.sys_start_addr = (uint8_t *)hdr +
- le32_to_cpu(hdr->header.ucode_array_offset_bytes);
+ sos_hdr = (const struct psp_firmware_header_v1_0 *)adev->psp.sos_fw->data;
+ adev->psp.sos_fw_version = le32_to_cpu(sos_hdr->header.ucode_version);
+ adev->psp.sos_feature_version = le32_to_cpu(sos_hdr->ucode_feature_version);
+ adev->psp.sos_bin_size = le32_to_cpu(sos_hdr->sos_size_bytes);
+ adev->psp.sys_bin_size = le32_to_cpu(sos_hdr->header.ucode_size_bytes) -
+ le32_to_cpu(sos_hdr->sos_size_bytes);
+ adev->psp.sys_start_addr = (uint8_t *)sos_hdr +
+ le32_to_cpu(sos_hdr->header.ucode_array_offset_bytes);
adev->psp.sos_start_addr = (uint8_t *)adev->psp.sys_start_addr +
- le32_to_cpu(hdr->sos_offset_bytes);
+ le32_to_cpu(sos_hdr->sos_offset_bytes);
+
+ snprintf(fw_name, sizeof(fw_name), "amdgpu/%s_ta.bin", chip_name);
+ err = request_firmware(&adev->psp.ta_fw, fw_name, adev->dev);
+ if (err)
+ goto out;
+
+ err = amdgpu_ucode_validate(adev->psp.ta_fw);
+ if (err)
+ goto out;
+
+ ta_hdr = (const struct ta_firmware_header_v1_0 *)adev->psp.ta_fw->data;
+ adev->psp.ta_xgmi_ucode_version = le32_to_cpu(ta_hdr->ta_xgmi_ucode_version);
+ adev->psp.ta_xgmi_ucode_size = le32_to_cpu(ta_hdr->ta_xgmi_size_bytes);
+ adev->psp.ta_xgmi_start_addr = (uint8_t *)ta_hdr +
+ le32_to_cpu(ta_hdr->header.ucode_array_offset_bytes);
+
return 0;
out:
if (err) {
@@ -167,7 +185,7 @@ static int psp_v11_0_bootloader_load_sysdrv(struct psp_context *psp)
/* Copy PSP System Driver binary to memory */
memcpy(psp->fw_pri_buf, psp->sys_start_addr, psp->sys_bin_size);
- /* Provide the sys driver to bootrom */
+ /* Provide the sys driver to bootloader */
WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36,
(uint32_t)(psp->fw_pri_mc_addr >> 20));
psp_gfxdrv_command_reg = 1 << 16;
@@ -208,7 +226,7 @@ static int psp_v11_0_bootloader_load_sos(struct psp_context *psp)
/* Copy Secure OS binary to PSP memory */
memcpy(psp->fw_pri_buf, psp->sos_start_addr, psp->sos_bin_size);
- /* Provide the PSP secure OS to bootrom */
+ /* Provide the PSP secure OS to bootloader */
WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36,
(uint32_t)(psp->fw_pri_mc_addr >> 20));
psp_gfxdrv_command_reg = 2 << 16;
@@ -552,24 +570,110 @@ static int psp_v11_0_mode1_reset(struct psp_context *psp)
static int psp_v11_0_xgmi_get_topology_info(struct psp_context *psp,
int number_devices, struct psp_xgmi_topology_info *topology)
{
+ struct ta_xgmi_shared_memory *xgmi_cmd;
+ struct ta_xgmi_cmd_get_topology_info_input *topology_info_input;
+ struct ta_xgmi_cmd_get_topology_info_output *topology_info_output;
+ int i;
+ int ret;
+
+ if (!topology || topology->num_nodes > TA_XGMI__MAX_CONNECTED_NODES)
+ return -EINVAL;
+
+ xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf;
+ memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory));
+
+ /* Fill in the shared memory with topology information as input */
+ topology_info_input = &xgmi_cmd->xgmi_in_message.get_topology_info;
+ xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_GET_TOPOLOGY_INFO;
+ topology_info_input->num_nodes = number_devices;
+
+ for (i = 0; i < topology_info_input->num_nodes; i++) {
+ topology_info_input->nodes[i].node_id = topology->nodes[i].node_id;
+ topology_info_input->nodes[i].num_hops = topology->nodes[i].num_hops;
+ topology_info_input->nodes[i].is_sharing_enabled = topology->nodes[i].is_sharing_enabled;
+ topology_info_input->nodes[i].sdma_engine = topology->nodes[i].sdma_engine;
+ }
+
+ /* Invoke xgmi ta to get the topology information */
+ ret = psp_xgmi_invoke(psp, TA_COMMAND_XGMI__GET_GET_TOPOLOGY_INFO);
+ if (ret)
+ return ret;
+
+ /* Read the output topology information from the shared memory */
+ topology_info_output = &xgmi_cmd->xgmi_out_message.get_topology_info;
+ topology->num_nodes = xgmi_cmd->xgmi_out_message.get_topology_info.num_nodes;
+ for (i = 0; i < topology->num_nodes; i++) {
+ topology->nodes[i].node_id = topology_info_output->nodes[i].node_id;
+ topology->nodes[i].num_hops = topology_info_output->nodes[i].num_hops;
+ topology->nodes[i].is_sharing_enabled = topology_info_output->nodes[i].is_sharing_enabled;
+ topology->nodes[i].sdma_engine = topology_info_output->nodes[i].sdma_engine;
+ }
+
return 0;
}
static int psp_v11_0_xgmi_set_topology_info(struct psp_context *psp,
int number_devices, struct psp_xgmi_topology_info *topology)
{
- return 0;
+ struct ta_xgmi_shared_memory *xgmi_cmd;
+ struct ta_xgmi_cmd_get_topology_info_input *topology_info_input;
+ int i;
+
+ if (!topology || topology->num_nodes > TA_XGMI__MAX_CONNECTED_NODES)
+ return -EINVAL;
+
+ xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf;
+ memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory));
+
+ topology_info_input = &xgmi_cmd->xgmi_in_message.get_topology_info;
+ xgmi_cmd->cmd_id = TA_COMMAND_XGMI__SET_TOPOLOGY_INFO;
+ topology_info_input->num_nodes = number_devices;
+
+ for (i = 0; i < topology_info_input->num_nodes; i++) {
+ topology_info_input->nodes[i].node_id = topology->nodes[i].node_id;
+ topology_info_input->nodes[i].num_hops = topology->nodes[i].num_hops;
+ topology_info_input->nodes[i].is_sharing_enabled = topology->nodes[i].is_sharing_enabled;
+ topology_info_input->nodes[i].sdma_engine = topology->nodes[i].sdma_engine;
+ }
+
+ /* Invoke xgmi ta to set topology information */
+ return psp_xgmi_invoke(psp, TA_COMMAND_XGMI__SET_TOPOLOGY_INFO);
}
static u64 psp_v11_0_xgmi_get_hive_id(struct psp_context *psp)
{
- u64 hive_id = 0;
+ struct ta_xgmi_shared_memory *xgmi_cmd;
+ int ret;
- /* Remove me when we can get correct hive_id through PSP */
- if (psp->adev->gmc.xgmi.num_physical_nodes)
- hive_id = 0x123456789abcdef;
+ xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf;
+ memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory));
- return hive_id;
+ xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_HIVE_ID;
+
+ /* Invoke xgmi ta to get hive id */
+ ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id);
+ if (ret)
+ return 0;
+ else
+ return xgmi_cmd->xgmi_out_message.get_hive_id.hive_id;
+}
+
+static u64 psp_v11_0_xgmi_get_node_id(struct psp_context *psp)
+{
+ struct ta_xgmi_shared_memory *xgmi_cmd;
+ int ret;
+
+ xgmi_cmd = (struct ta_xgmi_shared_memory*)psp->xgmi_context.xgmi_shared_buf;
+ memset(xgmi_cmd, 0, sizeof(struct ta_xgmi_shared_memory));
+
+ xgmi_cmd->cmd_id = TA_COMMAND_XGMI__GET_NODE_ID;
+
+ /* Invoke xgmi ta to get the node id */
+ ret = psp_xgmi_invoke(psp, xgmi_cmd->cmd_id);
+ if (ret)
+ return 0;
+ else
+ return xgmi_cmd->xgmi_out_message.get_node_id.node_id;
}
static const struct psp_funcs psp_v11_0_funcs = {
@@ -587,6 +691,7 @@ static const struct psp_funcs psp_v11_0_funcs = {
.xgmi_get_topology_info = psp_v11_0_xgmi_get_topology_info,
.xgmi_set_topology_info = psp_v11_0_xgmi_set_topology_info,
.xgmi_get_hive_id = psp_v11_0_xgmi_get_hive_id,
+ .xgmi_get_node_id = psp_v11_0_xgmi_get_node_id,
};
void psp_v11_0_set_psp_funcs(struct psp_context *psp)
diff --git a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
index e1ebf77..9cea0bb 100644
--- a/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
+++ b/drivers/gpu/drm/amd/amdgpu/psp_v3_1.c
@@ -194,7 +194,7 @@ static int psp_v3_1_bootloader_load_sysdrv(struct psp_context *psp)
/* Copy PSP System Driver binary to memory */
memcpy(psp->fw_pri_buf, psp->sys_start_addr, psp->sys_bin_size);
- /* Provide the sys driver to bootrom */
+ /* Provide the sys driver to bootloader */
WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36,
(uint32_t)(psp->fw_pri_mc_addr >> 20));
psp_gfxdrv_command_reg = 1 << 16;
@@ -254,7 +254,7 @@ static int psp_v3_1_bootloader_load_sos(struct psp_context *psp)
/* Copy Secure OS binary to PSP memory */
memcpy(psp->fw_pri_buf, psp->sos_start_addr, psp->sos_bin_size);
- /* Provide the PSP secure OS to bootrom */
+ /* Provide the PSP secure OS to bootloader */
WREG32_SOC15(MP0, 0, mmMP0_SMN_C2PMSG_36,
(uint32_t)(psp->fw_pri_mc_addr >> 20));
psp_gfxdrv_command_reg = 2 << 16;
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index 2d4770e..9f3cb2a 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -225,7 +225,7 @@ static void sdma_v2_4_ring_set_wptr(struct amdgpu_ring *ring)
static void sdma_v2_4_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
{
- struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
+ struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
int i;
for (i = 0; i < count; i++)
@@ -245,9 +245,12 @@ static void sdma_v2_4_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
* Schedule an IB in the DMA ring (VI).
*/
static void sdma_v2_4_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
/* IB packet must end on a 8 DW boundary */
sdma_v2_4_ring_insert_nop(ring, (10 - (lower_32_bits(ring->wptr) & 7)) % 8);
@@ -349,8 +352,8 @@ static void sdma_v2_4_gfx_stop(struct amdgpu_device *adev)
ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0);
WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
}
- sdma0->ready = false;
- sdma1->ready = false;
+ sdma0->sched.ready = false;
+ sdma1->sched.ready = false;
}
/**
@@ -471,17 +474,15 @@ static int sdma_v2_4_gfx_resume(struct amdgpu_device *adev)
/* enable DMA IBs */
WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
- ring->ready = true;
+ ring->sched.ready = true;
}
sdma_v2_4_enable(adev, true);
for (i = 0; i < adev->sdma.num_instances; i++) {
ring = &adev->sdma.instance[i].ring;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
return r;
- }
if (adev->mman.buffer_funcs_ring == ring)
amdgpu_ttm_set_buffer_funcs_status(adev, true);
@@ -550,21 +551,16 @@ static int sdma_v2_4_ring_test_ring(struct amdgpu_ring *ring)
u64 gpu_addr;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
r = amdgpu_ring_alloc(ring, 5);
- if (r) {
- DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
- amdgpu_device_wb_free(adev, index);
- return r;
- }
+ if (r)
+ goto error_free_wb;
amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR));
@@ -581,15 +577,11 @@ static int sdma_v2_4_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n", ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
- amdgpu_device_wb_free(adev, index);
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+error_free_wb:
+ amdgpu_device_wb_free(adev, index);
return r;
}
@@ -612,20 +604,16 @@ static int sdma_v2_4_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 256, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err0;
- }
ib.ptr[0] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR);
@@ -644,21 +632,16 @@ static int sdma_v2_4_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out\n");
r = -ETIMEDOUT;
goto err1;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto err1;
}
tmp = le32_to_cpu(adev->wb.wb[index]);
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ if (tmp == 0xDEADBEEF)
r = 0;
- } else {
- DRM_ERROR("amdgpu: ib test failed (0x%08X)\n", tmp);
+ else
r = -EINVAL;
- }
err1:
amdgpu_ib_free(adev, &ib, NULL);
@@ -760,7 +743,7 @@ static void sdma_v2_4_vm_set_pte_pde(struct amdgpu_ib *ib, uint64_t pe,
*/
static void sdma_v2_4_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib)
{
- struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
+ struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
u32 pad_count;
int i;
@@ -1105,8 +1088,14 @@ static int sdma_v2_4_process_illegal_inst_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
+ u8 instance_id, queue_id;
+
DRM_ERROR("Illegal instruction in SDMA command stream\n");
- schedule_work(&adev->reset_work);
+ instance_id = (entry->ring_id & 0x3) >> 0;
+ queue_id = (entry->ring_id & 0xc) >> 2;
+
+ if (instance_id <= 1 && queue_id == 0)
+ drm_sched_fault(&adev->sdma.instance[instance_id].ring.sched);
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 6fb3eda..b6a25f9 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -399,7 +399,7 @@ static void sdma_v3_0_ring_set_wptr(struct amdgpu_ring *ring)
static void sdma_v3_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
{
- struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
+ struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
int i;
for (i = 0; i < count; i++)
@@ -419,9 +419,12 @@ static void sdma_v3_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
* Schedule an IB in the DMA ring (VI).
*/
static void sdma_v3_0_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
/* IB packet must end on a 8 DW boundary */
sdma_v3_0_ring_insert_nop(ring, (10 - (lower_32_bits(ring->wptr) & 7)) % 8);
@@ -523,8 +526,8 @@ static void sdma_v3_0_gfx_stop(struct amdgpu_device *adev)
ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0);
WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
}
- sdma0->ready = false;
- sdma1->ready = false;
+ sdma0->sched.ready = false;
+ sdma1->sched.ready = false;
}
/**
@@ -739,7 +742,7 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev)
/* enable DMA IBs */
WREG32(mmSDMA0_GFX_IB_CNTL + sdma_offsets[i], ib_cntl);
- ring->ready = true;
+ ring->sched.ready = true;
}
/* unhalt the MEs */
@@ -749,11 +752,9 @@ static int sdma_v3_0_gfx_resume(struct amdgpu_device *adev)
for (i = 0; i < adev->sdma.num_instances; i++) {
ring = &adev->sdma.instance[i].ring;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
return r;
- }
if (adev->mman.buffer_funcs_ring == ring)
amdgpu_ttm_set_buffer_funcs_status(adev, true);
@@ -822,21 +823,16 @@ static int sdma_v3_0_ring_test_ring(struct amdgpu_ring *ring)
u64 gpu_addr;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
r = amdgpu_ring_alloc(ring, 5);
- if (r) {
- DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
- amdgpu_device_wb_free(adev, index);
- return r;
- }
+ if (r)
+ goto error_free_wb;
amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR));
@@ -853,15 +849,11 @@ static int sdma_v3_0_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n", ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
- amdgpu_device_wb_free(adev, index);
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+error_free_wb:
+ amdgpu_device_wb_free(adev, index);
return r;
}
@@ -884,20 +876,16 @@ static int sdma_v3_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 256, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err0;
- }
ib.ptr[0] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR);
@@ -916,21 +904,16 @@ static int sdma_v3_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out\n");
r = -ETIMEDOUT;
goto err1;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto err1;
}
tmp = le32_to_cpu(adev->wb.wb[index]);
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ if (tmp == 0xDEADBEEF)
r = 0;
- } else {
- DRM_ERROR("amdgpu: ib test failed (0x%08X)\n", tmp);
+ else
r = -EINVAL;
- }
err1:
amdgpu_ib_free(adev, &ib, NULL);
dma_fence_put(f);
@@ -1031,7 +1014,7 @@ static void sdma_v3_0_vm_set_pte_pde(struct amdgpu_ib *ib, uint64_t pe,
*/
static void sdma_v3_0_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib)
{
- struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
+ struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
u32 pad_count;
int i;
@@ -1440,8 +1423,14 @@ static int sdma_v3_0_process_illegal_inst_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
+ u8 instance_id, queue_id;
+
DRM_ERROR("Illegal instruction in SDMA command stream\n");
- schedule_work(&adev->reset_work);
+ instance_id = (entry->ring_id & 0x3) >> 0;
+ queue_id = (entry->ring_id & 0xc) >> 2;
+
+ if (instance_id <= 1 && queue_id == 0)
+ drm_sched_fault(&adev->sdma.instance[instance_id].ring.sched);
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
index 7a8c917..f4490cd 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v4_0.c
@@ -54,6 +54,11 @@ MODULE_FIRMWARE("amdgpu/raven2_sdma.bin");
#define SDMA0_POWER_CNTL__ON_OFF_CONDITION_HOLD_TIME_MASK 0x000000F8L
#define SDMA0_POWER_CNTL__ON_OFF_STATUS_DURATION_TIME_MASK 0xFC000000L
+#define WREG32_SDMA(instance, offset, value) \
+ WREG32(sdma_v4_0_get_reg_offset(adev, (instance), (offset)), value)
+#define RREG32_SDMA(instance, offset) \
+ RREG32(sdma_v4_0_get_reg_offset(adev, (instance), (offset)))
+
static void sdma_v4_0_set_ring_funcs(struct amdgpu_device *adev);
static void sdma_v4_0_set_buffer_funcs(struct amdgpu_device *adev);
static void sdma_v4_0_set_vm_pte_funcs(struct amdgpu_device *adev);
@@ -367,16 +372,11 @@ static uint64_t sdma_v4_0_ring_get_wptr(struct amdgpu_ring *ring)
wptr = READ_ONCE(*((u64 *)&adev->wb.wb[ring->wptr_offs]));
DRM_DEBUG("wptr/doorbell before shift == 0x%016llx\n", wptr);
} else {
- u32 lowbit, highbit;
-
- lowbit = RREG32(sdma_v4_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR)) >> 2;
- highbit = RREG32(sdma_v4_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR_HI)) >> 2;
-
- DRM_DEBUG("wptr [%i]high== 0x%08x low==0x%08x\n",
- ring->me, highbit, lowbit);
- wptr = highbit;
+ wptr = RREG32_SDMA(ring->me, mmSDMA0_GFX_RB_WPTR_HI);
wptr = wptr << 32;
- wptr |= lowbit;
+ wptr |= RREG32_SDMA(ring->me, mmSDMA0_GFX_RB_WPTR);
+ DRM_DEBUG("wptr before shift [%i] wptr == 0x%016llx\n",
+ ring->me, wptr);
}
return wptr >> 2;
@@ -417,14 +417,67 @@ static void sdma_v4_0_ring_set_wptr(struct amdgpu_ring *ring)
lower_32_bits(ring->wptr << 2),
ring->me,
upper_32_bits(ring->wptr << 2));
- WREG32(sdma_v4_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr << 2));
- WREG32(sdma_v4_0_get_reg_offset(adev, ring->me, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr << 2));
+ WREG32_SDMA(ring->me, mmSDMA0_GFX_RB_WPTR,
+ lower_32_bits(ring->wptr << 2));
+ WREG32_SDMA(ring->me, mmSDMA0_GFX_RB_WPTR_HI,
+ upper_32_bits(ring->wptr << 2));
+ }
+}
+
+/**
+ * sdma_v4_0_page_ring_get_wptr - get the current write pointer
+ *
+ * @ring: amdgpu ring pointer
+ *
+ * Get the current wptr from the hardware (VEGA10+).
+ */
+static uint64_t sdma_v4_0_page_ring_get_wptr(struct amdgpu_ring *ring)
+{
+ struct amdgpu_device *adev = ring->adev;
+ u64 wptr;
+
+ if (ring->use_doorbell) {
+ /* XXX check if swapping is necessary on BE */
+ wptr = READ_ONCE(*((u64 *)&adev->wb.wb[ring->wptr_offs]));
+ } else {
+ wptr = RREG32_SDMA(ring->me, mmSDMA0_PAGE_RB_WPTR_HI);
+ wptr = wptr << 32;
+ wptr |= RREG32_SDMA(ring->me, mmSDMA0_PAGE_RB_WPTR);
+ }
+
+ return wptr >> 2;
+}
+
+/**
+ * sdma_v4_0_ring_set_wptr - commit the write pointer
+ *
+ * @ring: amdgpu ring pointer
+ *
+ * Write the wptr back to the hardware (VEGA10+).
+ */
+static void sdma_v4_0_page_ring_set_wptr(struct amdgpu_ring *ring)
+{
+ struct amdgpu_device *adev = ring->adev;
+
+ if (ring->use_doorbell) {
+ u64 *wb = (u64 *)&adev->wb.wb[ring->wptr_offs];
+
+ /* XXX check if swapping is necessary on BE */
+ WRITE_ONCE(*wb, (ring->wptr << 2));
+ WDOORBELL64(ring->doorbell_index, ring->wptr << 2);
+ } else {
+ uint64_t wptr = ring->wptr << 2;
+
+ WREG32_SDMA(ring->me, mmSDMA0_PAGE_RB_WPTR,
+ lower_32_bits(wptr));
+ WREG32_SDMA(ring->me, mmSDMA0_PAGE_RB_WPTR_HI,
+ upper_32_bits(wptr));
}
}
static void sdma_v4_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
{
- struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
+ struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
int i;
for (i = 0; i < count; i++)
@@ -444,9 +497,12 @@ static void sdma_v4_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
* Schedule an IB in the DMA ring (VEGA10).
*/
static void sdma_v4_0_ring_emit_ib(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
/* IB packet must end on a 8 DW boundary */
sdma_v4_0_ring_insert_nop(ring, (10 - (lower_32_bits(ring->wptr) & 7)) % 8);
@@ -568,16 +624,16 @@ static void sdma_v4_0_gfx_stop(struct amdgpu_device *adev)
amdgpu_ttm_set_buffer_funcs_status(adev, false);
for (i = 0; i < adev->sdma.num_instances; i++) {
- rb_cntl = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
+ rb_cntl = RREG32_SDMA(i, mmSDMA0_GFX_RB_CNTL);
rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 0);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
- ib_cntl = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL));
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_CNTL, rb_cntl);
+ ib_cntl = RREG32_SDMA(i, mmSDMA0_GFX_IB_CNTL);
ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 0);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);
+ WREG32_SDMA(i, mmSDMA0_GFX_IB_CNTL, ib_cntl);
}
- sdma0->ready = false;
- sdma1->ready = false;
+ sdma0->sched.ready = false;
+ sdma1->sched.ready = false;
}
/**
@@ -593,6 +649,39 @@ static void sdma_v4_0_rlc_stop(struct amdgpu_device *adev)
}
/**
+ * sdma_v4_0_page_stop - stop the page async dma engines
+ *
+ * @adev: amdgpu_device pointer
+ *
+ * Stop the page async dma ring buffers (VEGA10).
+ */
+static void sdma_v4_0_page_stop(struct amdgpu_device *adev)
+{
+ struct amdgpu_ring *sdma0 = &adev->sdma.instance[0].page;
+ struct amdgpu_ring *sdma1 = &adev->sdma.instance[1].page;
+ u32 rb_cntl, ib_cntl;
+ int i;
+
+ if ((adev->mman.buffer_funcs_ring == sdma0) ||
+ (adev->mman.buffer_funcs_ring == sdma1))
+ amdgpu_ttm_set_buffer_funcs_status(adev, false);
+
+ for (i = 0; i < adev->sdma.num_instances; i++) {
+ rb_cntl = RREG32_SDMA(i, mmSDMA0_PAGE_RB_CNTL);
+ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_PAGE_RB_CNTL,
+ RB_ENABLE, 0);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_CNTL, rb_cntl);
+ ib_cntl = RREG32_SDMA(i, mmSDMA0_PAGE_IB_CNTL);
+ ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_PAGE_IB_CNTL,
+ IB_ENABLE, 0);
+ WREG32_SDMA(i, mmSDMA0_PAGE_IB_CNTL, ib_cntl);
+ }
+
+ sdma0->sched.ready = false;
+ sdma1->sched.ready = false;
+}
+
+/**
* sdma_v_0_ctx_switch_enable - stop the async dma engines context switch
*
* @adev: amdgpu_device pointer
@@ -630,18 +719,15 @@ static void sdma_v4_0_ctx_switch_enable(struct amdgpu_device *adev, bool enable)
}
for (i = 0; i < adev->sdma.num_instances; i++) {
- f32_cntl = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_CNTL));
+ f32_cntl = RREG32_SDMA(i, mmSDMA0_CNTL);
f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_CNTL,
AUTO_CTXSW_ENABLE, enable ? 1 : 0);
if (enable && amdgpu_sdma_phase_quantum) {
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_PHASE0_QUANTUM),
- phase_quantum);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_PHASE1_QUANTUM),
- phase_quantum);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_PHASE2_QUANTUM),
- phase_quantum);
+ WREG32_SDMA(i, mmSDMA0_PHASE0_QUANTUM, phase_quantum);
+ WREG32_SDMA(i, mmSDMA0_PHASE1_QUANTUM, phase_quantum);
+ WREG32_SDMA(i, mmSDMA0_PHASE2_QUANTUM, phase_quantum);
}
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_CNTL), f32_cntl);
+ WREG32_SDMA(i, mmSDMA0_CNTL, f32_cntl);
}
}
@@ -662,156 +748,217 @@ static void sdma_v4_0_enable(struct amdgpu_device *adev, bool enable)
if (enable == false) {
sdma_v4_0_gfx_stop(adev);
sdma_v4_0_rlc_stop(adev);
+ if (adev->sdma.has_page_queue)
+ sdma_v4_0_page_stop(adev);
}
for (i = 0; i < adev->sdma.num_instances; i++) {
- f32_cntl = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL));
+ f32_cntl = RREG32_SDMA(i, mmSDMA0_F32_CNTL);
f32_cntl = REG_SET_FIELD(f32_cntl, SDMA0_F32_CNTL, HALT, enable ? 0 : 1);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), f32_cntl);
+ WREG32_SDMA(i, mmSDMA0_F32_CNTL, f32_cntl);
}
}
/**
+ * sdma_v4_0_rb_cntl - get parameters for rb_cntl
+ */
+static uint32_t sdma_v4_0_rb_cntl(struct amdgpu_ring *ring, uint32_t rb_cntl)
+{
+ /* Set ring buffer size in dwords */
+ uint32_t rb_bufsz = order_base_2(ring->ring_size / 4);
+
+ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz);
+#ifdef __BIG_ENDIAN
+ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1);
+ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL,
+ RPTR_WRITEBACK_SWAP_ENABLE, 1);
+#endif
+ return rb_cntl;
+}
+
+/**
* sdma_v4_0_gfx_resume - setup and start the async dma engines
*
* @adev: amdgpu_device pointer
+ * @i: instance to resume
*
* Set up the gfx DMA ring buffers and enable them (VEGA10).
* Returns 0 for success, error for failure.
*/
-static int sdma_v4_0_gfx_resume(struct amdgpu_device *adev)
+static void sdma_v4_0_gfx_resume(struct amdgpu_device *adev, unsigned int i)
{
- struct amdgpu_ring *ring;
+ struct amdgpu_ring *ring = &adev->sdma.instance[i].ring;
u32 rb_cntl, ib_cntl, wptr_poll_cntl;
- u32 rb_bufsz;
u32 wb_offset;
u32 doorbell;
u32 doorbell_offset;
- u32 temp;
u64 wptr_gpu_addr;
- int i, r;
- for (i = 0; i < adev->sdma.num_instances; i++) {
- ring = &adev->sdma.instance[i].ring;
- wb_offset = (ring->rptr_offs * 4);
+ wb_offset = (ring->rptr_offs * 4);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL), 0);
+ rb_cntl = RREG32_SDMA(i, mmSDMA0_GFX_RB_CNTL);
+ rb_cntl = sdma_v4_0_rb_cntl(ring, rb_cntl);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_CNTL, rb_cntl);
- /* Set ring buffer size in dwords */
- rb_bufsz = order_base_2(ring->ring_size / 4);
- rb_cntl = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL));
- rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SIZE, rb_bufsz);
-#ifdef __BIG_ENDIAN
- rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_SWAP_ENABLE, 1);
- rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL,
- RPTR_WRITEBACK_SWAP_ENABLE, 1);
-#endif
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
+ /* Initialize the ring buffer's read and write pointers */
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_RPTR, 0);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_RPTR_HI, 0);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR, 0);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_HI, 0);
- /* Initialize the ring buffer's read and write pointers */
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR), 0);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_HI), 0);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), 0);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), 0);
+ /* set the wb address whether it's enabled or not */
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_RPTR_ADDR_HI,
+ upper_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFF);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_RPTR_ADDR_LO,
+ lower_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFC);
- /* set the wb address whether it's enabled or not */
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_HI),
- upper_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFF);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_RPTR_ADDR_LO),
- lower_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFC);
+ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL,
+ RPTR_WRITEBACK_ENABLE, 1);
- rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_BASE, ring->gpu_addr >> 8);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_BASE_HI, ring->gpu_addr >> 40);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE), ring->gpu_addr >> 8);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_BASE_HI), ring->gpu_addr >> 40);
+ ring->wptr = 0;
- ring->wptr = 0;
+ /* before programing wptr to a less value, need set minor_ptr_update first */
+ WREG32_SDMA(i, mmSDMA0_GFX_MINOR_PTR_UPDATE, 1);
- /* before programing wptr to a less value, need set minor_ptr_update first */
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 1);
+ doorbell = RREG32_SDMA(i, mmSDMA0_GFX_DOORBELL);
+ doorbell_offset = RREG32_SDMA(i, mmSDMA0_GFX_DOORBELL_OFFSET);
- if (!amdgpu_sriov_vf(adev)) { /* only bare-metal use register write for wptr */
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR), lower_32_bits(ring->wptr) << 2);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_HI), upper_32_bits(ring->wptr) << 2);
- }
-
- doorbell = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL));
- doorbell_offset = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET));
-
- if (ring->use_doorbell) {
- doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 1);
- doorbell_offset = REG_SET_FIELD(doorbell_offset, SDMA0_GFX_DOORBELL_OFFSET,
+ doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE,
+ ring->use_doorbell);
+ doorbell_offset = REG_SET_FIELD(doorbell_offset,
+ SDMA0_GFX_DOORBELL_OFFSET,
OFFSET, ring->doorbell_index);
- } else {
- doorbell = REG_SET_FIELD(doorbell, SDMA0_GFX_DOORBELL, ENABLE, 0);
- }
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL), doorbell);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_DOORBELL_OFFSET), doorbell_offset);
- adev->nbio_funcs->sdma_doorbell_range(adev, i, ring->use_doorbell,
- ring->doorbell_index);
+ WREG32_SDMA(i, mmSDMA0_GFX_DOORBELL, doorbell);
+ WREG32_SDMA(i, mmSDMA0_GFX_DOORBELL_OFFSET, doorbell_offset);
+ adev->nbio_funcs->sdma_doorbell_range(adev, i, ring->use_doorbell,
+ ring->doorbell_index);
- if (amdgpu_sriov_vf(adev))
- sdma_v4_0_ring_set_wptr(ring);
+ sdma_v4_0_ring_set_wptr(ring);
- /* set minor_ptr_update to 0 after wptr programed */
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_MINOR_PTR_UPDATE), 0);
+ /* set minor_ptr_update to 0 after wptr programed */
+ WREG32_SDMA(i, mmSDMA0_GFX_MINOR_PTR_UPDATE, 0);
- /* set utc l1 enable flag always to 1 */
- temp = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_CNTL));
- temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_CNTL), temp);
+ /* setup the wptr shadow polling */
+ wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO,
+ lower_32_bits(wptr_gpu_addr));
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI,
+ upper_32_bits(wptr_gpu_addr));
+ wptr_poll_cntl = RREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL);
+ wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl,
+ SDMA0_GFX_RB_WPTR_POLL_CNTL,
+ F32_POLL_ENABLE, amdgpu_sriov_vf(adev));
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL, wptr_poll_cntl);
- if (!amdgpu_sriov_vf(adev)) {
- /* unhalt engine */
- temp = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL));
- temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_F32_CNTL), temp);
- }
+ /* enable DMA RB */
+ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1);
+ WREG32_SDMA(i, mmSDMA0_GFX_RB_CNTL, rb_cntl);
- /* setup the wptr shadow polling */
- wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_LO),
- lower_32_bits(wptr_gpu_addr));
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_ADDR_HI),
- upper_32_bits(wptr_gpu_addr));
- wptr_poll_cntl = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL));
- if (amdgpu_sriov_vf(adev))
- wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, SDMA0_GFX_RB_WPTR_POLL_CNTL, F32_POLL_ENABLE, 1);
- else
- wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl, SDMA0_GFX_RB_WPTR_POLL_CNTL, F32_POLL_ENABLE, 0);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_WPTR_POLL_CNTL), wptr_poll_cntl);
-
- /* enable DMA RB */
- rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_GFX_RB_CNTL, RB_ENABLE, 1);
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_RB_CNTL), rb_cntl);
-
- ib_cntl = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL));
- ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 1);
+ ib_cntl = RREG32_SDMA(i, mmSDMA0_GFX_IB_CNTL);
+ ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_ENABLE, 1);
#ifdef __BIG_ENDIAN
- ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_SWAP_ENABLE, 1);
+ ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_GFX_IB_CNTL, IB_SWAP_ENABLE, 1);
#endif
- /* enable DMA IBs */
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_GFX_IB_CNTL), ib_cntl);
+ /* enable DMA IBs */
+ WREG32_SDMA(i, mmSDMA0_GFX_IB_CNTL, ib_cntl);
- ring->ready = true;
+ ring->sched.ready = true;
+}
- if (amdgpu_sriov_vf(adev)) { /* bare-metal sequence doesn't need below to lines */
- sdma_v4_0_ctx_switch_enable(adev, true);
- sdma_v4_0_enable(adev, true);
- }
+/**
+ * sdma_v4_0_page_resume - setup and start the async dma engines
+ *
+ * @adev: amdgpu_device pointer
+ * @i: instance to resume
+ *
+ * Set up the page DMA ring buffers and enable them (VEGA10).
+ * Returns 0 for success, error for failure.
+ */
+static void sdma_v4_0_page_resume(struct amdgpu_device *adev, unsigned int i)
+{
+ struct amdgpu_ring *ring = &adev->sdma.instance[i].page;
+ u32 rb_cntl, ib_cntl, wptr_poll_cntl;
+ u32 wb_offset;
+ u32 doorbell;
+ u32 doorbell_offset;
+ u64 wptr_gpu_addr;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
- return r;
- }
+ wb_offset = (ring->rptr_offs * 4);
- if (adev->mman.buffer_funcs_ring == ring)
- amdgpu_ttm_set_buffer_funcs_status(adev, true);
+ rb_cntl = RREG32_SDMA(i, mmSDMA0_PAGE_RB_CNTL);
+ rb_cntl = sdma_v4_0_rb_cntl(ring, rb_cntl);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_CNTL, rb_cntl);
- }
+ /* Initialize the ring buffer's read and write pointers */
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_RPTR, 0);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_RPTR_HI, 0);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR, 0);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_HI, 0);
- return 0;
+ /* set the wb address whether it's enabled or not */
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_RPTR_ADDR_HI,
+ upper_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFF);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_RPTR_ADDR_LO,
+ lower_32_bits(adev->wb.gpu_addr + wb_offset) & 0xFFFFFFFC);
+
+ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_PAGE_RB_CNTL,
+ RPTR_WRITEBACK_ENABLE, 1);
+
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_BASE, ring->gpu_addr >> 8);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_BASE_HI, ring->gpu_addr >> 40);
+
+ ring->wptr = 0;
+
+ /* before programing wptr to a less value, need set minor_ptr_update first */
+ WREG32_SDMA(i, mmSDMA0_PAGE_MINOR_PTR_UPDATE, 1);
+
+ doorbell = RREG32_SDMA(i, mmSDMA0_PAGE_DOORBELL);
+ doorbell_offset = RREG32_SDMA(i, mmSDMA0_PAGE_DOORBELL_OFFSET);
+
+ doorbell = REG_SET_FIELD(doorbell, SDMA0_PAGE_DOORBELL, ENABLE,
+ ring->use_doorbell);
+ doorbell_offset = REG_SET_FIELD(doorbell_offset,
+ SDMA0_PAGE_DOORBELL_OFFSET,
+ OFFSET, ring->doorbell_index);
+ WREG32_SDMA(i, mmSDMA0_PAGE_DOORBELL, doorbell);
+ WREG32_SDMA(i, mmSDMA0_PAGE_DOORBELL_OFFSET, doorbell_offset);
+ /* TODO: enable doorbell support */
+ /*adev->nbio_funcs->sdma_doorbell_range(adev, i, ring->use_doorbell,
+ ring->doorbell_index);*/
+
+ sdma_v4_0_ring_set_wptr(ring);
+
+ /* set minor_ptr_update to 0 after wptr programed */
+ WREG32_SDMA(i, mmSDMA0_PAGE_MINOR_PTR_UPDATE, 0);
+
+ /* setup the wptr shadow polling */
+ wptr_gpu_addr = adev->wb.gpu_addr + (ring->wptr_offs * 4);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_ADDR_LO,
+ lower_32_bits(wptr_gpu_addr));
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_ADDR_HI,
+ upper_32_bits(wptr_gpu_addr));
+ wptr_poll_cntl = RREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL);
+ wptr_poll_cntl = REG_SET_FIELD(wptr_poll_cntl,
+ SDMA0_PAGE_RB_WPTR_POLL_CNTL,
+ F32_POLL_ENABLE, amdgpu_sriov_vf(adev));
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_WPTR_POLL_CNTL, wptr_poll_cntl);
+
+ /* enable DMA RB */
+ rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_PAGE_RB_CNTL, RB_ENABLE, 1);
+ WREG32_SDMA(i, mmSDMA0_PAGE_RB_CNTL, rb_cntl);
+
+ ib_cntl = RREG32_SDMA(i, mmSDMA0_PAGE_IB_CNTL);
+ ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_PAGE_IB_CNTL, IB_ENABLE, 1);
+#ifdef __BIG_ENDIAN
+ ib_cntl = REG_SET_FIELD(ib_cntl, SDMA0_PAGE_IB_CNTL, IB_SWAP_ENABLE, 1);
+#endif
+ /* enable DMA IBs */
+ WREG32_SDMA(i, mmSDMA0_PAGE_IB_CNTL, ib_cntl);
+
+ ring->sched.ready = true;
}
static void
@@ -922,12 +1069,14 @@ static int sdma_v4_0_load_microcode(struct amdgpu_device *adev)
(adev->sdma.instance[i].fw->data +
le32_to_cpu(hdr->header.ucode_array_offset_bytes));
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_UCODE_ADDR), 0);
+ WREG32_SDMA(i, mmSDMA0_UCODE_ADDR, 0);
for (j = 0; j < fw_size; j++)
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_UCODE_DATA), le32_to_cpup(fw_data++));
+ WREG32_SDMA(i, mmSDMA0_UCODE_DATA,
+ le32_to_cpup(fw_data++));
- WREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_UCODE_ADDR), adev->sdma.instance[i].fw_version);
+ WREG32_SDMA(i, mmSDMA0_UCODE_ADDR,
+ adev->sdma.instance[i].fw_version);
}
return 0;
@@ -943,33 +1092,78 @@ static int sdma_v4_0_load_microcode(struct amdgpu_device *adev)
*/
static int sdma_v4_0_start(struct amdgpu_device *adev)
{
- int r = 0;
+ struct amdgpu_ring *ring;
+ int i, r;
if (amdgpu_sriov_vf(adev)) {
sdma_v4_0_ctx_switch_enable(adev, false);
sdma_v4_0_enable(adev, false);
+ } else {
- /* set RB registers */
- r = sdma_v4_0_gfx_resume(adev);
- return r;
+ if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
+ r = sdma_v4_0_load_microcode(adev);
+ if (r)
+ return r;
+ }
+
+ /* unhalt the MEs */
+ sdma_v4_0_enable(adev, true);
+ /* enable sdma ring preemption */
+ sdma_v4_0_ctx_switch_enable(adev, true);
}
- if (adev->firmware.load_type != AMDGPU_FW_LOAD_PSP) {
- r = sdma_v4_0_load_microcode(adev);
+ /* start the gfx rings and rlc compute queues */
+ for (i = 0; i < adev->sdma.num_instances; i++) {
+ uint32_t temp;
+
+ WREG32_SDMA(i, mmSDMA0_SEM_WAIT_FAIL_TIMER_CNTL, 0);
+ sdma_v4_0_gfx_resume(adev, i);
+ if (adev->sdma.has_page_queue)
+ sdma_v4_0_page_resume(adev, i);
+
+ /* set utc l1 enable flag always to 1 */
+ temp = RREG32_SDMA(i, mmSDMA0_CNTL);
+ temp = REG_SET_FIELD(temp, SDMA0_CNTL, UTC_L1_ENABLE, 1);
+ WREG32_SDMA(i, mmSDMA0_CNTL, temp);
+
+ if (!amdgpu_sriov_vf(adev)) {
+ /* unhalt engine */
+ temp = RREG32_SDMA(i, mmSDMA0_F32_CNTL);
+ temp = REG_SET_FIELD(temp, SDMA0_F32_CNTL, HALT, 0);
+ WREG32_SDMA(i, mmSDMA0_F32_CNTL, temp);
+ }
+ }
+
+ if (amdgpu_sriov_vf(adev)) {
+ sdma_v4_0_ctx_switch_enable(adev, true);
+ sdma_v4_0_enable(adev, true);
+ } else {
+ r = sdma_v4_0_rlc_resume(adev);
if (r)
return r;
}
- /* unhalt the MEs */
- sdma_v4_0_enable(adev, true);
- /* enable sdma ring preemption */
- sdma_v4_0_ctx_switch_enable(adev, true);
+ for (i = 0; i < adev->sdma.num_instances; i++) {
+ ring = &adev->sdma.instance[i].ring;
- /* start the gfx rings and rlc compute queues */
- r = sdma_v4_0_gfx_resume(adev);
- if (r)
- return r;
- r = sdma_v4_0_rlc_resume(adev);
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
+ return r;
+
+ if (adev->sdma.has_page_queue) {
+ struct amdgpu_ring *page = &adev->sdma.instance[i].page;
+
+ r = amdgpu_ring_test_helper(page);
+ if (r)
+ return r;
+
+ if (adev->mman.buffer_funcs_ring == page)
+ amdgpu_ttm_set_buffer_funcs_status(adev, true);
+ }
+
+ if (adev->mman.buffer_funcs_ring == ring)
+ amdgpu_ttm_set_buffer_funcs_status(adev, true);
+ }
return r;
}
@@ -993,21 +1187,16 @@ static int sdma_v4_0_ring_test_ring(struct amdgpu_ring *ring)
u64 gpu_addr;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
r = amdgpu_ring_alloc(ring, 5);
- if (r) {
- DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
- amdgpu_device_wb_free(adev, index);
- return r;
- }
+ if (r)
+ goto error_free_wb;
amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR));
@@ -1024,15 +1213,11 @@ static int sdma_v4_0_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n", ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
- amdgpu_device_wb_free(adev, index);
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+error_free_wb:
+ amdgpu_device_wb_free(adev, index);
return r;
}
@@ -1055,20 +1240,16 @@ static int sdma_v4_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
u64 gpu_addr;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 256, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err0;
- }
ib.ptr[0] = SDMA_PKT_HEADER_OP(SDMA_OP_WRITE) |
SDMA_PKT_HEADER_SUB_OP(SDMA_SUBOP_WRITE_LINEAR);
@@ -1087,21 +1268,17 @@ static int sdma_v4_0_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out\n");
r = -ETIMEDOUT;
goto err1;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto err1;
}
tmp = le32_to_cpu(adev->wb.wb[index]);
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ if (tmp == 0xDEADBEEF)
r = 0;
- } else {
- DRM_ERROR("amdgpu: ib test failed (0x%08X)\n", tmp);
+ else
r = -EINVAL;
- }
+
err1:
amdgpu_ib_free(adev, &ib, NULL);
dma_fence_put(f);
@@ -1206,7 +1383,7 @@ static void sdma_v4_0_vm_set_pte_pde(struct amdgpu_ib *ib,
*/
static void sdma_v4_0_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib)
{
- struct amdgpu_sdma_instance *sdma = amdgpu_get_sdma_instance(ring);
+ struct amdgpu_sdma_instance *sdma = amdgpu_sdma_get_instance_from_ring(ring);
u32 pad_count;
int i;
@@ -1276,10 +1453,18 @@ static int sdma_v4_0_early_init(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
- if (adev->asic_type == CHIP_RAVEN)
+ if (adev->asic_type == CHIP_RAVEN) {
adev->sdma.num_instances = 1;
- else
+ adev->sdma.has_page_queue = false;
+ } else {
adev->sdma.num_instances = 2;
+ /* TODO: Page queue breaks driver reload under SRIOV */
+ if ((adev->asic_type == CHIP_VEGA10) && amdgpu_sriov_vf((adev)))
+ adev->sdma.has_page_queue = false;
+ else if (adev->asic_type != CHIP_VEGA20 &&
+ adev->asic_type != CHIP_VEGA12)
+ adev->sdma.has_page_queue = true;
+ }
sdma_v4_0_set_ring_funcs(adev);
sdma_v4_0_set_buffer_funcs(adev);
@@ -1340,6 +1525,21 @@ static int sdma_v4_0_sw_init(void *handle)
AMDGPU_SDMA_IRQ_TRAP1);
if (r)
return r;
+
+ if (adev->sdma.has_page_queue) {
+ ring = &adev->sdma.instance[i].page;
+ ring->ring_obj = NULL;
+ ring->use_doorbell = false;
+
+ sprintf(ring->name, "page%d", i);
+ r = amdgpu_ring_init(adev, ring, 1024,
+ &adev->sdma.trap_irq,
+ (i == 0) ?
+ AMDGPU_SDMA_IRQ_TRAP0 :
+ AMDGPU_SDMA_IRQ_TRAP1);
+ if (r)
+ return r;
+ }
}
return r;
@@ -1350,8 +1550,11 @@ static int sdma_v4_0_sw_fini(void *handle)
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
int i;
- for (i = 0; i < adev->sdma.num_instances; i++)
+ for (i = 0; i < adev->sdma.num_instances; i++) {
amdgpu_ring_fini(&adev->sdma.instance[i].ring);
+ if (adev->sdma.has_page_queue)
+ amdgpu_ring_fini(&adev->sdma.instance[i].page);
+ }
for (i = 0; i < adev->sdma.num_instances; i++) {
release_firmware(adev->sdma.instance[i].fw);
@@ -1414,7 +1617,7 @@ static bool sdma_v4_0_is_idle(void *handle)
u32 i;
for (i = 0; i < adev->sdma.num_instances; i++) {
- u32 tmp = RREG32(sdma_v4_0_get_reg_offset(adev, i, mmSDMA0_STATUS_REG));
+ u32 tmp = RREG32_SDMA(i, mmSDMA0_STATUS_REG);
if (!(tmp & SDMA0_STATUS_REG__IDLE_MASK))
return false;
@@ -1430,8 +1633,8 @@ static int sdma_v4_0_wait_for_idle(void *handle)
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
for (i = 0; i < adev->usec_timeout; i++) {
- sdma0 = RREG32(sdma_v4_0_get_reg_offset(adev, 0, mmSDMA0_STATUS_REG));
- sdma1 = RREG32(sdma_v4_0_get_reg_offset(adev, 1, mmSDMA0_STATUS_REG));
+ sdma0 = RREG32_SDMA(0, mmSDMA0_STATUS_REG);
+ sdma1 = RREG32_SDMA(1, mmSDMA0_STATUS_REG);
if (sdma0 & sdma1 & SDMA0_STATUS_REG__IDLE_MASK)
return 0;
@@ -1452,16 +1655,13 @@ static int sdma_v4_0_set_trap_irq_state(struct amdgpu_device *adev,
unsigned type,
enum amdgpu_interrupt_state state)
{
+ unsigned int instance = (type == AMDGPU_SDMA_IRQ_TRAP0) ? 0 : 1;
u32 sdma_cntl;
- u32 reg_offset = (type == AMDGPU_SDMA_IRQ_TRAP0) ?
- sdma_v4_0_get_reg_offset(adev, 0, mmSDMA0_CNTL) :
- sdma_v4_0_get_reg_offset(adev, 1, mmSDMA0_CNTL);
-
- sdma_cntl = RREG32(reg_offset);
+ sdma_cntl = RREG32_SDMA(instance, mmSDMA0_CNTL);
sdma_cntl = REG_SET_FIELD(sdma_cntl, SDMA0_CNTL, TRAP_ENABLE,
state == AMDGPU_IRQ_STATE_ENABLE ? 1 : 0);
- WREG32(reg_offset, sdma_cntl);
+ WREG32_SDMA(instance, mmSDMA0_CNTL, sdma_cntl);
return 0;
}
@@ -1470,39 +1670,32 @@ static int sdma_v4_0_process_trap_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
+ uint32_t instance;
+
DRM_DEBUG("IH: SDMA trap\n");
switch (entry->client_id) {
case SOC15_IH_CLIENTID_SDMA0:
- switch (entry->ring_id) {
- case 0:
- amdgpu_fence_process(&adev->sdma.instance[0].ring);
- break;
- case 1:
- /* XXX compute */
- break;
- case 2:
- /* XXX compute */
- break;
- case 3:
- /* XXX page queue*/
- break;
- }
+ instance = 0;
break;
case SOC15_IH_CLIENTID_SDMA1:
- switch (entry->ring_id) {
- case 0:
- amdgpu_fence_process(&adev->sdma.instance[1].ring);
- break;
- case 1:
- /* XXX compute */
- break;
- case 2:
- /* XXX compute */
- break;
- case 3:
- /* XXX page queue*/
- break;
- }
+ instance = 1;
+ break;
+ default:
+ return 0;
+ }
+
+ switch (entry->ring_id) {
+ case 0:
+ amdgpu_fence_process(&adev->sdma.instance[instance].ring);
+ break;
+ case 1:
+ /* XXX compute */
+ break;
+ case 2:
+ /* XXX compute */
+ break;
+ case 3:
+ amdgpu_fence_process(&adev->sdma.instance[instance].page);
break;
}
return 0;
@@ -1512,12 +1705,29 @@ static int sdma_v4_0_process_illegal_inst_irq(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
struct amdgpu_iv_entry *entry)
{
+ int instance;
+
DRM_ERROR("Illegal instruction in SDMA command stream\n");
- schedule_work(&adev->reset_work);
+
+ switch (entry->client_id) {
+ case SOC15_IH_CLIENTID_SDMA0:
+ instance = 0;
+ break;
+ case SOC15_IH_CLIENTID_SDMA1:
+ instance = 1;
+ break;
+ default:
+ return 0;
+ }
+
+ switch (entry->ring_id) {
+ case 0:
+ drm_sched_fault(&adev->sdma.instance[instance].ring.sched);
+ break;
+ }
return 0;
}
-
static void sdma_v4_0_update_medium_grain_clock_gating(
struct amdgpu_device *adev,
bool enable)
@@ -1730,6 +1940,38 @@ static const struct amdgpu_ring_funcs sdma_v4_0_ring_funcs = {
.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
};
+static const struct amdgpu_ring_funcs sdma_v4_0_page_ring_funcs = {
+ .type = AMDGPU_RING_TYPE_SDMA,
+ .align_mask = 0xf,
+ .nop = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP),
+ .support_64bit_ptrs = true,
+ .vmhub = AMDGPU_MMHUB,
+ .get_rptr = sdma_v4_0_ring_get_rptr,
+ .get_wptr = sdma_v4_0_page_ring_get_wptr,
+ .set_wptr = sdma_v4_0_page_ring_set_wptr,
+ .emit_frame_size =
+ 6 + /* sdma_v4_0_ring_emit_hdp_flush */
+ 3 + /* hdp invalidate */
+ 6 + /* sdma_v4_0_ring_emit_pipeline_sync */
+ /* sdma_v4_0_ring_emit_vm_flush */
+ SOC15_FLUSH_GPU_TLB_NUM_WREG * 3 +
+ SOC15_FLUSH_GPU_TLB_NUM_REG_WAIT * 6 +
+ 10 + 10 + 10, /* sdma_v4_0_ring_emit_fence x3 for user fence, vm fence */
+ .emit_ib_size = 7 + 6, /* sdma_v4_0_ring_emit_ib */
+ .emit_ib = sdma_v4_0_ring_emit_ib,
+ .emit_fence = sdma_v4_0_ring_emit_fence,
+ .emit_pipeline_sync = sdma_v4_0_ring_emit_pipeline_sync,
+ .emit_vm_flush = sdma_v4_0_ring_emit_vm_flush,
+ .emit_hdp_flush = sdma_v4_0_ring_emit_hdp_flush,
+ .test_ring = sdma_v4_0_ring_test_ring,
+ .test_ib = sdma_v4_0_ring_test_ib,
+ .insert_nop = sdma_v4_0_ring_insert_nop,
+ .pad_ib = sdma_v4_0_ring_pad_ib,
+ .emit_wreg = sdma_v4_0_ring_emit_wreg,
+ .emit_reg_wait = sdma_v4_0_ring_emit_reg_wait,
+ .emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
+};
+
static void sdma_v4_0_set_ring_funcs(struct amdgpu_device *adev)
{
int i;
@@ -1737,6 +1979,10 @@ static void sdma_v4_0_set_ring_funcs(struct amdgpu_device *adev)
for (i = 0; i < adev->sdma.num_instances; i++) {
adev->sdma.instance[i].ring.funcs = &sdma_v4_0_ring_funcs;
adev->sdma.instance[i].ring.me = i;
+ if (adev->sdma.has_page_queue) {
+ adev->sdma.instance[i].page.funcs = &sdma_v4_0_page_ring_funcs;
+ adev->sdma.instance[i].page.me = i;
+ }
}
}
@@ -1818,7 +2064,10 @@ static const struct amdgpu_buffer_funcs sdma_v4_0_buffer_funcs = {
static void sdma_v4_0_set_buffer_funcs(struct amdgpu_device *adev)
{
adev->mman.buffer_funcs = &sdma_v4_0_buffer_funcs;
- adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].ring;
+ if (adev->sdma.has_page_queue)
+ adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].page;
+ else
+ adev->mman.buffer_funcs_ring = &adev->sdma.instance[0].ring;
}
static const struct amdgpu_vm_pte_funcs sdma_v4_0_vm_pte_funcs = {
@@ -1836,7 +2085,10 @@ static void sdma_v4_0_set_vm_pte_funcs(struct amdgpu_device *adev)
adev->vm_manager.vm_pte_funcs = &sdma_v4_0_vm_pte_funcs;
for (i = 0; i < adev->sdma.num_instances; i++) {
- sched = &adev->sdma.instance[i].ring.sched;
+ if (adev->sdma.has_page_queue)
+ sched = &adev->sdma.instance[i].page.sched;
+ else
+ sched = &adev->sdma.instance[i].ring.sched;
adev->vm_manager.vm_pte_rqs[i] =
&sched->sched_rq[DRM_SCHED_PRIORITY_KERNEL];
}
diff --git a/drivers/gpu/drm/amd/amdgpu/si_dma.c b/drivers/gpu/drm/amd/amdgpu/si_dma.c
index adbaea6..b6e4731 100644
--- a/drivers/gpu/drm/amd/amdgpu/si_dma.c
+++ b/drivers/gpu/drm/amd/amdgpu/si_dma.c
@@ -61,9 +61,11 @@ static void si_dma_ring_set_wptr(struct amdgpu_ring *ring)
}
static void si_dma_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
/* The indirect buffer packet must end on an 8 DW boundary in the DMA ring.
* Pad as necessary with NOPs.
*/
@@ -122,7 +124,7 @@ static void si_dma_stop(struct amdgpu_device *adev)
if (adev->mman.buffer_funcs_ring == ring)
amdgpu_ttm_set_buffer_funcs_status(adev, false);
- ring->ready = false;
+ ring->sched.ready = false;
}
}
@@ -175,13 +177,11 @@ static int si_dma_start(struct amdgpu_device *adev)
WREG32(DMA_RB_WPTR + sdma_offsets[i], lower_32_bits(ring->wptr) << 2);
WREG32(DMA_RB_CNTL + sdma_offsets[i], rb_cntl | DMA_RB_ENABLE);
- ring->ready = true;
+ ring->sched.ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
return r;
- }
if (adev->mman.buffer_funcs_ring == ring)
amdgpu_ttm_set_buffer_funcs_status(adev, true);
@@ -209,21 +209,16 @@ static int si_dma_ring_test_ring(struct amdgpu_ring *ring)
u64 gpu_addr;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%d) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
r = amdgpu_ring_alloc(ring, 4);
- if (r) {
- DRM_ERROR("amdgpu: dma failed to lock ring %d (%d).\n", ring->idx, r);
- amdgpu_device_wb_free(adev, index);
- return r;
- }
+ if (r)
+ goto error_free_wb;
amdgpu_ring_write(ring, DMA_PACKET(DMA_PACKET_WRITE, 0, 0, 0, 1));
amdgpu_ring_write(ring, lower_32_bits(gpu_addr));
@@ -238,15 +233,11 @@ static int si_dma_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n", ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
- amdgpu_device_wb_free(adev, index);
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+error_free_wb:
+ amdgpu_device_wb_free(adev, index);
return r;
}
@@ -269,20 +260,16 @@ static int si_dma_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = amdgpu_device_wb_get(adev, &index);
- if (r) {
- dev_err(adev->dev, "(%ld) failed to allocate wb slot\n", r);
+ if (r)
return r;
- }
gpu_addr = adev->wb.gpu_addr + (index * 4);
tmp = 0xCAFEDEAD;
adev->wb.wb[index] = cpu_to_le32(tmp);
memset(&ib, 0, sizeof(ib));
r = amdgpu_ib_get(adev, NULL, 256, &ib);
- if (r) {
- DRM_ERROR("amdgpu: failed to get ib (%ld).\n", r);
+ if (r)
goto err0;
- }
ib.ptr[0] = DMA_PACKET(DMA_PACKET_WRITE, 0, 0, 0, 1);
ib.ptr[1] = lower_32_bits(gpu_addr);
@@ -295,21 +282,16 @@ static int si_dma_ring_test_ib(struct amdgpu_ring *ring, long timeout)
r = dma_fence_wait_timeout(f, false, timeout);
if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out\n");
r = -ETIMEDOUT;
goto err1;
} else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
goto err1;
}
tmp = le32_to_cpu(adev->wb.wb[index]);
- if (tmp == 0xDEADBEEF) {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ if (tmp == 0xDEADBEEF)
r = 0;
- } else {
- DRM_ERROR("amdgpu: ib test failed (0x%08X)\n", tmp);
+ else
r = -EINVAL;
- }
err1:
amdgpu_ib_free(adev, &ib, NULL);
@@ -658,15 +640,6 @@ static int si_dma_process_trap_irq(struct amdgpu_device *adev,
return 0;
}
-static int si_dma_process_illegal_inst_irq(struct amdgpu_device *adev,
- struct amdgpu_irq_src *source,
- struct amdgpu_iv_entry *entry)
-{
- DRM_ERROR("Illegal instruction in SDMA command stream\n");
- schedule_work(&adev->reset_work);
- return 0;
-}
-
static int si_dma_set_clockgating_state(void *handle,
enum amd_clockgating_state state)
{
@@ -781,15 +754,10 @@ static const struct amdgpu_irq_src_funcs si_dma_trap_irq_funcs = {
.process = si_dma_process_trap_irq,
};
-static const struct amdgpu_irq_src_funcs si_dma_illegal_inst_irq_funcs = {
- .process = si_dma_process_illegal_inst_irq,
-};
-
static void si_dma_set_irq_funcs(struct amdgpu_device *adev)
{
adev->sdma.trap_irq.num_types = AMDGPU_SDMA_IRQ_LAST;
adev->sdma.trap_irq.funcs = &si_dma_trap_irq_funcs;
- adev->sdma.illegal_inst_irq.funcs = &si_dma_illegal_inst_irq_funcs;
}
/**
diff --git a/drivers/gpu/drm/amd/amdgpu/ta_xgmi_if.h b/drivers/gpu/drm/amd/amdgpu/ta_xgmi_if.h
new file mode 100644
index 0000000..ac2c27b
--- /dev/null
+++ b/drivers/gpu/drm/amd/amdgpu/ta_xgmi_if.h
@@ -0,0 +1,130 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef _TA_XGMI_IF_H
+#define _TA_XGMI_IF_H
+
+/* Responses have bit 31 set */
+#define RSP_ID_MASK (1U << 31)
+#define RSP_ID(cmdId) (((uint32_t)(cmdId)) | RSP_ID_MASK)
+
+enum ta_command_xgmi {
+ TA_COMMAND_XGMI__INITIALIZE = 0x00,
+ TA_COMMAND_XGMI__GET_NODE_ID = 0x01,
+ TA_COMMAND_XGMI__GET_HIVE_ID = 0x02,
+ TA_COMMAND_XGMI__GET_GET_TOPOLOGY_INFO = 0x03,
+ TA_COMMAND_XGMI__SET_TOPOLOGY_INFO = 0x04
+};
+
+/* XGMI related enumerations */
+/**********************************************************/;
+enum ta_xgmi_connected_nodes {
+ TA_XGMI__MAX_CONNECTED_NODES = 64
+};
+
+enum ta_xgmi_status {
+ TA_XGMI_STATUS__SUCCESS = 0x00,
+ TA_XGMI_STATUS__GENERIC_FAILURE = 0x01,
+ TA_XGMI_STATUS__NULL_POINTER = 0x02,
+ TA_XGMI_STATUS__INVALID_PARAMETER = 0x03,
+ TA_XGMI_STATUS__NOT_INITIALIZED = 0x04,
+ TA_XGMI_STATUS__INVALID_NODE_NUM = 0x05,
+ TA_XGMI_STATUS__INVALID_NODE_ID = 0x06,
+ TA_XGMI_STATUS__INVALID_TOPOLOGY = 0x07,
+ TA_XGMI_STATUS__FAILED_ID_GEN = 0x08,
+ TA_XGMI_STATUS__FAILED_TOPOLOGY_INIT = 0x09,
+ TA_XGMI_STATUS__SET_SHARING_ERROR = 0x0A
+};
+
+enum ta_xgmi_assigned_sdma_engine {
+ TA_XGMI_ASSIGNED_SDMA_ENGINE__NOT_ASSIGNED = -1,
+ TA_XGMI_ASSIGNED_SDMA_ENGINE__SDMA0 = 0,
+ TA_XGMI_ASSIGNED_SDMA_ENGINE__SDMA1 = 1,
+ TA_XGMI_ASSIGNED_SDMA_ENGINE__SDMA2 = 2,
+ TA_XGMI_ASSIGNED_SDMA_ENGINE__SDMA3 = 3,
+ TA_XGMI_ASSIGNED_SDMA_ENGINE__SDMA4 = 4,
+ TA_XGMI_ASSIGNED_SDMA_ENGINE__SDMA5 = 5
+};
+
+/* input/output structures for XGMI commands */
+/**********************************************************/
+struct ta_xgmi_node_info {
+ uint64_t node_id;
+ uint8_t num_hops;
+ uint8_t is_sharing_enabled;
+ enum ta_xgmi_assigned_sdma_engine sdma_engine;
+};
+
+struct ta_xgmi_cmd_initialize_output {
+ uint32_t status;
+};
+
+struct ta_xgmi_cmd_get_node_id_output {
+ uint64_t node_id;
+};
+
+struct ta_xgmi_cmd_get_hive_id_output {
+ uint64_t hive_id;
+};
+
+struct ta_xgmi_cmd_get_topology_info_input {
+ uint32_t num_nodes;
+ struct ta_xgmi_node_info nodes[TA_XGMI__MAX_CONNECTED_NODES];
+};
+
+struct ta_xgmi_cmd_get_topology_info_output {
+ uint32_t num_nodes;
+ struct ta_xgmi_node_info nodes[TA_XGMI__MAX_CONNECTED_NODES];
+};
+
+struct ta_xgmi_cmd_set_topology_info_input {
+ uint32_t num_nodes;
+ struct ta_xgmi_node_info nodes[TA_XGMI__MAX_CONNECTED_NODES];
+};
+
+/**********************************************************/
+/* Common input structure for XGMI callbacks */
+union ta_xgmi_cmd_input {
+ struct ta_xgmi_cmd_get_topology_info_input get_topology_info;
+ struct ta_xgmi_cmd_set_topology_info_input set_topology_info;
+};
+
+/* Common output structure for XGMI callbacks */
+union ta_xgmi_cmd_output {
+ struct ta_xgmi_cmd_initialize_output initialize;
+ struct ta_xgmi_cmd_get_node_id_output get_node_id;
+ struct ta_xgmi_cmd_get_hive_id_output get_hive_id;
+ struct ta_xgmi_cmd_get_topology_info_output get_topology_info;
+};
+/**********************************************************/
+
+struct ta_xgmi_shared_memory {
+ uint32_t cmd_id;
+ uint32_t resp_id;
+ enum ta_xgmi_status xgmi_status;
+ uint32_t reserved;
+ union ta_xgmi_cmd_input xgmi_in_message;
+ union ta_xgmi_cmd_output xgmi_out_message;
+};
+
+#endif //_TA_XGMI_IF_H
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
index 1fc17bf..90bbcee 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
@@ -162,12 +162,9 @@ static int uvd_v4_2_hw_init(void *handle)
uvd_v4_2_enable_mgcg(adev, true);
amdgpu_asic_set_uvd_clocks(adev, 10000, 10000);
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
r = amdgpu_ring_alloc(ring, 10);
if (r) {
@@ -218,7 +215,7 @@ static int uvd_v4_2_hw_fini(void *handle)
if (RREG32(mmUVD_STATUS) != 0)
uvd_v4_2_stop(adev);
- ring->ready = false;
+ ring->sched.ready = false;
return 0;
}
@@ -484,11 +481,9 @@ static int uvd_v4_2_ring_test_ring(struct amdgpu_ring *ring)
WREG32(mmUVD_CONTEXT_ID, 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n",
- ring->idx, r);
+ if (r)
return r;
- }
+
amdgpu_ring_write(ring, PACKET0(mmUVD_CONTEXT_ID, 0));
amdgpu_ring_write(ring, 0xDEADBEEF);
amdgpu_ring_commit(ring);
@@ -499,14 +494,9 @@ static int uvd_v4_2_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
return r;
}
@@ -519,8 +509,9 @@ static int uvd_v4_2_ring_test_ring(struct amdgpu_ring *ring)
* Write ring commands to execute the indirect buffer
*/
static void uvd_v4_2_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
amdgpu_ring_write(ring, PACKET0(mmUVD_RBC_IB_BASE, 0));
amdgpu_ring_write(ring, ib->gpu_addr);
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
index fde6ad5..1c5e127 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
@@ -158,12 +158,9 @@ static int uvd_v5_0_hw_init(void *handle)
uvd_v5_0_set_clockgating_state(adev, AMD_CG_STATE_UNGATE);
uvd_v5_0_enable_mgcg(adev, true);
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
r = amdgpu_ring_alloc(ring, 10);
if (r) {
@@ -215,7 +212,7 @@ static int uvd_v5_0_hw_fini(void *handle)
if (RREG32(mmUVD_STATUS) != 0)
uvd_v5_0_stop(adev);
- ring->ready = false;
+ ring->sched.ready = false;
return 0;
}
@@ -500,11 +497,8 @@ static int uvd_v5_0_ring_test_ring(struct amdgpu_ring *ring)
WREG32(mmUVD_CONTEXT_ID, 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n",
- ring->idx, r);
+ if (r)
return r;
- }
amdgpu_ring_write(ring, PACKET0(mmUVD_CONTEXT_ID, 0));
amdgpu_ring_write(ring, 0xDEADBEEF);
amdgpu_ring_commit(ring);
@@ -515,14 +509,9 @@ static int uvd_v5_0_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
return r;
}
@@ -535,8 +524,9 @@ static int uvd_v5_0_ring_test_ring(struct amdgpu_ring *ring)
* Write ring commands to execute the indirect buffer
*/
static void uvd_v5_0_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
amdgpu_ring_write(ring, PACKET0(mmUVD_LMI_RBC_IB_64BIT_BAR_LOW, 0));
amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
index 7a5b402..f184842 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
@@ -175,11 +175,8 @@ static int uvd_v6_0_enc_ring_test_ring(struct amdgpu_ring *ring)
int r;
r = amdgpu_ring_alloc(ring, 16);
- if (r) {
- DRM_ERROR("amdgpu: uvd enc failed to lock ring %d (%d).\n",
- ring->idx, r);
+ if (r)
return r;
- }
amdgpu_ring_write(ring, HEVC_ENC_CMD_END);
amdgpu_ring_commit(ring);
@@ -189,14 +186,8 @@ static int uvd_v6_0_enc_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed\n",
- ring->idx);
+ if (i >= adev->usec_timeout)
r = -ETIMEDOUT;
- }
return r;
}
@@ -336,31 +327,24 @@ static int uvd_v6_0_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = uvd_v6_0_enc_get_create_msg(ring, 1, NULL);
- if (r) {
- DRM_ERROR("amdgpu: failed to get create msg (%ld).\n", r);
+ if (r)
goto error;
- }
r = uvd_v6_0_enc_get_destroy_msg(ring, 1, &fence);
- if (r) {
- DRM_ERROR("amdgpu: failed to get destroy ib (%ld).\n", r);
+ if (r)
goto error;
- }
r = dma_fence_wait_timeout(fence, false, timeout);
- if (r == 0) {
- DRM_ERROR("amdgpu: IB test timed out.\n");
+ if (r == 0)
r = -ETIMEDOUT;
- } else if (r < 0) {
- DRM_ERROR("amdgpu: fence wait failed (%ld).\n", r);
- } else {
- DRM_DEBUG("ib test on ring %d succeeded\n", ring->idx);
+ else if (r > 0)
r = 0;
- }
+
error:
dma_fence_put(fence);
return r;
}
+
static int uvd_v6_0_early_init(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -476,12 +460,9 @@ static int uvd_v6_0_hw_init(void *handle)
uvd_v6_0_set_clockgating_state(adev, AMD_CG_STATE_UNGATE);
uvd_v6_0_enable_mgcg(adev, true);
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
r = amdgpu_ring_alloc(ring, 10);
if (r) {
@@ -513,12 +494,9 @@ static int uvd_v6_0_hw_init(void *handle)
if (uvd_v6_0_enc_support(adev)) {
for (i = 0; i < adev->uvd.num_enc_rings; ++i) {
ring = &adev->uvd.inst->ring_enc[i];
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
}
}
@@ -548,7 +526,7 @@ static int uvd_v6_0_hw_fini(void *handle)
if (RREG32(mmUVD_STATUS) != 0)
uvd_v6_0_stop(adev);
- ring->ready = false;
+ ring->sched.ready = false;
return 0;
}
@@ -969,11 +947,9 @@ static int uvd_v6_0_ring_test_ring(struct amdgpu_ring *ring)
WREG32(mmUVD_CONTEXT_ID, 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: cp failed to lock ring %d (%d).\n",
- ring->idx, r);
+ if (r)
return r;
- }
+
amdgpu_ring_write(ring, PACKET0(mmUVD_CONTEXT_ID, 0));
amdgpu_ring_write(ring, 0xDEADBEEF);
amdgpu_ring_commit(ring);
@@ -984,14 +960,9 @@ static int uvd_v6_0_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("ring test on %d succeeded in %d usecs\n",
- ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: ring %d test failed (0x%08X)\n",
- ring->idx, tmp);
- r = -EINVAL;
- }
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
return r;
}
@@ -1004,9 +975,12 @@ static int uvd_v6_0_ring_test_ring(struct amdgpu_ring *ring)
* Write ring commands to execute the indirect buffer
*/
static void uvd_v6_0_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
amdgpu_ring_write(ring, PACKET0(mmUVD_LMI_RBC_IB_VMID, 0));
amdgpu_ring_write(ring, vmid);
@@ -1027,8 +1001,12 @@ static void uvd_v6_0_ring_emit_ib(struct amdgpu_ring *ring,
* Write enc ring commands to execute the indirect buffer
*/
static void uvd_v6_0_enc_ring_emit_ib(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib, unsigned int vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
amdgpu_ring_write(ring, HEVC_ENC_CMD_IB_VM);
amdgpu_ring_write(ring, vmid);
amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
index 58b39af..8a45959 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v7_0.c
@@ -183,11 +183,8 @@ static int uvd_v7_0_enc_ring_test_ring(struct amdgpu_ring *ring)
return 0;
r = amdgpu_ring_alloc(ring, 16);
- if (r) {
- DRM_ERROR("amdgpu: uvd enc failed to lock (%d)ring %d (%d).\n",
- ring->me, ring->idx, r);
+ if (r)
return r;
- }
amdgpu_ring_write(ring, HEVC_ENC_CMD_END);
amdgpu_ring_commit(ring);
@@ -197,14 +194,8 @@ static int uvd_v7_0_enc_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("(%d)ring test on %d succeeded in %d usecs\n",
- ring->me, ring->idx, i);
- } else {
- DRM_ERROR("amdgpu: (%d)ring %d test failed\n",
- ring->me, ring->idx);
+ if (i >= adev->usec_timeout)
r = -ETIMEDOUT;
- }
return r;
}
@@ -343,27 +334,19 @@ static int uvd_v7_0_enc_ring_test_ib(struct amdgpu_ring *ring, long timeout)
long r;
r = uvd_v7_0_enc_get_create_msg(ring, 1, NULL);
- if (r) {
- DRM_ERROR("amdgpu: (%d)failed to get create msg (%ld).\n", ring->me, r);
+ if (r)
goto error;
- }
r = uvd_v7_0_enc_get_destroy_msg(ring, 1, &fence);
- if (r) {
- DRM_ERROR("amdgpu: (%d)failed to get destroy ib (%ld).\n", ring->me, r);
+ if (r)
goto error;
- }
r = dma_fence_wait_timeout(fence, false, timeout);
- if (r == 0) {
- DRM_ERROR("amdgpu: (%d)IB test timed out.\n", ring->me);
+ if (r == 0)
r = -ETIMEDOUT;
- } else if (r < 0) {
- DRM_ERROR("amdgpu: (%d)fence wait failed (%ld).\n", ring->me, r);
- } else {
- DRM_DEBUG("ib test on (%d)ring %d succeeded\n", ring->me, ring->idx);
+ else if (r > 0)
r = 0;
- }
+
error:
dma_fence_put(fence);
return r;
@@ -540,12 +523,9 @@ static int uvd_v7_0_hw_init(void *handle)
ring = &adev->uvd.inst[j].ring;
if (!amdgpu_sriov_vf(adev)) {
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
r = amdgpu_ring_alloc(ring, 10);
if (r) {
@@ -582,12 +562,9 @@ static int uvd_v7_0_hw_init(void *handle)
for (i = 0; i < adev->uvd.num_enc_rings; ++i) {
ring = &adev->uvd.inst[j].ring_enc[i];
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
}
}
done:
@@ -619,7 +596,7 @@ static int uvd_v7_0_hw_fini(void *handle)
for (i = 0; i < adev->uvd.num_uvd_inst; ++i) {
if (adev->uvd.harvest_config & (1 << i))
continue;
- adev->uvd.inst[i].ring.ready = false;
+ adev->uvd.inst[i].ring.sched.ready = false;
}
return 0;
@@ -1235,11 +1212,9 @@ static int uvd_v7_0_ring_test_ring(struct amdgpu_ring *ring)
WREG32_SOC15(UVD, ring->me, mmUVD_CONTEXT_ID, 0xCAFEDEAD);
r = amdgpu_ring_alloc(ring, 3);
- if (r) {
- DRM_ERROR("amdgpu: (%d)cp failed to lock ring %d (%d).\n",
- ring->me, ring->idx, r);
+ if (r)
return r;
- }
+
amdgpu_ring_write(ring,
PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_CONTEXT_ID), 0));
amdgpu_ring_write(ring, 0xDEADBEEF);
@@ -1251,14 +1226,9 @@ static int uvd_v7_0_ring_test_ring(struct amdgpu_ring *ring)
DRM_UDELAY(1);
}
- if (i < adev->usec_timeout) {
- DRM_DEBUG("(%d)ring test on %d succeeded in %d usecs\n",
- ring->me, ring->idx, i);
- } else {
- DRM_ERROR("(%d)amdgpu: ring %d test failed (0x%08X)\n",
- ring->me, ring->idx, tmp);
- r = -EINVAL;
- }
+ if (i >= adev->usec_timeout)
+ r = -ETIMEDOUT;
+
return r;
}
@@ -1300,10 +1270,12 @@ static int uvd_v7_0_ring_patch_cs_in_place(struct amdgpu_cs_parser *p,
* Write ring commands to execute the indirect buffer
*/
static void uvd_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
+ struct amdgpu_job *job,
struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ bool ctx_switch)
{
struct amdgpu_device *adev = ring->adev;
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
amdgpu_ring_write(ring,
PACKET0(SOC15_REG_OFFSET(UVD, ring->me, mmUVD_LMI_RBC_IB_VMID), 0));
@@ -1329,8 +1301,12 @@ static void uvd_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
* Write enc ring commands to execute the indirect buffer
*/
static void uvd_v7_0_enc_ring_emit_ib(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib, unsigned int vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
amdgpu_ring_write(ring, HEVC_ENC_CMD_IB_VM);
amdgpu_ring_write(ring, vmid);
amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
index ea28828..bed78a7 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
@@ -463,15 +463,11 @@ static int vce_v2_0_hw_init(void *handle)
amdgpu_asic_set_vce_clocks(adev, 10000, 10000);
vce_v2_0_enable_mgcg(adev, true, false);
- for (i = 0; i < adev->vce.num_rings; i++)
- adev->vce.ring[i].ready = false;
for (i = 0; i < adev->vce.num_rings; i++) {
- r = amdgpu_ring_test_ring(&adev->vce.ring[i]);
+ r = amdgpu_ring_test_helper(&adev->vce.ring[i]);
if (r)
return r;
- else
- adev->vce.ring[i].ready = true;
}
DRM_INFO("VCE initialized successfully.\n");
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index 6dbd397..3e84840 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -474,15 +474,10 @@ static int vce_v3_0_hw_init(void *handle)
amdgpu_asic_set_vce_clocks(adev, 10000, 10000);
- for (i = 0; i < adev->vce.num_rings; i++)
- adev->vce.ring[i].ready = false;
-
for (i = 0; i < adev->vce.num_rings; i++) {
- r = amdgpu_ring_test_ring(&adev->vce.ring[i]);
+ r = amdgpu_ring_test_helper(&adev->vce.ring[i]);
if (r)
return r;
- else
- adev->vce.ring[i].ready = true;
}
DRM_INFO("VCE initialized successfully.\n");
@@ -838,8 +833,12 @@ static void vce_v3_0_get_clockgating_state(void *handle, u32 *flags)
}
static void vce_v3_0_ring_emit_ib(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib, unsigned int vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
amdgpu_ring_write(ring, VCE_CMD_IB_VM);
amdgpu_ring_write(ring, vmid);
amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
index 1c94718..0054ba1 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
@@ -519,15 +519,10 @@ static int vce_v4_0_hw_init(void *handle)
if (r)
return r;
- for (i = 0; i < adev->vce.num_rings; i++)
- adev->vce.ring[i].ready = false;
-
for (i = 0; i < adev->vce.num_rings; i++) {
- r = amdgpu_ring_test_ring(&adev->vce.ring[i]);
+ r = amdgpu_ring_test_helper(&adev->vce.ring[i]);
if (r)
return r;
- else
- adev->vce.ring[i].ready = true;
}
DRM_INFO("VCE initialized successfully.\n");
@@ -549,7 +544,7 @@ static int vce_v4_0_hw_fini(void *handle)
}
for (i = 0; i < adev->vce.num_rings; i++)
- adev->vce.ring[i].ready = false;
+ adev->vce.ring[i].sched.ready = false;
return 0;
}
@@ -951,9 +946,11 @@ static int vce_v4_0_set_powergating_state(void *handle,
}
#endif
-static void vce_v4_0_ring_emit_ib(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib, unsigned int vmid, bool ctx_switch)
+static void vce_v4_0_ring_emit_ib(struct amdgpu_ring *ring, struct amdgpu_job *job,
+ struct amdgpu_ib *ib, bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
amdgpu_ring_write(ring, VCE_CMD_IB_VM);
amdgpu_ring_write(ring, vmid);
amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
index eae9092..c1a0350 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v1_0.c
@@ -176,30 +176,22 @@ static int vcn_v1_0_hw_init(void *handle)
struct amdgpu_ring *ring = &adev->vcn.ring_dec;
int i, r;
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
for (i = 0; i < adev->vcn.num_enc_rings; ++i) {
ring = &adev->vcn.ring_enc[i];
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ ring->sched.ready = true;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
}
ring = &adev->vcn.ring_jpeg;
- ring->ready = true;
- r = amdgpu_ring_test_ring(ring);
- if (r) {
- ring->ready = false;
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
goto done;
- }
done:
if (!r)
@@ -224,7 +216,7 @@ static int vcn_v1_0_hw_fini(void *handle)
if (RREG32_SOC15(VCN, 0, mmUVD_STATUS))
vcn_v1_0_stop(adev);
- ring->ready = false;
+ ring->sched.ready = false;
return 0;
}
@@ -1366,10 +1358,12 @@ static void vcn_v1_0_dec_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u64
* Write ring commands to execute the indirect buffer
*/
static void vcn_v1_0_dec_ring_emit_ib(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
struct amdgpu_device *adev = ring->adev;
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
amdgpu_ring_write(ring,
PACKET0(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_RBC_IB_VMID), 0));
@@ -1524,8 +1518,12 @@ static void vcn_v1_0_enc_ring_insert_end(struct amdgpu_ring *ring)
* Write enc ring commands to execute the indirect buffer
*/
static void vcn_v1_0_enc_ring_emit_ib(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib, unsigned int vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
+
amdgpu_ring_write(ring, VCN_ENC_CMD_IB);
amdgpu_ring_write(ring, vmid);
amdgpu_ring_write(ring, lower_32_bits(ib->gpu_addr));
@@ -1725,10 +1723,12 @@ static void vcn_v1_0_jpeg_ring_emit_fence(struct amdgpu_ring *ring, u64 addr, u6
* Write ring commands to execute the indirect buffer.
*/
static void vcn_v1_0_jpeg_ring_emit_ib(struct amdgpu_ring *ring,
- struct amdgpu_ib *ib,
- unsigned vmid, bool ctx_switch)
+ struct amdgpu_job *job,
+ struct amdgpu_ib *ib,
+ bool ctx_switch)
{
struct amdgpu_device *adev = ring->adev;
+ unsigned vmid = AMDGPU_JOB_GET_VMID(job);
amdgpu_ring_write(ring,
PACKETJ(SOC15_REG_OFFSET(UVD, 0, mmUVD_LMI_JRBC_IB_VMID), 0, 0, PACKETJ_TYPE0));
diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
index a99f717..a0fda6f 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c
@@ -129,7 +129,7 @@ static int vega10_ih_irq_init(struct amdgpu_device *adev)
else
wptr_off = adev->wb.gpu_addr + (adev->irq.ih.wptr_offs * 4);
WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_LO, lower_32_bits(wptr_off));
- WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFF);
+ WREG32_SOC15(OSSSYS, 0, mmIH_RB_WPTR_ADDR_HI, upper_32_bits(wptr_off) & 0xFFFF);
/* set rptr, wptr to 0 */
WREG32_SOC15(OSSSYS, 0, mmIH_RB_RPTR, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
index 2d44735..d13fc4f 100644
--- a/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
+++ b/drivers/gpu/drm/amd/amdgpu/vega20_reg_init.c
@@ -49,6 +49,7 @@ int vega20_reg_base_init(struct amdgpu_device *adev)
adev->reg_offset[SMUIO_HWIP][i] = (uint32_t *)(&(SMUIO_BASE.instance[i]));
adev->reg_offset[NBIF_HWIP][i] = (uint32_t *)(&(NBIO_BASE.instance[i]));
adev->reg_offset[THM_HWIP][i] = (uint32_t *)(&(THM_BASE.instance[i]));
+ adev->reg_offset[CLK_HWIP][i] = (uint32_t *)(&(CLK_BASE.instance[i]));
}
return 0;
}
diff --git a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
index 5d2475d..177d1e5 100644
--- a/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
+++ b/drivers/gpu/drm/amd/amdkfd/cik_event_interrupt.c
@@ -23,6 +23,7 @@
#include "kfd_priv.h"
#include "kfd_events.h"
#include "cik_int.h"
+#include "amdgpu_amdkfd.h"
static bool cik_event_interrupt_isr(struct kfd_dev *dev,
const uint32_t *ih_ring_entry,
@@ -107,7 +108,7 @@ static void cik_event_interrupt_wq(struct kfd_dev *dev,
kfd_process_vm_fault(dev->dqm, pasid);
memset(&info, 0, sizeof(info));
- dev->kfd2kgd->get_vm_fault_info(dev->kgd, &info);
+ amdgpu_amdkfd_gpuvm_get_vm_fault_info(dev->kgd, &info);
if (!info.page_addr && !info.status)
return;
diff --git a/drivers/gpu/drm/amd/amdkfd/cik_regs.h b/drivers/gpu/drm/amd/amdkfd/cik_regs.h
index 37ce6dd..8e2a166 100644
--- a/drivers/gpu/drm/amd/amdkfd/cik_regs.h
+++ b/drivers/gpu/drm/amd/amdkfd/cik_regs.h
@@ -68,6 +68,4 @@
#define GRBM_GFX_INDEX 0x30800
-#define ATC_VMID_PASID_MAPPING_VALID (1U << 31)
-
#endif
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 14d5b5f..5f4062b 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -37,6 +37,7 @@
#include "kfd_priv.h"
#include "kfd_device_queue_manager.h"
#include "kfd_dbgmgr.h"
+#include "amdgpu_amdkfd.h"
static long kfd_ioctl(struct file *, unsigned int, unsigned long);
static int kfd_open(struct inode *, struct file *);
@@ -834,8 +835,7 @@ static int kfd_ioctl_get_clock_counters(struct file *filep,
dev = kfd_device_by_id(args->gpu_id);
if (dev)
/* Reading GPU clock counter from KGD */
- args->gpu_clock_counter =
- dev->kfd2kgd->get_gpu_clock_counter(dev->kgd);
+ args->gpu_clock_counter = amdgpu_amdkfd_get_gpu_clock_counter(dev->kgd);
else
/* Node without GPU resource */
args->gpu_clock_counter = 0;
@@ -1042,7 +1042,7 @@ static int kfd_ioctl_create_event(struct file *filp, struct kfd_process *p,
}
mutex_unlock(&p->mutex);
- err = kfd->kfd2kgd->map_gtt_bo_to_kernel(kfd->kgd,
+ err = amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(kfd->kgd,
mem, &kern_addr, &size);
if (err) {
pr_err("Failed to map event page to kernel\n");
@@ -1240,7 +1240,7 @@ bool kfd_dev_is_large_bar(struct kfd_dev *dev)
if (dev->device_info->needs_iommu_device)
return false;
- dev->kfd2kgd->get_local_mem_info(dev->kgd, &mem_info);
+ amdgpu_amdkfd_get_local_mem_info(dev->kgd, &mem_info);
if (mem_info.local_mem_size_private == 0 &&
mem_info.local_mem_size_public > 0)
return true;
@@ -1281,7 +1281,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep,
goto err_unlock;
}
- err = dev->kfd2kgd->alloc_memory_of_gpu(
+ err = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(
dev->kgd, args->va_addr, args->size,
pdd->vm, (struct kgd_mem **) &mem, &offset,
flags);
@@ -1303,7 +1303,7 @@ static int kfd_ioctl_alloc_memory_of_gpu(struct file *filep,
return 0;
err_free:
- dev->kfd2kgd->free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem);
+ amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem);
err_unlock:
mutex_unlock(&p->mutex);
return err;
@@ -1338,7 +1338,8 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep,
goto err_unlock;
}
- ret = dev->kfd2kgd->free_memory_of_gpu(dev->kgd, (struct kgd_mem *)mem);
+ ret = amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd,
+ (struct kgd_mem *)mem);
/* If freeing the buffer failed, leave the handle in place for
* clean-up during process tear-down.
@@ -1418,7 +1419,7 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
err = PTR_ERR(peer_pdd);
goto get_mem_obj_from_handle_failed;
}
- err = peer->kfd2kgd->map_memory_to_gpu(
+ err = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(
peer->kgd, (struct kgd_mem *)mem, peer_pdd->vm);
if (err) {
pr_err("Failed to map to gpu %d/%d\n",
@@ -1430,7 +1431,7 @@ static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
mutex_unlock(&p->mutex);
- err = dev->kfd2kgd->sync_memory(dev->kgd, (struct kgd_mem *) mem, true);
+ err = amdgpu_amdkfd_gpuvm_sync_memory(dev->kgd, (struct kgd_mem *) mem, true);
if (err) {
pr_debug("Sync memory failed, wait interrupted by user signal\n");
goto sync_memory_failed;
@@ -1525,7 +1526,7 @@ static int kfd_ioctl_unmap_memory_from_gpu(struct file *filep,
err = -ENODEV;
goto get_mem_obj_from_handle_failed;
}
- err = dev->kfd2kgd->unmap_memory_to_gpu(
+ err = amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(
peer->kgd, (struct kgd_mem *)mem, peer_pdd->vm);
if (err) {
pr_err("Failed to unmap from gpu %d/%d\n",
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
index 56412b0..3783d12 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_crat.c
@@ -26,6 +26,7 @@
#include "kfd_priv.h"
#include "kfd_topology.h"
#include "kfd_iommu.h"
+#include "amdgpu_amdkfd.h"
/* GPU Processor ID base for dGPUs for which VCRAT needs to be created.
* GPU processor ID are expressed with Bit[31]=1.
@@ -753,12 +754,10 @@ int kfd_create_crat_image_acpi(void **crat_image, size_t *size)
return -ENODATA;
}
- pcrat_image = kmalloc(crat_table->length, GFP_KERNEL);
+ pcrat_image = kmemdup(crat_table, crat_table->length, GFP_KERNEL);
if (!pcrat_image)
return -ENOMEM;
- memcpy(pcrat_image, crat_table, crat_table->length);
-
*crat_image = pcrat_image;
*size = crat_table->length;
@@ -1161,7 +1160,7 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
cu->flags |= CRAT_CU_FLAGS_GPU_PRESENT;
cu->proximity_domain = proximity_domain;
- kdev->kfd2kgd->get_cu_info(kdev->kgd, &cu_info);
+ amdgpu_amdkfd_get_cu_info(kdev->kgd, &cu_info);
cu->num_simd_per_cu = cu_info.simd_per_cu;
cu->num_simd_cores = cu_info.simd_per_cu * cu_info.cu_active_number;
cu->max_waves_simd = cu_info.max_waves_per_simd;
@@ -1192,7 +1191,7 @@ static int kfd_create_vcrat_image_gpu(void *pcrat_image,
* report the total FB size (public+private) as a single
* private heap.
*/
- kdev->kfd2kgd->get_local_mem_info(kdev->kgd, &local_mem_info);
+ amdgpu_amdkfd_get_local_mem_info(kdev->kgd, &local_mem_info);
sub_type_hdr = (typeof(sub_type_hdr))((char *)sub_type_hdr +
sub_type_hdr->length);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
index a9f18ea..c004647 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c
@@ -28,6 +28,7 @@
#include "kfd_pm4_headers_vi.h"
#include "cwsr_trap_handler.h"
#include "kfd_iommu.h"
+#include "amdgpu_amdkfd.h"
#define MQD_SIZE_ALIGNED 768
@@ -478,7 +479,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
/* add another 512KB for all other allocations on gart (HPD, fences) */
size += 512 * 1024;
- if (kfd->kfd2kgd->init_gtt_mem_allocation(
+ if (amdgpu_amdkfd_alloc_gtt_mem(
kfd->kgd, size, &kfd->gtt_mem,
&kfd->gtt_start_gpu_addr, &kfd->gtt_start_cpu_ptr,
false)) {
@@ -552,7 +553,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd,
kfd_doorbell_error:
kfd_gtt_sa_fini(kfd);
kfd_gtt_sa_init_error:
- kfd->kfd2kgd->free_gtt_mem(kfd->kgd, kfd->gtt_mem);
+ amdgpu_amdkfd_free_gtt_mem(kfd->kgd, kfd->gtt_mem);
dev_err(kfd_device,
"device %x:%x NOT added due to errors\n",
kfd->pdev->vendor, kfd->pdev->device);
@@ -569,7 +570,7 @@ void kgd2kfd_device_exit(struct kfd_dev *kfd)
kfd_topology_remove_device(kfd);
kfd_doorbell_fini(kfd);
kfd_gtt_sa_fini(kfd);
- kfd->kfd2kgd->free_gtt_mem(kfd->kgd, kfd->gtt_mem);
+ amdgpu_amdkfd_free_gtt_mem(kfd->kgd, kfd->gtt_mem);
}
kfree(kfd);
@@ -681,6 +682,7 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
{
uint32_t patched_ihre[KFD_MAX_RING_ENTRY_SIZE];
bool is_patched = false;
+ unsigned long flags;
if (!kfd->init_complete)
return;
@@ -690,7 +692,7 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
return;
}
- spin_lock(&kfd->interrupt_lock);
+ spin_lock_irqsave(&kfd->interrupt_lock, flags);
if (kfd->interrupts_active
&& interrupt_is_wanted(kfd, ih_ring_entry,
@@ -699,7 +701,7 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry)
is_patched ? patched_ihre : ih_ring_entry))
queue_work(kfd->ih_wq, &kfd->interrupt_work);
- spin_unlock(&kfd->interrupt_lock);
+ spin_unlock_irqrestore(&kfd->interrupt_lock, flags);
}
int kgd2kfd_quiesce_mm(struct mm_struct *mm)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
index a3b9339..fb9d66e 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c
@@ -33,6 +33,7 @@
#include "kfd_mqd_manager.h"
#include "cik_regs.h"
#include "kfd_kernel_queue.h"
+#include "amdgpu_amdkfd.h"
/* Size of the per-pipe EOP queue */
#define CIK_HPD_EOP_BYTES_LOG2 11
@@ -219,7 +220,7 @@ static int flush_texture_cache_nocpsch(struct kfd_dev *kdev,
if (ret)
return ret;
- return kdev->kfd2kgd->submit_ib(kdev->kgd, KGD_ENGINE_MEC1, qpd->vmid,
+ return amdgpu_amdkfd_submit_ib(kdev->kgd, KGD_ENGINE_MEC1, qpd->vmid,
qpd->ib_base, (uint32_t *)qpd->ib_kaddr,
pmf->release_mem_size / sizeof(uint32_t));
}
@@ -672,7 +673,7 @@ static int restore_process_queues_nocpsch(struct device_queue_manager *dqm,
pdd = qpd_to_pdd(qpd);
/* Retrieve PD base */
- pd_base = dqm->dev->kfd2kgd->get_process_page_dir(pdd->vm);
+ pd_base = amdgpu_amdkfd_gpuvm_get_process_page_dir(pdd->vm);
dqm_lock(dqm);
if (WARN_ON_ONCE(!qpd->evicted)) /* already restored, do nothing */
@@ -743,7 +744,7 @@ static int restore_process_queues_cpsch(struct device_queue_manager *dqm,
pdd = qpd_to_pdd(qpd);
/* Retrieve PD base */
- pd_base = dqm->dev->kfd2kgd->get_process_page_dir(pdd->vm);
+ pd_base = amdgpu_amdkfd_gpuvm_get_process_page_dir(pdd->vm);
dqm_lock(dqm);
if (WARN_ON_ONCE(!qpd->evicted)) /* already restored, do nothing */
@@ -793,7 +794,7 @@ static int register_process(struct device_queue_manager *dqm,
pdd = qpd_to_pdd(qpd);
/* Retrieve PD base */
- pd_base = dqm->dev->kfd2kgd->get_process_page_dir(pdd->vm);
+ pd_base = amdgpu_amdkfd_gpuvm_get_process_page_dir(pdd->vm);
dqm_lock(dqm);
list_add(&n->list, &dqm->queues);
@@ -805,7 +806,7 @@ static int register_process(struct device_queue_manager *dqm,
retval = dqm->asic_ops.update_qpd(dqm, qpd);
if (dqm->processes_count++ == 0)
- dqm->dev->kfd2kgd->set_compute_idle(dqm->dev->kgd, false);
+ amdgpu_amdkfd_set_compute_idle(dqm->dev->kgd, false);
dqm_unlock(dqm);
@@ -829,7 +830,7 @@ static int unregister_process(struct device_queue_manager *dqm,
list_del(&cur->list);
kfree(cur);
if (--dqm->processes_count == 0)
- dqm->dev->kfd2kgd->set_compute_idle(
+ amdgpu_amdkfd_set_compute_idle(
dqm->dev->kgd, true);
goto out;
}
@@ -845,15 +846,8 @@ static int
set_pasid_vmid_mapping(struct device_queue_manager *dqm, unsigned int pasid,
unsigned int vmid)
{
- uint32_t pasid_mapping;
-
- pasid_mapping = (pasid == 0) ? 0 :
- (uint32_t)pasid |
- ATC_VMID_PASID_MAPPING_VALID;
-
return dqm->dev->kfd2kgd->set_pasid_vmid_mapping(
- dqm->dev->kgd, pasid_mapping,
- vmid);
+ dqm->dev->kgd, pasid, vmid);
}
static void init_interrupts(struct device_queue_manager *dqm)
@@ -1796,7 +1790,7 @@ static void kfd_process_hw_exception(struct work_struct *work)
{
struct device_queue_manager *dqm = container_of(work,
struct device_queue_manager, hw_exception_work);
- dqm->dev->kfd2kgd->gpu_recover(dqm->dev->kgd);
+ amdgpu_amdkfd_gpu_reset(dqm->dev->kgd);
}
#if defined(CONFIG_DEBUG_FS)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
index e33019a..6910028 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c
@@ -22,6 +22,7 @@
*/
#include "kfd_mqd_manager.h"
+#include "amdgpu_amdkfd.h"
struct mqd_manager *mqd_manager_init(enum KFD_MQD_TYPE type,
struct kfd_dev *dev)
@@ -58,7 +59,7 @@ void mqd_symmetrically_map_cu_mask(struct mqd_manager *mm,
uint32_t cu_per_sh[4] = {0};
int i, se, cu = 0;
- mm->dev->kfd2kgd->get_cu_info(mm->dev->kgd, &cu_info);
+ amdgpu_amdkfd_get_cu_info(mm->dev->kgd, &cu_info);
if (cu_mask_count > cu_info.cu_active_number)
cu_mask_count = cu_info.cu_active_number;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
index f381c1c..9dbba60 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v9.c
@@ -30,6 +30,7 @@
#include "gc/gc_9_0_offset.h"
#include "gc/gc_9_0_sh_mask.h"
#include "sdma0/sdma0_4_0_sh_mask.h"
+#include "amdgpu_amdkfd.h"
static inline struct v9_mqd *get_mqd(void *mqd)
{
@@ -83,7 +84,7 @@ static int init_mqd(struct mqd_manager *mm, void **mqd,
*mqd_mem_obj = kzalloc(sizeof(struct kfd_mem_obj), GFP_KERNEL);
if (!*mqd_mem_obj)
return -ENOMEM;
- retval = kfd->kfd2kgd->init_gtt_mem_allocation(kfd->kgd,
+ retval = amdgpu_amdkfd_alloc_gtt_mem(kfd->kgd,
ALIGN(q->ctl_stack_size, PAGE_SIZE) +
ALIGN(sizeof(struct v9_mqd), PAGE_SIZE),
&((*mqd_mem_obj)->gtt_mem),
@@ -250,7 +251,7 @@ static void uninit_mqd(struct mqd_manager *mm, void *mqd,
struct kfd_dev *kfd = mm->dev;
if (mqd_mem_obj->gtt_mem) {
- kfd->kfd2kgd->free_gtt_mem(kfd->kgd, mqd_mem_obj->gtt_mem);
+ amdgpu_amdkfd_free_gtt_mem(kfd->kgd, mqd_mem_obj->gtt_mem);
kfree(mqd_mem_obj);
} else {
kfd_gtt_sa_free(mm->dev, mqd_mem_obj);
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
index 15fff44..33b08ff 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_pasid.c
@@ -22,6 +22,7 @@
#include <linux/types.h>
#include "kfd_priv.h"
+#include "amdgpu_ids.h"
static unsigned int pasid_bits = 16;
static const struct kfd2kgd_calls *kfd2kgd;
@@ -71,7 +72,7 @@ unsigned int kfd_pasid_alloc(void)
return false;
}
- r = kfd2kgd->alloc_pasid(pasid_bits);
+ r = amdgpu_pasid_alloc(pasid_bits);
return r > 0 ? r : 0;
}
@@ -79,5 +80,5 @@ unsigned int kfd_pasid_alloc(void)
void kfd_pasid_free(unsigned int pasid)
{
if (kfd2kgd)
- kfd2kgd->free_pasid(pasid);
+ amdgpu_pasid_free(pasid);
}
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
index 53ff86d..dec8e64 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h
@@ -507,6 +507,7 @@ struct qcm_process_device {
* All the memory management data should be here too
*/
uint64_t gds_context_area;
+ /* Contains page table flags such as AMDGPU_PTE_VALID since gfx9 */
uint64_t page_table_base;
uint32_t sh_mem_config;
uint32_t sh_mem_bases;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
index 0039e45..80b36e8 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c
@@ -31,6 +31,7 @@
#include <linux/compat.h>
#include <linux/mman.h>
#include <linux/file.h>
+#include "amdgpu_amdkfd.h"
struct mm_struct;
@@ -100,8 +101,8 @@ static void kfd_process_free_gpuvm(struct kgd_mem *mem,
{
struct kfd_dev *dev = pdd->dev;
- dev->kfd2kgd->unmap_memory_to_gpu(dev->kgd, mem, pdd->vm);
- dev->kfd2kgd->free_memory_of_gpu(dev->kgd, mem);
+ amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(dev->kgd, mem, pdd->vm);
+ amdgpu_amdkfd_gpuvm_free_memory_of_gpu(dev->kgd, mem);
}
/* kfd_process_alloc_gpuvm - Allocate GPU VM for the KFD process
@@ -119,16 +120,16 @@ static int kfd_process_alloc_gpuvm(struct kfd_process_device *pdd,
int handle;
int err;
- err = kdev->kfd2kgd->alloc_memory_of_gpu(kdev->kgd, gpu_va, size,
+ err = amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu(kdev->kgd, gpu_va, size,
pdd->vm, &mem, NULL, flags);
if (err)
goto err_alloc_mem;
- err = kdev->kfd2kgd->map_memory_to_gpu(kdev->kgd, mem, pdd->vm);
+ err = amdgpu_amdkfd_gpuvm_map_memory_to_gpu(kdev->kgd, mem, pdd->vm);
if (err)
goto err_map_mem;
- err = kdev->kfd2kgd->sync_memory(kdev->kgd, mem, true);
+ err = amdgpu_amdkfd_gpuvm_sync_memory(kdev->kgd, mem, true);
if (err) {
pr_debug("Sync memory failed, wait interrupted by user signal\n");
goto sync_memory_failed;
@@ -147,7 +148,7 @@ static int kfd_process_alloc_gpuvm(struct kfd_process_device *pdd,
}
if (kptr) {
- err = kdev->kfd2kgd->map_gtt_bo_to_kernel(kdev->kgd,
+ err = amdgpu_amdkfd_gpuvm_map_gtt_bo_to_kernel(kdev->kgd,
(struct kgd_mem *)mem, kptr, NULL);
if (err) {
pr_debug("Map GTT BO to kernel failed\n");
@@ -165,7 +166,7 @@ static int kfd_process_alloc_gpuvm(struct kfd_process_device *pdd,
return err;
err_map_mem:
- kdev->kfd2kgd->free_memory_of_gpu(kdev->kgd, mem);
+ amdgpu_amdkfd_gpuvm_free_memory_of_gpu(kdev->kgd, mem);
err_alloc_mem:
*kptr = NULL;
return err;
@@ -296,11 +297,11 @@ static void kfd_process_device_free_bos(struct kfd_process_device *pdd)
per_device_list) {
if (!peer_pdd->vm)
continue;
- peer_pdd->dev->kfd2kgd->unmap_memory_to_gpu(
+ amdgpu_amdkfd_gpuvm_unmap_memory_from_gpu(
peer_pdd->dev->kgd, mem, peer_pdd->vm);
}
- pdd->dev->kfd2kgd->free_memory_of_gpu(pdd->dev->kgd, mem);
+ amdgpu_amdkfd_gpuvm_free_memory_of_gpu(pdd->dev->kgd, mem);
kfd_process_device_remove_obj_handle(pdd, id);
}
}
@@ -323,11 +324,12 @@ static void kfd_process_destroy_pdds(struct kfd_process *p)
pdd->dev->id, p->pasid);
if (pdd->drm_file) {
- pdd->dev->kfd2kgd->release_process_vm(pdd->dev->kgd, pdd->vm);
+ amdgpu_amdkfd_gpuvm_release_process_vm(
+ pdd->dev->kgd, pdd->vm);
fput(pdd->drm_file);
}
else if (pdd->vm)
- pdd->dev->kfd2kgd->destroy_process_vm(
+ amdgpu_amdkfd_gpuvm_destroy_process_vm(
pdd->dev->kgd, pdd->vm);
list_del(&pdd->per_device_list);
@@ -688,12 +690,12 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd,
dev = pdd->dev;
if (drm_file)
- ret = dev->kfd2kgd->acquire_process_vm(
+ ret = amdgpu_amdkfd_gpuvm_acquire_process_vm(
dev->kgd, drm_file, p->pasid,
&pdd->vm, &p->kgd_process_info, &p->ef);
else
- ret = dev->kfd2kgd->create_process_vm(
- dev->kgd, p->pasid, &pdd->vm, &p->kgd_process_info, &p->ef);
+ ret = amdgpu_amdkfd_gpuvm_create_process_vm(dev->kgd, p->pasid,
+ &pdd->vm, &p->kgd_process_info, &p->ef);
if (ret) {
pr_err("Failed to create process VM object\n");
return ret;
@@ -714,7 +716,7 @@ int kfd_process_device_init_vm(struct kfd_process_device *pdd,
err_reserve_ib_mem:
kfd_process_device_free_bos(pdd);
if (!drm_file)
- dev->kfd2kgd->destroy_process_vm(dev->kgd, pdd->vm);
+ amdgpu_amdkfd_gpuvm_destroy_process_vm(dev->kgd, pdd->vm);
pdd->vm = NULL;
return ret;
@@ -972,7 +974,7 @@ static void restore_process_worker(struct work_struct *work)
*/
p->last_restore_timestamp = get_jiffies_64();
- ret = pdd->dev->kfd2kgd->restore_process_bos(p->kgd_process_info,
+ ret = amdgpu_amdkfd_gpuvm_restore_process_bos(p->kgd_process_info,
&p->ef);
if (ret) {
pr_debug("Failed to restore BOs of pasid %d, retry after %d ms\n",
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
index e3843c5..c73b4ff 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c
@@ -36,6 +36,7 @@
#include "kfd_topology.h"
#include "kfd_device_queue_manager.h"
#include "kfd_iommu.h"
+#include "amdgpu_amdkfd.h"
/* topology_device_list - Master list of all topology devices */
static struct list_head topology_device_list;
@@ -1052,7 +1053,7 @@ static uint32_t kfd_generate_gpu_id(struct kfd_dev *gpu)
if (!gpu)
return 0;
- gpu->kfd2kgd->get_local_mem_info(gpu->kgd, &local_mem_info);
+ amdgpu_amdkfd_get_local_mem_info(gpu->kgd, &local_mem_info);
local_mem_size = local_mem_info.local_mem_size_private +
local_mem_info.local_mem_size_public;
@@ -1118,8 +1119,7 @@ static void kfd_fill_mem_clk_max_info(struct kfd_topology_device *dev)
* for APUs - If CRAT from ACPI reports more than one bank, then
* all the banks will report the same mem_clk_max information
*/
- dev->gpu->kfd2kgd->get_local_mem_info(dev->gpu->kgd,
- &local_mem_info);
+ amdgpu_amdkfd_get_local_mem_info(dev->gpu->kgd, &local_mem_info);
list_for_each_entry(mem, &dev->mem_props, list)
mem->mem_clk_max = local_mem_info.mem_clk_max;
@@ -1240,7 +1240,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
* needed for the topology
*/
- dev->gpu->kfd2kgd->get_cu_info(dev->gpu->kgd, &cu_info);
+ amdgpu_amdkfd_get_cu_info(dev->gpu->kgd, &cu_info);
dev->node_props.simd_arrays_per_engine =
cu_info.num_shader_arrays_per_engine;
@@ -1249,7 +1249,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu)
dev->node_props.location_id = PCI_DEVID(gpu->pdev->bus->number,
gpu->pdev->devfn);
dev->node_props.max_engine_clk_fcompute =
- dev->gpu->kfd2kgd->get_max_engine_clock_in_mhz(dev->gpu->kgd);
+ amdgpu_amdkfd_get_max_engine_clock_in_mhz(dev->gpu->kgd);
dev->node_props.max_engine_clk_ccompute =
cpufreq_quick_get_max(0) / 1000;
dev->node_props.drm_render_minor =
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
index dd688cf..aa43bb25 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c
@@ -76,6 +76,16 @@
#define FIRMWARE_RAVEN_DMCU "amdgpu/raven_dmcu.bin"
MODULE_FIRMWARE(FIRMWARE_RAVEN_DMCU);
+/**
+ * DOC: overview
+ *
+ * The AMDgpu display manager, **amdgpu_dm** (or even simpler,
+ * **dm**) sits between DRM and DC. It acts as a liason, converting DRM
+ * requests into DC requests, and DC responses into DRM responses.
+ *
+ * The root control structure is &struct amdgpu_display_manager.
+ */
+
/* basic init/fini API */
static int amdgpu_dm_init(struct amdgpu_device *adev);
static void amdgpu_dm_fini(struct amdgpu_device *adev);
@@ -95,7 +105,7 @@ static void
amdgpu_dm_update_connector_after_detect(struct amdgpu_dm_connector *aconnector);
static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
- struct amdgpu_plane *aplane,
+ struct drm_plane *plane,
unsigned long possible_crtcs);
static int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm,
struct drm_plane *plane,
@@ -379,11 +389,6 @@ static void amdgpu_dm_fbc_init(struct drm_connector *connector)
}
-/*
- * Init display KMS
- *
- * Returns 0 on success
- */
static int amdgpu_dm_init(struct amdgpu_device *adev)
{
struct dc_init_data init_data;
@@ -429,6 +434,9 @@ static int amdgpu_dm_init(struct amdgpu_device *adev)
adev->asic_type < CHIP_RAVEN)
init_data.flags.gpu_vm_support = true;
+ if (amdgpu_dc_feature_mask & DC_FBC_MASK)
+ init_data.flags.fbc_support = true;
+
/* Display Core create. */
adev->dm.dc = dc_create(&init_data);
@@ -660,6 +668,26 @@ static void s3_handle_mst(struct drm_device *dev, bool suspend)
drm_modeset_unlock(&dev->mode_config.connection_mutex);
}
+/**
+ * dm_hw_init() - Initialize DC device
+ * @handle: The base driver device containing the amdpgu_dm device.
+ *
+ * Initialize the &struct amdgpu_display_manager device. This involves calling
+ * the initializers of each DM component, then populating the struct with them.
+ *
+ * Although the function implies hardware initialization, both hardware and
+ * software are initialized here. Splitting them out to their relevant init
+ * hooks is a future TODO item.
+ *
+ * Some notable things that are initialized here:
+ *
+ * - Display Core, both software and hardware
+ * - DC modules that we need (freesync and color management)
+ * - DRM software states
+ * - Interrupt sources and handlers
+ * - Vblank support
+ * - Debug FS entries, if enabled
+ */
static int dm_hw_init(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -670,6 +698,14 @@ static int dm_hw_init(void *handle)
return 0;
}
+/**
+ * dm_hw_fini() - Teardown DC device
+ * @handle: The base driver device containing the amdpgu_dm device.
+ *
+ * Teardown components within &struct amdgpu_display_manager that require
+ * cleanup. This involves cleaning up the DRM device, DC, and any modules that
+ * were loaded. Also flush IRQ workqueues and disable them.
+ */
static int dm_hw_fini(void *handle)
{
struct amdgpu_device *adev = (struct amdgpu_device *)handle;
@@ -895,6 +931,16 @@ static int dm_resume(void *handle)
return ret;
}
+/**
+ * DOC: DM Lifecycle
+ *
+ * DM (and consequently DC) is registered in the amdgpu base driver as a IP
+ * block. When CONFIG_DRM_AMD_DC is enabled, the DM device IP block is added to
+ * the base driver's device list to be initialized and torn down accordingly.
+ *
+ * The functions to do so are provided as hooks in &struct amd_ip_funcs.
+ */
+
static const struct amd_ip_funcs amdgpu_dm_funcs = {
.name = "dm",
.early_init = dm_early_init,
@@ -962,6 +1008,12 @@ dm_atomic_state_alloc_free(struct drm_atomic_state *state)
kfree(dm_state);
}
+/**
+ * DOC: atomic
+ *
+ * *WIP*
+ */
+
static const struct drm_mode_config_funcs amdgpu_dm_mode_funcs = {
.fb_create = amdgpu_display_user_framebuffer_create,
.output_poll_changed = drm_fb_helper_output_poll_changed,
@@ -1524,15 +1576,23 @@ static int amdgpu_dm_backlight_update_status(struct backlight_device *bd)
{
struct amdgpu_display_manager *dm = bl_get_data(bd);
+ /* backlight_pwm_u16_16 parameter is in unsigned 32 bit, 16 bit integer
+ * and 16 bit fractional, where 1.0 is max backlight value.
+ * bd->props.brightness is 8 bit format and needs to be converted by
+ * scaling via copy lower byte to upper byte of 16 bit value.
+ */
+ uint32_t brightness = bd->props.brightness * 0x101;
+
/*
* PWM interperts 0 as 100% rather than 0% because of HW
- * limitation for level 0.So limiting minimum brightness level
+ * limitation for level 0. So limiting minimum brightness level
* to 1.
*/
if (bd->props.brightness < 1)
- return 1;
+ brightness = 0x101;
+
if (dc_link_set_backlight_level(dm->backlight_link,
- bd->props.brightness, 0, 0))
+ brightness, 0, 0))
return 0;
else
return 1;
@@ -1584,18 +1644,18 @@ static int initialize_plane(struct amdgpu_display_manager *dm,
struct amdgpu_mode_info *mode_info,
int plane_id)
{
- struct amdgpu_plane *plane;
+ struct drm_plane *plane;
unsigned long possible_crtcs;
int ret = 0;
- plane = kzalloc(sizeof(struct amdgpu_plane), GFP_KERNEL);
+ plane = kzalloc(sizeof(struct drm_plane), GFP_KERNEL);
mode_info->planes[plane_id] = plane;
if (!plane) {
DRM_ERROR("KMS: Failed to allocate plane\n");
return -ENOMEM;
}
- plane->base.type = mode_info->plane_type[plane_id];
+ plane->type = mode_info->plane_type[plane_id];
/*
* HACK: IGT tests expect that each plane can only have
@@ -1686,7 +1746,7 @@ static int amdgpu_dm_initialize_drm_device(struct amdgpu_device *adev)
}
for (i = 0; i < dm->dc->caps.max_streams; i++)
- if (amdgpu_dm_crtc_init(dm, &mode_info->planes[i]->base, i)) {
+ if (amdgpu_dm_crtc_init(dm, mode_info->planes[i], i)) {
DRM_ERROR("KMS: Failed to initialize crtc\n");
goto fail;
}
@@ -2707,18 +2767,11 @@ create_stream_for_sink(struct amdgpu_dm_connector *aconnector,
drm_connector = &aconnector->base;
if (!aconnector->dc_sink) {
- /*
- * Create dc_sink when necessary to MST
- * Don't apply fake_sink to MST
- */
- if (aconnector->mst_port) {
- dm_dp_mst_dc_sink_create(drm_connector);
- return stream;
+ if (!aconnector->mst_port) {
+ sink = create_fake_sink(aconnector);
+ if (!sink)
+ return stream;
}
-
- sink = create_fake_sink(aconnector);
- if (!sink)
- return stream;
} else {
sink = aconnector->dc_sink;
}
@@ -3307,7 +3360,7 @@ void dm_drm_plane_destroy_state(struct drm_plane *plane,
static const struct drm_plane_funcs dm_plane_funcs = {
.update_plane = drm_atomic_helper_update_plane,
.disable_plane = drm_atomic_helper_disable_plane,
- .destroy = drm_plane_cleanup,
+ .destroy = drm_primary_helper_destroy,
.reset = dm_drm_plane_reset,
.atomic_duplicate_state = dm_drm_plane_duplicate_state,
.atomic_destroy_state = dm_drm_plane_destroy_state,
@@ -3468,49 +3521,49 @@ static const u32 cursor_formats[] = {
};
static int amdgpu_dm_plane_init(struct amdgpu_display_manager *dm,
- struct amdgpu_plane *aplane,
+ struct drm_plane *plane,
unsigned long possible_crtcs)
{
int res = -EPERM;
- switch (aplane->base.type) {
+ switch (plane->type) {
case DRM_PLANE_TYPE_PRIMARY:
res = drm_universal_plane_init(
dm->adev->ddev,
- &aplane->base,
+ plane,
possible_crtcs,
&dm_plane_funcs,
rgb_formats,
ARRAY_SIZE(rgb_formats),
- NULL, aplane->base.type, NULL);
+ NULL, plane->type, NULL);
break;
case DRM_PLANE_TYPE_OVERLAY:
res = drm_universal_plane_init(
dm->adev->ddev,
- &aplane->base,
+ plane,
possible_crtcs,
&dm_plane_funcs,
yuv_formats,
ARRAY_SIZE(yuv_formats),
- NULL, aplane->base.type, NULL);
+ NULL, plane->type, NULL);
break;
case DRM_PLANE_TYPE_CURSOR:
res = drm_universal_plane_init(
dm->adev->ddev,
- &aplane->base,
+ plane,
possible_crtcs,
&dm_plane_funcs,
cursor_formats,
ARRAY_SIZE(cursor_formats),
- NULL, aplane->base.type, NULL);
+ NULL, plane->type, NULL);
break;
}
- drm_plane_helper_add(&aplane->base, &dm_plane_helper_funcs);
+ drm_plane_helper_add(plane, &dm_plane_helper_funcs);
/* Create (reset) the plane state */
- if (aplane->base.funcs->reset)
- aplane->base.funcs->reset(&aplane->base);
+ if (plane->funcs->reset)
+ plane->funcs->reset(plane);
return res;
@@ -3521,7 +3574,7 @@ static int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm,
uint32_t crtc_index)
{
struct amdgpu_crtc *acrtc = NULL;
- struct amdgpu_plane *cursor_plane;
+ struct drm_plane *cursor_plane;
int res = -ENOMEM;
@@ -3529,7 +3582,7 @@ static int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm,
if (!cursor_plane)
goto fail;
- cursor_plane->base.type = DRM_PLANE_TYPE_CURSOR;
+ cursor_plane->type = DRM_PLANE_TYPE_CURSOR;
res = amdgpu_dm_plane_init(dm, cursor_plane, 0);
acrtc = kzalloc(sizeof(struct amdgpu_crtc), GFP_KERNEL);
@@ -3540,7 +3593,7 @@ static int amdgpu_dm_crtc_init(struct amdgpu_display_manager *dm,
dm->ddev,
&acrtc->base,
plane,
- &cursor_plane->base,
+ cursor_plane,
&amdgpu_dm_crtc_funcs, NULL);
if (res)
@@ -3779,12 +3832,12 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm,
case DRM_MODE_CONNECTOR_HDMIA:
aconnector->base.polled = DRM_CONNECTOR_POLL_HPD;
aconnector->base.ycbcr_420_allowed =
- link->link_enc->features.ycbcr420_supported ? true : false;
+ link->link_enc->features.hdmi_ycbcr420_supported ? true : false;
break;
case DRM_MODE_CONNECTOR_DisplayPort:
aconnector->base.polled = DRM_CONNECTOR_POLL_HPD;
aconnector->base.ycbcr_420_allowed =
- link->link_enc->features.ycbcr420_supported ? true : false;
+ link->link_enc->features.dp_ycbcr420_supported ? true : false;
break;
case DRM_MODE_CONNECTOR_DVID:
aconnector->base.polled = DRM_CONNECTOR_POLL_HPD;
@@ -4542,6 +4595,14 @@ static int amdgpu_dm_atomic_commit(struct drm_device *dev,
/*TODO Handle EINTR, reenable IRQ*/
}
+/**
+ * amdgpu_dm_atomic_commit_tail() - AMDgpu DM's commit tail implementation.
+ * @state: The atomic state to commit
+ *
+ * This will tell DC to commit the constructed DC state from atomic_check,
+ * programming the hardware. Any failures here implies a hardware failure, since
+ * atomic check should have filtered anything non-kosher.
+ */
static void amdgpu_dm_atomic_commit_tail(struct drm_atomic_state *state)
{
struct drm_device *dev = state->dev;
@@ -5313,6 +5374,12 @@ enum surface_update_type dm_determine_update_type_for_commit(struct dc *dc, stru
struct dc_stream_update stream_update;
enum surface_update_type update_type = UPDATE_TYPE_FAST;
+ if (!updates || !surface) {
+ DRM_ERROR("Plane or surface update failed to allocate");
+ /* Set type to FULL to avoid crashing in DC*/
+ update_type = UPDATE_TYPE_FULL;
+ goto ret;
+ }
for_each_oldnew_crtc_in_state(state, crtc, old_crtc_state, new_crtc_state, i) {
new_dm_crtc_state = to_dm_crtc_state(new_crtc_state);
@@ -5388,6 +5455,31 @@ enum surface_update_type dm_determine_update_type_for_commit(struct dc *dc, stru
return update_type;
}
+/**
+ * amdgpu_dm_atomic_check() - Atomic check implementation for AMDgpu DM.
+ * @dev: The DRM device
+ * @state: The atomic state to commit
+ *
+ * Validate that the given atomic state is programmable by DC into hardware.
+ * This involves constructing a &struct dc_state reflecting the new hardware
+ * state we wish to commit, then querying DC to see if it is programmable. It's
+ * important not to modify the existing DC state. Otherwise, atomic_check
+ * may unexpectedly commit hardware changes.
+ *
+ * When validating the DC state, it's important that the right locks are
+ * acquired. For full updates case which removes/adds/updates streams on one
+ * CRTC while flipping on another CRTC, acquiring global lock will guarantee
+ * that any such full update commit will wait for completion of any outstanding
+ * flip using DRMs synchronization events. See
+ * dm_determine_update_type_for_commit()
+ *
+ * Note that DM adds the affected connectors for all CRTCs in state, when that
+ * might not seem necessary. This is because DC stream creation requires the
+ * DC sink, which is tied to the DRM connector state. Cleaning this up should
+ * be possible but non-trivial - a possible TODO item.
+ *
+ * Return: -Error code if validation failed.
+ */
static int amdgpu_dm_atomic_check(struct drm_device *dev,
struct drm_atomic_state *state)
{
@@ -5490,15 +5582,6 @@ static int amdgpu_dm_atomic_check(struct drm_device *dev,
lock_and_validation_needed = true;
}
- /*
- * For full updates case when
- * removing/adding/updating streams on one CRTC while flipping
- * on another CRTC,
- * acquiring global lock will guarantee that any such full
- * update commit
- * will wait for completion of any outstanding flip using DRMs
- * synchronization events.
- */
update_type = dm_determine_update_type_for_commit(dc, state);
if (overall_update_type < update_type)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
index 978b34a..d696064 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h
@@ -59,49 +59,100 @@ struct common_irq_params {
enum dc_irq_source irq_src;
};
+/**
+ * struct irq_list_head - Linked-list for low context IRQ handlers.
+ *
+ * @head: The list_head within &struct handler_data
+ * @work: A work_struct containing the deferred handler work
+ */
struct irq_list_head {
struct list_head head;
/* In case this interrupt needs post-processing, 'work' will be queued*/
struct work_struct work;
};
+/**
+ * struct dm_compressor_info - Buffer info used by frame buffer compression
+ * @cpu_addr: MMIO cpu addr
+ * @bo_ptr: Pointer to the buffer object
+ * @gpu_addr: MMIO gpu addr
+ */
struct dm_comressor_info {
void *cpu_addr;
struct amdgpu_bo *bo_ptr;
uint64_t gpu_addr;
};
+/**
+ * struct amdgpu_display_manager - Central amdgpu display manager device
+ *
+ * @dc: Display Core control structure
+ * @adev: AMDGPU base driver structure
+ * @ddev: DRM base driver structure
+ * @display_indexes_num: Max number of display streams supported
+ * @irq_handler_list_table_lock: Synchronizes access to IRQ tables
+ * @backlight_dev: Backlight control device
+ * @cached_state: Caches device atomic state for suspend/resume
+ * @compressor: Frame buffer compression buffer. See &struct dm_comressor_info
+ */
struct amdgpu_display_manager {
+
struct dc *dc;
+
+ /**
+ * @cgs_device:
+ *
+ * The Common Graphics Services device. It provides an interface for
+ * accessing registers.
+ */
struct cgs_device *cgs_device;
- struct amdgpu_device *adev; /*AMD base driver*/
- struct drm_device *ddev; /*DRM base driver*/
+ struct amdgpu_device *adev;
+ struct drm_device *ddev;
u16 display_indexes_num;
- /*
- * 'irq_source_handler_table' holds a list of handlers
- * per (DAL) IRQ source.
+ /**
+ * @irq_handler_list_low_tab:
*
- * Each IRQ source may need to be handled at different contexts.
- * By 'context' we mean, for example:
- * - The ISR context, which is the direct interrupt handler.
- * - The 'deferred' context - this is the post-processing of the
- * interrupt, but at a lower priority.
+ * Low priority IRQ handler table.
+ *
+ * It is a n*m table consisting of n IRQ sources, and m handlers per IRQ
+ * source. Low priority IRQ handlers are deferred to a workqueue to be
+ * processed. Hence, they can sleep.
*
* Note that handlers are called in the same order as they were
* registered (FIFO).
*/
struct irq_list_head irq_handler_list_low_tab[DAL_IRQ_SOURCES_NUMBER];
+
+ /**
+ * @irq_handler_list_high_tab:
+ *
+ * High priority IRQ handler table.
+ *
+ * It is a n*m table, same as &irq_handler_list_low_tab. However,
+ * handlers in this table are not deferred and are called immediately.
+ */
struct list_head irq_handler_list_high_tab[DAL_IRQ_SOURCES_NUMBER];
+ /**
+ * @pflip_params:
+ *
+ * Page flip IRQ parameters, passed to registered handlers when
+ * triggered.
+ */
struct common_irq_params
pflip_params[DC_IRQ_SOURCE_PFLIP_LAST - DC_IRQ_SOURCE_PFLIP_FIRST + 1];
+ /**
+ * @vblank_params:
+ *
+ * Vertical blanking IRQ parameters, passed to registered handlers when
+ * triggered.
+ */
struct common_irq_params
vblank_params[DC_IRQ_SOURCE_VBLANK6 - DC_IRQ_SOURCE_VBLANK1 + 1];
- /* this spin lock synchronizes access to 'irq_handler_list_table' */
spinlock_t irq_handler_list_table_lock;
struct backlight_device *backlight_dev;
@@ -110,9 +161,6 @@ struct amdgpu_display_manager {
struct mod_freesync *freesync_module;
- /**
- * Caches device atomic state for suspend/resume
- */
struct drm_atomic_state *cached_state;
struct dm_comressor_info compressor;
@@ -160,8 +208,6 @@ struct amdgpu_dm_connector {
struct mutex hpd_lock;
bool fake_enable;
-
- bool mst_connected;
};
#define to_amdgpu_dm_connector(x) container_of(x, struct amdgpu_dm_connector, base)
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
index be19e68..216e48c 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c
@@ -164,7 +164,7 @@ int amdgpu_dm_set_regamma_lut(struct dm_crtc_state *crtc)
*/
stream->out_transfer_func->type = TF_TYPE_DISTRIBUTED_POINTS;
ret = mod_color_calculate_regamma_params(stream->out_transfer_func,
- gamma, true, adev->asic_type <= CHIP_RAVEN);
+ gamma, true, adev->asic_type <= CHIP_RAVEN, NULL);
dc_gamma_release(&gamma);
if (!ret) {
stream->out_transfer_func->type = old_type;
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
index 01fc571..f088ac5 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_crc.c
@@ -75,6 +75,11 @@ int amdgpu_dm_crtc_set_crc_source(struct drm_crtc *crtc, const char *src_name)
return -EINVAL;
}
+ if (!stream_state) {
+ DRM_ERROR("No stream state for CRTC%d\n", crtc->index);
+ return -EINVAL;
+ }
+
/* When enabling CRC, we should also disable dithering. */
if (source == AMDGPU_DM_PIPE_CRC_SOURCE_AUTO) {
if (dc_stream_configure_crc(stream_state->ctx->dc,
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
index a212178..cd10f77 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c
@@ -32,16 +32,55 @@
#include "amdgpu_dm.h"
#include "amdgpu_dm_irq.h"
+/**
+ * DOC: overview
+ *
+ * DM provides another layer of IRQ management on top of what the base driver
+ * already provides. This is something that could be cleaned up, and is a
+ * future TODO item.
+ *
+ * The base driver provides IRQ source registration with DRM, handler
+ * registration into the base driver's IRQ table, and a handler callback
+ * amdgpu_irq_handler(), with which DRM calls on interrupts. This generic
+ * handler looks up the IRQ table, and calls the respective
+ * &amdgpu_irq_src_funcs.process hookups.
+ *
+ * What DM provides on top are two IRQ tables specifically for top-half and
+ * bottom-half IRQ handling, with the bottom-half implementing workqueues:
+ *
+ * - &amdgpu_display_manager.irq_handler_list_high_tab
+ * - &amdgpu_display_manager.irq_handler_list_low_tab
+ *
+ * They override the base driver's IRQ table, and the effect can be seen
+ * in the hooks that DM provides for &amdgpu_irq_src_funcs.process. They
+ * are all set to the DM generic handler amdgpu_dm_irq_handler(), which looks up
+ * DM's IRQ tables. However, in order for base driver to recognize this hook, DM
+ * still needs to register the IRQ with the base driver. See
+ * dce110_register_irq_handlers() and dcn10_register_irq_handlers().
+ *
+ * To expose DC's hardware interrupt toggle to the base driver, DM implements
+ * &amdgpu_irq_src_funcs.set hooks. Base driver calls it through
+ * amdgpu_irq_update() to enable or disable the interrupt.
+ */
+
/******************************************************************************
* Private declarations.
*****************************************************************************/
+/**
+ * struct amdgpu_dm_irq_handler_data - Data for DM interrupt handlers.
+ *
+ * @list: Linked list entry referencing the next/previous handler
+ * @handler: Handler function
+ * @handler_arg: Argument passed to the handler when triggered
+ * @dm: DM which this handler belongs to
+ * @irq_source: DC interrupt source that this handler is registered for
+ */
struct amdgpu_dm_irq_handler_data {
struct list_head list;
interrupt_handler handler;
void *handler_arg;
- /* DM which this handler belongs to */
struct amdgpu_display_manager *dm;
/* DAL irq source which registered for this interrupt. */
enum dc_irq_source irq_source;
@@ -68,7 +107,7 @@ static void init_handler_common_data(struct amdgpu_dm_irq_handler_data *hcd,
}
/**
- * dm_irq_work_func - Handle an IRQ outside of the interrupt handler proper.
+ * dm_irq_work_func() - Handle an IRQ outside of the interrupt handler proper.
*
* @work: work struct
*/
@@ -99,8 +138,8 @@ static void dm_irq_work_func(struct work_struct *work)
* (The most common use is HPD interrupt) */
}
-/**
- * Remove a handler and return a pointer to hander list from which the
+/*
+ * Remove a handler and return a pointer to handler list from which the
* handler was removed.
*/
static struct list_head *remove_irq_handler(struct amdgpu_device *adev,
@@ -203,6 +242,24 @@ static bool validate_irq_unregistration_params(enum dc_irq_source irq_source,
* Note: caller is responsible for input validation.
*****************************************************************************/
+/**
+ * amdgpu_dm_irq_register_interrupt() - Register a handler within DM.
+ * @adev: The base driver device containing the DM device.
+ * @int_params: Interrupt parameters containing the source, and handler context
+ * @ih: Function pointer to the interrupt handler to register
+ * @handler_args: Arguments passed to the handler when the interrupt occurs
+ *
+ * Register an interrupt handler for the given IRQ source, under the given
+ * context. The context can either be high or low. High context handlers are
+ * executed directly within ISR context, while low context is executed within a
+ * workqueue, thereby allowing operations that sleep.
+ *
+ * Registered handlers are called in a FIFO manner, i.e. the most recently
+ * registered handler will be called first.
+ *
+ * Return: Handler data &struct amdgpu_dm_irq_handler_data containing the IRQ
+ * source, handler function, and args
+ */
void *amdgpu_dm_irq_register_interrupt(struct amdgpu_device *adev,
struct dc_interrupt_params *int_params,
void (*ih)(void *),
@@ -261,6 +318,15 @@ void *amdgpu_dm_irq_register_interrupt(struct amdgpu_device *adev,
return handler_data;
}
+/**
+ * amdgpu_dm_irq_unregister_interrupt() - Remove a handler from the DM IRQ table
+ * @adev: The base driver device containing the DM device
+ * @irq_source: IRQ source to remove the given handler from
+ * @ih: Function pointer to the interrupt handler to unregister
+ *
+ * Go through both low and high context IRQ tables, and find the given handler
+ * for the given irq source. If found, remove it. Otherwise, do nothing.
+ */
void amdgpu_dm_irq_unregister_interrupt(struct amdgpu_device *adev,
enum dc_irq_source irq_source,
void *ih)
@@ -295,6 +361,20 @@ void amdgpu_dm_irq_unregister_interrupt(struct amdgpu_device *adev,
}
}
+/**
+ * amdgpu_dm_irq_init() - Initialize DM IRQ management
+ * @adev: The base driver device containing the DM device
+ *
+ * Initialize DM's high and low context IRQ tables.
+ *
+ * The N by M table contains N IRQ sources, with M
+ * &struct amdgpu_dm_irq_handler_data hooked together in a linked list. The
+ * list_heads are initialized here. When an interrupt n is triggered, all m
+ * handlers are called in sequence, FIFO according to registration order.
+ *
+ * The low context table requires special steps to initialize, since handlers
+ * will be deferred to a workqueue. See &struct irq_list_head.
+ */
int amdgpu_dm_irq_init(struct amdgpu_device *adev)
{
int src;
@@ -317,7 +397,12 @@ int amdgpu_dm_irq_init(struct amdgpu_device *adev)
return 0;
}
-/* DM IRQ and timer resource release */
+/**
+ * amdgpu_dm_irq_fini() - Tear down DM IRQ management
+ * @adev: The base driver device containing the DM device
+ *
+ * Flush all work within the low context IRQ table.
+ */
void amdgpu_dm_irq_fini(struct amdgpu_device *adev)
{
int src;
@@ -414,7 +499,7 @@ int amdgpu_dm_irq_resume_late(struct amdgpu_device *adev)
return 0;
}
-/**
+/*
* amdgpu_dm_irq_schedule_work - schedule all work items registered for the
* "irq_source".
*/
@@ -439,8 +524,9 @@ static void amdgpu_dm_irq_schedule_work(struct amdgpu_device *adev,
}
-/** amdgpu_dm_irq_immediate_work
- * Callback high irq work immediately, don't send to work queue
+/*
+ * amdgpu_dm_irq_immediate_work
+ * Callback high irq work immediately, don't send to work queue
*/
static void amdgpu_dm_irq_immediate_work(struct amdgpu_device *adev,
enum dc_irq_source irq_source)
@@ -467,11 +553,14 @@ static void amdgpu_dm_irq_immediate_work(struct amdgpu_device *adev,
DM_IRQ_TABLE_UNLOCK(adev, irq_table_flags);
}
-/*
- * amdgpu_dm_irq_handler
+/**
+ * amdgpu_dm_irq_handler - Generic DM IRQ handler
+ * @adev: amdgpu base driver device containing the DM device
+ * @source: Unused
+ * @entry: Data about the triggered interrupt
*
- * Generic IRQ handler, calls all registered high irq work immediately, and
- * schedules work for low irq
+ * Calls all registered high irq work immediately, and schedules work for low
+ * irq. The DM IRQ table is used to find the corresponding handlers.
*/
static int amdgpu_dm_irq_handler(struct amdgpu_device *adev,
struct amdgpu_irq_src *source,
@@ -613,7 +702,7 @@ void amdgpu_dm_set_irq_funcs(struct amdgpu_device *adev)
adev->hpd_irq.funcs = &dm_hpd_irq_funcs;
}
-/*
+/**
* amdgpu_dm_hpd_init - hpd setup callback.
*
* @adev: amdgpu_device pointer
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
index 03601d7..d02c32a 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.c
@@ -205,40 +205,6 @@ static const struct drm_connector_funcs dm_dp_mst_connector_funcs = {
.atomic_get_property = amdgpu_dm_connector_atomic_get_property
};
-void dm_dp_mst_dc_sink_create(struct drm_connector *connector)
-{
- struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
- struct dc_sink *dc_sink;
- struct dc_sink_init_data init_params = {
- .link = aconnector->dc_link,
- .sink_signal = SIGNAL_TYPE_DISPLAY_PORT_MST };
-
- /* FIXME none of this is safe. we shouldn't touch aconnector here in
- * atomic_check
- */
-
- /*
- * TODO: Need to further figure out why ddc.algo is NULL while MST port exists
- */
- if (!aconnector->port || !aconnector->port->aux.ddc.algo)
- return;
-
- ASSERT(aconnector->edid);
-
- dc_sink = dc_link_add_remote_sink(
- aconnector->dc_link,
- (uint8_t *)aconnector->edid,
- (aconnector->edid->extensions + 1) * EDID_LENGTH,
- &init_params);
-
- dc_sink->priv = aconnector;
- aconnector->dc_sink = dc_sink;
-
- if (aconnector->dc_sink)
- amdgpu_dm_update_freesync_caps(
- connector, aconnector->edid);
-}
-
static int dm_dp_mst_get_modes(struct drm_connector *connector)
{
struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
@@ -319,12 +285,7 @@ dm_dp_create_fake_mst_encoder(struct amdgpu_dm_connector *connector)
struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_encoder *amdgpu_encoder;
struct drm_encoder *encoder;
- const struct drm_connector_helper_funcs *connector_funcs =
- connector->base.helper_private;
- struct drm_encoder *enc_master =
- connector_funcs->best_encoder(&connector->base);
- DRM_DEBUG_KMS("enc master is %p\n", enc_master);
amdgpu_encoder = kzalloc(sizeof(*amdgpu_encoder), GFP_KERNEL);
if (!amdgpu_encoder)
return NULL;
@@ -354,25 +315,6 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_dm_connector *aconnector;
struct drm_connector *connector;
- struct drm_connector_list_iter conn_iter;
-
- drm_connector_list_iter_begin(dev, &conn_iter);
- drm_for_each_connector_iter(connector, &conn_iter) {
- aconnector = to_amdgpu_dm_connector(connector);
- if (aconnector->mst_port == master
- && !aconnector->port) {
- DRM_INFO("DM_MST: reusing connector: %p [id: %d] [master: %p]\n",
- aconnector, connector->base.id, aconnector->mst_port);
-
- aconnector->port = port;
- drm_connector_set_path_property(connector, pathprop);
-
- drm_connector_list_iter_end(&conn_iter);
- aconnector->mst_connected = true;
- return &aconnector->base;
- }
- }
- drm_connector_list_iter_end(&conn_iter);
aconnector = kzalloc(sizeof(*aconnector), GFP_KERNEL);
if (!aconnector)
@@ -421,8 +363,6 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
*/
amdgpu_dm_connector_funcs_reset(connector);
- aconnector->mst_connected = true;
-
DRM_INFO("DM_MST: added connector: %p [id: %d] [master: %p]\n",
aconnector, connector->base.id, aconnector->mst_port);
@@ -434,6 +374,9 @@ dm_dp_add_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
static void dm_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
struct drm_connector *connector)
{
+ struct amdgpu_dm_connector *master = container_of(mgr, struct amdgpu_dm_connector, mst_mgr);
+ struct drm_device *dev = master->base.dev;
+ struct amdgpu_device *adev = dev->dev_private;
struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
DRM_INFO("DM_MST: Disabling connector: %p [id: %d] [master: %p]\n",
@@ -447,7 +390,10 @@ static void dm_dp_destroy_mst_connector(struct drm_dp_mst_topology_mgr *mgr,
aconnector->dc_sink = NULL;
}
- aconnector->mst_connected = false;
+ drm_connector_unregister(connector);
+ if (adev->mode_info.rfbdev)
+ drm_fb_helper_remove_one_connector(&adev->mode_info.rfbdev->helper, connector);
+ drm_connector_put(connector);
}
static void dm_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr)
@@ -458,18 +404,10 @@ static void dm_dp_mst_hotplug(struct drm_dp_mst_topology_mgr *mgr)
drm_kms_helper_hotplug_event(dev);
}
-static void dm_dp_mst_link_status_reset(struct drm_connector *connector)
-{
- mutex_lock(&connector->dev->mode_config.mutex);
- drm_connector_set_link_status_property(connector, DRM_MODE_LINK_STATUS_BAD);
- mutex_unlock(&connector->dev->mode_config.mutex);
-}
-
static void dm_dp_mst_register_connector(struct drm_connector *connector)
{
struct drm_device *dev = connector->dev;
struct amdgpu_device *adev = dev->dev_private;
- struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector);
if (adev->mode_info.rfbdev)
drm_fb_helper_add_one_connector(&adev->mode_info.rfbdev->helper, connector);
@@ -477,9 +415,6 @@ static void dm_dp_mst_register_connector(struct drm_connector *connector)
DRM_ERROR("adev->mode_info.rfbdev is NULL\n");
drm_connector_register(connector);
-
- if (aconnector->mst_connected)
- dm_dp_mst_link_status_reset(connector);
}
static const struct drm_dp_mst_topology_cbs dm_mst_cbs = {
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
index 8cf51da..2da851b 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_mst_types.h
@@ -31,6 +31,5 @@ struct amdgpu_dm_connector;
void amdgpu_dm_initialize_dp_connector(struct amdgpu_display_manager *dm,
struct amdgpu_dm_connector *aconnector);
-void dm_dp_mst_dc_sink_create(struct drm_connector *connector);
#endif
diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
index 12001a0..9d2d698 100644
--- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
+++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_pp_smu.c
@@ -485,11 +485,11 @@ void pp_rv_set_display_requirement(struct pp_smu *pp,
return;
clock.clock_type = amd_pp_dcf_clock;
- clock.clock_freq_in_khz = req->hard_min_dcefclk_khz;
+ clock.clock_freq_in_khz = req->hard_min_dcefclk_mhz * 1000;
pp_funcs->display_clock_voltage_request(pp_handle, &clock);
clock.clock_type = amd_pp_f_clock;
- clock.clock_freq_in_khz = req->hard_min_fclk_khz;
+ clock.clock_freq_in_khz = req->hard_min_fclk_mhz * 1000;
pp_funcs->display_clock_voltage_request(pp_handle, &clock);
}
@@ -518,13 +518,13 @@ void pp_rv_set_wm_ranges(struct pp_smu *pp,
wm_dce_clocks[i].wm_set_id =
ranges->reader_wm_sets[i].wm_inst;
wm_dce_clocks[i].wm_max_dcfclk_clk_in_khz =
- ranges->reader_wm_sets[i].max_drain_clk_khz;
+ ranges->reader_wm_sets[i].max_drain_clk_mhz * 1000;
wm_dce_clocks[i].wm_min_dcfclk_clk_in_khz =
- ranges->reader_wm_sets[i].min_drain_clk_khz;
+ ranges->reader_wm_sets[i].min_drain_clk_mhz * 1000;
wm_dce_clocks[i].wm_max_mem_clk_in_khz =
- ranges->reader_wm_sets[i].max_fill_clk_khz;
+ ranges->reader_wm_sets[i].max_fill_clk_mhz * 1000;
wm_dce_clocks[i].wm_min_mem_clk_in_khz =
- ranges->reader_wm_sets[i].min_fill_clk_khz;
+ ranges->reader_wm_sets[i].min_fill_clk_mhz * 1000;
}
for (i = 0; i < wm_with_clock_ranges.num_wm_mcif_sets; i++) {
@@ -534,13 +534,13 @@ void pp_rv_set_wm_ranges(struct pp_smu *pp,
wm_soc_clocks[i].wm_set_id =
ranges->writer_wm_sets[i].wm_inst;
wm_soc_clocks[i].wm_max_socclk_clk_in_khz =
- ranges->writer_wm_sets[i].max_fill_clk_khz;
+ ranges->writer_wm_sets[i].max_fill_clk_mhz * 1000;
wm_soc_clocks[i].wm_min_socclk_clk_in_khz =
- ranges->writer_wm_sets[i].min_fill_clk_khz;
+ ranges->writer_wm_sets[i].min_fill_clk_mhz * 1000;
wm_soc_clocks[i].wm_max_mem_clk_in_khz =
- ranges->writer_wm_sets[i].max_drain_clk_khz;
+ ranges->writer_wm_sets[i].max_drain_clk_mhz * 1000;
wm_soc_clocks[i].wm_min_mem_clk_in_khz =
- ranges->writer_wm_sets[i].min_drain_clk_khz;
+ ranges->writer_wm_sets[i].min_drain_clk_mhz * 1000;
}
pp_funcs->set_watermarks_for_clocks_ranges(pp_handle, &wm_with_clock_ranges);
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
index 0e1dc1b..c2ab026 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c
@@ -2030,7 +2030,7 @@ static uint32_t get_src_obj_list(struct bios_parser *bp, ATOM_OBJECT *object,
static struct device_id device_type_from_device_id(uint16_t device_id)
{
- struct device_id result_device_id;
+ struct device_id result_device_id = {0};
switch (device_id) {
case ATOM_DEVICE_LCD1_SUPPORT:
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
index ff764da..751bb61 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c
@@ -1884,6 +1884,8 @@ static const struct dc_vbios_funcs vbios_funcs = {
.is_accelerated_mode = bios_parser_is_accelerated_mode,
+ .is_active_display = bios_is_active_display,
+
.set_scratch_critical_state = bios_parser_set_scratch_critical_state,
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c
index d458947..fdda8aa 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.c
@@ -88,3 +88,96 @@ uint32_t bios_get_vga_enabled_displays(
return active_disp;
}
+bool bios_is_active_display(
+ struct dc_bios *bios,
+ enum signal_type signal,
+ const struct connector_device_tag_info *device_tag)
+{
+ uint32_t active = 0;
+ uint32_t connected = 0;
+ uint32_t bios_scratch_0 = 0;
+ uint32_t bios_scratch_3 = 0;
+
+ switch (signal) {
+ case SIGNAL_TYPE_DVI_SINGLE_LINK:
+ case SIGNAL_TYPE_DVI_DUAL_LINK:
+ case SIGNAL_TYPE_HDMI_TYPE_A:
+ case SIGNAL_TYPE_DISPLAY_PORT:
+ case SIGNAL_TYPE_DISPLAY_PORT_MST:
+ {
+ if (device_tag->dev_id.device_type == DEVICE_TYPE_DFP) {
+ switch (device_tag->dev_id.enum_id) {
+ case 1:
+ {
+ active = ATOM_S3_DFP1_ACTIVE;
+ connected = 0x0008; //ATOM_DISPLAY_DFP1_CONNECT
+ }
+ break;
+
+ case 2:
+ {
+ active = ATOM_S3_DFP2_ACTIVE;
+ connected = 0x0080; //ATOM_DISPLAY_DFP2_CONNECT
+ }
+ break;
+
+ case 3:
+ {
+ active = ATOM_S3_DFP3_ACTIVE;
+ connected = 0x0200; //ATOM_DISPLAY_DFP3_CONNECT
+ }
+ break;
+
+ case 4:
+ {
+ active = ATOM_S3_DFP4_ACTIVE;
+ connected = 0x0400; //ATOM_DISPLAY_DFP4_CONNECT
+ }
+ break;
+
+ case 5:
+ {
+ active = ATOM_S3_DFP5_ACTIVE;
+ connected = 0x0800; //ATOM_DISPLAY_DFP5_CONNECT
+ }
+ break;
+
+ case 6:
+ {
+ active = ATOM_S3_DFP6_ACTIVE;
+ connected = 0x0040; //ATOM_DISPLAY_DFP6_CONNECT
+ }
+ break;
+
+ default:
+ break;
+ }
+ }
+ }
+ break;
+
+ case SIGNAL_TYPE_LVDS:
+ case SIGNAL_TYPE_EDP:
+ {
+ active = ATOM_S3_LCD1_ACTIVE;
+ connected = 0x0002; //ATOM_DISPLAY_LCD1_CONNECT
+ }
+ break;
+
+ default:
+ break;
+ }
+
+
+ if (bios->regs->BIOS_SCRATCH_0) /*follow up with other asic, todo*/
+ bios_scratch_0 = REG_READ(BIOS_SCRATCH_0);
+ if (bios->regs->BIOS_SCRATCH_3) /*follow up with other asic, todo*/
+ bios_scratch_3 = REG_READ(BIOS_SCRATCH_3);
+
+ bios_scratch_3 &= ATOM_S3_DEVICE_ACTIVE_MASK;
+ if ((active & bios_scratch_3) && (connected & bios_scratch_0))
+ return true;
+
+ return false;
+}
+
diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h
index 75a29e6..f33cac2 100644
--- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h
+++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h
@@ -35,6 +35,10 @@ bool bios_is_accelerated_mode(struct dc_bios *bios);
void bios_set_scratch_acc_mode_change(struct dc_bios *bios);
void bios_set_scratch_critical_state(struct dc_bios *bios, bool state);
uint32_t bios_get_vga_enabled_displays(struct dc_bios *bios);
+bool bios_is_active_display(
+ struct dc_bios *bios,
+ enum signal_type signal,
+ const struct connector_device_tag_info *device_tag);
#define GET_IMAGE(type, offset) ((type *) bios_get_image(&bp->base, offset, sizeof(type)))
diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
index 3208188..43e4a2b 100644
--- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
+++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
@@ -1423,27 +1423,27 @@ void dcn_bw_notify_pplib_of_wm_ranges(struct dc *dc)
ranges.num_reader_wm_sets = WM_SET_COUNT;
ranges.num_writer_wm_sets = WM_SET_COUNT;
ranges.reader_wm_sets[0].wm_inst = WM_A;
- ranges.reader_wm_sets[0].min_drain_clk_khz = min_dcfclk_khz;
- ranges.reader_wm_sets[0].max_drain_clk_khz = overdrive;
- ranges.reader_wm_sets[0].min_fill_clk_khz = min_fclk_khz;
- ranges.reader_wm_sets[0].max_fill_clk_khz = overdrive;
+ ranges.reader_wm_sets[0].min_drain_clk_mhz = min_dcfclk_khz / 1000;
+ ranges.reader_wm_sets[0].max_drain_clk_mhz = overdrive / 1000;
+ ranges.reader_wm_sets[0].min_fill_clk_mhz = min_fclk_khz / 1000;
+ ranges.reader_wm_sets[0].max_fill_clk_mhz = overdrive / 1000;
ranges.writer_wm_sets[0].wm_inst = WM_A;
- ranges.writer_wm_sets[0].min_fill_clk_khz = socclk_khz;
- ranges.writer_wm_sets[0].max_fill_clk_khz = overdrive;
- ranges.writer_wm_sets[0].min_drain_clk_khz = min_fclk_khz;
- ranges.writer_wm_sets[0].max_drain_clk_khz = overdrive;
+ ranges.writer_wm_sets[0].min_fill_clk_mhz = socclk_khz / 1000;
+ ranges.writer_wm_sets[0].max_fill_clk_mhz = overdrive / 1000;
+ ranges.writer_wm_sets[0].min_drain_clk_mhz = min_fclk_khz / 1000;
+ ranges.writer_wm_sets[0].max_drain_clk_mhz = overdrive / 1000;
if (dc->debug.pplib_wm_report_mode == WM_REPORT_OVERRIDE) {
ranges.reader_wm_sets[0].wm_inst = WM_A;
- ranges.reader_wm_sets[0].min_drain_clk_khz = 300000;
- ranges.reader_wm_sets[0].max_drain_clk_khz = 5000000;
- ranges.reader_wm_sets[0].min_fill_clk_khz = 800000;
- ranges.reader_wm_sets[0].max_fill_clk_khz = 5000000;
+ ranges.reader_wm_sets[0].min_drain_clk_mhz = 300;
+ ranges.reader_wm_sets[0].max_drain_clk_mhz = 5000;
+ ranges.reader_wm_sets[0].min_fill_clk_mhz = 800;
+ ranges.reader_wm_sets[0].max_fill_clk_mhz = 5000;
ranges.writer_wm_sets[0].wm_inst = WM_A;
- ranges.writer_wm_sets[0].min_fill_clk_khz = 200000;
- ranges.writer_wm_sets[0].max_fill_clk_khz = 5000000;
- ranges.writer_wm_sets[0].min_drain_clk_khz = 800000;
- ranges.writer_wm_sets[0].max_drain_clk_khz = 5000000;
+ ranges.writer_wm_sets[0].min_fill_clk_mhz = 200;
+ ranges.writer_wm_sets[0].max_fill_clk_mhz = 5000;
+ ranges.writer_wm_sets[0].min_drain_clk_mhz = 800;
+ ranges.writer_wm_sets[0].max_drain_clk_mhz = 5000;
}
ranges.reader_wm_sets[1] = ranges.writer_wm_sets[0];
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c
index 7c491c9..3279e26 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc.c
@@ -391,9 +391,11 @@ bool dc_stream_program_csc_matrix(struct dc *dc, struct dc_stream_state *stream)
== stream) {
pipes = &dc->current_state->res_ctx.pipe_ctx[i];
- dc->hwss.program_csc_matrix(pipes,
- stream->output_color_space,
- stream->csc_color_matrix.matrix);
+ dc->hwss.program_output_csc(dc,
+ pipes,
+ stream->output_color_space,
+ stream->csc_color_matrix.matrix,
+ pipes->plane_res.hubp->opp_id);
ret = true;
}
}
@@ -941,7 +943,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
if (!dcb->funcs->is_accelerated_mode(dcb))
dc->hwss.enable_accelerated_mode(dc, context);
- dc->hwss.set_bandwidth(dc, context, false);
+ dc->hwss.prepare_bandwidth(dc, context);
/* re-program planes for existing stream, in case we need to
* free up plane resource for later use
@@ -957,8 +959,6 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
}
/* Program hardware */
- dc->hwss.ready_shared_resources(dc, context);
-
for (i = 0; i < dc->res_pool->pipe_count; i++) {
pipe = &context->res_ctx.pipe_ctx[i];
dc->hwss.wait_for_mpcc_disconnect(dc, dc->res_pool, pipe);
@@ -1012,7 +1012,7 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
dc_enable_stereo(dc, context, dc_streams, context->stream_count);
/* pplib is notified if disp_num changed */
- dc->hwss.set_bandwidth(dc, context, true);
+ dc->hwss.optimize_bandwidth(dc, context);
dc_release_state(dc->current_state);
@@ -1020,8 +1020,6 @@ static enum dc_status dc_commit_state_no_check(struct dc *dc, struct dc_state *c
dc_retain_state(dc->current_state);
- dc->hwss.optimize_shared_resources(dc);
-
return result;
}
@@ -1063,7 +1061,7 @@ bool dc_post_update_surfaces_to_stream(struct dc *dc)
dc->optimized_required = false;
- dc->hwss.set_bandwidth(dc, context, true);
+ dc->hwss.optimize_bandwidth(dc, context);
return true;
}
@@ -1369,35 +1367,6 @@ static struct dc_stream_status *stream_get_status(
static const enum surface_update_type update_surface_trace_level = UPDATE_TYPE_FULL;
-static void notify_display_count_to_smu(
- struct dc *dc,
- struct dc_state *context)
-{
- int i, display_count;
- struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu;
-
- /*
- * if function pointer not set up, this message is
- * sent as part of pplib_apply_display_requirements.
- * So just return.
- */
- if (!pp_smu || !pp_smu->set_display_count)
- return;
-
- display_count = 0;
- for (i = 0; i < context->stream_count; i++) {
- const struct dc_stream_state *stream = context->streams[i];
-
- /* only notify active stream */
- if (stream->dpms_off)
- continue;
-
- display_count++;
- }
-
- pp_smu->set_display_count(&pp_smu->pp_smu, display_count);
-}
-
static void commit_planes_do_stream_update(struct dc *dc,
struct dc_stream_state *stream,
struct dc_stream_update *stream_update,
@@ -1422,7 +1391,6 @@ static void commit_planes_do_stream_update(struct dc *dc,
stream_update->adjust->v_total_max);
if (stream_update->periodic_fn_vsync_delta &&
- pipe_ctx->stream_res.tg &&
pipe_ctx->stream_res.tg->funcs->program_vline_interrupt)
pipe_ctx->stream_res.tg->funcs->program_vline_interrupt(
pipe_ctx->stream_res.tg, &pipe_ctx->stream->timing,
@@ -1448,19 +1416,13 @@ static void commit_planes_do_stream_update(struct dc *dc,
if (stream_update->dpms_off) {
if (*stream_update->dpms_off) {
core_link_disable_stream(pipe_ctx, KEEP_ACQUIRED_RESOURCE);
- dc->hwss.pplib_apply_display_requirements(
- dc, dc->current_state);
- notify_display_count_to_smu(dc, dc->current_state);
+ dc->hwss.optimize_bandwidth(dc, dc->current_state);
} else {
- dc->hwss.pplib_apply_display_requirements(
- dc, dc->current_state);
- notify_display_count_to_smu(dc, dc->current_state);
+ dc->hwss.prepare_bandwidth(dc, dc->current_state);
core_link_enable_stream(dc->current_state, pipe_ctx);
}
}
-
-
if (stream_update->abm_level && pipe_ctx->stream_res.abm) {
if (pipe_ctx->stream_res.tg->funcs->is_blanked) {
// if otg funcs defined check if blanked before programming
@@ -1487,7 +1449,7 @@ static void commit_planes_for_stream(struct dc *dc,
struct pipe_ctx *top_pipe_to_program = NULL;
if (update_type == UPDATE_TYPE_FULL) {
- dc->hwss.set_bandwidth(dc, context, false);
+ dc->hwss.prepare_bandwidth(dc, context);
context_clock_trace(dc, context);
}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
index e1ebdf7..73d0495 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_debug.c
@@ -311,7 +311,7 @@ void context_timing_trace(
{
int i;
struct dc *core_dc = dc;
- int h_pos[MAX_PIPES], v_pos[MAX_PIPES];
+ int h_pos[MAX_PIPES] = {0}, v_pos[MAX_PIPES] = {0};
struct crtc_position position;
unsigned int underlay_idx = core_dc->res_pool->underlay_pipe_index;
DC_LOGGER_INIT(dc->ctx->logger);
@@ -322,8 +322,7 @@ void context_timing_trace(
/* get_position() returns CRTC vertical/horizontal counter
* hence not applicable for underlay pipe
*/
- if (pipe_ctx->stream == NULL
- || pipe_ctx->pipe_idx == underlay_idx)
+ if (pipe_ctx->stream == NULL || pipe_ctx->pipe_idx == underlay_idx)
continue;
pipe_ctx->stream_res.tg->funcs->get_position(pipe_ctx->stream_res.tg, &position);
@@ -333,7 +332,7 @@ void context_timing_trace(
for (i = 0; i < core_dc->res_pool->pipe_count; i++) {
struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[i];
- if (pipe_ctx->stream == NULL)
+ if (pipe_ctx->stream == NULL || pipe_ctx->pipe_idx == underlay_idx)
continue;
TIMING_TRACE("OTG_%d H_tot:%d V_tot:%d H_pos:%d V_pos:%d\n",
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
index fb04a4a..7ee9c03 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c
@@ -1357,28 +1357,13 @@ static enum dc_status enable_link_dp(
struct dc_link *link = stream->sink->link;
struct dc_link_settings link_settings = {0};
enum dp_panel_mode panel_mode;
- enum dc_link_rate max_link_rate = LINK_RATE_HIGH2;
/* get link settings for video mode timing */
decide_link_settings(stream, &link_settings);
- /* raise clock state for HBR3 if required. Confirmed with HW DCE/DPCS
- * logic for HBR3 still needs Nominal (0.8V) on VDDC rail
- */
- if (link->link_enc->features.flags.bits.IS_HBR3_CAPABLE)
- max_link_rate = LINK_RATE_HIGH3;
-
- if (link_settings.link_rate == max_link_rate) {
- struct dc_clocks clocks = state->bw.dcn.clk;
-
- /* dce/dcn compat, do not update dispclk */
- clocks.dispclk_khz = 0;
- /* 27mhz = 27000000hz= 27000khz */
- clocks.phyclk_khz = link_settings.link_rate * 27000;
-
- state->dis_clk->funcs->update_clocks(
- state->dis_clk, &clocks, false);
- }
+ pipe_ctx->stream_res.pix_clk_params.requested_sym_clk =
+ link_settings.link_rate * LINK_RATE_REF_FREQ_IN_KHZ;
+ state->dccg->funcs->update_clocks(state->dccg, state, false);
dp_enable_link_phy(
link,
@@ -1722,7 +1707,7 @@ static void write_i2c_retimer_setting(
i2c_success = i2c_write(pipe_ctx, slave_address,
buffer, sizeof(buffer));
RETIMER_REDRIVER_INFO("retimer write to slave_address = 0x%x,\
- offset = 0x%d, reg_val = 0x%d, i2c_success = %d\n",
+ offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n",
slave_address, buffer[0], buffer[1], i2c_success?1:0);
if (!i2c_success)
/* Write failure */
@@ -1734,7 +1719,7 @@ static void write_i2c_retimer_setting(
i2c_success = i2c_write(pipe_ctx, slave_address,
buffer, sizeof(buffer));
RETIMER_REDRIVER_INFO("retimer write to slave_address = 0x%x,\
- offset = 0x%d, reg_val = 0x%d, i2c_success = %d\n",
+ offset = 0x%x, reg_val = 0x%x, i2c_success = %d\n",
slave_address, buffer[0], buffer[1], i2c_success?1:0);
if (!i2c_success)
/* Write failure */
@@ -2156,14 +2141,16 @@ int dc_link_get_backlight_level(const struct dc_link *link)
{
struct abm *abm = link->ctx->dc->res_pool->abm;
- if (abm == NULL || abm->funcs->get_current_backlight_8_bit == NULL)
+ if (abm == NULL || abm->funcs->get_current_backlight == NULL)
return DC_ERROR_UNEXPECTED;
- return (int) abm->funcs->get_current_backlight_8_bit(abm);
+ return (int) abm->funcs->get_current_backlight(abm);
}
-bool dc_link_set_backlight_level(const struct dc_link *link, uint32_t level,
- uint32_t frame_ramp, const struct dc_stream_state *stream)
+bool dc_link_set_backlight_level(const struct dc_link *link,
+ uint32_t backlight_pwm_u16_16,
+ uint32_t frame_ramp,
+ const struct dc_stream_state *stream)
{
struct dc *core_dc = link->ctx->dc;
struct abm *abm = core_dc->res_pool->abm;
@@ -2175,19 +2162,17 @@ bool dc_link_set_backlight_level(const struct dc_link *link, uint32_t level,
if ((dmcu == NULL) ||
(abm == NULL) ||
- (abm->funcs->set_backlight_level == NULL))
+ (abm->funcs->set_backlight_level_pwm == NULL))
return false;
- if (stream) {
- if (stream->bl_pwm_level == EDP_BACKLIGHT_RAMP_DISABLE_LEVEL)
- frame_ramp = 0;
-
- ((struct dc_stream_state *)stream)->bl_pwm_level = level;
- }
+ if (stream)
+ ((struct dc_stream_state *)stream)->bl_pwm_level =
+ backlight_pwm_u16_16;
use_smooth_brightness = dmcu->funcs->is_dmcu_initialized(dmcu);
- DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n", level, level);
+ DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n",
+ backlight_pwm_u16_16, backlight_pwm_u16_16);
if (dc_is_embedded_signal(link->connector_signal)) {
if (stream != NULL) {
@@ -2204,9 +2189,9 @@ bool dc_link_set_backlight_level(const struct dc_link *link, uint32_t level,
1;
}
}
- abm->funcs->set_backlight_level(
+ abm->funcs->set_backlight_level_pwm(
abm,
- level,
+ backlight_pwm_u16_16,
frame_ramp,
controller_id,
use_smooth_brightness);
@@ -2220,7 +2205,7 @@ bool dc_link_set_abm_disable(const struct dc_link *link)
struct dc *core_dc = link->ctx->dc;
struct abm *abm = core_dc->res_pool->abm;
- if ((abm == NULL) || (abm->funcs->set_backlight_level == NULL))
+ if ((abm == NULL) || (abm->funcs->set_backlight_level_pwm == NULL))
return false;
abm->funcs->set_abm_immediate_disable(abm);
@@ -2609,6 +2594,10 @@ void core_link_enable_stream(
core_dc->hwss.unblank_stream(pipe_ctx,
&pipe_ctx->stream->sink->link->cur_link_settings);
+ dc_link_set_backlight_level(pipe_ctx->stream->sink->link,
+ pipe_ctx->stream->bl_pwm_level,
+ 0,
+ pipe_ctx->stream);
}
}
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
index b6fe29b..fc65b00 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c
@@ -499,8 +499,13 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx)
pipe_ctx->top_pipe->plane_state == pipe_ctx->plane_state;
bool flip_vert_scan_dir = false, flip_horz_scan_dir = false;
+
/*
- * Need to calculate the scan direction for viewport to properly determine offset
+ * We need take horizontal mirror into account. On an unrotated surface this means
+ * that the viewport offset is actually the offset from the other side of source
+ * image so we have to subtract the right edge of the viewport from the right edge of
+ * the source window. Similar to mirror we need to take into account how offset is
+ * affected for 270/180 rotations
*/
if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_180) {
flip_vert_scan_dir = true;
@@ -510,6 +515,9 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx)
else if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270)
flip_horz_scan_dir = true;
+ if (pipe_ctx->plane_state->horizontal_mirror)
+ flip_horz_scan_dir = !flip_horz_scan_dir;
+
if (stream->view_format == VIEW_3D_FORMAT_SIDE_BY_SIDE ||
stream->view_format == VIEW_3D_FORMAT_TOP_AND_BOTTOM) {
pri_split = false;
@@ -540,45 +548,27 @@ static void calculate_viewport(struct pipe_ctx *pipe_ctx)
plane_state->clip_rect.y + plane_state->clip_rect.height - clip.y ;
/* offset = surf_src.ofs + (clip.ofs - surface->dst_rect.ofs) * scl_ratio
+ * note: surf_src.ofs should be added after rotation/mirror offset direction
+ * adjustment since it is already in viewport space
* num_pixels = clip.num_pix * scl_ratio
*/
- data->viewport.x = surf_src.x + (clip.x - plane_state->dst_rect.x) *
+ data->viewport.x = (clip.x - plane_state->dst_rect.x) *
surf_src.width / plane_state->dst_rect.width;
data->viewport.width = clip.width *
surf_src.width / plane_state->dst_rect.width;
- data->viewport.y = surf_src.y + (clip.y - plane_state->dst_rect.y) *
+ data->viewport.y = (clip.y - plane_state->dst_rect.y) *
surf_src.height / plane_state->dst_rect.height;
data->viewport.height = clip.height *
surf_src.height / plane_state->dst_rect.height;
- /* To transfer the x, y to correct coordinate on mirror image (camera).
- * deg 0 : transfer x,
- * deg 90 : don't need to transfer,
- * deg180 : transfer y,
- * deg270 : transfer x and y.
- * To transfer the x, y to correct coordinate on non-mirror image (video).
- * deg 0 : don't need to transfer,
- * deg 90 : transfer y,
- * deg180 : transfer x and y,
- * deg270 : transfer x.
- */
- if (pipe_ctx->plane_state->horizontal_mirror) {
- if (flip_horz_scan_dir && !flip_vert_scan_dir) {
- data->viewport.y = surf_src.height - data->viewport.y - data->viewport.height;
- data->viewport.x = surf_src.width - data->viewport.x - data->viewport.width;
- } else if (flip_horz_scan_dir && flip_vert_scan_dir)
- data->viewport.y = surf_src.height - data->viewport.y - data->viewport.height;
- else {
- if (!flip_horz_scan_dir && !flip_vert_scan_dir)
- data->viewport.x = surf_src.width - data->viewport.x - data->viewport.width;
- }
- } else {
- if (flip_horz_scan_dir)
- data->viewport.x = surf_src.width - data->viewport.x - data->viewport.width;
- if (flip_vert_scan_dir)
- data->viewport.y = surf_src.height - data->viewport.y - data->viewport.height;
- }
+ if (flip_vert_scan_dir)
+ data->viewport.y = surf_src.height - data->viewport.y - data->viewport.height;
+ if (flip_horz_scan_dir)
+ data->viewport.x = surf_src.width - data->viewport.x - data->viewport.width;
+
+ data->viewport.x += surf_src.x;
+ data->viewport.y += surf_src.y;
/* Round down, compensate in init */
data->viewport_c.x = data->viewport.x / vpc_div;
@@ -773,22 +763,15 @@ static void calculate_inits_and_adj_vp(struct pipe_ctx *pipe_ctx, struct rect *r
else if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270)
flip_horz_scan_dir = true;
+ if (pipe_ctx->plane_state->horizontal_mirror)
+ flip_horz_scan_dir = !flip_horz_scan_dir;
+
if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 ||
pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270) {
rect_swap_helper(&src);
rect_swap_helper(&data->viewport_c);
rect_swap_helper(&data->viewport);
-
- if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_270 &&
- pipe_ctx->plane_state->horizontal_mirror) {
- flip_vert_scan_dir = true;
- }
- if (pipe_ctx->plane_state->rotation == ROTATION_ANGLE_90 &&
- pipe_ctx->plane_state->horizontal_mirror) {
- flip_vert_scan_dir = false;
- }
- } else if (pipe_ctx->plane_state->horizontal_mirror)
- flip_horz_scan_dir = !flip_horz_scan_dir;
+ }
/*
* Init calculated according to formula:
@@ -1115,9 +1098,6 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx)
pipe_ctx->plane_res.scl_data.format = convert_pixel_format_to_dalsurface(
pipe_ctx->plane_state->format);
- if (pipe_ctx->stream->timing.flags.INTERLACE)
- pipe_ctx->stream->dst.height *= 2;
-
calculate_scaling_ratios(pipe_ctx);
calculate_viewport(pipe_ctx);
@@ -1138,9 +1118,6 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx)
pipe_ctx->plane_res.scl_data.h_active = timing->h_addressable + timing->h_border_left + timing->h_border_right;
pipe_ctx->plane_res.scl_data.v_active = timing->v_addressable + timing->v_border_top + timing->v_border_bottom;
- if (pipe_ctx->stream->timing.flags.INTERLACE)
- pipe_ctx->plane_res.scl_data.v_active *= 2;
-
/* Taps calculations */
if (pipe_ctx->plane_res.xfm != NULL)
@@ -1185,9 +1162,6 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx)
plane_state->dst_rect.x,
plane_state->dst_rect.y);
- if (pipe_ctx->stream->timing.flags.INTERLACE)
- pipe_ctx->stream->dst.height /= 2;
-
return res;
}
@@ -2071,7 +2045,7 @@ void dc_resource_state_construct(
const struct dc *dc,
struct dc_state *dst_ctx)
{
- dst_ctx->dis_clk = dc->res_pool->dccg;
+ dst_ctx->dccg = dc->res_pool->clk_mgr;
}
enum dc_status dc_validate_global_state(
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
index 2ac848a1..e113439 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_stream.c
@@ -106,6 +106,7 @@ static void construct(struct dc_stream_state *stream,
stream->out_transfer_func = dc_create_transfer_func();
stream->out_transfer_func->type = TF_TYPE_BYPASS;
+ stream->out_transfer_func->ctx = stream->ctx;
}
static void destruct(struct dc_stream_state *stream)
diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
index 8fb3aef..c60c9b4 100644
--- a/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
+++ b/drivers/gpu/drm/amd/display/dc/core/dc_surface.c
@@ -44,6 +44,7 @@ static void construct(struct dc_context *ctx, struct dc_plane_state *plane_state
plane_state->in_transfer_func = dc_create_transfer_func();
plane_state->in_transfer_func->type = TF_TYPE_BYPASS;
+ plane_state->in_transfer_func->ctx = ctx;
}
static void destruct(struct dc_plane_state *plane_state)
diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h
index 1995271..d16a20c 100644
--- a/drivers/gpu/drm/amd/display/dc/dc.h
+++ b/drivers/gpu/drm/amd/display/dc/dc.h
@@ -38,7 +38,7 @@
#include "inc/compressor.h"
#include "dml/display_mode_lib.h"
-#define DC_VER "3.1.68"
+#define DC_VER "3.2.04"
#define MAX_SURFACES 3
#define MAX_STREAMS 6
@@ -169,6 +169,7 @@ struct link_training_settings;
struct dc_config {
bool gpu_vm_support;
bool disable_disp_pll_sharing;
+ bool fbc_support;
};
enum visual_confirm {
@@ -249,8 +250,6 @@ struct dc_debug_options {
bool disable_dmcu;
bool disable_psr;
bool force_abm_enable;
- bool disable_hbup_pg;
- bool disable_dpp_pg;
bool disable_stereo_support;
bool vsr_support;
bool performance_trace;
@@ -304,11 +303,6 @@ struct dc {
struct hw_sequencer_funcs hwss;
struct dce_hwseq *hwseq;
- /* temp store of dm_pp_display_configuration
- * to compare to see if display config changed
- */
- struct dm_pp_display_configuration prev_display_config;
-
bool optimized_required;
/* FBC compressor */
diff --git a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h
index 8130b95..a8b3ced 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_bios_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_bios_types.h
@@ -86,6 +86,10 @@ struct dc_vbios_funcs {
bool (*is_accelerated_mode)(
struct dc_bios *bios);
+ bool (*is_active_display)(
+ struct dc_bios *bios,
+ enum signal_type signal,
+ const struct connector_device_tag_info *device_tag);
void (*set_scratch_critical_state)(
struct dc_bios *bios,
bool state);
@@ -141,6 +145,7 @@ struct dc_vbios_funcs {
};
struct bios_registers {
+ uint32_t BIOS_SCRATCH_0;
uint32_t BIOS_SCRATCH_3;
uint32_t BIOS_SCRATCH_6;
};
diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h
index 3bfdccc..8738f27 100644
--- a/drivers/gpu/drm/amd/display/dc/dc_link.h
+++ b/drivers/gpu/drm/amd/display/dc/dc_link.h
@@ -138,9 +138,14 @@ static inline struct dc_link *dc_get_link_at_index(struct dc *dc, uint32_t link_
return dc->links[link_index];
}
-/* Set backlight level of an embedded panel (eDP, LVDS). */
-bool dc_link_set_backlight_level(const struct dc_link *dc_link, uint32_t level,
- uint32_t frame_ramp, const struct dc_stream_state *stream);
+/* Set backlight level of an embedded panel (eDP, LVDS).
+ * backlight_pwm_u16_16 is unsigned 32 bit with 16 bit integer
+ * and 16 bit fractional, where 1.0 is max backlight value.
+ */
+bool dc_link_set_backlight_level(const struct dc_link *dc_link,
+ uint32_t backlight_pwm_u16_16,
+ uint32_t frame_ramp,
+ const struct dc_stream_state *stream);
int dc_link_get_backlight_level(const struct dc_link *dc_link);
diff --git a/drivers/gpu/drm/amd/display/dc/dce/Makefile b/drivers/gpu/drm/amd/display/dc/dce/Makefile
index 8f7f0e8..6d7b64a 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dce/Makefile
@@ -28,7 +28,7 @@
DCE = dce_audio.o dce_stream_encoder.o dce_link_encoder.o dce_hwseq.o \
dce_mem_input.o dce_clock_source.o dce_scl_filters.o dce_transform.o \
-dce_clocks.o dce_opp.o dce_dmcu.o dce_abm.o dce_ipp.o dce_aux.o \
+dce_clk_mgr.o dce_opp.o dce_dmcu.o dce_abm.o dce_ipp.o dce_aux.o \
dce_i2c.o dce_i2c_hw.o dce_i2c_sw.o
AMD_DAL_DCE = $(addprefix $(AMDDALPATH)/dc/dce/,$(DCE))
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c
index 29294db..2a342ea 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_abm.c
@@ -54,7 +54,7 @@
#define MCP_DISABLE_ABM_IMMEDIATELY 255
-static unsigned int get_current_backlight_16_bit(struct dce_abm *abm_dce)
+static unsigned int calculate_16_bit_backlight_from_pwm(struct dce_abm *abm_dce)
{
uint64_t current_backlight;
uint32_t round_result;
@@ -103,45 +103,21 @@ static unsigned int get_current_backlight_16_bit(struct dce_abm *abm_dce)
return (uint32_t)(current_backlight);
}
-static void driver_set_backlight_level(struct dce_abm *abm_dce, uint32_t level)
+static void driver_set_backlight_level(struct dce_abm *abm_dce,
+ uint32_t backlight_pwm_u16_16)
{
- uint32_t backlight_24bit;
- uint32_t backlight_17bit;
uint32_t backlight_16bit;
uint32_t masked_pwm_period;
- uint8_t rounding_bit;
uint8_t bit_count;
uint64_t active_duty_cycle;
uint32_t pwm_period_bitcnt;
/*
- * 1. Convert 8-bit value to 17 bit U1.16 format
- * (1 integer, 16 fractional bits)
- */
-
- /* 1.1 multiply 8 bit value by 0x10101 to get a 24 bit value,
- * effectively multiplying value by 256/255
- * eg. for a level of 0xEF, backlight_24bit = 0xEF * 0x10101 = 0xEFEFEF
- */
- backlight_24bit = level * 0x10101;
-
- /* 1.2 The upper 16 bits of the 24 bit value is the fraction, lower 8
- * used for rounding, take most significant bit of fraction for
- * rounding, e.g. for 0xEFEFEF, rounding bit is 1
- */
- rounding_bit = (backlight_24bit >> 7) & 1;
-
- /* 1.3 Add the upper 16 bits of the 24 bit value with the rounding bit
- * resulting in a 17 bit value e.g. 0xEFF0 = (0xEFEFEF >> 8) + 1
- */
- backlight_17bit = (backlight_24bit >> 8) + rounding_bit;
-
- /*
- * 2. Find 16 bit backlight active duty cycle, where 0 <= backlight
+ * 1. Find 16 bit backlight active duty cycle, where 0 <= backlight
* active duty cycle <= backlight period
*/
- /* 2.1 Apply bitmask for backlight period value based on value of BITCNT
+ /* 1.1 Apply bitmask for backlight period value based on value of BITCNT
*/
REG_GET_2(BL_PWM_PERIOD_CNTL,
BL_PWM_PERIOD_BITCNT, &pwm_period_bitcnt,
@@ -155,13 +131,13 @@ static void driver_set_backlight_level(struct dce_abm *abm_dce, uint32_t level)
/* e.g. maskedPwmPeriod = 0x24 when bitCount is 6 */
masked_pwm_period = masked_pwm_period & ((1 << bit_count) - 1);
- /* 2.2 Calculate integer active duty cycle required upper 16 bits
+ /* 1.2 Calculate integer active duty cycle required upper 16 bits
* contain integer component, lower 16 bits contain fractional component
* of active duty cycle e.g. 0x21BDC0 = 0xEFF0 * 0x24
*/
- active_duty_cycle = backlight_17bit * masked_pwm_period;
+ active_duty_cycle = backlight_pwm_u16_16 * masked_pwm_period;
- /* 2.3 Calculate 16 bit active duty cycle from integer and fractional
+ /* 1.3 Calculate 16 bit active duty cycle from integer and fractional
* components shift by bitCount then mask 16 bits and add rounding bit
* from MSB of fraction e.g. 0x86F7 = ((0x21BDC0 >> 6) & 0xFFF) + 0
*/
@@ -170,23 +146,23 @@ static void driver_set_backlight_level(struct dce_abm *abm_dce, uint32_t level)
backlight_16bit += (active_duty_cycle >> (bit_count - 1)) & 0x1;
/*
- * 3. Program register with updated value
+ * 2. Program register with updated value
*/
- /* 3.1 Lock group 2 backlight registers */
+ /* 2.1 Lock group 2 backlight registers */
REG_UPDATE_2(BL_PWM_GRP1_REG_LOCK,
BL_PWM_GRP1_IGNORE_MASTER_LOCK_EN, 1,
BL_PWM_GRP1_REG_LOCK, 1);
- // 3.2 Write new active duty cycle
+ // 2.2 Write new active duty cycle
REG_UPDATE(BL_PWM_CNTL, BL_ACTIVE_INT_FRAC_CNT, backlight_16bit);
- /* 3.3 Unlock group 2 backlight registers */
+ /* 2.3 Unlock group 2 backlight registers */
REG_UPDATE(BL_PWM_GRP1_REG_LOCK,
BL_PWM_GRP1_REG_LOCK, 0);
- /* 5.4.4 Wait for pending bit to be cleared */
+ /* 3 Wait for pending bit to be cleared */
REG_WAIT(BL_PWM_GRP1_REG_LOCK,
BL_PWM_GRP1_REG_UPDATE_PENDING, 0,
1, 10000);
@@ -194,16 +170,21 @@ static void driver_set_backlight_level(struct dce_abm *abm_dce, uint32_t level)
static void dmcu_set_backlight_level(
struct dce_abm *abm_dce,
- uint32_t level,
+ uint32_t backlight_pwm_u16_16,
uint32_t frame_ramp,
uint32_t controller_id)
{
- unsigned int backlight_16_bit = (level * 0x10101) >> 8;
- unsigned int backlight_17_bit = backlight_16_bit +
- (((backlight_16_bit & 0x80) >> 7) & 1);
+ unsigned int backlight_8_bit = 0;
uint32_t rampingBoundary = 0xFFFF;
uint32_t s2;
+ if (backlight_pwm_u16_16 & 0x10000)
+ // Check for max backlight condition
+ backlight_8_bit = 0xFF;
+ else
+ // Take MSB of fractional part since backlight is not max
+ backlight_8_bit = (backlight_pwm_u16_16 >> 8) & 0xFF;
+
/* set ramping boundary */
REG_WRITE(MASTER_COMM_DATA_REG1, rampingBoundary);
@@ -220,7 +201,7 @@ static void dmcu_set_backlight_level(
0, 1, 80000);
/* setDMCUParam_BL */
- REG_UPDATE(BL1_PWM_USER_LEVEL, BL1_PWM_USER_LEVEL, backlight_17_bit);
+ REG_UPDATE(BL1_PWM_USER_LEVEL, BL1_PWM_USER_LEVEL, backlight_pwm_u16_16);
/* write ramp */
if (controller_id == 0)
@@ -237,9 +218,9 @@ static void dmcu_set_backlight_level(
s2 = REG_READ(BIOS_SCRATCH_2);
s2 &= ~ATOM_S2_CURRENT_BL_LEVEL_MASK;
- level &= (ATOM_S2_CURRENT_BL_LEVEL_MASK >>
+ backlight_8_bit &= (ATOM_S2_CURRENT_BL_LEVEL_MASK >>
ATOM_S2_CURRENT_BL_LEVEL_SHIFT);
- s2 |= (level << ATOM_S2_CURRENT_BL_LEVEL_SHIFT);
+ s2 |= (backlight_8_bit << ATOM_S2_CURRENT_BL_LEVEL_SHIFT);
REG_WRITE(BIOS_SCRATCH_2, s2);
}
@@ -247,7 +228,7 @@ static void dmcu_set_backlight_level(
static void dce_abm_init(struct abm *abm)
{
struct dce_abm *abm_dce = TO_DCE_ABM(abm);
- unsigned int backlight = get_current_backlight_16_bit(abm_dce);
+ unsigned int backlight = calculate_16_bit_backlight_from_pwm(abm_dce);
REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x103);
REG_WRITE(DC_ABM1_HG_SAMPLE_RATE, 0x101);
@@ -284,12 +265,26 @@ static void dce_abm_init(struct abm *abm)
ABM1_BL_REG_READ_MISSED_FRAME_CLEAR, 1);
}
-static unsigned int dce_abm_get_current_backlight_8_bit(struct abm *abm)
+static unsigned int dce_abm_get_current_backlight(struct abm *abm)
{
struct dce_abm *abm_dce = TO_DCE_ABM(abm);
unsigned int backlight = REG_READ(BL1_PWM_CURRENT_ABM_LEVEL);
- return (backlight >> 8);
+ /* return backlight in hardware format which is unsigned 17 bits, with
+ * 1 bit integer and 16 bit fractional
+ */
+ return backlight;
+}
+
+static unsigned int dce_abm_get_target_backlight(struct abm *abm)
+{
+ struct dce_abm *abm_dce = TO_DCE_ABM(abm);
+ unsigned int backlight = REG_READ(BL1_PWM_TARGET_ABM_LEVEL);
+
+ /* return backlight in hardware format which is unsigned 17 bits, with
+ * 1 bit integer and 16 bit fractional
+ */
+ return backlight;
}
static bool dce_abm_set_level(struct abm *abm, uint32_t level)
@@ -396,9 +391,9 @@ static bool dce_abm_init_backlight(struct abm *abm)
return true;
}
-static bool dce_abm_set_backlight_level(
+static bool dce_abm_set_backlight_level_pwm(
struct abm *abm,
- unsigned int backlight_level,
+ unsigned int backlight_pwm_u16_16,
unsigned int frame_ramp,
unsigned int controller_id,
bool use_smooth_brightness)
@@ -406,16 +401,16 @@ static bool dce_abm_set_backlight_level(
struct dce_abm *abm_dce = TO_DCE_ABM(abm);
DC_LOG_BACKLIGHT("New Backlight level: %d (0x%X)\n",
- backlight_level, backlight_level);
+ backlight_pwm_u16_16, backlight_pwm_u16_16);
/* If DMCU is in reset state, DMCU is uninitialized */
if (use_smooth_brightness)
dmcu_set_backlight_level(abm_dce,
- backlight_level,
+ backlight_pwm_u16_16,
frame_ramp,
controller_id);
else
- driver_set_backlight_level(abm_dce, backlight_level);
+ driver_set_backlight_level(abm_dce, backlight_pwm_u16_16);
return true;
}
@@ -424,8 +419,9 @@ static const struct abm_funcs dce_funcs = {
.abm_init = dce_abm_init,
.set_abm_level = dce_abm_set_level,
.init_backlight = dce_abm_init_backlight,
- .set_backlight_level = dce_abm_set_backlight_level,
- .get_current_backlight_8_bit = dce_abm_get_current_backlight_8_bit,
+ .set_backlight_level_pwm = dce_abm_set_backlight_level_pwm,
+ .get_current_backlight = dce_abm_get_current_backlight,
+ .get_target_backlight = dce_abm_get_target_backlight,
.set_abm_immediate_disable = dce_abm_immediate_disable
};
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c
new file mode 100644
index 0000000..9a28a04
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.c
@@ -0,0 +1,879 @@
+/*
+ * Copyright 2012-16 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#include "dce_clk_mgr.h"
+
+#include "reg_helper.h"
+#include "dmcu.h"
+#include "core_types.h"
+#include "dal_asic_id.h"
+
+#define TO_DCE_CLK_MGR(clocks)\
+ container_of(clocks, struct dce_clk_mgr, base)
+
+#define REG(reg) \
+ (clk_mgr_dce->regs->reg)
+
+#undef FN
+#define FN(reg_name, field_name) \
+ clk_mgr_dce->clk_mgr_shift->field_name, clk_mgr_dce->clk_mgr_mask->field_name
+
+#define CTX \
+ clk_mgr_dce->base.ctx
+#define DC_LOGGER \
+ clk_mgr->ctx->logger
+
+/* Max clock values for each state indexed by "enum clocks_state": */
+static const struct state_dependent_clocks dce80_max_clks_by_state[] = {
+/* ClocksStateInvalid - should not be used */
+{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
+/* ClocksStateUltraLow - not expected to be used for DCE 8.0 */
+{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
+/* ClocksStateLow */
+{ .display_clk_khz = 352000, .pixel_clk_khz = 330000},
+/* ClocksStateNominal */
+{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 },
+/* ClocksStatePerformance */
+{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 } };
+
+static const struct state_dependent_clocks dce110_max_clks_by_state[] = {
+/*ClocksStateInvalid - should not be used*/
+{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
+/*ClocksStateUltraLow - currently by HW design team not supposed to be used*/
+{ .display_clk_khz = 352000, .pixel_clk_khz = 330000 },
+/*ClocksStateLow*/
+{ .display_clk_khz = 352000, .pixel_clk_khz = 330000 },
+/*ClocksStateNominal*/
+{ .display_clk_khz = 467000, .pixel_clk_khz = 400000 },
+/*ClocksStatePerformance*/
+{ .display_clk_khz = 643000, .pixel_clk_khz = 400000 } };
+
+static const struct state_dependent_clocks dce112_max_clks_by_state[] = {
+/*ClocksStateInvalid - should not be used*/
+{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
+/*ClocksStateUltraLow - currently by HW design team not supposed to be used*/
+{ .display_clk_khz = 389189, .pixel_clk_khz = 346672 },
+/*ClocksStateLow*/
+{ .display_clk_khz = 459000, .pixel_clk_khz = 400000 },
+/*ClocksStateNominal*/
+{ .display_clk_khz = 667000, .pixel_clk_khz = 600000 },
+/*ClocksStatePerformance*/
+{ .display_clk_khz = 1132000, .pixel_clk_khz = 600000 } };
+
+static const struct state_dependent_clocks dce120_max_clks_by_state[] = {
+/*ClocksStateInvalid - should not be used*/
+{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
+/*ClocksStateUltraLow - currently by HW design team not supposed to be used*/
+{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
+/*ClocksStateLow*/
+{ .display_clk_khz = 460000, .pixel_clk_khz = 400000 },
+/*ClocksStateNominal*/
+{ .display_clk_khz = 670000, .pixel_clk_khz = 600000 },
+/*ClocksStatePerformance*/
+{ .display_clk_khz = 1133000, .pixel_clk_khz = 600000 } };
+
+static int dentist_get_divider_from_did(int did)
+{
+ if (did < DENTIST_BASE_DID_1)
+ did = DENTIST_BASE_DID_1;
+ if (did > DENTIST_MAX_DID)
+ did = DENTIST_MAX_DID;
+
+ if (did < DENTIST_BASE_DID_2) {
+ return DENTIST_DIVIDER_RANGE_1_START + DENTIST_DIVIDER_RANGE_1_STEP
+ * (did - DENTIST_BASE_DID_1);
+ } else if (did < DENTIST_BASE_DID_3) {
+ return DENTIST_DIVIDER_RANGE_2_START + DENTIST_DIVIDER_RANGE_2_STEP
+ * (did - DENTIST_BASE_DID_2);
+ } else if (did < DENTIST_BASE_DID_4) {
+ return DENTIST_DIVIDER_RANGE_3_START + DENTIST_DIVIDER_RANGE_3_STEP
+ * (did - DENTIST_BASE_DID_3);
+ } else {
+ return DENTIST_DIVIDER_RANGE_4_START + DENTIST_DIVIDER_RANGE_4_STEP
+ * (did - DENTIST_BASE_DID_4);
+ }
+}
+
+/* SW will adjust DP REF Clock average value for all purposes
+ * (DP DTO / DP Audio DTO and DP GTC)
+ if clock is spread for all cases:
+ -if SS enabled on DP Ref clock and HW de-spreading enabled with SW
+ calculations for DS_INCR/DS_MODULO (this is planned to be default case)
+ -if SS enabled on DP Ref clock and HW de-spreading enabled with HW
+ calculations (not planned to be used, but average clock should still
+ be valid)
+ -if SS enabled on DP Ref clock and HW de-spreading disabled
+ (should not be case with CIK) then SW should program all rates
+ generated according to average value (case as with previous ASICs)
+ */
+static int clk_mgr_adjust_dp_ref_freq_for_ss(struct dce_clk_mgr *clk_mgr_dce, int dp_ref_clk_khz)
+{
+ if (clk_mgr_dce->ss_on_dprefclk && clk_mgr_dce->dprefclk_ss_divider != 0) {
+ struct fixed31_32 ss_percentage = dc_fixpt_div_int(
+ dc_fixpt_from_fraction(clk_mgr_dce->dprefclk_ss_percentage,
+ clk_mgr_dce->dprefclk_ss_divider), 200);
+ struct fixed31_32 adj_dp_ref_clk_khz;
+
+ ss_percentage = dc_fixpt_sub(dc_fixpt_one, ss_percentage);
+ adj_dp_ref_clk_khz = dc_fixpt_mul_int(ss_percentage, dp_ref_clk_khz);
+ dp_ref_clk_khz = dc_fixpt_floor(adj_dp_ref_clk_khz);
+ }
+ return dp_ref_clk_khz;
+}
+
+static int dce_get_dp_ref_freq_khz(struct clk_mgr *clk_mgr)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+ int dprefclk_wdivider;
+ int dprefclk_src_sel;
+ int dp_ref_clk_khz = 600000;
+ int target_div;
+
+ /* ASSERT DP Reference Clock source is from DFS*/
+ REG_GET(DPREFCLK_CNTL, DPREFCLK_SRC_SEL, &dprefclk_src_sel);
+ ASSERT(dprefclk_src_sel == 0);
+
+ /* Read the mmDENTIST_DISPCLK_CNTL to get the currently
+ * programmed DID DENTIST_DPREFCLK_WDIVIDER*/
+ REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DPREFCLK_WDIVIDER, &dprefclk_wdivider);
+
+ /* Convert DENTIST_DPREFCLK_WDIVIDERto actual divider*/
+ target_div = dentist_get_divider_from_did(dprefclk_wdivider);
+
+ /* Calculate the current DFS clock, in kHz.*/
+ dp_ref_clk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR
+ * clk_mgr_dce->dentist_vco_freq_khz) / target_div;
+
+ return clk_mgr_adjust_dp_ref_freq_for_ss(clk_mgr_dce, dp_ref_clk_khz);
+}
+
+int dce12_get_dp_ref_freq_khz(struct clk_mgr *clk_mgr)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+
+ return clk_mgr_adjust_dp_ref_freq_for_ss(clk_mgr_dce, clk_mgr_dce->dprefclk_khz);
+}
+
+/* unit: in_khz before mode set, get pixel clock from context. ASIC register
+ * may not be programmed yet
+ */
+static uint32_t get_max_pixel_clock_for_all_paths(struct dc_state *context)
+{
+ uint32_t max_pix_clk = 0;
+ int i;
+
+ for (i = 0; i < MAX_PIPES; i++) {
+ struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
+
+ if (pipe_ctx->stream == NULL)
+ continue;
+
+ /* do not check under lay */
+ if (pipe_ctx->top_pipe)
+ continue;
+
+ if (pipe_ctx->stream_res.pix_clk_params.requested_pix_clk > max_pix_clk)
+ max_pix_clk = pipe_ctx->stream_res.pix_clk_params.requested_pix_clk;
+
+ /* raise clock state for HBR3/2 if required. Confirmed with HW DCE/DPCS
+ * logic for HBR3 still needs Nominal (0.8V) on VDDC rail
+ */
+ if (dc_is_dp_signal(pipe_ctx->stream->signal) &&
+ pipe_ctx->stream_res.pix_clk_params.requested_sym_clk > max_pix_clk)
+ max_pix_clk = pipe_ctx->stream_res.pix_clk_params.requested_sym_clk;
+ }
+
+ return max_pix_clk;
+}
+
+static enum dm_pp_clocks_state dce_get_required_clocks_state(
+ struct clk_mgr *clk_mgr,
+ struct dc_state *context)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+ int i;
+ enum dm_pp_clocks_state low_req_clk;
+ int max_pix_clk = get_max_pixel_clock_for_all_paths(context);
+
+ /* Iterate from highest supported to lowest valid state, and update
+ * lowest RequiredState with the lowest state that satisfies
+ * all required clocks
+ */
+ for (i = clk_mgr_dce->max_clks_state; i >= DM_PP_CLOCKS_STATE_ULTRA_LOW; i--)
+ if (context->bw.dce.dispclk_khz >
+ clk_mgr_dce->max_clks_by_state[i].display_clk_khz
+ || max_pix_clk >
+ clk_mgr_dce->max_clks_by_state[i].pixel_clk_khz)
+ break;
+
+ low_req_clk = i + 1;
+ if (low_req_clk > clk_mgr_dce->max_clks_state) {
+ /* set max clock state for high phyclock, invalid on exceeding display clock */
+ if (clk_mgr_dce->max_clks_by_state[clk_mgr_dce->max_clks_state].display_clk_khz
+ < context->bw.dce.dispclk_khz)
+ low_req_clk = DM_PP_CLOCKS_STATE_INVALID;
+ else
+ low_req_clk = clk_mgr_dce->max_clks_state;
+ }
+
+ return low_req_clk;
+}
+
+static int dce_set_clock(
+ struct clk_mgr *clk_mgr,
+ int requested_clk_khz)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+ struct bp_pixel_clock_parameters pxl_clk_params = { 0 };
+ struct dc_bios *bp = clk_mgr->ctx->dc_bios;
+ int actual_clock = requested_clk_khz;
+ struct dmcu *dmcu = clk_mgr_dce->base.ctx->dc->res_pool->dmcu;
+
+ /* Make sure requested clock isn't lower than minimum threshold*/
+ if (requested_clk_khz > 0)
+ requested_clk_khz = max(requested_clk_khz,
+ clk_mgr_dce->dentist_vco_freq_khz / 64);
+
+ /* Prepare to program display clock*/
+ pxl_clk_params.target_pixel_clock = requested_clk_khz;
+ pxl_clk_params.pll_id = CLOCK_SOURCE_ID_DFS;
+
+ if (clk_mgr_dce->dfs_bypass_active)
+ pxl_clk_params.flags.SET_DISPCLK_DFS_BYPASS = true;
+
+ bp->funcs->program_display_engine_pll(bp, &pxl_clk_params);
+
+ if (clk_mgr_dce->dfs_bypass_active) {
+ /* Cache the fixed display clock*/
+ clk_mgr_dce->dfs_bypass_disp_clk =
+ pxl_clk_params.dfs_bypass_display_clock;
+ actual_clock = pxl_clk_params.dfs_bypass_display_clock;
+ }
+
+ /* from power down, we need mark the clock state as ClocksStateNominal
+ * from HWReset, so when resume we will call pplib voltage regulator.*/
+ if (requested_clk_khz == 0)
+ clk_mgr_dce->cur_min_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
+
+ dmcu->funcs->set_psr_wait_loop(dmcu, actual_clock / 1000 / 7);
+
+ return actual_clock;
+}
+
+int dce112_set_clock(struct clk_mgr *clk_mgr, int requested_clk_khz)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+ struct bp_set_dce_clock_parameters dce_clk_params;
+ struct dc_bios *bp = clk_mgr->ctx->dc_bios;
+ struct dc *core_dc = clk_mgr->ctx->dc;
+ struct dmcu *dmcu = core_dc->res_pool->dmcu;
+ int actual_clock = requested_clk_khz;
+ /* Prepare to program display clock*/
+ memset(&dce_clk_params, 0, sizeof(dce_clk_params));
+
+ /* Make sure requested clock isn't lower than minimum threshold*/
+ if (requested_clk_khz > 0)
+ requested_clk_khz = max(requested_clk_khz,
+ clk_mgr_dce->dentist_vco_freq_khz / 62);
+
+ dce_clk_params.target_clock_frequency = requested_clk_khz;
+ dce_clk_params.pll_id = CLOCK_SOURCE_ID_DFS;
+ dce_clk_params.clock_type = DCECLOCK_TYPE_DISPLAY_CLOCK;
+
+ bp->funcs->set_dce_clock(bp, &dce_clk_params);
+ actual_clock = dce_clk_params.target_clock_frequency;
+
+ /* from power down, we need mark the clock state as ClocksStateNominal
+ * from HWReset, so when resume we will call pplib voltage regulator.*/
+ if (requested_clk_khz == 0)
+ clk_mgr_dce->cur_min_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
+
+ /*Program DP ref Clock*/
+ /*VBIOS will determine DPREFCLK frequency, so we don't set it*/
+ dce_clk_params.target_clock_frequency = 0;
+ dce_clk_params.clock_type = DCECLOCK_TYPE_DPREFCLK;
+ if (!ASICREV_IS_VEGA20_P(clk_mgr->ctx->asic_id.hw_internal_rev))
+ dce_clk_params.flags.USE_GENLOCK_AS_SOURCE_FOR_DPREFCLK =
+ (dce_clk_params.pll_id ==
+ CLOCK_SOURCE_COMBO_DISPLAY_PLL0);
+ else
+ dce_clk_params.flags.USE_GENLOCK_AS_SOURCE_FOR_DPREFCLK = false;
+
+ bp->funcs->set_dce_clock(bp, &dce_clk_params);
+
+ if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) {
+ if (clk_mgr_dce->dfs_bypass_disp_clk != actual_clock)
+ dmcu->funcs->set_psr_wait_loop(dmcu,
+ actual_clock / 1000 / 7);
+ }
+
+ clk_mgr_dce->dfs_bypass_disp_clk = actual_clock;
+ return actual_clock;
+}
+
+static void dce_clock_read_integrated_info(struct dce_clk_mgr *clk_mgr_dce)
+{
+ struct dc_debug_options *debug = &clk_mgr_dce->base.ctx->dc->debug;
+ struct dc_bios *bp = clk_mgr_dce->base.ctx->dc_bios;
+ struct integrated_info info = { { { 0 } } };
+ struct dc_firmware_info fw_info = { { 0 } };
+ int i;
+
+ if (bp->integrated_info)
+ info = *bp->integrated_info;
+
+ clk_mgr_dce->dentist_vco_freq_khz = info.dentist_vco_freq;
+ if (clk_mgr_dce->dentist_vco_freq_khz == 0) {
+ bp->funcs->get_firmware_info(bp, &fw_info);
+ clk_mgr_dce->dentist_vco_freq_khz =
+ fw_info.smu_gpu_pll_output_freq;
+ if (clk_mgr_dce->dentist_vco_freq_khz == 0)
+ clk_mgr_dce->dentist_vco_freq_khz = 3600000;
+ }
+
+ /*update the maximum display clock for each power state*/
+ for (i = 0; i < NUMBER_OF_DISP_CLK_VOLTAGE; ++i) {
+ enum dm_pp_clocks_state clk_state = DM_PP_CLOCKS_STATE_INVALID;
+
+ switch (i) {
+ case 0:
+ clk_state = DM_PP_CLOCKS_STATE_ULTRA_LOW;
+ break;
+
+ case 1:
+ clk_state = DM_PP_CLOCKS_STATE_LOW;
+ break;
+
+ case 2:
+ clk_state = DM_PP_CLOCKS_STATE_NOMINAL;
+ break;
+
+ case 3:
+ clk_state = DM_PP_CLOCKS_STATE_PERFORMANCE;
+ break;
+
+ default:
+ clk_state = DM_PP_CLOCKS_STATE_INVALID;
+ break;
+ }
+
+ /*Do not allow bad VBIOS/SBIOS to override with invalid values,
+ * check for > 100MHz*/
+ if (info.disp_clk_voltage[i].max_supported_clk >= 100000)
+ clk_mgr_dce->max_clks_by_state[clk_state].display_clk_khz =
+ info.disp_clk_voltage[i].max_supported_clk;
+ }
+
+ if (!debug->disable_dfs_bypass && bp->integrated_info)
+ if (bp->integrated_info->gpu_cap_info & DFS_BYPASS_ENABLE)
+ clk_mgr_dce->dfs_bypass_enabled = true;
+}
+
+void dce_clock_read_ss_info(struct dce_clk_mgr *clk_mgr_dce)
+{
+ struct dc_bios *bp = clk_mgr_dce->base.ctx->dc_bios;
+ int ss_info_num = bp->funcs->get_ss_entry_number(
+ bp, AS_SIGNAL_TYPE_GPU_PLL);
+
+ if (ss_info_num) {
+ struct spread_spectrum_info info = { { 0 } };
+ enum bp_result result = bp->funcs->get_spread_spectrum_info(
+ bp, AS_SIGNAL_TYPE_GPU_PLL, 0, &info);
+
+ /* Based on VBIOS, VBIOS will keep entry for GPU PLL SS
+ * even if SS not enabled and in that case
+ * SSInfo.spreadSpectrumPercentage !=0 would be sign
+ * that SS is enabled
+ */
+ if (result == BP_RESULT_OK &&
+ info.spread_spectrum_percentage != 0) {
+ clk_mgr_dce->ss_on_dprefclk = true;
+ clk_mgr_dce->dprefclk_ss_divider = info.spread_percentage_divider;
+
+ if (info.type.CENTER_MODE == 0) {
+ /* TODO: Currently for DP Reference clock we
+ * need only SS percentage for
+ * downspread */
+ clk_mgr_dce->dprefclk_ss_percentage =
+ info.spread_spectrum_percentage;
+ }
+
+ return;
+ }
+
+ result = bp->funcs->get_spread_spectrum_info(
+ bp, AS_SIGNAL_TYPE_DISPLAY_PORT, 0, &info);
+
+ /* Based on VBIOS, VBIOS will keep entry for DPREFCLK SS
+ * even if SS not enabled and in that case
+ * SSInfo.spreadSpectrumPercentage !=0 would be sign
+ * that SS is enabled
+ */
+ if (result == BP_RESULT_OK &&
+ info.spread_spectrum_percentage != 0) {
+ clk_mgr_dce->ss_on_dprefclk = true;
+ clk_mgr_dce->dprefclk_ss_divider = info.spread_percentage_divider;
+
+ if (info.type.CENTER_MODE == 0) {
+ /* Currently for DP Reference clock we
+ * need only SS percentage for
+ * downspread */
+ clk_mgr_dce->dprefclk_ss_percentage =
+ info.spread_spectrum_percentage;
+ }
+ }
+ }
+}
+
+void dce110_fill_display_configs(
+ const struct dc_state *context,
+ struct dm_pp_display_configuration *pp_display_cfg)
+{
+ int j;
+ int num_cfgs = 0;
+
+ for (j = 0; j < context->stream_count; j++) {
+ int k;
+
+ const struct dc_stream_state *stream = context->streams[j];
+ struct dm_pp_single_disp_config *cfg =
+ &pp_display_cfg->disp_configs[num_cfgs];
+ const struct pipe_ctx *pipe_ctx = NULL;
+
+ for (k = 0; k < MAX_PIPES; k++)
+ if (stream == context->res_ctx.pipe_ctx[k].stream) {
+ pipe_ctx = &context->res_ctx.pipe_ctx[k];
+ break;
+ }
+
+ ASSERT(pipe_ctx != NULL);
+
+ /* only notify active stream */
+ if (stream->dpms_off)
+ continue;
+
+ num_cfgs++;
+ cfg->signal = pipe_ctx->stream->signal;
+ cfg->pipe_idx = pipe_ctx->stream_res.tg->inst;
+ cfg->src_height = stream->src.height;
+ cfg->src_width = stream->src.width;
+ cfg->ddi_channel_mapping =
+ stream->sink->link->ddi_channel_mapping.raw;
+ cfg->transmitter =
+ stream->sink->link->link_enc->transmitter;
+ cfg->link_settings.lane_count =
+ stream->sink->link->cur_link_settings.lane_count;
+ cfg->link_settings.link_rate =
+ stream->sink->link->cur_link_settings.link_rate;
+ cfg->link_settings.link_spread =
+ stream->sink->link->cur_link_settings.link_spread;
+ cfg->sym_clock = stream->phy_pix_clk;
+ /* Round v_refresh*/
+ cfg->v_refresh = stream->timing.pix_clk_khz * 1000;
+ cfg->v_refresh /= stream->timing.h_total;
+ cfg->v_refresh = (cfg->v_refresh + stream->timing.v_total / 2)
+ / stream->timing.v_total;
+ }
+
+ pp_display_cfg->display_count = num_cfgs;
+}
+
+static uint32_t dce110_get_min_vblank_time_us(const struct dc_state *context)
+{
+ uint8_t j;
+ uint32_t min_vertical_blank_time = -1;
+
+ for (j = 0; j < context->stream_count; j++) {
+ struct dc_stream_state *stream = context->streams[j];
+ uint32_t vertical_blank_in_pixels = 0;
+ uint32_t vertical_blank_time = 0;
+
+ vertical_blank_in_pixels = stream->timing.h_total *
+ (stream->timing.v_total
+ - stream->timing.v_addressable);
+
+ vertical_blank_time = vertical_blank_in_pixels
+ * 1000 / stream->timing.pix_clk_khz;
+
+ if (min_vertical_blank_time > vertical_blank_time)
+ min_vertical_blank_time = vertical_blank_time;
+ }
+
+ return min_vertical_blank_time;
+}
+
+static int determine_sclk_from_bounding_box(
+ const struct dc *dc,
+ int required_sclk)
+{
+ int i;
+
+ /*
+ * Some asics do not give us sclk levels, so we just report the actual
+ * required sclk
+ */
+ if (dc->sclk_lvls.num_levels == 0)
+ return required_sclk;
+
+ for (i = 0; i < dc->sclk_lvls.num_levels; i++) {
+ if (dc->sclk_lvls.clocks_in_khz[i] >= required_sclk)
+ return dc->sclk_lvls.clocks_in_khz[i];
+ }
+ /*
+ * even maximum level could not satisfy requirement, this
+ * is unexpected at this stage, should have been caught at
+ * validation time
+ */
+ ASSERT(0);
+ return dc->sclk_lvls.clocks_in_khz[dc->sclk_lvls.num_levels - 1];
+}
+
+static void dce_pplib_apply_display_requirements(
+ struct dc *dc,
+ struct dc_state *context)
+{
+ struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
+
+ pp_display_cfg->avail_mclk_switch_time_us = dce110_get_min_vblank_time_us(context);
+
+ dce110_fill_display_configs(context, pp_display_cfg);
+
+ if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) != 0)
+ dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
+}
+
+static void dce11_pplib_apply_display_requirements(
+ struct dc *dc,
+ struct dc_state *context)
+{
+ struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
+
+ pp_display_cfg->all_displays_in_sync =
+ context->bw.dce.all_displays_in_sync;
+ pp_display_cfg->nb_pstate_switch_disable =
+ context->bw.dce.nbp_state_change_enable == false;
+ pp_display_cfg->cpu_cc6_disable =
+ context->bw.dce.cpuc_state_change_enable == false;
+ pp_display_cfg->cpu_pstate_disable =
+ context->bw.dce.cpup_state_change_enable == false;
+ pp_display_cfg->cpu_pstate_separation_time =
+ context->bw.dce.blackout_recovery_time_us;
+
+ pp_display_cfg->min_memory_clock_khz = context->bw.dce.yclk_khz
+ / MEMORY_TYPE_MULTIPLIER_CZ;
+
+ pp_display_cfg->min_engine_clock_khz = determine_sclk_from_bounding_box(
+ dc,
+ context->bw.dce.sclk_khz);
+
+ pp_display_cfg->min_engine_clock_deep_sleep_khz
+ = context->bw.dce.sclk_deep_sleep_khz;
+
+ pp_display_cfg->avail_mclk_switch_time_us =
+ dce110_get_min_vblank_time_us(context);
+ /* TODO: dce11.2*/
+ pp_display_cfg->avail_mclk_switch_time_in_disp_active_us = 0;
+
+ pp_display_cfg->disp_clk_khz = dc->res_pool->clk_mgr->clks.dispclk_khz;
+
+ dce110_fill_display_configs(context, pp_display_cfg);
+
+ /* TODO: is this still applicable?*/
+ if (pp_display_cfg->display_count == 1) {
+ const struct dc_crtc_timing *timing =
+ &context->streams[0]->timing;
+
+ pp_display_cfg->crtc_index =
+ pp_display_cfg->disp_configs[0].pipe_idx;
+ pp_display_cfg->line_time_in_us = timing->h_total * 1000 / timing->pix_clk_khz;
+ }
+
+ if (memcmp(&dc->current_state->pp_display_cfg, pp_display_cfg, sizeof(*pp_display_cfg)) != 0)
+ dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
+}
+
+static void dce_update_clocks(struct clk_mgr *clk_mgr,
+ struct dc_state *context,
+ bool safe_to_lower)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+ struct dm_pp_power_level_change_request level_change_req;
+ int unpatched_disp_clk = context->bw.dce.dispclk_khz;
+
+ /*TODO: W/A for dal3 linux, investigate why this works */
+ if (!clk_mgr_dce->dfs_bypass_active)
+ context->bw.dce.dispclk_khz = context->bw.dce.dispclk_khz * 115 / 100;
+
+ level_change_req.power_level = dce_get_required_clocks_state(clk_mgr, context);
+ /* get max clock state from PPLIB */
+ if ((level_change_req.power_level < clk_mgr_dce->cur_min_clks_state && safe_to_lower)
+ || level_change_req.power_level > clk_mgr_dce->cur_min_clks_state) {
+ if (dm_pp_apply_power_level_change_request(clk_mgr->ctx, &level_change_req))
+ clk_mgr_dce->cur_min_clks_state = level_change_req.power_level;
+ }
+
+ if (should_set_clock(safe_to_lower, context->bw.dce.dispclk_khz, clk_mgr->clks.dispclk_khz)) {
+ context->bw.dce.dispclk_khz = dce_set_clock(clk_mgr, context->bw.dce.dispclk_khz);
+ clk_mgr->clks.dispclk_khz = context->bw.dce.dispclk_khz;
+ }
+ dce_pplib_apply_display_requirements(clk_mgr->ctx->dc, context);
+
+ context->bw.dce.dispclk_khz = unpatched_disp_clk;
+}
+
+static void dce11_update_clocks(struct clk_mgr *clk_mgr,
+ struct dc_state *context,
+ bool safe_to_lower)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+ struct dm_pp_power_level_change_request level_change_req;
+
+ level_change_req.power_level = dce_get_required_clocks_state(clk_mgr, context);
+ /* get max clock state from PPLIB */
+ if ((level_change_req.power_level < clk_mgr_dce->cur_min_clks_state && safe_to_lower)
+ || level_change_req.power_level > clk_mgr_dce->cur_min_clks_state) {
+ if (dm_pp_apply_power_level_change_request(clk_mgr->ctx, &level_change_req))
+ clk_mgr_dce->cur_min_clks_state = level_change_req.power_level;
+ }
+
+ if (should_set_clock(safe_to_lower, context->bw.dce.dispclk_khz, clk_mgr->clks.dispclk_khz)) {
+ context->bw.dce.dispclk_khz = dce_set_clock(clk_mgr, context->bw.dce.dispclk_khz);
+ clk_mgr->clks.dispclk_khz = context->bw.dce.dispclk_khz;
+ }
+ dce11_pplib_apply_display_requirements(clk_mgr->ctx->dc, context);
+}
+
+static void dce112_update_clocks(struct clk_mgr *clk_mgr,
+ struct dc_state *context,
+ bool safe_to_lower)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+ struct dm_pp_power_level_change_request level_change_req;
+
+ level_change_req.power_level = dce_get_required_clocks_state(clk_mgr, context);
+ /* get max clock state from PPLIB */
+ if ((level_change_req.power_level < clk_mgr_dce->cur_min_clks_state && safe_to_lower)
+ || level_change_req.power_level > clk_mgr_dce->cur_min_clks_state) {
+ if (dm_pp_apply_power_level_change_request(clk_mgr->ctx, &level_change_req))
+ clk_mgr_dce->cur_min_clks_state = level_change_req.power_level;
+ }
+
+ if (should_set_clock(safe_to_lower, context->bw.dce.dispclk_khz, clk_mgr->clks.dispclk_khz)) {
+ context->bw.dce.dispclk_khz = dce112_set_clock(clk_mgr, context->bw.dce.dispclk_khz);
+ clk_mgr->clks.dispclk_khz = context->bw.dce.dispclk_khz;
+ }
+ dce11_pplib_apply_display_requirements(clk_mgr->ctx->dc, context);
+}
+
+static void dce12_update_clocks(struct clk_mgr *clk_mgr,
+ struct dc_state *context,
+ bool safe_to_lower)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(clk_mgr);
+ struct dm_pp_clock_for_voltage_req clock_voltage_req = {0};
+ int max_pix_clk = get_max_pixel_clock_for_all_paths(context);
+ int unpatched_disp_clk = context->bw.dce.dispclk_khz;
+
+ /*TODO: W/A for dal3 linux, investigate why this works */
+ if (!clk_mgr_dce->dfs_bypass_active)
+ context->bw.dce.dispclk_khz = context->bw.dce.dispclk_khz * 115 / 100;
+
+ if (should_set_clock(safe_to_lower, context->bw.dce.dispclk_khz, clk_mgr->clks.dispclk_khz)) {
+ clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DISPLAY_CLK;
+ clock_voltage_req.clocks_in_khz = context->bw.dce.dispclk_khz;
+ context->bw.dce.dispclk_khz = dce112_set_clock(clk_mgr, context->bw.dce.dispclk_khz);
+ clk_mgr->clks.dispclk_khz = context->bw.dce.dispclk_khz;
+
+ dm_pp_apply_clock_for_voltage_request(clk_mgr->ctx, &clock_voltage_req);
+ }
+
+ if (should_set_clock(safe_to_lower, max_pix_clk, clk_mgr->clks.phyclk_khz)) {
+ clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DISPLAYPHYCLK;
+ clock_voltage_req.clocks_in_khz = max_pix_clk;
+ clk_mgr->clks.phyclk_khz = max_pix_clk;
+
+ dm_pp_apply_clock_for_voltage_request(clk_mgr->ctx, &clock_voltage_req);
+ }
+ dce11_pplib_apply_display_requirements(clk_mgr->ctx->dc, context);
+
+ context->bw.dce.dispclk_khz = unpatched_disp_clk;
+}
+
+static const struct clk_mgr_funcs dce120_funcs = {
+ .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
+ .update_clocks = dce12_update_clocks
+};
+
+static const struct clk_mgr_funcs dce112_funcs = {
+ .get_dp_ref_clk_frequency = dce_get_dp_ref_freq_khz,
+ .update_clocks = dce112_update_clocks
+};
+
+static const struct clk_mgr_funcs dce110_funcs = {
+ .get_dp_ref_clk_frequency = dce_get_dp_ref_freq_khz,
+ .update_clocks = dce11_update_clocks,
+};
+
+static const struct clk_mgr_funcs dce_funcs = {
+ .get_dp_ref_clk_frequency = dce_get_dp_ref_freq_khz,
+ .update_clocks = dce_update_clocks
+};
+
+static void dce_clk_mgr_construct(
+ struct dce_clk_mgr *clk_mgr_dce,
+ struct dc_context *ctx,
+ const struct clk_mgr_registers *regs,
+ const struct clk_mgr_shift *clk_shift,
+ const struct clk_mgr_mask *clk_mask)
+{
+ struct clk_mgr *base = &clk_mgr_dce->base;
+ struct dm_pp_static_clock_info static_clk_info = {0};
+
+ base->ctx = ctx;
+ base->funcs = &dce_funcs;
+
+ clk_mgr_dce->regs = regs;
+ clk_mgr_dce->clk_mgr_shift = clk_shift;
+ clk_mgr_dce->clk_mgr_mask = clk_mask;
+
+ clk_mgr_dce->dfs_bypass_disp_clk = 0;
+
+ clk_mgr_dce->dprefclk_ss_percentage = 0;
+ clk_mgr_dce->dprefclk_ss_divider = 1000;
+ clk_mgr_dce->ss_on_dprefclk = false;
+
+
+ if (dm_pp_get_static_clocks(ctx, &static_clk_info))
+ clk_mgr_dce->max_clks_state = static_clk_info.max_clocks_state;
+ else
+ clk_mgr_dce->max_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
+ clk_mgr_dce->cur_min_clks_state = DM_PP_CLOCKS_STATE_INVALID;
+
+ dce_clock_read_integrated_info(clk_mgr_dce);
+ dce_clock_read_ss_info(clk_mgr_dce);
+}
+
+struct clk_mgr *dce_clk_mgr_create(
+ struct dc_context *ctx,
+ const struct clk_mgr_registers *regs,
+ const struct clk_mgr_shift *clk_shift,
+ const struct clk_mgr_mask *clk_mask)
+{
+ struct dce_clk_mgr *clk_mgr_dce = kzalloc(sizeof(*clk_mgr_dce), GFP_KERNEL);
+
+ if (clk_mgr_dce == NULL) {
+ BREAK_TO_DEBUGGER();
+ return NULL;
+ }
+
+ memcpy(clk_mgr_dce->max_clks_by_state,
+ dce80_max_clks_by_state,
+ sizeof(dce80_max_clks_by_state));
+
+ dce_clk_mgr_construct(
+ clk_mgr_dce, ctx, regs, clk_shift, clk_mask);
+
+ return &clk_mgr_dce->base;
+}
+
+struct clk_mgr *dce110_clk_mgr_create(
+ struct dc_context *ctx,
+ const struct clk_mgr_registers *regs,
+ const struct clk_mgr_shift *clk_shift,
+ const struct clk_mgr_mask *clk_mask)
+{
+ struct dce_clk_mgr *clk_mgr_dce = kzalloc(sizeof(*clk_mgr_dce), GFP_KERNEL);
+
+ if (clk_mgr_dce == NULL) {
+ BREAK_TO_DEBUGGER();
+ return NULL;
+ }
+
+ memcpy(clk_mgr_dce->max_clks_by_state,
+ dce110_max_clks_by_state,
+ sizeof(dce110_max_clks_by_state));
+
+ dce_clk_mgr_construct(
+ clk_mgr_dce, ctx, regs, clk_shift, clk_mask);
+
+ clk_mgr_dce->base.funcs = &dce110_funcs;
+
+ return &clk_mgr_dce->base;
+}
+
+struct clk_mgr *dce112_clk_mgr_create(
+ struct dc_context *ctx,
+ const struct clk_mgr_registers *regs,
+ const struct clk_mgr_shift *clk_shift,
+ const struct clk_mgr_mask *clk_mask)
+{
+ struct dce_clk_mgr *clk_mgr_dce = kzalloc(sizeof(*clk_mgr_dce), GFP_KERNEL);
+
+ if (clk_mgr_dce == NULL) {
+ BREAK_TO_DEBUGGER();
+ return NULL;
+ }
+
+ memcpy(clk_mgr_dce->max_clks_by_state,
+ dce112_max_clks_by_state,
+ sizeof(dce112_max_clks_by_state));
+
+ dce_clk_mgr_construct(
+ clk_mgr_dce, ctx, regs, clk_shift, clk_mask);
+
+ clk_mgr_dce->base.funcs = &dce112_funcs;
+
+ return &clk_mgr_dce->base;
+}
+
+struct clk_mgr *dce120_clk_mgr_create(struct dc_context *ctx)
+{
+ struct dce_clk_mgr *clk_mgr_dce = kzalloc(sizeof(*clk_mgr_dce), GFP_KERNEL);
+
+ if (clk_mgr_dce == NULL) {
+ BREAK_TO_DEBUGGER();
+ return NULL;
+ }
+
+ memcpy(clk_mgr_dce->max_clks_by_state,
+ dce120_max_clks_by_state,
+ sizeof(dce120_max_clks_by_state));
+
+ dce_clk_mgr_construct(
+ clk_mgr_dce, ctx, NULL, NULL, NULL);
+
+ clk_mgr_dce->dprefclk_khz = 600000;
+ clk_mgr_dce->base.funcs = &dce120_funcs;
+
+ return &clk_mgr_dce->base;
+}
+
+void dce_clk_mgr_destroy(struct clk_mgr **clk_mgr)
+{
+ struct dce_clk_mgr *clk_mgr_dce = TO_DCE_CLK_MGR(*clk_mgr);
+
+ kfree(clk_mgr_dce);
+ *clk_mgr = NULL;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h
new file mode 100644
index 0000000..0460777
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_clk_mgr.h
@@ -0,0 +1,168 @@
+/*
+ * Copyright 2012-16 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+
+#ifndef _DCE_CLK_MGR_H_
+#define _DCE_CLK_MGR_H_
+
+#include "clk_mgr.h"
+#include "dccg.h"
+
+#define MEMORY_TYPE_MULTIPLIER_CZ 4
+
+#define CLK_COMMON_REG_LIST_DCE_BASE() \
+ .DPREFCLK_CNTL = mmDPREFCLK_CNTL, \
+ .DENTIST_DISPCLK_CNTL = mmDENTIST_DISPCLK_CNTL
+
+#define CLK_COMMON_REG_LIST_DCN_BASE() \
+ SR(DENTIST_DISPCLK_CNTL)
+
+#define CLK_SF(reg_name, field_name, post_fix)\
+ .field_name = reg_name ## __ ## field_name ## post_fix
+
+#define CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh) \
+ CLK_SF(DPREFCLK_CNTL, DPREFCLK_SRC_SEL, mask_sh), \
+ CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DPREFCLK_WDIVIDER, mask_sh)
+
+#define CLK_COMMON_MASK_SH_LIST_DCN_COMMON_BASE(mask_sh) \
+ CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh),\
+ CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh)
+
+#define CLK_REG_FIELD_LIST(type) \
+ type DPREFCLK_SRC_SEL; \
+ type DENTIST_DPREFCLK_WDIVIDER; \
+ type DENTIST_DISPCLK_WDIVIDER; \
+ type DENTIST_DISPCLK_CHG_DONE;
+
+struct clk_mgr_shift {
+ CLK_REG_FIELD_LIST(uint8_t)
+};
+
+struct clk_mgr_mask {
+ CLK_REG_FIELD_LIST(uint32_t)
+};
+
+struct clk_mgr_registers {
+ uint32_t DPREFCLK_CNTL;
+ uint32_t DENTIST_DISPCLK_CNTL;
+};
+
+struct state_dependent_clocks {
+ int display_clk_khz;
+ int pixel_clk_khz;
+};
+
+struct dce_clk_mgr {
+ struct clk_mgr base;
+ const struct clk_mgr_registers *regs;
+ const struct clk_mgr_shift *clk_mgr_shift;
+ const struct clk_mgr_mask *clk_mgr_mask;
+
+ struct dccg *dccg;
+
+ struct state_dependent_clocks max_clks_by_state[DM_PP_CLOCKS_MAX_STATES];
+
+ int dentist_vco_freq_khz;
+
+ /* Cache the status of DFS-bypass feature*/
+ bool dfs_bypass_enabled;
+ /* True if the DFS-bypass feature is enabled and active. */
+ bool dfs_bypass_active;
+ /* Cache the display clock returned by VBIOS if DFS-bypass is enabled.
+ * This is basically "Crystal Frequency In KHz" (XTALIN) frequency */
+ int dfs_bypass_disp_clk;
+
+ /* Flag for Enabled SS on DPREFCLK */
+ bool ss_on_dprefclk;
+ /* DPREFCLK SS percentage (if down-spread enabled) */
+ int dprefclk_ss_percentage;
+ /* DPREFCLK SS percentage Divider (100 or 1000) */
+ int dprefclk_ss_divider;
+ int dprefclk_khz;
+
+ enum dm_pp_clocks_state max_clks_state;
+ enum dm_pp_clocks_state cur_min_clks_state;
+};
+
+/* Starting DID for each range */
+enum dentist_base_divider_id {
+ DENTIST_BASE_DID_1 = 0x08,
+ DENTIST_BASE_DID_2 = 0x40,
+ DENTIST_BASE_DID_3 = 0x60,
+ DENTIST_BASE_DID_4 = 0x7e,
+ DENTIST_MAX_DID = 0x7f
+};
+
+/* Starting point and step size for each divider range.*/
+enum dentist_divider_range {
+ DENTIST_DIVIDER_RANGE_1_START = 8, /* 2.00 */
+ DENTIST_DIVIDER_RANGE_1_STEP = 1, /* 0.25 */
+ DENTIST_DIVIDER_RANGE_2_START = 64, /* 16.00 */
+ DENTIST_DIVIDER_RANGE_2_STEP = 2, /* 0.50 */
+ DENTIST_DIVIDER_RANGE_3_START = 128, /* 32.00 */
+ DENTIST_DIVIDER_RANGE_3_STEP = 4, /* 1.00 */
+ DENTIST_DIVIDER_RANGE_4_START = 248, /* 62.00 */
+ DENTIST_DIVIDER_RANGE_4_STEP = 264, /* 66.00 */
+ DENTIST_DIVIDER_RANGE_SCALE_FACTOR = 4
+};
+
+static inline bool should_set_clock(bool safe_to_lower, int calc_clk, int cur_clk)
+{
+ return ((safe_to_lower && calc_clk < cur_clk) || calc_clk > cur_clk);
+}
+
+void dce_clock_read_ss_info(struct dce_clk_mgr *dccg_dce);
+
+int dce12_get_dp_ref_freq_khz(struct clk_mgr *dccg);
+
+void dce110_fill_display_configs(
+ const struct dc_state *context,
+ struct dm_pp_display_configuration *pp_display_cfg);
+
+int dce112_set_clock(struct clk_mgr *dccg, int requested_clk_khz);
+
+struct clk_mgr *dce_clk_mgr_create(
+ struct dc_context *ctx,
+ const struct clk_mgr_registers *regs,
+ const struct clk_mgr_shift *clk_shift,
+ const struct clk_mgr_mask *clk_mask);
+
+struct clk_mgr *dce110_clk_mgr_create(
+ struct dc_context *ctx,
+ const struct clk_mgr_registers *regs,
+ const struct clk_mgr_shift *clk_shift,
+ const struct clk_mgr_mask *clk_mask);
+
+struct clk_mgr *dce112_clk_mgr_create(
+ struct dc_context *ctx,
+ const struct clk_mgr_registers *regs,
+ const struct clk_mgr_shift *clk_shift,
+ const struct clk_mgr_mask *clk_mask);
+
+struct clk_mgr *dce120_clk_mgr_create(struct dc_context *ctx);
+
+void dce_clk_mgr_destroy(struct clk_mgr **clk_mgr);
+
+#endif /* _DCE_CLK_MGR_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.c b/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.c
deleted file mode 100644
index d89a097..0000000
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.c
+++ /dev/null
@@ -1,947 +0,0 @@
-/*
- * Copyright 2012-16 Advanced Micro Devices, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: AMD
- *
- */
-
-#include "dce_clocks.h"
-#include "dm_services.h"
-#include "reg_helper.h"
-#include "fixed31_32.h"
-#include "bios_parser_interface.h"
-#include "dc.h"
-#include "dmcu.h"
-#if defined(CONFIG_DRM_AMD_DC_DCN1_0)
-#include "dcn_calcs.h"
-#endif
-#include "core_types.h"
-#include "dc_types.h"
-#include "dal_asic_id.h"
-
-#define TO_DCE_CLOCKS(clocks)\
- container_of(clocks, struct dce_dccg, base)
-
-#define REG(reg) \
- (clk_dce->regs->reg)
-
-#undef FN
-#define FN(reg_name, field_name) \
- clk_dce->clk_shift->field_name, clk_dce->clk_mask->field_name
-
-#define CTX \
- clk_dce->base.ctx
-#define DC_LOGGER \
- clk->ctx->logger
-
-/* Max clock values for each state indexed by "enum clocks_state": */
-static const struct state_dependent_clocks dce80_max_clks_by_state[] = {
-/* ClocksStateInvalid - should not be used */
-{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
-/* ClocksStateUltraLow - not expected to be used for DCE 8.0 */
-{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
-/* ClocksStateLow */
-{ .display_clk_khz = 352000, .pixel_clk_khz = 330000},
-/* ClocksStateNominal */
-{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 },
-/* ClocksStatePerformance */
-{ .display_clk_khz = 600000, .pixel_clk_khz = 400000 } };
-
-static const struct state_dependent_clocks dce110_max_clks_by_state[] = {
-/*ClocksStateInvalid - should not be used*/
-{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
-/*ClocksStateUltraLow - currently by HW design team not supposed to be used*/
-{ .display_clk_khz = 352000, .pixel_clk_khz = 330000 },
-/*ClocksStateLow*/
-{ .display_clk_khz = 352000, .pixel_clk_khz = 330000 },
-/*ClocksStateNominal*/
-{ .display_clk_khz = 467000, .pixel_clk_khz = 400000 },
-/*ClocksStatePerformance*/
-{ .display_clk_khz = 643000, .pixel_clk_khz = 400000 } };
-
-static const struct state_dependent_clocks dce112_max_clks_by_state[] = {
-/*ClocksStateInvalid - should not be used*/
-{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
-/*ClocksStateUltraLow - currently by HW design team not supposed to be used*/
-{ .display_clk_khz = 389189, .pixel_clk_khz = 346672 },
-/*ClocksStateLow*/
-{ .display_clk_khz = 459000, .pixel_clk_khz = 400000 },
-/*ClocksStateNominal*/
-{ .display_clk_khz = 667000, .pixel_clk_khz = 600000 },
-/*ClocksStatePerformance*/
-{ .display_clk_khz = 1132000, .pixel_clk_khz = 600000 } };
-
-static const struct state_dependent_clocks dce120_max_clks_by_state[] = {
-/*ClocksStateInvalid - should not be used*/
-{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
-/*ClocksStateUltraLow - currently by HW design team not supposed to be used*/
-{ .display_clk_khz = 0, .pixel_clk_khz = 0 },
-/*ClocksStateLow*/
-{ .display_clk_khz = 460000, .pixel_clk_khz = 400000 },
-/*ClocksStateNominal*/
-{ .display_clk_khz = 670000, .pixel_clk_khz = 600000 },
-/*ClocksStatePerformance*/
-{ .display_clk_khz = 1133000, .pixel_clk_khz = 600000 } };
-
-/* Starting DID for each range */
-enum dentist_base_divider_id {
- DENTIST_BASE_DID_1 = 0x08,
- DENTIST_BASE_DID_2 = 0x40,
- DENTIST_BASE_DID_3 = 0x60,
- DENTIST_BASE_DID_4 = 0x7e,
- DENTIST_MAX_DID = 0x7f
-};
-
-/* Starting point and step size for each divider range.*/
-enum dentist_divider_range {
- DENTIST_DIVIDER_RANGE_1_START = 8, /* 2.00 */
- DENTIST_DIVIDER_RANGE_1_STEP = 1, /* 0.25 */
- DENTIST_DIVIDER_RANGE_2_START = 64, /* 16.00 */
- DENTIST_DIVIDER_RANGE_2_STEP = 2, /* 0.50 */
- DENTIST_DIVIDER_RANGE_3_START = 128, /* 32.00 */
- DENTIST_DIVIDER_RANGE_3_STEP = 4, /* 1.00 */
- DENTIST_DIVIDER_RANGE_4_START = 248, /* 62.00 */
- DENTIST_DIVIDER_RANGE_4_STEP = 264, /* 66.00 */
- DENTIST_DIVIDER_RANGE_SCALE_FACTOR = 4
-};
-
-static int dentist_get_divider_from_did(int did)
-{
- if (did < DENTIST_BASE_DID_1)
- did = DENTIST_BASE_DID_1;
- if (did > DENTIST_MAX_DID)
- did = DENTIST_MAX_DID;
-
- if (did < DENTIST_BASE_DID_2) {
- return DENTIST_DIVIDER_RANGE_1_START + DENTIST_DIVIDER_RANGE_1_STEP
- * (did - DENTIST_BASE_DID_1);
- } else if (did < DENTIST_BASE_DID_3) {
- return DENTIST_DIVIDER_RANGE_2_START + DENTIST_DIVIDER_RANGE_2_STEP
- * (did - DENTIST_BASE_DID_2);
- } else if (did < DENTIST_BASE_DID_4) {
- return DENTIST_DIVIDER_RANGE_3_START + DENTIST_DIVIDER_RANGE_3_STEP
- * (did - DENTIST_BASE_DID_3);
- } else {
- return DENTIST_DIVIDER_RANGE_4_START + DENTIST_DIVIDER_RANGE_4_STEP
- * (did - DENTIST_BASE_DID_4);
- }
-}
-
-/* SW will adjust DP REF Clock average value for all purposes
- * (DP DTO / DP Audio DTO and DP GTC)
- if clock is spread for all cases:
- -if SS enabled on DP Ref clock and HW de-spreading enabled with SW
- calculations for DS_INCR/DS_MODULO (this is planned to be default case)
- -if SS enabled on DP Ref clock and HW de-spreading enabled with HW
- calculations (not planned to be used, but average clock should still
- be valid)
- -if SS enabled on DP Ref clock and HW de-spreading disabled
- (should not be case with CIK) then SW should program all rates
- generated according to average value (case as with previous ASICs)
- */
-static int dccg_adjust_dp_ref_freq_for_ss(struct dce_dccg *clk_dce, int dp_ref_clk_khz)
-{
- if (clk_dce->ss_on_dprefclk && clk_dce->dprefclk_ss_divider != 0) {
- struct fixed31_32 ss_percentage = dc_fixpt_div_int(
- dc_fixpt_from_fraction(clk_dce->dprefclk_ss_percentage,
- clk_dce->dprefclk_ss_divider), 200);
- struct fixed31_32 adj_dp_ref_clk_khz;
-
- ss_percentage = dc_fixpt_sub(dc_fixpt_one, ss_percentage);
- adj_dp_ref_clk_khz = dc_fixpt_mul_int(ss_percentage, dp_ref_clk_khz);
- dp_ref_clk_khz = dc_fixpt_floor(adj_dp_ref_clk_khz);
- }
- return dp_ref_clk_khz;
-}
-
-static int dce_get_dp_ref_freq_khz(struct dccg *clk)
-{
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(clk);
- int dprefclk_wdivider;
- int dprefclk_src_sel;
- int dp_ref_clk_khz = 600000;
- int target_div;
-
- /* ASSERT DP Reference Clock source is from DFS*/
- REG_GET(DPREFCLK_CNTL, DPREFCLK_SRC_SEL, &dprefclk_src_sel);
- ASSERT(dprefclk_src_sel == 0);
-
- /* Read the mmDENTIST_DISPCLK_CNTL to get the currently
- * programmed DID DENTIST_DPREFCLK_WDIVIDER*/
- REG_GET(DENTIST_DISPCLK_CNTL, DENTIST_DPREFCLK_WDIVIDER, &dprefclk_wdivider);
-
- /* Convert DENTIST_DPREFCLK_WDIVIDERto actual divider*/
- target_div = dentist_get_divider_from_did(dprefclk_wdivider);
-
- /* Calculate the current DFS clock, in kHz.*/
- dp_ref_clk_khz = (DENTIST_DIVIDER_RANGE_SCALE_FACTOR
- * clk_dce->dentist_vco_freq_khz) / target_div;
-
- return dccg_adjust_dp_ref_freq_for_ss(clk_dce, dp_ref_clk_khz);
-}
-
-static int dce12_get_dp_ref_freq_khz(struct dccg *clk)
-{
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(clk);
-
- return dccg_adjust_dp_ref_freq_for_ss(clk_dce, clk_dce->dprefclk_khz);
-}
-
-static enum dm_pp_clocks_state dce_get_required_clocks_state(
- struct dccg *clk,
- struct dc_clocks *req_clocks)
-{
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(clk);
- int i;
- enum dm_pp_clocks_state low_req_clk;
-
- /* Iterate from highest supported to lowest valid state, and update
- * lowest RequiredState with the lowest state that satisfies
- * all required clocks
- */
- for (i = clk->max_clks_state; i >= DM_PP_CLOCKS_STATE_ULTRA_LOW; i--)
- if (req_clocks->dispclk_khz >
- clk_dce->max_clks_by_state[i].display_clk_khz
- || req_clocks->phyclk_khz >
- clk_dce->max_clks_by_state[i].pixel_clk_khz)
- break;
-
- low_req_clk = i + 1;
- if (low_req_clk > clk->max_clks_state) {
- /* set max clock state for high phyclock, invalid on exceeding display clock */
- if (clk_dce->max_clks_by_state[clk->max_clks_state].display_clk_khz
- < req_clocks->dispclk_khz)
- low_req_clk = DM_PP_CLOCKS_STATE_INVALID;
- else
- low_req_clk = clk->max_clks_state;
- }
-
- return low_req_clk;
-}
-
-static int dce_set_clock(
- struct dccg *clk,
- int requested_clk_khz)
-{
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(clk);
- struct bp_pixel_clock_parameters pxl_clk_params = { 0 };
- struct dc_bios *bp = clk->ctx->dc_bios;
- int actual_clock = requested_clk_khz;
-
- /* Make sure requested clock isn't lower than minimum threshold*/
- if (requested_clk_khz > 0)
- requested_clk_khz = max(requested_clk_khz,
- clk_dce->dentist_vco_freq_khz / 64);
-
- /* Prepare to program display clock*/
- pxl_clk_params.target_pixel_clock = requested_clk_khz;
- pxl_clk_params.pll_id = CLOCK_SOURCE_ID_DFS;
-
- if (clk_dce->dfs_bypass_active)
- pxl_clk_params.flags.SET_DISPCLK_DFS_BYPASS = true;
-
- bp->funcs->program_display_engine_pll(bp, &pxl_clk_params);
-
- if (clk_dce->dfs_bypass_active) {
- /* Cache the fixed display clock*/
- clk_dce->dfs_bypass_disp_clk =
- pxl_clk_params.dfs_bypass_display_clock;
- actual_clock = pxl_clk_params.dfs_bypass_display_clock;
- }
-
- /* from power down, we need mark the clock state as ClocksStateNominal
- * from HWReset, so when resume we will call pplib voltage regulator.*/
- if (requested_clk_khz == 0)
- clk->cur_min_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
- return actual_clock;
-}
-
-static int dce_psr_set_clock(
- struct dccg *clk,
- int requested_clk_khz)
-{
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(clk);
- struct dc_context *ctx = clk_dce->base.ctx;
- struct dc *core_dc = ctx->dc;
- struct dmcu *dmcu = core_dc->res_pool->dmcu;
- int actual_clk_khz = requested_clk_khz;
-
- actual_clk_khz = dce_set_clock(clk, requested_clk_khz);
-
- dmcu->funcs->set_psr_wait_loop(dmcu, actual_clk_khz / 1000 / 7);
- return actual_clk_khz;
-}
-
-static int dce112_set_clock(
- struct dccg *clk,
- int requested_clk_khz)
-{
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(clk);
- struct bp_set_dce_clock_parameters dce_clk_params;
- struct dc_bios *bp = clk->ctx->dc_bios;
- struct dc *core_dc = clk->ctx->dc;
- struct dmcu *dmcu = core_dc->res_pool->dmcu;
- int actual_clock = requested_clk_khz;
- /* Prepare to program display clock*/
- memset(&dce_clk_params, 0, sizeof(dce_clk_params));
-
- /* Make sure requested clock isn't lower than minimum threshold*/
- if (requested_clk_khz > 0)
- requested_clk_khz = max(requested_clk_khz,
- clk_dce->dentist_vco_freq_khz / 62);
-
- dce_clk_params.target_clock_frequency = requested_clk_khz;
- dce_clk_params.pll_id = CLOCK_SOURCE_ID_DFS;
- dce_clk_params.clock_type = DCECLOCK_TYPE_DISPLAY_CLOCK;
-
- bp->funcs->set_dce_clock(bp, &dce_clk_params);
- actual_clock = dce_clk_params.target_clock_frequency;
-
- /* from power down, we need mark the clock state as ClocksStateNominal
- * from HWReset, so when resume we will call pplib voltage regulator.*/
- if (requested_clk_khz == 0)
- clk->cur_min_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
-
- /*Program DP ref Clock*/
- /*VBIOS will determine DPREFCLK frequency, so we don't set it*/
- dce_clk_params.target_clock_frequency = 0;
- dce_clk_params.clock_type = DCECLOCK_TYPE_DPREFCLK;
- if (!ASICREV_IS_VEGA20_P(clk->ctx->asic_id.hw_internal_rev))
- dce_clk_params.flags.USE_GENLOCK_AS_SOURCE_FOR_DPREFCLK =
- (dce_clk_params.pll_id ==
- CLOCK_SOURCE_COMBO_DISPLAY_PLL0);
- else
- dce_clk_params.flags.USE_GENLOCK_AS_SOURCE_FOR_DPREFCLK = false;
-
- bp->funcs->set_dce_clock(bp, &dce_clk_params);
-
- if (!IS_FPGA_MAXIMUS_DC(core_dc->ctx->dce_environment)) {
- if (clk_dce->dfs_bypass_disp_clk != actual_clock)
- dmcu->funcs->set_psr_wait_loop(dmcu,
- actual_clock / 1000 / 7);
- }
-
- clk_dce->dfs_bypass_disp_clk = actual_clock;
- return actual_clock;
-}
-
-static void dce_clock_read_integrated_info(struct dce_dccg *clk_dce)
-{
- struct dc_debug_options *debug = &clk_dce->base.ctx->dc->debug;
- struct dc_bios *bp = clk_dce->base.ctx->dc_bios;
- struct integrated_info info = { { { 0 } } };
- struct dc_firmware_info fw_info = { { 0 } };
- int i;
-
- if (bp->integrated_info)
- info = *bp->integrated_info;
-
- clk_dce->dentist_vco_freq_khz = info.dentist_vco_freq;
- if (clk_dce->dentist_vco_freq_khz == 0) {
- bp->funcs->get_firmware_info(bp, &fw_info);
- clk_dce->dentist_vco_freq_khz =
- fw_info.smu_gpu_pll_output_freq;
- if (clk_dce->dentist_vco_freq_khz == 0)
- clk_dce->dentist_vco_freq_khz = 3600000;
- }
-
- /*update the maximum display clock for each power state*/
- for (i = 0; i < NUMBER_OF_DISP_CLK_VOLTAGE; ++i) {
- enum dm_pp_clocks_state clk_state = DM_PP_CLOCKS_STATE_INVALID;
-
- switch (i) {
- case 0:
- clk_state = DM_PP_CLOCKS_STATE_ULTRA_LOW;
- break;
-
- case 1:
- clk_state = DM_PP_CLOCKS_STATE_LOW;
- break;
-
- case 2:
- clk_state = DM_PP_CLOCKS_STATE_NOMINAL;
- break;
-
- case 3:
- clk_state = DM_PP_CLOCKS_STATE_PERFORMANCE;
- break;
-
- default:
- clk_state = DM_PP_CLOCKS_STATE_INVALID;
- break;
- }
-
- /*Do not allow bad VBIOS/SBIOS to override with invalid values,
- * check for > 100MHz*/
- if (info.disp_clk_voltage[i].max_supported_clk >= 100000)
- clk_dce->max_clks_by_state[clk_state].display_clk_khz =
- info.disp_clk_voltage[i].max_supported_clk;
- }
-
- if (!debug->disable_dfs_bypass && bp->integrated_info)
- if (bp->integrated_info->gpu_cap_info & DFS_BYPASS_ENABLE)
- clk_dce->dfs_bypass_enabled = true;
-}
-
-static void dce_clock_read_ss_info(struct dce_dccg *clk_dce)
-{
- struct dc_bios *bp = clk_dce->base.ctx->dc_bios;
- int ss_info_num = bp->funcs->get_ss_entry_number(
- bp, AS_SIGNAL_TYPE_GPU_PLL);
-
- if (ss_info_num) {
- struct spread_spectrum_info info = { { 0 } };
- enum bp_result result = bp->funcs->get_spread_spectrum_info(
- bp, AS_SIGNAL_TYPE_GPU_PLL, 0, &info);
-
- /* Based on VBIOS, VBIOS will keep entry for GPU PLL SS
- * even if SS not enabled and in that case
- * SSInfo.spreadSpectrumPercentage !=0 would be sign
- * that SS is enabled
- */
- if (result == BP_RESULT_OK &&
- info.spread_spectrum_percentage != 0) {
- clk_dce->ss_on_dprefclk = true;
- clk_dce->dprefclk_ss_divider = info.spread_percentage_divider;
-
- if (info.type.CENTER_MODE == 0) {
- /* TODO: Currently for DP Reference clock we
- * need only SS percentage for
- * downspread */
- clk_dce->dprefclk_ss_percentage =
- info.spread_spectrum_percentage;
- }
-
- return;
- }
-
- result = bp->funcs->get_spread_spectrum_info(
- bp, AS_SIGNAL_TYPE_DISPLAY_PORT, 0, &info);
-
- /* Based on VBIOS, VBIOS will keep entry for DPREFCLK SS
- * even if SS not enabled and in that case
- * SSInfo.spreadSpectrumPercentage !=0 would be sign
- * that SS is enabled
- */
- if (result == BP_RESULT_OK &&
- info.spread_spectrum_percentage != 0) {
- clk_dce->ss_on_dprefclk = true;
- clk_dce->dprefclk_ss_divider = info.spread_percentage_divider;
-
- if (info.type.CENTER_MODE == 0) {
- /* Currently for DP Reference clock we
- * need only SS percentage for
- * downspread */
- clk_dce->dprefclk_ss_percentage =
- info.spread_spectrum_percentage;
- }
- }
- }
-}
-
-static inline bool should_set_clock(bool safe_to_lower, int calc_clk, int cur_clk)
-{
- return ((safe_to_lower && calc_clk < cur_clk) || calc_clk > cur_clk);
-}
-
-static void dce12_update_clocks(struct dccg *dccg,
- struct dc_clocks *new_clocks,
- bool safe_to_lower)
-{
- struct dm_pp_clock_for_voltage_req clock_voltage_req = {0};
-
- /* TODO: Investigate why this is needed to fix display corruption. */
- new_clocks->dispclk_khz = new_clocks->dispclk_khz * 115 / 100;
-
- if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, dccg->clks.dispclk_khz)) {
- clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DISPLAY_CLK;
- clock_voltage_req.clocks_in_khz = new_clocks->dispclk_khz;
- new_clocks->dispclk_khz = dccg->funcs->set_dispclk(dccg, new_clocks->dispclk_khz);
- dccg->clks.dispclk_khz = new_clocks->dispclk_khz;
-
- dm_pp_apply_clock_for_voltage_request(dccg->ctx, &clock_voltage_req);
- }
-
- if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, dccg->clks.phyclk_khz)) {
- clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DISPLAYPHYCLK;
- clock_voltage_req.clocks_in_khz = new_clocks->phyclk_khz;
- dccg->clks.phyclk_khz = new_clocks->phyclk_khz;
-
- dm_pp_apply_clock_for_voltage_request(dccg->ctx, &clock_voltage_req);
- }
-}
-
-#ifdef CONFIG_DRM_AMD_DC_DCN1_0
-static int dcn1_determine_dppclk_threshold(struct dccg *dccg, struct dc_clocks *new_clocks)
-{
- bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz;
- bool dispclk_increase = new_clocks->dispclk_khz > dccg->clks.dispclk_khz;
- int disp_clk_threshold = new_clocks->max_supported_dppclk_khz;
- bool cur_dpp_div = dccg->clks.dispclk_khz > dccg->clks.dppclk_khz;
-
- /* increase clock, looking for div is 0 for current, request div is 1*/
- if (dispclk_increase) {
- /* already divided by 2, no need to reach target clk with 2 steps*/
- if (cur_dpp_div)
- return new_clocks->dispclk_khz;
-
- /* request disp clk is lower than maximum supported dpp clk,
- * no need to reach target clk with two steps.
- */
- if (new_clocks->dispclk_khz <= disp_clk_threshold)
- return new_clocks->dispclk_khz;
-
- /* target dpp clk not request divided by 2, still within threshold */
- if (!request_dpp_div)
- return new_clocks->dispclk_khz;
-
- } else {
- /* decrease clock, looking for current dppclk divided by 2,
- * request dppclk not divided by 2.
- */
-
- /* current dpp clk not divided by 2, no need to ramp*/
- if (!cur_dpp_div)
- return new_clocks->dispclk_khz;
-
- /* current disp clk is lower than current maximum dpp clk,
- * no need to ramp
- */
- if (dccg->clks.dispclk_khz <= disp_clk_threshold)
- return new_clocks->dispclk_khz;
-
- /* request dpp clk need to be divided by 2 */
- if (request_dpp_div)
- return new_clocks->dispclk_khz;
- }
-
- return disp_clk_threshold;
-}
-
-static void dcn1_ramp_up_dispclk_with_dpp(struct dccg *dccg, struct dc_clocks *new_clocks)
-{
- struct dc *dc = dccg->ctx->dc;
- int dispclk_to_dpp_threshold = dcn1_determine_dppclk_threshold(dccg, new_clocks);
- bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz;
- int i;
-
- /* set disp clk to dpp clk threshold */
- dccg->funcs->set_dispclk(dccg, dispclk_to_dpp_threshold);
-
- /* update request dpp clk division option */
- for (i = 0; i < dc->res_pool->pipe_count; i++) {
- struct pipe_ctx *pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i];
-
- if (!pipe_ctx->plane_state)
- continue;
-
- pipe_ctx->plane_res.dpp->funcs->dpp_dppclk_control(
- pipe_ctx->plane_res.dpp,
- request_dpp_div,
- true);
- }
-
- /* If target clk not same as dppclk threshold, set to target clock */
- if (dispclk_to_dpp_threshold != new_clocks->dispclk_khz)
- dccg->funcs->set_dispclk(dccg, new_clocks->dispclk_khz);
-
- dccg->clks.dispclk_khz = new_clocks->dispclk_khz;
- dccg->clks.dppclk_khz = new_clocks->dppclk_khz;
- dccg->clks.max_supported_dppclk_khz = new_clocks->max_supported_dppclk_khz;
-}
-
-static void dcn1_update_clocks(struct dccg *dccg,
- struct dc_clocks *new_clocks,
- bool safe_to_lower)
-{
- struct dc *dc = dccg->ctx->dc;
- struct pp_smu_display_requirement_rv *smu_req_cur =
- &dc->res_pool->pp_smu_req;
- struct pp_smu_display_requirement_rv smu_req = *smu_req_cur;
- struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu;
- struct dm_pp_clock_for_voltage_req clock_voltage_req = {0};
- bool send_request_to_increase = false;
- bool send_request_to_lower = false;
-
- if (new_clocks->phyclk_khz)
- smu_req.display_count = 1;
- else
- smu_req.display_count = 0;
-
- if (new_clocks->dispclk_khz > dccg->clks.dispclk_khz
- || new_clocks->phyclk_khz > dccg->clks.phyclk_khz
- || new_clocks->fclk_khz > dccg->clks.fclk_khz
- || new_clocks->dcfclk_khz > dccg->clks.dcfclk_khz)
- send_request_to_increase = true;
-
- if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, dccg->clks.phyclk_khz)) {
- dccg->clks.phyclk_khz = new_clocks->phyclk_khz;
-
- send_request_to_lower = true;
- }
-
- if (should_set_clock(safe_to_lower, new_clocks->fclk_khz, dccg->clks.fclk_khz)) {
- dccg->clks.fclk_khz = new_clocks->fclk_khz;
- clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_FCLK;
- clock_voltage_req.clocks_in_khz = new_clocks->fclk_khz;
- smu_req.hard_min_fclk_khz = new_clocks->fclk_khz;
-
- dm_pp_apply_clock_for_voltage_request(dccg->ctx, &clock_voltage_req);
- send_request_to_lower = true;
- }
-
- if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, dccg->clks.dcfclk_khz)) {
- dccg->clks.dcfclk_khz = new_clocks->dcfclk_khz;
- smu_req.hard_min_dcefclk_khz = new_clocks->dcfclk_khz;
-
- send_request_to_lower = true;
- }
-
- if (should_set_clock(safe_to_lower,
- new_clocks->dcfclk_deep_sleep_khz, dccg->clks.dcfclk_deep_sleep_khz)) {
- dccg->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz;
- smu_req.min_deep_sleep_dcefclk_mhz = new_clocks->dcfclk_deep_sleep_khz;
-
- send_request_to_lower = true;
- }
-
- /* make sure dcf clk is before dpp clk to
- * make sure we have enough voltage to run dpp clk
- */
- if (send_request_to_increase) {
- /*use dcfclk to request voltage*/
- clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DCFCLK;
- clock_voltage_req.clocks_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks);
- dm_pp_apply_clock_for_voltage_request(dccg->ctx, &clock_voltage_req);
- if (pp_smu->set_display_requirement)
- pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
- }
-
- /* dcn1 dppclk is tied to dispclk */
- /* program dispclk on = as a w/a for sleep resume clock ramping issues */
- if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, dccg->clks.dispclk_khz)
- || new_clocks->dispclk_khz == dccg->clks.dispclk_khz) {
- dcn1_ramp_up_dispclk_with_dpp(dccg, new_clocks);
- dccg->clks.dispclk_khz = new_clocks->dispclk_khz;
-
- send_request_to_lower = true;
- }
-
- if (!send_request_to_increase && send_request_to_lower) {
- /*use dcfclk to request voltage*/
- clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DCFCLK;
- clock_voltage_req.clocks_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks);
- dm_pp_apply_clock_for_voltage_request(dccg->ctx, &clock_voltage_req);
- if (pp_smu->set_display_requirement)
- pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
- }
-
-
- *smu_req_cur = smu_req;
-}
-#endif
-
-static void dce_update_clocks(struct dccg *dccg,
- struct dc_clocks *new_clocks,
- bool safe_to_lower)
-{
- struct dm_pp_power_level_change_request level_change_req;
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(dccg);
-
- /* TODO: Investigate why this is needed to fix display corruption. */
- if (!clk_dce->dfs_bypass_active)
- new_clocks->dispclk_khz = new_clocks->dispclk_khz * 115 / 100;
-
- level_change_req.power_level = dce_get_required_clocks_state(dccg, new_clocks);
- /* get max clock state from PPLIB */
- if ((level_change_req.power_level < dccg->cur_min_clks_state && safe_to_lower)
- || level_change_req.power_level > dccg->cur_min_clks_state) {
- if (dm_pp_apply_power_level_change_request(dccg->ctx, &level_change_req))
- dccg->cur_min_clks_state = level_change_req.power_level;
- }
-
- if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, dccg->clks.dispclk_khz)) {
- new_clocks->dispclk_khz = dccg->funcs->set_dispclk(dccg, new_clocks->dispclk_khz);
- dccg->clks.dispclk_khz = new_clocks->dispclk_khz;
- }
-}
-
-static bool dce_update_dfs_bypass(
- struct dccg *dccg,
- struct dc *dc,
- struct dc_state *context,
- int requested_clock_khz)
-{
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(dccg);
- struct resource_context *res_ctx = &context->res_ctx;
- enum signal_type signal_type = SIGNAL_TYPE_NONE;
- bool was_active = clk_dce->dfs_bypass_active;
- int i;
-
- /* Disable DFS bypass by default. */
- clk_dce->dfs_bypass_active = false;
-
- /* Check that DFS bypass is available. */
- if (!clk_dce->dfs_bypass_enabled)
- goto update;
-
- /* Check if the requested display clock is below the threshold. */
- if (requested_clock_khz >= 400000)
- goto update;
-
- /* DFS-bypass should only be enabled on single stream setups */
- if (context->stream_count != 1)
- goto update;
-
- /* Check that the stream's signal type is an embedded panel */
- for (i = 0; i < dc->res_pool->pipe_count; i++) {
- if (res_ctx->pipe_ctx[i].stream) {
- struct pipe_ctx *pipe_ctx = &res_ctx->pipe_ctx[i];
-
- signal_type = pipe_ctx->stream->sink->link->connector_signal;
- break;
- }
- }
-
- if (signal_type == SIGNAL_TYPE_EDP ||
- signal_type == SIGNAL_TYPE_LVDS)
- clk_dce->dfs_bypass_active = true;
-
-update:
- /* Update the clock state. We don't need to respect safe_to_lower
- * because DFS bypass should always be greater than the current
- * display clock frequency.
- */
- if (was_active != clk_dce->dfs_bypass_active) {
- dccg->clks.dispclk_khz =
- dccg->funcs->set_dispclk(dccg, dccg->clks.dispclk_khz);
- return true;
- }
-
- return false;
-}
-
-#ifdef CONFIG_DRM_AMD_DC_DCN1_0
-static const struct display_clock_funcs dcn1_funcs = {
- .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
- .set_dispclk = dce112_set_clock,
- .update_clocks = dcn1_update_clocks
-};
-#endif
-
-static const struct display_clock_funcs dce120_funcs = {
- .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
- .set_dispclk = dce112_set_clock,
- .update_clocks = dce12_update_clocks
-};
-
-static const struct display_clock_funcs dce112_funcs = {
- .get_dp_ref_clk_frequency = dce_get_dp_ref_freq_khz,
- .set_dispclk = dce112_set_clock,
- .update_clocks = dce_update_clocks
-};
-
-static const struct display_clock_funcs dce110_funcs = {
- .get_dp_ref_clk_frequency = dce_get_dp_ref_freq_khz,
- .set_dispclk = dce_psr_set_clock,
- .update_clocks = dce_update_clocks,
- .update_dfs_bypass = dce_update_dfs_bypass
-};
-
-static const struct display_clock_funcs dce_funcs = {
- .get_dp_ref_clk_frequency = dce_get_dp_ref_freq_khz,
- .set_dispclk = dce_set_clock,
- .update_clocks = dce_update_clocks
-};
-
-static void dce_dccg_construct(
- struct dce_dccg *clk_dce,
- struct dc_context *ctx,
- const struct dccg_registers *regs,
- const struct dccg_shift *clk_shift,
- const struct dccg_mask *clk_mask)
-{
- struct dccg *base = &clk_dce->base;
-
- base->ctx = ctx;
- base->funcs = &dce_funcs;
-
- clk_dce->regs = regs;
- clk_dce->clk_shift = clk_shift;
- clk_dce->clk_mask = clk_mask;
-
- clk_dce->dfs_bypass_disp_clk = 0;
-
- clk_dce->dprefclk_ss_percentage = 0;
- clk_dce->dprefclk_ss_divider = 1000;
- clk_dce->ss_on_dprefclk = false;
-
- base->max_clks_state = DM_PP_CLOCKS_STATE_NOMINAL;
- base->cur_min_clks_state = DM_PP_CLOCKS_STATE_INVALID;
-
- dce_clock_read_integrated_info(clk_dce);
- dce_clock_read_ss_info(clk_dce);
-}
-
-struct dccg *dce_dccg_create(
- struct dc_context *ctx,
- const struct dccg_registers *regs,
- const struct dccg_shift *clk_shift,
- const struct dccg_mask *clk_mask)
-{
- struct dce_dccg *clk_dce = kzalloc(sizeof(*clk_dce), GFP_KERNEL);
-
- if (clk_dce == NULL) {
- BREAK_TO_DEBUGGER();
- return NULL;
- }
-
- memcpy(clk_dce->max_clks_by_state,
- dce80_max_clks_by_state,
- sizeof(dce80_max_clks_by_state));
-
- dce_dccg_construct(
- clk_dce, ctx, regs, clk_shift, clk_mask);
-
- return &clk_dce->base;
-}
-
-struct dccg *dce110_dccg_create(
- struct dc_context *ctx,
- const struct dccg_registers *regs,
- const struct dccg_shift *clk_shift,
- const struct dccg_mask *clk_mask)
-{
- struct dce_dccg *clk_dce = kzalloc(sizeof(*clk_dce), GFP_KERNEL);
-
- if (clk_dce == NULL) {
- BREAK_TO_DEBUGGER();
- return NULL;
- }
-
- memcpy(clk_dce->max_clks_by_state,
- dce110_max_clks_by_state,
- sizeof(dce110_max_clks_by_state));
-
- dce_dccg_construct(
- clk_dce, ctx, regs, clk_shift, clk_mask);
-
- clk_dce->base.funcs = &dce110_funcs;
-
- return &clk_dce->base;
-}
-
-struct dccg *dce112_dccg_create(
- struct dc_context *ctx,
- const struct dccg_registers *regs,
- const struct dccg_shift *clk_shift,
- const struct dccg_mask *clk_mask)
-{
- struct dce_dccg *clk_dce = kzalloc(sizeof(*clk_dce), GFP_KERNEL);
-
- if (clk_dce == NULL) {
- BREAK_TO_DEBUGGER();
- return NULL;
- }
-
- memcpy(clk_dce->max_clks_by_state,
- dce112_max_clks_by_state,
- sizeof(dce112_max_clks_by_state));
-
- dce_dccg_construct(
- clk_dce, ctx, regs, clk_shift, clk_mask);
-
- clk_dce->base.funcs = &dce112_funcs;
-
- return &clk_dce->base;
-}
-
-struct dccg *dce120_dccg_create(struct dc_context *ctx)
-{
- struct dce_dccg *clk_dce = kzalloc(sizeof(*clk_dce), GFP_KERNEL);
-
- if (clk_dce == NULL) {
- BREAK_TO_DEBUGGER();
- return NULL;
- }
-
- memcpy(clk_dce->max_clks_by_state,
- dce120_max_clks_by_state,
- sizeof(dce120_max_clks_by_state));
-
- dce_dccg_construct(
- clk_dce, ctx, NULL, NULL, NULL);
-
- clk_dce->dprefclk_khz = 600000;
- clk_dce->base.funcs = &dce120_funcs;
-
- return &clk_dce->base;
-}
-
-#ifdef CONFIG_DRM_AMD_DC_DCN1_0
-struct dccg *dcn1_dccg_create(struct dc_context *ctx)
-{
- struct dc_debug_options *debug = &ctx->dc->debug;
- struct dc_bios *bp = ctx->dc_bios;
- struct dc_firmware_info fw_info = { { 0 } };
- struct dce_dccg *clk_dce = kzalloc(sizeof(*clk_dce), GFP_KERNEL);
-
- if (clk_dce == NULL) {
- BREAK_TO_DEBUGGER();
- return NULL;
- }
-
- clk_dce->base.ctx = ctx;
- clk_dce->base.funcs = &dcn1_funcs;
-
- clk_dce->dfs_bypass_disp_clk = 0;
-
- clk_dce->dprefclk_ss_percentage = 0;
- clk_dce->dprefclk_ss_divider = 1000;
- clk_dce->ss_on_dprefclk = false;
-
- clk_dce->dprefclk_khz = 600000;
- if (bp->integrated_info)
- clk_dce->dentist_vco_freq_khz = bp->integrated_info->dentist_vco_freq;
- if (clk_dce->dentist_vco_freq_khz == 0) {
- bp->funcs->get_firmware_info(bp, &fw_info);
- clk_dce->dentist_vco_freq_khz = fw_info.smu_gpu_pll_output_freq;
- if (clk_dce->dentist_vco_freq_khz == 0)
- clk_dce->dentist_vco_freq_khz = 3600000;
- }
-
- if (!debug->disable_dfs_bypass && bp->integrated_info)
- if (bp->integrated_info->gpu_cap_info & DFS_BYPASS_ENABLE)
- clk_dce->dfs_bypass_enabled = true;
-
- dce_clock_read_ss_info(clk_dce);
-
- return &clk_dce->base;
-}
-#endif
-
-void dce_dccg_destroy(struct dccg **dccg)
-{
- struct dce_dccg *clk_dce = TO_DCE_CLOCKS(*dccg);
-
- kfree(clk_dce);
- *dccg = NULL;
-}
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.h b/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.h
deleted file mode 100644
index 34fdb38..0000000
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_clocks.h
+++ /dev/null
@@ -1,123 +0,0 @@
-/*
- * Copyright 2012-16 Advanced Micro Devices, Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: AMD
- *
- */
-
-
-#ifndef _DCE_CLOCKS_H_
-#define _DCE_CLOCKS_H_
-
-#include "display_clock.h"
-
-#define CLK_COMMON_REG_LIST_DCE_BASE() \
- .DPREFCLK_CNTL = mmDPREFCLK_CNTL, \
- .DENTIST_DISPCLK_CNTL = mmDENTIST_DISPCLK_CNTL
-
-#define CLK_COMMON_REG_LIST_DCN_BASE() \
- SR(DENTIST_DISPCLK_CNTL)
-
-#define CLK_SF(reg_name, field_name, post_fix)\
- .field_name = reg_name ## __ ## field_name ## post_fix
-
-#define CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(mask_sh) \
- CLK_SF(DPREFCLK_CNTL, DPREFCLK_SRC_SEL, mask_sh), \
- CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DPREFCLK_WDIVIDER, mask_sh)
-
-#define CLK_COMMON_MASK_SH_LIST_DCN_COMMON_BASE(mask_sh) \
- CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_WDIVIDER, mask_sh),\
- CLK_SF(DENTIST_DISPCLK_CNTL, DENTIST_DISPCLK_CHG_DONE, mask_sh)
-
-#define CLK_REG_FIELD_LIST(type) \
- type DPREFCLK_SRC_SEL; \
- type DENTIST_DPREFCLK_WDIVIDER; \
- type DENTIST_DISPCLK_WDIVIDER; \
- type DENTIST_DISPCLK_CHG_DONE;
-
-struct dccg_shift {
- CLK_REG_FIELD_LIST(uint8_t)
-};
-
-struct dccg_mask {
- CLK_REG_FIELD_LIST(uint32_t)
-};
-
-struct dccg_registers {
- uint32_t DPREFCLK_CNTL;
- uint32_t DENTIST_DISPCLK_CNTL;
-};
-
-struct dce_dccg {
- struct dccg base;
- const struct dccg_registers *regs;
- const struct dccg_shift *clk_shift;
- const struct dccg_mask *clk_mask;
-
- struct state_dependent_clocks max_clks_by_state[DM_PP_CLOCKS_MAX_STATES];
-
- int dentist_vco_freq_khz;
-
- /* Cache the status of DFS-bypass feature*/
- bool dfs_bypass_enabled;
- /* True if the DFS-bypass feature is enabled and active. */
- bool dfs_bypass_active;
- /* Cache the display clock returned by VBIOS if DFS-bypass is enabled.
- * This is basically "Crystal Frequency In KHz" (XTALIN) frequency */
- int dfs_bypass_disp_clk;
-
- /* Flag for Enabled SS on DPREFCLK */
- bool ss_on_dprefclk;
- /* DPREFCLK SS percentage (if down-spread enabled) */
- int dprefclk_ss_percentage;
- /* DPREFCLK SS percentage Divider (100 or 1000) */
- int dprefclk_ss_divider;
- int dprefclk_khz;
-};
-
-
-struct dccg *dce_dccg_create(
- struct dc_context *ctx,
- const struct dccg_registers *regs,
- const struct dccg_shift *clk_shift,
- const struct dccg_mask *clk_mask);
-
-struct dccg *dce110_dccg_create(
- struct dc_context *ctx,
- const struct dccg_registers *regs,
- const struct dccg_shift *clk_shift,
- const struct dccg_mask *clk_mask);
-
-struct dccg *dce112_dccg_create(
- struct dc_context *ctx,
- const struct dccg_registers *regs,
- const struct dccg_shift *clk_shift,
- const struct dccg_mask *clk_mask);
-
-struct dccg *dce120_dccg_create(struct dc_context *ctx);
-
-#ifdef CONFIG_DRM_AMD_DC_DCN1_0
-struct dccg *dcn1_dccg_create(struct dc_context *ctx);
-#endif
-
-void dce_dccg_destroy(struct dccg **dccg);
-
-#endif /* _DCE_CLOCKS_H_ */
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h
index 64dc753..c83a7f0 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_hwseq.h
@@ -233,6 +233,16 @@ struct dce_hwseq_registers {
uint32_t DOMAIN5_PG_CONFIG;
uint32_t DOMAIN6_PG_CONFIG;
uint32_t DOMAIN7_PG_CONFIG;
+ uint32_t DOMAIN8_PG_CONFIG;
+ uint32_t DOMAIN9_PG_CONFIG;
+ uint32_t DOMAIN10_PG_CONFIG;
+ uint32_t DOMAIN11_PG_CONFIG;
+ uint32_t DOMAIN16_PG_CONFIG;
+ uint32_t DOMAIN17_PG_CONFIG;
+ uint32_t DOMAIN18_PG_CONFIG;
+ uint32_t DOMAIN19_PG_CONFIG;
+ uint32_t DOMAIN20_PG_CONFIG;
+ uint32_t DOMAIN21_PG_CONFIG;
uint32_t DOMAIN0_PG_STATUS;
uint32_t DOMAIN1_PG_STATUS;
uint32_t DOMAIN2_PG_STATUS;
@@ -241,6 +251,16 @@ struct dce_hwseq_registers {
uint32_t DOMAIN5_PG_STATUS;
uint32_t DOMAIN6_PG_STATUS;
uint32_t DOMAIN7_PG_STATUS;
+ uint32_t DOMAIN8_PG_STATUS;
+ uint32_t DOMAIN9_PG_STATUS;
+ uint32_t DOMAIN10_PG_STATUS;
+ uint32_t DOMAIN11_PG_STATUS;
+ uint32_t DOMAIN16_PG_STATUS;
+ uint32_t DOMAIN17_PG_STATUS;
+ uint32_t DOMAIN18_PG_STATUS;
+ uint32_t DOMAIN19_PG_STATUS;
+ uint32_t DOMAIN20_PG_STATUS;
+ uint32_t DOMAIN21_PG_STATUS;
uint32_t DIO_MEM_PWR_CTRL;
uint32_t DCCG_GATE_DISABLE_CNTL;
uint32_t DCCG_GATE_DISABLE_CNTL2;
@@ -262,6 +282,8 @@ struct dce_hwseq_registers {
uint32_t D2VGA_CONTROL;
uint32_t D3VGA_CONTROL;
uint32_t D4VGA_CONTROL;
+ uint32_t D5VGA_CONTROL;
+ uint32_t D6VGA_CONTROL;
uint32_t VGA_TEST_CONTROL;
/* MMHUB registers. read only. temporary hack */
uint32_t VM_CONTEXT0_PAGE_TABLE_BASE_ADDR_HI32;
@@ -489,6 +511,26 @@ struct dce_hwseq_registers {
type DOMAIN6_POWER_GATE; \
type DOMAIN7_POWER_FORCEON; \
type DOMAIN7_POWER_GATE; \
+ type DOMAIN8_POWER_FORCEON; \
+ type DOMAIN8_POWER_GATE; \
+ type DOMAIN9_POWER_FORCEON; \
+ type DOMAIN9_POWER_GATE; \
+ type DOMAIN10_POWER_FORCEON; \
+ type DOMAIN10_POWER_GATE; \
+ type DOMAIN11_POWER_FORCEON; \
+ type DOMAIN11_POWER_GATE; \
+ type DOMAIN16_POWER_FORCEON; \
+ type DOMAIN16_POWER_GATE; \
+ type DOMAIN17_POWER_FORCEON; \
+ type DOMAIN17_POWER_GATE; \
+ type DOMAIN18_POWER_FORCEON; \
+ type DOMAIN18_POWER_GATE; \
+ type DOMAIN19_POWER_FORCEON; \
+ type DOMAIN19_POWER_GATE; \
+ type DOMAIN20_POWER_FORCEON; \
+ type DOMAIN20_POWER_GATE; \
+ type DOMAIN21_POWER_FORCEON; \
+ type DOMAIN21_POWER_GATE; \
type DOMAIN0_PGFSM_PWR_STATUS; \
type DOMAIN1_PGFSM_PWR_STATUS; \
type DOMAIN2_PGFSM_PWR_STATUS; \
@@ -497,6 +539,16 @@ struct dce_hwseq_registers {
type DOMAIN5_PGFSM_PWR_STATUS; \
type DOMAIN6_PGFSM_PWR_STATUS; \
type DOMAIN7_PGFSM_PWR_STATUS; \
+ type DOMAIN8_PGFSM_PWR_STATUS; \
+ type DOMAIN9_PGFSM_PWR_STATUS; \
+ type DOMAIN10_PGFSM_PWR_STATUS; \
+ type DOMAIN11_PGFSM_PWR_STATUS; \
+ type DOMAIN16_PGFSM_PWR_STATUS; \
+ type DOMAIN17_PGFSM_PWR_STATUS; \
+ type DOMAIN18_PGFSM_PWR_STATUS; \
+ type DOMAIN19_PGFSM_PWR_STATUS; \
+ type DOMAIN20_PGFSM_PWR_STATUS; \
+ type DOMAIN21_PGFSM_PWR_STATUS; \
type DCFCLK_GATE_DIS; \
type DCHUBBUB_GLOBAL_TIMER_REFDIV; \
type VGA_TEST_ENABLE; \
diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
index 366bc8c..3e18ea8 100644
--- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c
@@ -645,7 +645,7 @@ static bool dce110_link_encoder_validate_hdmi_output(
return false;
/* DCE11 HW does not support 420 */
- if (!enc110->base.features.ycbcr420_supported &&
+ if (!enc110->base.features.hdmi_ycbcr420_supported &&
crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
return false;
diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c
index 74c05e8..bc50a8e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.c
@@ -105,74 +105,18 @@ bool dce100_enable_display_power_gating(
return false;
}
-static void dce100_pplib_apply_display_requirements(
- struct dc *dc,
- struct dc_state *context)
-{
- struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
-
- pp_display_cfg->avail_mclk_switch_time_us =
- dce110_get_min_vblank_time_us(context);
- /*pp_display_cfg->min_memory_clock_khz = context->bw.dce.yclk_khz
- / MEMORY_TYPE_MULTIPLIER;*/
-
- dce110_fill_display_configs(context, pp_display_cfg);
-
- if (memcmp(&dc->prev_display_config, pp_display_cfg, sizeof(
- struct dm_pp_display_configuration)) != 0)
- dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
-
- dc->prev_display_config = *pp_display_cfg;
-}
-
-/* unit: in_khz before mode set, get pixel clock from context. ASIC register
- * may not be programmed yet
- */
-static uint32_t get_max_pixel_clock_for_all_paths(
- struct dc *dc,
- struct dc_state *context)
-{
- uint32_t max_pix_clk = 0;
- int i;
-
- for (i = 0; i < MAX_PIPES; i++) {
- struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
-
- if (pipe_ctx->stream == NULL)
- continue;
-
- /* do not check under lay */
- if (pipe_ctx->top_pipe)
- continue;
-
- if (pipe_ctx->stream_res.pix_clk_params.requested_pix_clk > max_pix_clk)
- max_pix_clk =
- pipe_ctx->stream_res.pix_clk_params.requested_pix_clk;
- }
- return max_pix_clk;
-}
-
-void dce100_set_bandwidth(
+void dce100_prepare_bandwidth(
struct dc *dc,
- struct dc_state *context,
- bool decrease_allowed)
+ struct dc_state *context)
{
- struct dc_clocks req_clks;
-
- req_clks.dispclk_khz = context->bw.dce.dispclk_khz * 115 / 100;
- req_clks.phyclk_khz = get_max_pixel_clock_for_all_paths(dc, context);
-
dce110_set_safe_displaymarks(&context->res_ctx, dc->res_pool);
- dc->res_pool->dccg->funcs->update_clocks(
- dc->res_pool->dccg,
- &req_clks,
- decrease_allowed);
-
- dce100_pplib_apply_display_requirements(dc, context);
+ dc->res_pool->clk_mgr->funcs->update_clocks(
+ dc->res_pool->clk_mgr,
+ context,
+ false);
}
-
/**************************************************************************/
void dce100_hw_sequencer_construct(struct dc *dc)
@@ -180,8 +124,7 @@ void dce100_hw_sequencer_construct(struct dc *dc)
dce110_hw_sequencer_construct(dc);
dc->hwss.enable_display_power_gating = dce100_enable_display_power_gating;
- dc->hwss.set_bandwidth = dce100_set_bandwidth;
- dc->hwss.pplib_apply_display_requirements =
- dce100_pplib_apply_display_requirements;
+ dc->hwss.prepare_bandwidth = dce100_prepare_bandwidth;
+ dc->hwss.optimize_bandwidth = dce100_prepare_bandwidth;
}
diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.h
index c6ec0ed..acd4185 100644
--- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_hw_sequencer.h
@@ -33,10 +33,9 @@ struct dc_state;
void dce100_hw_sequencer_construct(struct dc *dc);
-void dce100_set_bandwidth(
+void dce100_prepare_bandwidth(
struct dc *dc,
- struct dc_state *context,
- bool decrease_allowed);
+ struct dc_state *context);
bool dce100_enable_display_power_gating(struct dc *dc, uint8_t controller_id,
struct dc_bios *dcb,
diff --git a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
index 14754a8..6ae51a5 100644
--- a/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce100/dce100_resource.c
@@ -36,11 +36,11 @@
#include "dce/dce_link_encoder.h"
#include "dce/dce_stream_encoder.h"
+#include "dce/dce_clk_mgr.h"
#include "dce/dce_mem_input.h"
#include "dce/dce_ipp.h"
#include "dce/dce_transform.h"
#include "dce/dce_opp.h"
-#include "dce/dce_clocks.h"
#include "dce/dce_clock_source.h"
#include "dce/dce_audio.h"
#include "dce/dce_hwseq.h"
@@ -137,15 +137,15 @@ static const struct dce110_timing_generator_offsets dce100_tg_offsets[] = {
.reg_name = mm ## block ## id ## _ ## reg_name
-static const struct dccg_registers disp_clk_regs = {
+static const struct clk_mgr_registers disp_clk_regs = {
CLK_COMMON_REG_LIST_DCE_BASE()
};
-static const struct dccg_shift disp_clk_shift = {
+static const struct clk_mgr_shift disp_clk_shift = {
CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(__SHIFT)
};
-static const struct dccg_mask disp_clk_mask = {
+static const struct clk_mgr_mask disp_clk_mask = {
CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(_MASK)
};
@@ -722,8 +722,8 @@ static void destruct(struct dce110_resource_pool *pool)
dce_aud_destroy(&pool->base.audios[i]);
}
- if (pool->base.dccg != NULL)
- dce_dccg_destroy(&pool->base.dccg);
+ if (pool->base.clk_mgr != NULL)
+ dce_clk_mgr_destroy(&pool->base.clk_mgr);
if (pool->base.abm != NULL)
dce_abm_destroy(&pool->base.abm);
@@ -767,7 +767,7 @@ bool dce100_validate_bandwidth(
if (at_least_one_pipe) {
/* TODO implement when needed but for now hardcode max value*/
context->bw.dce.dispclk_khz = 681000;
- context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER;
+ context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
} else {
context->bw.dce.dispclk_khz = 0;
context->bw.dce.yclk_khz = 0;
@@ -860,7 +860,6 @@ static bool construct(
struct dc_context *ctx = dc->ctx;
struct dc_firmware_info info;
struct dc_bios *bp;
- struct dm_pp_static_clock_info static_clk_info = {0};
ctx->dc_bios->regs = &bios_regs;
@@ -908,11 +907,11 @@ static bool construct(
}
}
- pool->base.dccg = dce_dccg_create(ctx,
+ pool->base.clk_mgr = dce_clk_mgr_create(ctx,
&disp_clk_regs,
&disp_clk_shift,
&disp_clk_mask);
- if (pool->base.dccg == NULL) {
+ if (pool->base.clk_mgr == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto res_create_fail;
@@ -938,12 +937,6 @@ static bool construct(
goto res_create_fail;
}
- /* get static clock information for PPLIB or firmware, save
- * max_clock_state
- */
- if (dm_pp_get_static_clocks(ctx, &static_clk_info))
- pool->base.dccg->max_clks_state =
- static_clk_info.max_clocks_state;
{
struct irq_service_init_data init_data;
init_data.ctx = dc->ctx;
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
index b75ede5..9724a17 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c
@@ -548,14 +548,14 @@ dce110_translate_regamma_to_hw_format(const struct dc_transfer_func *output_tf,
regamma_params->hw_points_num = hw_points;
- i = 1;
- for (k = 0; k < 16 && i < 16; k++) {
+ k = 0;
+ for (i = 1; i < 16; i++) {
if (seg_distr[k] != -1) {
regamma_params->arr_curve_points[k].segments_num = seg_distr[k];
regamma_params->arr_curve_points[i].offset =
regamma_params->arr_curve_points[k].offset + (1 << seg_distr[k]);
}
- i++;
+ k++;
}
if (seg_distr[k] != -1)
@@ -1085,7 +1085,6 @@ void dce110_unblank_stream(struct pipe_ctx *pipe_ctx,
if (link->local_sink && link->local_sink->sink_signal == SIGNAL_TYPE_EDP) {
link->dc->hwss.edp_backlight_control(link, true);
- stream->bl_pwm_level = EDP_BACKLIGHT_RAMP_DISABLE_LEVEL;
}
}
void dce110_blank_stream(struct pipe_ctx *pipe_ctx)
@@ -1192,8 +1191,8 @@ static void build_audio_output(
if (pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT ||
pipe_ctx->stream->signal == SIGNAL_TYPE_DISPLAY_PORT_MST) {
audio_output->pll_info.dp_dto_source_clock_in_khz =
- state->dis_clk->funcs->get_dp_ref_clk_frequency(
- state->dis_clk);
+ state->dccg->funcs->get_dp_ref_clk_frequency(
+ state->dccg);
}
audio_output->pll_info.feed_back_divider =
@@ -1547,6 +1546,7 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
int i;
struct dc_link *edp_link_to_turnoff = NULL;
struct dc_link *edp_link = get_link_for_edp(dc);
+ struct dc_bios *bios = dc->ctx->dc_bios;
bool can_edp_fast_boot_optimize = false;
bool apply_edp_fast_boot_optimization = false;
@@ -1573,6 +1573,20 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context)
if (context->streams[i]->signal == SIGNAL_TYPE_EDP) {
context->streams[i]->apply_edp_fast_boot_optimization = true;
apply_edp_fast_boot_optimization = true;
+
+ /* When after S4 and S5, vbios may post edp and previous dpms_off
+ * doesn't make sense.
+ * Update dpms_off state to align hw and sw state via check
+ * vBios scratch register.
+ */
+ if (bios->funcs->is_active_display) {
+ const struct connector_device_tag_info *device_tag = &(edp_link->device_tag);
+
+ if (bios->funcs->is_active_display(bios,
+ context->streams[i]->signal,
+ device_tag))
+ context->streams[i]->dpms_off = false;
+ }
}
}
}
@@ -1736,41 +1750,18 @@ static void set_static_screen_control(struct pipe_ctx **pipe_ctx,
if (events->force_trigger)
value |= 0x1;
- value |= 0x84;
+ if (num_pipes) {
+ struct dc *dc = pipe_ctx[0]->stream->ctx->dc;
+
+ if (dc->fbc_compressor)
+ value |= 0x84;
+ }
for (i = 0; i < num_pipes; i++)
pipe_ctx[i]->stream_res.tg->funcs->
set_static_screen_control(pipe_ctx[i]->stream_res.tg, value);
}
-/* unit: in_khz before mode set, get pixel clock from context. ASIC register
- * may not be programmed yet
- */
-static uint32_t get_max_pixel_clock_for_all_paths(
- struct dc *dc,
- struct dc_state *context)
-{
- uint32_t max_pix_clk = 0;
- int i;
-
- for (i = 0; i < MAX_PIPES; i++) {
- struct pipe_ctx *pipe_ctx = &context->res_ctx.pipe_ctx[i];
-
- if (pipe_ctx->stream == NULL)
- continue;
-
- /* do not check under lay */
- if (pipe_ctx->top_pipe)
- continue;
-
- if (pipe_ctx->stream_res.pix_clk_params.requested_pix_clk > max_pix_clk)
- max_pix_clk =
- pipe_ctx->stream_res.pix_clk_params.requested_pix_clk;
- }
-
- return max_pix_clk;
-}
-
/*
* Check if FBC can be enabled
*/
@@ -2380,191 +2371,33 @@ static void init_hw(struct dc *dc)
}
-void dce110_fill_display_configs(
- const struct dc_state *context,
- struct dm_pp_display_configuration *pp_display_cfg)
-{
- int j;
- int num_cfgs = 0;
- for (j = 0; j < context->stream_count; j++) {
- int k;
-
- const struct dc_stream_state *stream = context->streams[j];
- struct dm_pp_single_disp_config *cfg =
- &pp_display_cfg->disp_configs[num_cfgs];
- const struct pipe_ctx *pipe_ctx = NULL;
-
- for (k = 0; k < MAX_PIPES; k++)
- if (stream == context->res_ctx.pipe_ctx[k].stream) {
- pipe_ctx = &context->res_ctx.pipe_ctx[k];
- break;
- }
-
- ASSERT(pipe_ctx != NULL);
-
- /* only notify active stream */
- if (stream->dpms_off)
- continue;
-
- num_cfgs++;
- cfg->signal = pipe_ctx->stream->signal;
- cfg->pipe_idx = pipe_ctx->stream_res.tg->inst;
- cfg->src_height = stream->src.height;
- cfg->src_width = stream->src.width;
- cfg->ddi_channel_mapping =
- stream->sink->link->ddi_channel_mapping.raw;
- cfg->transmitter =
- stream->sink->link->link_enc->transmitter;
- cfg->link_settings.lane_count =
- stream->sink->link->cur_link_settings.lane_count;
- cfg->link_settings.link_rate =
- stream->sink->link->cur_link_settings.link_rate;
- cfg->link_settings.link_spread =
- stream->sink->link->cur_link_settings.link_spread;
- cfg->sym_clock = stream->phy_pix_clk;
- /* Round v_refresh*/
- cfg->v_refresh = stream->timing.pix_clk_khz * 1000;
- cfg->v_refresh /= stream->timing.h_total;
- cfg->v_refresh = (cfg->v_refresh + stream->timing.v_total / 2)
- / stream->timing.v_total;
- }
-
- pp_display_cfg->display_count = num_cfgs;
-}
-
-uint32_t dce110_get_min_vblank_time_us(const struct dc_state *context)
-{
- uint8_t j;
- uint32_t min_vertical_blank_time = -1;
-
- for (j = 0; j < context->stream_count; j++) {
- struct dc_stream_state *stream = context->streams[j];
- uint32_t vertical_blank_in_pixels = 0;
- uint32_t vertical_blank_time = 0;
-
- vertical_blank_in_pixels = stream->timing.h_total *
- (stream->timing.v_total
- - stream->timing.v_addressable);
-
- vertical_blank_time = vertical_blank_in_pixels
- * 1000 / stream->timing.pix_clk_khz;
-
- if (min_vertical_blank_time > vertical_blank_time)
- min_vertical_blank_time = vertical_blank_time;
- }
-
- return min_vertical_blank_time;
-}
-
-static int determine_sclk_from_bounding_box(
- const struct dc *dc,
- int required_sclk)
-{
- int i;
-
- /*
- * Some asics do not give us sclk levels, so we just report the actual
- * required sclk
- */
- if (dc->sclk_lvls.num_levels == 0)
- return required_sclk;
-
- for (i = 0; i < dc->sclk_lvls.num_levels; i++) {
- if (dc->sclk_lvls.clocks_in_khz[i] >= required_sclk)
- return dc->sclk_lvls.clocks_in_khz[i];
- }
- /*
- * even maximum level could not satisfy requirement, this
- * is unexpected at this stage, should have been caught at
- * validation time
- */
- ASSERT(0);
- return dc->sclk_lvls.clocks_in_khz[dc->sclk_lvls.num_levels - 1];
-}
-
-static void pplib_apply_display_requirements(
- struct dc *dc,
- struct dc_state *context)
-{
- struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
-
- pp_display_cfg->all_displays_in_sync =
- context->bw.dce.all_displays_in_sync;
- pp_display_cfg->nb_pstate_switch_disable =
- context->bw.dce.nbp_state_change_enable == false;
- pp_display_cfg->cpu_cc6_disable =
- context->bw.dce.cpuc_state_change_enable == false;
- pp_display_cfg->cpu_pstate_disable =
- context->bw.dce.cpup_state_change_enable == false;
- pp_display_cfg->cpu_pstate_separation_time =
- context->bw.dce.blackout_recovery_time_us;
-
- pp_display_cfg->min_memory_clock_khz = context->bw.dce.yclk_khz
- / MEMORY_TYPE_MULTIPLIER;
-
- pp_display_cfg->min_engine_clock_khz = determine_sclk_from_bounding_box(
- dc,
- context->bw.dce.sclk_khz);
-
- pp_display_cfg->min_engine_clock_deep_sleep_khz
- = context->bw.dce.sclk_deep_sleep_khz;
-
- pp_display_cfg->avail_mclk_switch_time_us =
- dce110_get_min_vblank_time_us(context);
- /* TODO: dce11.2*/
- pp_display_cfg->avail_mclk_switch_time_in_disp_active_us = 0;
-
- pp_display_cfg->disp_clk_khz = dc->res_pool->dccg->clks.dispclk_khz;
-
- dce110_fill_display_configs(context, pp_display_cfg);
-
- /* TODO: is this still applicable?*/
- if (pp_display_cfg->display_count == 1) {
- const struct dc_crtc_timing *timing =
- &context->streams[0]->timing;
-
- pp_display_cfg->crtc_index =
- pp_display_cfg->disp_configs[0].pipe_idx;
- pp_display_cfg->line_time_in_us = timing->h_total * 1000
- / timing->pix_clk_khz;
- }
-
- if (memcmp(&dc->prev_display_config, pp_display_cfg, sizeof(
- struct dm_pp_display_configuration)) != 0)
- dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
-
- dc->prev_display_config = *pp_display_cfg;
-}
-
-static void dce110_set_bandwidth(
+void dce110_prepare_bandwidth(
struct dc *dc,
- struct dc_state *context,
- bool decrease_allowed)
+ struct dc_state *context)
{
- struct dc_clocks req_clks;
- struct dccg *dccg = dc->res_pool->dccg;
+ struct clk_mgr *dccg = dc->res_pool->clk_mgr;
- req_clks.dispclk_khz = context->bw.dce.dispclk_khz;
- req_clks.phyclk_khz = get_max_pixel_clock_for_all_paths(dc, context);
-
- if (decrease_allowed)
- dce110_set_displaymarks(dc, context);
- else
- dce110_set_safe_displaymarks(&context->res_ctx, dc->res_pool);
-
- if (dccg->funcs->update_dfs_bypass)
- dccg->funcs->update_dfs_bypass(
- dccg,
- dc,
- context,
- req_clks.dispclk_khz);
+ dce110_set_safe_displaymarks(&context->res_ctx, dc->res_pool);
dccg->funcs->update_clocks(
dccg,
- &req_clks,
- decrease_allowed);
- pplib_apply_display_requirements(dc, context);
+ context,
+ false);
+}
+
+void dce110_optimize_bandwidth(
+ struct dc *dc,
+ struct dc_state *context)
+{
+ struct clk_mgr *dccg = dc->res_pool->clk_mgr;
+
+ dce110_set_displaymarks(dc, context);
+
+ dccg->funcs->update_clocks(
+ dccg,
+ context,
+ true);
}
static void dce110_program_front_end_for_pipe(
@@ -2769,28 +2602,6 @@ static void dce110_wait_for_mpcc_disconnect(
/* do nothing*/
}
-static void program_csc_matrix(struct pipe_ctx *pipe_ctx,
- enum dc_color_space colorspace,
- uint16_t *matrix)
-{
- int i;
- struct out_csc_color_matrix tbl_entry;
-
- if (pipe_ctx->stream->csc_color_matrix.enable_adjustment
- == true) {
- enum dc_color_space color_space =
- pipe_ctx->stream->output_color_space;
-
- //uint16_t matrix[12];
- for (i = 0; i < 12; i++)
- tbl_entry.regval[i] = pipe_ctx->stream->csc_color_matrix.matrix[i];
-
- tbl_entry.color_space = color_space;
- //tbl_entry.regval = matrix;
- pipe_ctx->plane_res.xfm->funcs->opp_set_csc_adjustment(pipe_ctx->plane_res.xfm, &tbl_entry);
- }
-}
-
void dce110_set_cursor_position(struct pipe_ctx *pipe_ctx)
{
struct dc_cursor_position pos_cpy = pipe_ctx->stream->cursor_position;
@@ -2839,13 +2650,8 @@ void dce110_set_cursor_attribute(struct pipe_ctx *pipe_ctx)
pipe_ctx->plane_res.xfm, attributes);
}
-static void ready_shared_resources(struct dc *dc, struct dc_state *context) {}
-
-static void optimize_shared_resources(struct dc *dc) {}
-
static const struct hw_sequencer_funcs dce110_funcs = {
.program_gamut_remap = program_gamut_remap,
- .program_csc_matrix = program_csc_matrix,
.init_hw = init_hw,
.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
.apply_ctx_for_surface = dce110_apply_ctx_for_surface,
@@ -2868,7 +2674,8 @@ static const struct hw_sequencer_funcs dce110_funcs = {
.enable_display_power_gating = dce110_enable_display_power_gating,
.disable_plane = dce110_power_down_fe,
.pipe_control_lock = dce_pipe_control_lock,
- .set_bandwidth = dce110_set_bandwidth,
+ .prepare_bandwidth = dce110_prepare_bandwidth,
+ .optimize_bandwidth = dce110_optimize_bandwidth,
.set_drr = set_drr,
.get_position = get_position,
.set_static_screen_control = set_static_screen_control,
@@ -2877,9 +2684,6 @@ static const struct hw_sequencer_funcs dce110_funcs = {
.setup_stereo = NULL,
.set_avmute = dce110_set_avmute,
.wait_for_mpcc_disconnect = dce110_wait_for_mpcc_disconnect,
- .ready_shared_resources = ready_shared_resources,
- .optimize_shared_resources = optimize_shared_resources,
- .pplib_apply_display_requirements = pplib_apply_display_requirements,
.edp_backlight_control = hwss_edp_backlight_control,
.edp_power_control = hwss_edp_power_control,
.edp_wait_for_hpd_ready = hwss_edp_wait_for_hpd_ready,
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h
index d6db3db..cd3e36d 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.h
@@ -40,7 +40,6 @@ enum dc_status dce110_apply_ctx_to_hw(
struct dc_state *context);
-
void dce110_enable_stream(struct pipe_ctx *pipe_ctx);
void dce110_disable_stream(struct pipe_ctx *pipe_ctx, int option);
@@ -64,11 +63,13 @@ void dce110_set_safe_displaymarks(
struct resource_context *res_ctx,
const struct resource_pool *pool);
-void dce110_fill_display_configs(
- const struct dc_state *context,
- struct dm_pp_display_configuration *pp_display_cfg);
+void dce110_prepare_bandwidth(
+ struct dc *dc,
+ struct dc_state *context);
-uint32_t dce110_get_min_vblank_time_us(const struct dc_state *context);
+void dce110_optimize_bandwidth(
+ struct dc *dc,
+ struct dc_state *context);
void dp_receiver_power_ctrl(struct dc_link *link, bool on);
diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
index e3624ca..e33d117 100644
--- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_resource.c
@@ -31,6 +31,7 @@
#include "resource.h"
#include "dce110/dce110_resource.h"
+#include "dce/dce_clk_mgr.h"
#include "include/irq_service_interface.h"
#include "dce/dce_audio.h"
#include "dce110/dce110_timing_generator.h"
@@ -45,7 +46,6 @@
#include "dce110/dce110_transform_v.h"
#include "dce/dce_opp.h"
#include "dce110/dce110_opp_v.h"
-#include "dce/dce_clocks.h"
#include "dce/dce_clock_source.h"
#include "dce/dce_hwseq.h"
#include "dce110/dce110_hw_sequencer.h"
@@ -148,15 +148,15 @@ static const struct dce110_timing_generator_offsets dce110_tg_offsets[] = {
#define SRI(reg_name, block, id)\
.reg_name = mm ## block ## id ## _ ## reg_name
-static const struct dccg_registers disp_clk_regs = {
+static const struct clk_mgr_registers disp_clk_regs = {
CLK_COMMON_REG_LIST_DCE_BASE()
};
-static const struct dccg_shift disp_clk_shift = {
+static const struct clk_mgr_shift disp_clk_shift = {
CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(__SHIFT)
};
-static const struct dccg_mask disp_clk_mask = {
+static const struct clk_mgr_mask disp_clk_mask = {
CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(_MASK)
};
@@ -760,8 +760,8 @@ static void destruct(struct dce110_resource_pool *pool)
if (pool->base.dmcu != NULL)
dce_dmcu_destroy(&pool->base.dmcu);
- if (pool->base.dccg != NULL)
- dce_dccg_destroy(&pool->base.dccg);
+ if (pool->base.clk_mgr != NULL)
+ dce_clk_mgr_destroy(&pool->base.clk_mgr);
if (pool->base.irqs != NULL) {
dal_irq_service_destroy(&pool->base.irqs);
@@ -1173,12 +1173,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc)
&clks);
dc->bw_vbios->low_yclk = bw_frc_to_fixed(
- clks.clocks_in_khz[0] * MEMORY_TYPE_MULTIPLIER, 1000);
+ clks.clocks_in_khz[0] * MEMORY_TYPE_MULTIPLIER_CZ, 1000);
dc->bw_vbios->mid_yclk = bw_frc_to_fixed(
- clks.clocks_in_khz[clks.num_levels>>1] * MEMORY_TYPE_MULTIPLIER,
+ clks.clocks_in_khz[clks.num_levels>>1] * MEMORY_TYPE_MULTIPLIER_CZ,
1000);
dc->bw_vbios->high_yclk = bw_frc_to_fixed(
- clks.clocks_in_khz[clks.num_levels-1] * MEMORY_TYPE_MULTIPLIER,
+ clks.clocks_in_khz[clks.num_levels-1] * MEMORY_TYPE_MULTIPLIER_CZ,
1000);
}
@@ -1201,7 +1201,6 @@ static bool construct(
struct dc_context *ctx = dc->ctx;
struct dc_firmware_info info;
struct dc_bios *bp;
- struct dm_pp_static_clock_info static_clk_info = {0};
ctx->dc_bios->regs = &bios_regs;
@@ -1257,11 +1256,11 @@ static bool construct(
}
}
- pool->base.dccg = dce110_dccg_create(ctx,
+ pool->base.clk_mgr = dce110_clk_mgr_create(ctx,
&disp_clk_regs,
&disp_clk_shift,
&disp_clk_mask);
- if (pool->base.dccg == NULL) {
+ if (pool->base.clk_mgr == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto res_create_fail;
@@ -1287,13 +1286,6 @@ static bool construct(
goto res_create_fail;
}
- /* get static clock information for PPLIB or firmware, save
- * max_clock_state
- */
- if (dm_pp_get_static_clocks(ctx, &static_clk_info))
- pool->base.dccg->max_clks_state =
- static_clk_info.max_clocks_state;
-
{
struct irq_service_init_data init_data;
init_data.ctx = dc->ctx;
@@ -1362,7 +1354,8 @@ static bool construct(
pool->base.sw_i2cs[i] = NULL;
}
- dc->fbc_compressor = dce110_compressor_create(ctx);
+ if (dc->config.fbc_support)
+ dc->fbc_compressor = dce110_compressor_create(ctx);
if (!underlay_create(ctx, &pool->base))
goto res_create_fail;
diff --git a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
index 3ce79c2..969d4e7 100644
--- a/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce112/dce112_resource.c
@@ -35,6 +35,7 @@
#include "irq/dce110/irq_service_dce110.h"
+#include "dce/dce_clk_mgr.h"
#include "dce/dce_mem_input.h"
#include "dce/dce_transform.h"
#include "dce/dce_link_encoder.h"
@@ -42,7 +43,6 @@
#include "dce/dce_audio.h"
#include "dce/dce_opp.h"
#include "dce/dce_ipp.h"
-#include "dce/dce_clocks.h"
#include "dce/dce_clock_source.h"
#include "dce/dce_hwseq.h"
@@ -148,15 +148,15 @@ static const struct dce110_timing_generator_offsets dce112_tg_offsets[] = {
.reg_name = mm ## block ## id ## _ ## reg_name
-static const struct dccg_registers disp_clk_regs = {
+static const struct clk_mgr_registers disp_clk_regs = {
CLK_COMMON_REG_LIST_DCE_BASE()
};
-static const struct dccg_shift disp_clk_shift = {
+static const struct clk_mgr_shift disp_clk_shift = {
CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(__SHIFT)
};
-static const struct dccg_mask disp_clk_mask = {
+static const struct clk_mgr_mask disp_clk_mask = {
CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(_MASK)
};
@@ -551,7 +551,8 @@ static struct transform *dce112_transform_create(
static const struct encoder_feature_support link_enc_feature = {
.max_hdmi_deep_color = COLOR_DEPTH_121212,
.max_hdmi_pixel_clock = 600000,
- .ycbcr420_supported = true,
+ .hdmi_ycbcr420_supported = true,
+ .dp_ycbcr420_supported = false,
.flags.bits.IS_HBR2_CAPABLE = true,
.flags.bits.IS_HBR3_CAPABLE = true,
.flags.bits.IS_TPS3_CAPABLE = true,
@@ -749,8 +750,8 @@ static void destruct(struct dce110_resource_pool *pool)
if (pool->base.dmcu != NULL)
dce_dmcu_destroy(&pool->base.dmcu);
- if (pool->base.dccg != NULL)
- dce_dccg_destroy(&pool->base.dccg);
+ if (pool->base.clk_mgr != NULL)
+ dce_clk_mgr_destroy(&pool->base.clk_mgr);
if (pool->base.irqs != NULL) {
dal_irq_service_destroy(&pool->base.irqs);
@@ -1015,12 +1016,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc)
&clks);
dc->bw_vbios->low_yclk = bw_frc_to_fixed(
- clks.clocks_in_khz[0] * MEMORY_TYPE_MULTIPLIER, 1000);
+ clks.clocks_in_khz[0] * MEMORY_TYPE_MULTIPLIER_CZ, 1000);
dc->bw_vbios->mid_yclk = bw_frc_to_fixed(
- clks.clocks_in_khz[clks.num_levels>>1] * MEMORY_TYPE_MULTIPLIER,
+ clks.clocks_in_khz[clks.num_levels>>1] * MEMORY_TYPE_MULTIPLIER_CZ,
1000);
dc->bw_vbios->high_yclk = bw_frc_to_fixed(
- clks.clocks_in_khz[clks.num_levels-1] * MEMORY_TYPE_MULTIPLIER,
+ clks.clocks_in_khz[clks.num_levels-1] * MEMORY_TYPE_MULTIPLIER_CZ,
1000);
return;
@@ -1056,12 +1057,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc)
* YCLK = UMACLK*m_memoryTypeMultiplier
*/
dc->bw_vbios->low_yclk = bw_frc_to_fixed(
- mem_clks.data[0].clocks_in_khz * MEMORY_TYPE_MULTIPLIER, 1000);
+ mem_clks.data[0].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, 1000);
dc->bw_vbios->mid_yclk = bw_frc_to_fixed(
- mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER,
+ mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ,
1000);
dc->bw_vbios->high_yclk = bw_frc_to_fixed(
- mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER,
+ mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ,
1000);
/* Now notify PPLib/SMU about which Watermarks sets they should select
@@ -1131,7 +1132,6 @@ static bool construct(
{
unsigned int i;
struct dc_context *ctx = dc->ctx;
- struct dm_pp_static_clock_info static_clk_info = {0};
ctx->dc_bios->regs = &bios_regs;
@@ -1199,11 +1199,11 @@ static bool construct(
}
}
- pool->base.dccg = dce112_dccg_create(ctx,
+ pool->base.clk_mgr = dce112_clk_mgr_create(ctx,
&disp_clk_regs,
&disp_clk_shift,
&disp_clk_mask);
- if (pool->base.dccg == NULL) {
+ if (pool->base.clk_mgr == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto res_create_fail;
@@ -1229,13 +1229,6 @@ static bool construct(
goto res_create_fail;
}
- /* get static clock information for PPLIB or firmware, save
- * max_clock_state
- */
- if (dm_pp_get_static_clocks(ctx, &static_clk_info))
- pool->base.dccg->max_clks_state =
- static_clk_info.max_clocks_state;
-
{
struct irq_service_init_data init_data;
init_data.ctx = dc->ctx;
diff --git a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
index 79ab5f9..f126966 100644
--- a/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce120/dce120_resource.c
@@ -31,6 +31,7 @@
#include "resource.h"
#include "include/irq_service_interface.h"
#include "dce120_resource.h"
+
#include "dce112/dce112_resource.h"
#include "dce110/dce110_resource.h"
@@ -39,7 +40,6 @@
#include "irq/dce120/irq_service_dce120.h"
#include "dce/dce_opp.h"
#include "dce/dce_clock_source.h"
-#include "dce/dce_clocks.h"
#include "dce/dce_ipp.h"
#include "dce/dce_mem_input.h"
@@ -47,6 +47,7 @@
#include "dce120/dce120_hw_sequencer.h"
#include "dce/dce_transform.h"
+#include "dce/dce_clk_mgr.h"
#include "dce/dce_audio.h"
#include "dce/dce_link_encoder.h"
#include "dce/dce_stream_encoder.h"
@@ -573,8 +574,8 @@ static void destruct(struct dce110_resource_pool *pool)
if (pool->base.dmcu != NULL)
dce_dmcu_destroy(&pool->base.dmcu);
- if (pool->base.dccg != NULL)
- dce_dccg_destroy(&pool->base.dccg);
+ if (pool->base.clk_mgr != NULL)
+ dce_clk_mgr_destroy(&pool->base.clk_mgr);
}
static void read_dce_straps(
@@ -606,7 +607,8 @@ static struct audio *create_audio(
static const struct encoder_feature_support link_enc_feature = {
.max_hdmi_deep_color = COLOR_DEPTH_121212,
.max_hdmi_pixel_clock = 600000,
- .ycbcr420_supported = true,
+ .hdmi_ycbcr420_supported = true,
+ .dp_ycbcr420_supported = false,
.flags.bits.IS_HBR2_CAPABLE = true,
.flags.bits.IS_HBR3_CAPABLE = true,
.flags.bits.IS_TPS3_CAPABLE = true,
@@ -834,12 +836,12 @@ static void bw_calcs_data_update_from_pplib(struct dc *dc)
* YCLK = UMACLK*m_memoryTypeMultiplier
*/
dc->bw_vbios->low_yclk = bw_frc_to_fixed(
- mem_clks.data[0].clocks_in_khz * MEMORY_TYPE_MULTIPLIER, 1000);
+ mem_clks.data[0].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ, 1000);
dc->bw_vbios->mid_yclk = bw_frc_to_fixed(
- mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER,
+ mem_clks.data[mem_clks.num_levels>>1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ,
1000);
dc->bw_vbios->high_yclk = bw_frc_to_fixed(
- mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER,
+ mem_clks.data[mem_clks.num_levels-1].clocks_in_khz * MEMORY_TYPE_MULTIPLIER_CZ,
1000);
/* Now notify PPLib/SMU about which Watermarks sets they should select
@@ -973,8 +975,8 @@ static bool construct(
}
}
- pool->base.dccg = dce120_dccg_create(ctx);
- if (pool->base.dccg == NULL) {
+ pool->base.clk_mgr = dce120_clk_mgr_create(ctx);
+ if (pool->base.clk_mgr == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto dccg_create_fail;
diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c
index 6c6a1a1..a60a90e 100644
--- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_hw_sequencer.c
@@ -76,6 +76,7 @@ void dce80_hw_sequencer_construct(struct dc *dc)
dc->hwss.enable_display_power_gating = dce100_enable_display_power_gating;
dc->hwss.pipe_control_lock = dce_pipe_control_lock;
- dc->hwss.set_bandwidth = dce100_set_bandwidth;
+ dc->hwss.prepare_bandwidth = dce100_prepare_bandwidth;
+ dc->hwss.optimize_bandwidth = dce100_prepare_bandwidth;
}
diff --git a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
index d68f951..6d40b3d 100644
--- a/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dce80/dce80_resource.c
@@ -37,6 +37,7 @@
#include "dce110/dce110_timing_generator.h"
#include "dce110/dce110_resource.h"
#include "dce80/dce80_timing_generator.h"
+#include "dce/dce_clk_mgr.h"
#include "dce/dce_mem_input.h"
#include "dce/dce_link_encoder.h"
#include "dce/dce_stream_encoder.h"
@@ -44,7 +45,6 @@
#include "dce/dce_ipp.h"
#include "dce/dce_transform.h"
#include "dce/dce_opp.h"
-#include "dce/dce_clocks.h"
#include "dce/dce_clock_source.h"
#include "dce/dce_audio.h"
#include "dce/dce_hwseq.h"
@@ -155,15 +155,15 @@ static const struct dce110_timing_generator_offsets dce80_tg_offsets[] = {
.reg_name = mm ## block ## id ## _ ## reg_name
-static const struct dccg_registers disp_clk_regs = {
+static const struct clk_mgr_registers disp_clk_regs = {
CLK_COMMON_REG_LIST_DCE_BASE()
};
-static const struct dccg_shift disp_clk_shift = {
+static const struct clk_mgr_shift disp_clk_shift = {
CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(__SHIFT)
};
-static const struct dccg_mask disp_clk_mask = {
+static const struct clk_mgr_mask disp_clk_mask = {
CLK_COMMON_MASK_SH_LIST_DCE_COMMON_BASE(_MASK)
};
@@ -779,8 +779,8 @@ static void destruct(struct dce110_resource_pool *pool)
}
}
- if (pool->base.dccg != NULL)
- dce_dccg_destroy(&pool->base.dccg);
+ if (pool->base.clk_mgr != NULL)
+ dce_clk_mgr_destroy(&pool->base.clk_mgr);
if (pool->base.irqs != NULL) {
dal_irq_service_destroy(&pool->base.irqs);
@@ -793,7 +793,7 @@ bool dce80_validate_bandwidth(
{
/* TODO implement when needed but for now hardcode max value*/
context->bw.dce.dispclk_khz = 681000;
- context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER;
+ context->bw.dce.yclk_khz = 250000 * MEMORY_TYPE_MULTIPLIER_CZ;
return true;
}
@@ -855,7 +855,6 @@ static bool dce80_construct(
struct dc_context *ctx = dc->ctx;
struct dc_firmware_info info;
struct dc_bios *bp;
- struct dm_pp_static_clock_info static_clk_info = {0};
ctx->dc_bios->regs = &bios_regs;
@@ -918,11 +917,11 @@ static bool dce80_construct(
}
}
- pool->base.dccg = dce_dccg_create(ctx,
+ pool->base.clk_mgr = dce_clk_mgr_create(ctx,
&disp_clk_regs,
&disp_clk_shift,
&disp_clk_mask);
- if (pool->base.dccg == NULL) {
+ if (pool->base.clk_mgr == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto res_create_fail;
@@ -948,10 +947,6 @@ static bool dce80_construct(
goto res_create_fail;
}
- if (dm_pp_get_static_clocks(ctx, &static_clk_info))
- pool->base.dccg->max_clks_state =
- static_clk_info.max_clocks_state;
-
{
struct irq_service_init_data init_data;
init_data.ctx = dc->ctx;
@@ -1065,7 +1060,6 @@ static bool dce81_construct(
struct dc_context *ctx = dc->ctx;
struct dc_firmware_info info;
struct dc_bios *bp;
- struct dm_pp_static_clock_info static_clk_info = {0};
ctx->dc_bios->regs = &bios_regs;
@@ -1128,11 +1122,11 @@ static bool dce81_construct(
}
}
- pool->base.dccg = dce_dccg_create(ctx,
+ pool->base.clk_mgr = dce_clk_mgr_create(ctx,
&disp_clk_regs,
&disp_clk_shift,
&disp_clk_mask);
- if (pool->base.dccg == NULL) {
+ if (pool->base.clk_mgr == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto res_create_fail;
@@ -1158,10 +1152,6 @@ static bool dce81_construct(
goto res_create_fail;
}
- if (dm_pp_get_static_clocks(ctx, &static_clk_info))
- pool->base.dccg->max_clks_state =
- static_clk_info.max_clocks_state;
-
{
struct irq_service_init_data init_data;
init_data.ctx = dc->ctx;
@@ -1275,7 +1265,6 @@ static bool dce83_construct(
struct dc_context *ctx = dc->ctx;
struct dc_firmware_info info;
struct dc_bios *bp;
- struct dm_pp_static_clock_info static_clk_info = {0};
ctx->dc_bios->regs = &bios_regs;
@@ -1334,11 +1323,11 @@ static bool dce83_construct(
}
}
- pool->base.dccg = dce_dccg_create(ctx,
+ pool->base.clk_mgr = dce_clk_mgr_create(ctx,
&disp_clk_regs,
&disp_clk_shift,
&disp_clk_mask);
- if (pool->base.dccg == NULL) {
+ if (pool->base.clk_mgr == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto res_create_fail;
@@ -1364,10 +1353,6 @@ static bool dce83_construct(
goto res_create_fail;
}
- if (dm_pp_get_static_clocks(ctx, &static_clk_info))
- pool->base.dccg->max_clks_state =
- static_clk_info.max_clocks_state;
-
{
struct irq_service_init_data init_data;
init_data.ctx = dc->ctx;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile
index 032f872..55f293c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/Makefile
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/Makefile
@@ -24,7 +24,7 @@
DCN10 = dcn10_resource.o dcn10_ipp.o dcn10_hw_sequencer.o dcn10_hw_sequencer_debug.o \
dcn10_dpp.o dcn10_opp.o dcn10_optc.o \
- dcn10_hubp.o dcn10_mpc.o \
+ dcn10_hubp.o dcn10_mpc.o dcn10_clk_mgr.o \
dcn10_dpp_dscl.o dcn10_dpp_cm.o dcn10_cm_common.o \
dcn10_hubbub.o dcn10_stream_encoder.o dcn10_link_encoder.o
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c
new file mode 100644
index 0000000..20f531d
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.c
@@ -0,0 +1,379 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#include "dcn10_clk_mgr.h"
+
+#include "reg_helper.h"
+#include "core_types.h"
+
+#define TO_DCE_CLK_MGR(clocks)\
+ container_of(clocks, struct dce_clk_mgr, base)
+
+#define REG(reg) \
+ (clk_mgr_dce->regs->reg)
+
+#undef FN
+#define FN(reg_name, field_name) \
+ clk_mgr_dce->clk_mgr_shift->field_name, clk_mgr_dce->clk_mgr_mask->field_name
+
+#define CTX \
+ clk_mgr_dce->base.ctx
+#define DC_LOGGER \
+ clk_mgr->ctx->logger
+
+void dcn1_pplib_apply_display_requirements(
+ struct dc *dc,
+ struct dc_state *context)
+{
+ struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
+
+ pp_display_cfg->min_engine_clock_khz = dc->res_pool->clk_mgr->clks.dcfclk_khz;
+ pp_display_cfg->min_memory_clock_khz = dc->res_pool->clk_mgr->clks.fclk_khz;
+ pp_display_cfg->min_engine_clock_deep_sleep_khz = dc->res_pool->clk_mgr->clks.dcfclk_deep_sleep_khz;
+ pp_display_cfg->min_dcfc_deep_sleep_clock_khz = dc->res_pool->clk_mgr->clks.dcfclk_deep_sleep_khz;
+ pp_display_cfg->min_dcfclock_khz = dc->res_pool->clk_mgr->clks.dcfclk_khz;
+ pp_display_cfg->disp_clk_khz = dc->res_pool->clk_mgr->clks.dispclk_khz;
+ dce110_fill_display_configs(context, pp_display_cfg);
+
+ dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
+}
+
+static int dcn1_determine_dppclk_threshold(struct clk_mgr *clk_mgr, struct dc_clocks *new_clocks)
+{
+ bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz;
+ bool dispclk_increase = new_clocks->dispclk_khz > clk_mgr->clks.dispclk_khz;
+ int disp_clk_threshold = new_clocks->max_supported_dppclk_khz;
+ bool cur_dpp_div = clk_mgr->clks.dispclk_khz > clk_mgr->clks.dppclk_khz;
+
+ /* increase clock, looking for div is 0 for current, request div is 1*/
+ if (dispclk_increase) {
+ /* already divided by 2, no need to reach target clk with 2 steps*/
+ if (cur_dpp_div)
+ return new_clocks->dispclk_khz;
+
+ /* request disp clk is lower than maximum supported dpp clk,
+ * no need to reach target clk with two steps.
+ */
+ if (new_clocks->dispclk_khz <= disp_clk_threshold)
+ return new_clocks->dispclk_khz;
+
+ /* target dpp clk not request divided by 2, still within threshold */
+ if (!request_dpp_div)
+ return new_clocks->dispclk_khz;
+
+ } else {
+ /* decrease clock, looking for current dppclk divided by 2,
+ * request dppclk not divided by 2.
+ */
+
+ /* current dpp clk not divided by 2, no need to ramp*/
+ if (!cur_dpp_div)
+ return new_clocks->dispclk_khz;
+
+ /* current disp clk is lower than current maximum dpp clk,
+ * no need to ramp
+ */
+ if (clk_mgr->clks.dispclk_khz <= disp_clk_threshold)
+ return new_clocks->dispclk_khz;
+
+ /* request dpp clk need to be divided by 2 */
+ if (request_dpp_div)
+ return new_clocks->dispclk_khz;
+ }
+
+ return disp_clk_threshold;
+}
+
+static void dcn1_ramp_up_dispclk_with_dpp(struct clk_mgr *clk_mgr, struct dc_clocks *new_clocks)
+{
+ struct dc *dc = clk_mgr->ctx->dc;
+ int dispclk_to_dpp_threshold = dcn1_determine_dppclk_threshold(clk_mgr, new_clocks);
+ bool request_dpp_div = new_clocks->dispclk_khz > new_clocks->dppclk_khz;
+ int i;
+
+ /* set disp clk to dpp clk threshold */
+ dce112_set_clock(clk_mgr, dispclk_to_dpp_threshold);
+
+ /* update request dpp clk division option */
+ for (i = 0; i < dc->res_pool->pipe_count; i++) {
+ struct pipe_ctx *pipe_ctx = &dc->current_state->res_ctx.pipe_ctx[i];
+
+ if (!pipe_ctx->plane_state)
+ continue;
+
+ pipe_ctx->plane_res.dpp->funcs->dpp_dppclk_control(
+ pipe_ctx->plane_res.dpp,
+ request_dpp_div,
+ true);
+ }
+
+ /* If target clk not same as dppclk threshold, set to target clock */
+ if (dispclk_to_dpp_threshold != new_clocks->dispclk_khz)
+ dce112_set_clock(clk_mgr, new_clocks->dispclk_khz);
+
+ clk_mgr->clks.dispclk_khz = new_clocks->dispclk_khz;
+ clk_mgr->clks.dppclk_khz = new_clocks->dppclk_khz;
+ clk_mgr->clks.max_supported_dppclk_khz = new_clocks->max_supported_dppclk_khz;
+}
+
+static int get_active_display_cnt(
+ struct dc *dc,
+ struct dc_state *context)
+{
+ int i, display_count;
+
+ display_count = 0;
+ for (i = 0; i < context->stream_count; i++) {
+ const struct dc_stream_state *stream = context->streams[i];
+
+ /*
+ * Only notify active stream or virtual stream.
+ * Need to notify virtual stream to work around
+ * headless case. HPD does not fire when system is in
+ * S0i2.
+ */
+ if (!stream->dpms_off || stream->signal == SIGNAL_TYPE_VIRTUAL)
+ display_count++;
+ }
+
+ return display_count;
+}
+
+static void notify_deep_sleep_dcfclk_to_smu(
+ struct pp_smu_funcs_rv *pp_smu, int min_dcef_deep_sleep_clk_khz)
+{
+ int min_dcef_deep_sleep_clk_mhz; //minimum required DCEF Deep Sleep clock in mhz
+ /*
+ * if function pointer not set up, this message is
+ * sent as part of pplib_apply_display_requirements.
+ * So just return.
+ */
+ if (!pp_smu || !pp_smu->set_min_deep_sleep_dcfclk)
+ return;
+
+ min_dcef_deep_sleep_clk_mhz = (min_dcef_deep_sleep_clk_khz + 999) / 1000; //Round up
+ pp_smu->set_min_deep_sleep_dcfclk(&pp_smu->pp_smu, min_dcef_deep_sleep_clk_mhz);
+}
+
+static void notify_hard_min_dcfclk_to_smu(
+ struct pp_smu_funcs_rv *pp_smu, int min_dcf_clk_khz)
+{
+ int min_dcf_clk_mhz; //minimum required DCF clock in mhz
+
+ /*
+ * if function pointer not set up, this message is
+ * sent as part of pplib_apply_display_requirements.
+ * So just return.
+ */
+ if (!pp_smu || !pp_smu->set_hard_min_dcfclk_by_freq)
+ return;
+
+ min_dcf_clk_mhz = min_dcf_clk_khz / 1000;
+
+ pp_smu->set_hard_min_dcfclk_by_freq(&pp_smu->pp_smu, min_dcf_clk_mhz);
+}
+
+static void notify_hard_min_fclk_to_smu(
+ struct pp_smu_funcs_rv *pp_smu, int min_f_clk_khz)
+{
+ int min_f_clk_mhz; //minimum required F clock in mhz
+
+ /*
+ * if function pointer not set up, this message is
+ * sent as part of pplib_apply_display_requirements.
+ * So just return.
+ */
+ if (!pp_smu || !pp_smu->set_hard_min_fclk_by_freq)
+ return;
+
+ min_f_clk_mhz = min_f_clk_khz / 1000;
+
+ pp_smu->set_hard_min_fclk_by_freq(&pp_smu->pp_smu, min_f_clk_mhz);
+}
+
+static void dcn1_update_clocks(struct clk_mgr *clk_mgr,
+ struct dc_state *context,
+ bool safe_to_lower)
+{
+ struct dc *dc = clk_mgr->ctx->dc;
+ struct dc_clocks *new_clocks = &context->bw.dcn.clk;
+ struct pp_smu_display_requirement_rv *smu_req_cur =
+ &dc->res_pool->pp_smu_req;
+ struct pp_smu_display_requirement_rv smu_req = *smu_req_cur;
+ struct pp_smu_funcs_rv *pp_smu = dc->res_pool->pp_smu;
+ struct dm_pp_clock_for_voltage_req clock_voltage_req = {0};
+ bool send_request_to_increase = false;
+ bool send_request_to_lower = false;
+ int display_count;
+
+ bool enter_display_off = false;
+
+ display_count = get_active_display_cnt(dc, context);
+
+ if (display_count == 0)
+ enter_display_off = true;
+
+ if (enter_display_off == safe_to_lower) {
+ /*
+ * Notify SMU active displays
+ * if function pointer not set up, this message is
+ * sent as part of pplib_apply_display_requirements.
+ */
+ if (pp_smu->set_display_count)
+ pp_smu->set_display_count(&pp_smu->pp_smu, display_count);
+ else
+ smu_req.display_count = display_count;
+
+ }
+
+ if (new_clocks->dispclk_khz > clk_mgr->clks.dispclk_khz
+ || new_clocks->phyclk_khz > clk_mgr->clks.phyclk_khz
+ || new_clocks->fclk_khz > clk_mgr->clks.fclk_khz
+ || new_clocks->dcfclk_khz > clk_mgr->clks.dcfclk_khz)
+ send_request_to_increase = true;
+
+ if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, clk_mgr->clks.phyclk_khz)) {
+ clk_mgr->clks.phyclk_khz = new_clocks->phyclk_khz;
+
+ send_request_to_lower = true;
+ }
+
+ // F Clock
+ if (should_set_clock(safe_to_lower, new_clocks->fclk_khz, clk_mgr->clks.fclk_khz)) {
+ clk_mgr->clks.fclk_khz = new_clocks->fclk_khz;
+ clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_FCLK;
+ clock_voltage_req.clocks_in_khz = new_clocks->fclk_khz;
+ smu_req.hard_min_fclk_mhz = new_clocks->fclk_khz / 1000;
+
+ notify_hard_min_fclk_to_smu(pp_smu, new_clocks->fclk_khz);
+
+ send_request_to_lower = true;
+ }
+
+ //DCF Clock
+ if (should_set_clock(safe_to_lower, new_clocks->dcfclk_khz, clk_mgr->clks.dcfclk_khz)) {
+ clk_mgr->clks.dcfclk_khz = new_clocks->dcfclk_khz;
+ smu_req.hard_min_dcefclk_mhz = new_clocks->dcfclk_khz / 1000;
+
+ send_request_to_lower = true;
+ }
+
+ if (should_set_clock(safe_to_lower,
+ new_clocks->dcfclk_deep_sleep_khz, clk_mgr->clks.dcfclk_deep_sleep_khz)) {
+ clk_mgr->clks.dcfclk_deep_sleep_khz = new_clocks->dcfclk_deep_sleep_khz;
+ smu_req.min_deep_sleep_dcefclk_mhz = new_clocks->dcfclk_deep_sleep_khz / 1000;
+
+ send_request_to_lower = true;
+ }
+
+ /* make sure dcf clk is before dpp clk to
+ * make sure we have enough voltage to run dpp clk
+ */
+ if (send_request_to_increase) {
+ /*use dcfclk to request voltage*/
+ clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DCFCLK;
+ clock_voltage_req.clocks_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks);
+
+ notify_hard_min_dcfclk_to_smu(pp_smu, clock_voltage_req.clocks_in_khz);
+
+ if (pp_smu->set_display_requirement)
+ pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
+
+ notify_deep_sleep_dcfclk_to_smu(pp_smu, clk_mgr->clks.dcfclk_deep_sleep_khz);
+ dcn1_pplib_apply_display_requirements(dc, context);
+ }
+
+ /* dcn1 dppclk is tied to dispclk */
+ /* program dispclk on = as a w/a for sleep resume clock ramping issues */
+ if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr->clks.dispclk_khz)
+ || new_clocks->dispclk_khz == clk_mgr->clks.dispclk_khz) {
+ dcn1_ramp_up_dispclk_with_dpp(clk_mgr, new_clocks);
+ clk_mgr->clks.dispclk_khz = new_clocks->dispclk_khz;
+
+ send_request_to_lower = true;
+ }
+
+ if (!send_request_to_increase && send_request_to_lower) {
+ /*use dcfclk to request voltage*/
+ clock_voltage_req.clk_type = DM_PP_CLOCK_TYPE_DCFCLK;
+ clock_voltage_req.clocks_in_khz = dcn_find_dcfclk_suits_all(dc, new_clocks);
+
+ notify_hard_min_dcfclk_to_smu(pp_smu, clock_voltage_req.clocks_in_khz);
+
+ if (pp_smu->set_display_requirement)
+ pp_smu->set_display_requirement(&pp_smu->pp_smu, &smu_req);
+
+ notify_deep_sleep_dcfclk_to_smu(pp_smu, clk_mgr->clks.dcfclk_deep_sleep_khz);
+ dcn1_pplib_apply_display_requirements(dc, context);
+ }
+
+
+ *smu_req_cur = smu_req;
+}
+
+static const struct clk_mgr_funcs dcn1_funcs = {
+ .get_dp_ref_clk_frequency = dce12_get_dp_ref_freq_khz,
+ .update_clocks = dcn1_update_clocks
+};
+
+struct clk_mgr *dcn1_clk_mgr_create(struct dc_context *ctx)
+{
+ struct dc_debug_options *debug = &ctx->dc->debug;
+ struct dc_bios *bp = ctx->dc_bios;
+ struct dc_firmware_info fw_info = { { 0 } };
+ struct dce_clk_mgr *clk_mgr_dce = kzalloc(sizeof(*clk_mgr_dce), GFP_KERNEL);
+
+ if (clk_mgr_dce == NULL) {
+ BREAK_TO_DEBUGGER();
+ return NULL;
+ }
+
+ clk_mgr_dce->base.ctx = ctx;
+ clk_mgr_dce->base.funcs = &dcn1_funcs;
+
+ clk_mgr_dce->dfs_bypass_disp_clk = 0;
+
+ clk_mgr_dce->dprefclk_ss_percentage = 0;
+ clk_mgr_dce->dprefclk_ss_divider = 1000;
+ clk_mgr_dce->ss_on_dprefclk = false;
+
+ clk_mgr_dce->dprefclk_khz = 600000;
+ if (bp->integrated_info)
+ clk_mgr_dce->dentist_vco_freq_khz = bp->integrated_info->dentist_vco_freq;
+ if (clk_mgr_dce->dentist_vco_freq_khz == 0) {
+ bp->funcs->get_firmware_info(bp, &fw_info);
+ clk_mgr_dce->dentist_vco_freq_khz = fw_info.smu_gpu_pll_output_freq;
+ if (clk_mgr_dce->dentist_vco_freq_khz == 0)
+ clk_mgr_dce->dentist_vco_freq_khz = 3600000;
+ }
+
+ if (!debug->disable_dfs_bypass && bp->integrated_info)
+ if (bp->integrated_info->gpu_cap_info & DFS_BYPASS_ENABLE)
+ clk_mgr_dce->dfs_bypass_enabled = true;
+
+ dce_clock_read_ss_info(clk_mgr_dce);
+
+ return &clk_mgr_dce->base;
+}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h
new file mode 100644
index 0000000..9dbaf65
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_clk_mgr.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#ifndef __DCN10_CLK_MGR_H__
+#define __DCN10_CLK_MGR_H__
+
+#include "../dce/dce_clk_mgr.h"
+
+void dcn1_pplib_apply_display_requirements(
+ struct dc *dc,
+ struct dc_state *context);
+
+struct clk_mgr *dcn1_clk_mgr_create(struct dc_context *ctx);
+
+#endif //__DCN10_CLK_MGR_H__
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c
index 5d95a99..3eea440 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c
@@ -71,39 +71,39 @@ void cm_helper_program_xfer_func(
unsigned int i = 0;
REG_SET_2(reg->start_cntl_b, 0,
- exp_region_start, params->arr_points[0].custom_float_x,
+ exp_region_start, params->corner_points[0].blue.custom_float_x,
exp_resion_start_segment, 0);
REG_SET_2(reg->start_cntl_g, 0,
- exp_region_start, params->arr_points[0].custom_float_x,
+ exp_region_start, params->corner_points[0].green.custom_float_x,
exp_resion_start_segment, 0);
REG_SET_2(reg->start_cntl_r, 0,
- exp_region_start, params->arr_points[0].custom_float_x,
+ exp_region_start, params->corner_points[0].red.custom_float_x,
exp_resion_start_segment, 0);
REG_SET(reg->start_slope_cntl_b, 0,
- field_region_linear_slope, params->arr_points[0].custom_float_slope);
+ field_region_linear_slope, params->corner_points[0].blue.custom_float_slope);
REG_SET(reg->start_slope_cntl_g, 0,
- field_region_linear_slope, params->arr_points[0].custom_float_slope);
+ field_region_linear_slope, params->corner_points[0].green.custom_float_slope);
REG_SET(reg->start_slope_cntl_r, 0,
- field_region_linear_slope, params->arr_points[0].custom_float_slope);
+ field_region_linear_slope, params->corner_points[0].red.custom_float_slope);
REG_SET(reg->start_end_cntl1_b, 0,
- field_region_end, params->arr_points[1].custom_float_x);
+ field_region_end, params->corner_points[1].blue.custom_float_x);
REG_SET_2(reg->start_end_cntl2_b, 0,
- field_region_end_slope, params->arr_points[1].custom_float_slope,
- field_region_end_base, params->arr_points[1].custom_float_y);
+ field_region_end_slope, params->corner_points[1].blue.custom_float_slope,
+ field_region_end_base, params->corner_points[1].blue.custom_float_y);
REG_SET(reg->start_end_cntl1_g, 0,
- field_region_end, params->arr_points[1].custom_float_x);
+ field_region_end, params->corner_points[1].green.custom_float_x);
REG_SET_2(reg->start_end_cntl2_g, 0,
- field_region_end_slope, params->arr_points[1].custom_float_slope,
- field_region_end_base, params->arr_points[1].custom_float_y);
+ field_region_end_slope, params->corner_points[1].green.custom_float_slope,
+ field_region_end_base, params->corner_points[1].green.custom_float_y);
REG_SET(reg->start_end_cntl1_r, 0,
- field_region_end, params->arr_points[1].custom_float_x);
+ field_region_end, params->corner_points[1].red.custom_float_x);
REG_SET_2(reg->start_end_cntl2_r, 0,
- field_region_end_slope, params->arr_points[1].custom_float_slope,
- field_region_end_base, params->arr_points[1].custom_float_y);
+ field_region_end_slope, params->corner_points[1].red.custom_float_slope,
+ field_region_end_base, params->corner_points[1].red.custom_float_y);
for (reg_region_cur = reg->region_start;
reg_region_cur <= reg->region_end;
@@ -127,7 +127,7 @@ void cm_helper_program_xfer_func(
bool cm_helper_convert_to_custom_float(
struct pwl_result_data *rgb_resulted,
- struct curve_points *arr_points,
+ struct curve_points3 *corner_points,
uint32_t hw_points_num,
bool fixpoint)
{
@@ -141,20 +141,53 @@ bool cm_helper_convert_to_custom_float(
fmt.mantissa_bits = 12;
fmt.sign = false;
- if (!convert_to_custom_float_format(arr_points[0].x, &fmt,
- &arr_points[0].custom_float_x)) {
+ /* corner_points[0] - beginning base, slope offset for R,G,B
+ * corner_points[1] - end base, slope offset for R,G,B
+ */
+ if (!convert_to_custom_float_format(corner_points[0].red.x, &fmt,
+ &corner_points[0].red.custom_float_x)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[0].green.x, &fmt,
+ &corner_points[0].green.custom_float_x)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[0].blue.x, &fmt,
+ &corner_points[0].blue.custom_float_x)) {
BREAK_TO_DEBUGGER();
return false;
}
- if (!convert_to_custom_float_format(arr_points[0].offset, &fmt,
- &arr_points[0].custom_float_offset)) {
+ if (!convert_to_custom_float_format(corner_points[0].red.offset, &fmt,
+ &corner_points[0].red.custom_float_offset)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[0].green.offset, &fmt,
+ &corner_points[0].green.custom_float_offset)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[0].blue.offset, &fmt,
+ &corner_points[0].blue.custom_float_offset)) {
BREAK_TO_DEBUGGER();
return false;
}
- if (!convert_to_custom_float_format(arr_points[0].slope, &fmt,
- &arr_points[0].custom_float_slope)) {
+ if (!convert_to_custom_float_format(corner_points[0].red.slope, &fmt,
+ &corner_points[0].red.custom_float_slope)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[0].green.slope, &fmt,
+ &corner_points[0].green.custom_float_slope)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[0].blue.slope, &fmt,
+ &corner_points[0].blue.custom_float_slope)) {
BREAK_TO_DEBUGGER();
return false;
}
@@ -162,22 +195,59 @@ bool cm_helper_convert_to_custom_float(
fmt.mantissa_bits = 10;
fmt.sign = false;
- if (!convert_to_custom_float_format(arr_points[1].x, &fmt,
- &arr_points[1].custom_float_x)) {
+ if (!convert_to_custom_float_format(corner_points[1].red.x, &fmt,
+ &corner_points[1].red.custom_float_x)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[1].green.x, &fmt,
+ &corner_points[1].green.custom_float_x)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[1].blue.x, &fmt,
+ &corner_points[1].blue.custom_float_x)) {
BREAK_TO_DEBUGGER();
return false;
}
- if (fixpoint == true)
- arr_points[1].custom_float_y = dc_fixpt_clamp_u0d14(arr_points[1].y);
- else if (!convert_to_custom_float_format(arr_points[1].y, &fmt,
- &arr_points[1].custom_float_y)) {
+ if (fixpoint == true) {
+ corner_points[1].red.custom_float_y =
+ dc_fixpt_clamp_u0d14(corner_points[1].red.y);
+ corner_points[1].green.custom_float_y =
+ dc_fixpt_clamp_u0d14(corner_points[1].green.y);
+ corner_points[1].blue.custom_float_y =
+ dc_fixpt_clamp_u0d14(corner_points[1].blue.y);
+ } else {
+ if (!convert_to_custom_float_format(corner_points[1].red.y,
+ &fmt, &corner_points[1].red.custom_float_y)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[1].green.y,
+ &fmt, &corner_points[1].green.custom_float_y)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[1].blue.y,
+ &fmt, &corner_points[1].blue.custom_float_y)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ }
+
+ if (!convert_to_custom_float_format(corner_points[1].red.slope, &fmt,
+ &corner_points[1].red.custom_float_slope)) {
BREAK_TO_DEBUGGER();
return false;
}
-
- if (!convert_to_custom_float_format(arr_points[1].slope, &fmt,
- &arr_points[1].custom_float_slope)) {
+ if (!convert_to_custom_float_format(corner_points[1].green.slope, &fmt,
+ &corner_points[1].green.custom_float_slope)) {
+ BREAK_TO_DEBUGGER();
+ return false;
+ }
+ if (!convert_to_custom_float_format(corner_points[1].blue.slope, &fmt,
+ &corner_points[1].blue.custom_float_slope)) {
BREAK_TO_DEBUGGER();
return false;
}
@@ -242,15 +312,10 @@ bool cm_helper_translate_curve_to_hw_format(
const struct dc_transfer_func *output_tf,
struct pwl_params *lut_params, bool fixpoint)
{
- struct curve_points *arr_points;
+ struct curve_points3 *corner_points;
struct pwl_result_data *rgb_resulted;
struct pwl_result_data *rgb;
struct pwl_result_data *rgb_plus_1;
- struct fixed31_32 y_r;
- struct fixed31_32 y_g;
- struct fixed31_32 y_b;
- struct fixed31_32 y1_min;
- struct fixed31_32 y3_max;
int32_t region_start, region_end;
int32_t i;
@@ -261,14 +326,14 @@ bool cm_helper_translate_curve_to_hw_format(
PERF_TRACE();
- arr_points = lut_params->arr_points;
+ corner_points = lut_params->corner_points;
rgb_resulted = lut_params->rgb_resulted;
hw_points = 0;
memset(lut_params, 0, sizeof(struct pwl_params));
memset(seg_distr, 0, sizeof(seg_distr));
- if (output_tf->tf == TRANSFER_FUNCTION_PQ) {
+ if (output_tf->tf == TRANSFER_FUNCTION_PQ || output_tf->tf == TRANSFER_FUNCTION_GAMMA22) {
/* 32 segments
* segments are from 2^-25 to 2^7
*/
@@ -327,31 +392,37 @@ bool cm_helper_translate_curve_to_hw_format(
rgb_resulted[hw_points - 1].green = output_tf->tf_pts.green[start_index];
rgb_resulted[hw_points - 1].blue = output_tf->tf_pts.blue[start_index];
- arr_points[0].x = dc_fixpt_pow(dc_fixpt_from_int(2),
+ // All 3 color channels have same x
+ corner_points[0].red.x = dc_fixpt_pow(dc_fixpt_from_int(2),
dc_fixpt_from_int(region_start));
- arr_points[1].x = dc_fixpt_pow(dc_fixpt_from_int(2),
+ corner_points[0].green.x = corner_points[0].red.x;
+ corner_points[0].blue.x = corner_points[0].red.x;
+
+ corner_points[1].red.x = dc_fixpt_pow(dc_fixpt_from_int(2),
dc_fixpt_from_int(region_end));
+ corner_points[1].green.x = corner_points[1].red.x;
+ corner_points[1].blue.x = corner_points[1].red.x;
- y_r = rgb_resulted[0].red;
- y_g = rgb_resulted[0].green;
- y_b = rgb_resulted[0].blue;
+ corner_points[0].red.y = rgb_resulted[0].red;
+ corner_points[0].green.y = rgb_resulted[0].green;
+ corner_points[0].blue.y = rgb_resulted[0].blue;
- y1_min = dc_fixpt_min(y_r, dc_fixpt_min(y_g, y_b));
-
- arr_points[0].y = y1_min;
- arr_points[0].slope = dc_fixpt_div(arr_points[0].y, arr_points[0].x);
- y_r = rgb_resulted[hw_points - 1].red;
- y_g = rgb_resulted[hw_points - 1].green;
- y_b = rgb_resulted[hw_points - 1].blue;
+ corner_points[0].red.slope = dc_fixpt_div(corner_points[0].red.y,
+ corner_points[0].red.x);
+ corner_points[0].green.slope = dc_fixpt_div(corner_points[0].green.y,
+ corner_points[0].green.x);
+ corner_points[0].blue.slope = dc_fixpt_div(corner_points[0].blue.y,
+ corner_points[0].blue.x);
/* see comment above, m_arrPoints[1].y should be the Y value for the
* region end (m_numOfHwPoints), not last HW point(m_numOfHwPoints - 1)
*/
- y3_max = dc_fixpt_max(y_r, dc_fixpt_max(y_g, y_b));
-
- arr_points[1].y = y3_max;
-
- arr_points[1].slope = dc_fixpt_zero;
+ corner_points[1].red.y = rgb_resulted[hw_points - 1].red;
+ corner_points[1].green.y = rgb_resulted[hw_points - 1].green;
+ corner_points[1].blue.y = rgb_resulted[hw_points - 1].blue;
+ corner_points[1].red.slope = dc_fixpt_zero;
+ corner_points[1].green.slope = dc_fixpt_zero;
+ corner_points[1].blue.slope = dc_fixpt_zero;
if (output_tf->tf == TRANSFER_FUNCTION_PQ) {
/* for PQ, we want to have a straight line from last HW X point,
@@ -360,9 +431,15 @@ bool cm_helper_translate_curve_to_hw_format(
const struct fixed31_32 end_value =
dc_fixpt_from_int(125);
- arr_points[1].slope = dc_fixpt_div(
- dc_fixpt_sub(dc_fixpt_one, arr_points[1].y),
- dc_fixpt_sub(end_value, arr_points[1].x));
+ corner_points[1].red.slope = dc_fixpt_div(
+ dc_fixpt_sub(dc_fixpt_one, corner_points[1].red.y),
+ dc_fixpt_sub(end_value, corner_points[1].red.x));
+ corner_points[1].green.slope = dc_fixpt_div(
+ dc_fixpt_sub(dc_fixpt_one, corner_points[1].green.y),
+ dc_fixpt_sub(end_value, corner_points[1].green.x));
+ corner_points[1].blue.slope = dc_fixpt_div(
+ dc_fixpt_sub(dc_fixpt_one, corner_points[1].blue.y),
+ dc_fixpt_sub(end_value, corner_points[1].blue.x));
}
lut_params->hw_points_num = hw_points;
@@ -411,7 +488,7 @@ bool cm_helper_translate_curve_to_hw_format(
++i;
}
cm_helper_convert_to_custom_float(rgb_resulted,
- lut_params->arr_points,
+ lut_params->corner_points,
hw_points, fixpoint);
return true;
@@ -424,15 +501,10 @@ bool cm_helper_translate_curve_to_degamma_hw_format(
const struct dc_transfer_func *output_tf,
struct pwl_params *lut_params)
{
- struct curve_points *arr_points;
+ struct curve_points3 *corner_points;
struct pwl_result_data *rgb_resulted;
struct pwl_result_data *rgb;
struct pwl_result_data *rgb_plus_1;
- struct fixed31_32 y_r;
- struct fixed31_32 y_g;
- struct fixed31_32 y_b;
- struct fixed31_32 y1_min;
- struct fixed31_32 y3_max;
int32_t region_start, region_end;
int32_t i;
@@ -443,7 +515,7 @@ bool cm_helper_translate_curve_to_degamma_hw_format(
PERF_TRACE();
- arr_points = lut_params->arr_points;
+ corner_points = lut_params->corner_points;
rgb_resulted = lut_params->rgb_resulted;
hw_points = 0;
@@ -489,31 +561,28 @@ bool cm_helper_translate_curve_to_degamma_hw_format(
rgb_resulted[hw_points - 1].green = output_tf->tf_pts.green[start_index];
rgb_resulted[hw_points - 1].blue = output_tf->tf_pts.blue[start_index];
- arr_points[0].x = dc_fixpt_pow(dc_fixpt_from_int(2),
+ corner_points[0].red.x = dc_fixpt_pow(dc_fixpt_from_int(2),
dc_fixpt_from_int(region_start));
- arr_points[1].x = dc_fixpt_pow(dc_fixpt_from_int(2),
+ corner_points[0].green.x = corner_points[0].red.x;
+ corner_points[0].blue.x = corner_points[0].red.x;
+ corner_points[1].red.x = dc_fixpt_pow(dc_fixpt_from_int(2),
dc_fixpt_from_int(region_end));
+ corner_points[1].green.x = corner_points[1].red.x;
+ corner_points[1].blue.x = corner_points[1].red.x;
- y_r = rgb_resulted[0].red;
- y_g = rgb_resulted[0].green;
- y_b = rgb_resulted[0].blue;
-
- y1_min = dc_fixpt_min(y_r, dc_fixpt_min(y_g, y_b));
-
- arr_points[0].y = y1_min;
- arr_points[0].slope = dc_fixpt_div(arr_points[0].y, arr_points[0].x);
- y_r = rgb_resulted[hw_points - 1].red;
- y_g = rgb_resulted[hw_points - 1].green;
- y_b = rgb_resulted[hw_points - 1].blue;
+ corner_points[0].red.y = rgb_resulted[0].red;
+ corner_points[0].green.y = rgb_resulted[0].green;
+ corner_points[0].blue.y = rgb_resulted[0].blue;
/* see comment above, m_arrPoints[1].y should be the Y value for the
* region end (m_numOfHwPoints), not last HW point(m_numOfHwPoints - 1)
*/
- y3_max = dc_fixpt_max(y_r, dc_fixpt_max(y_g, y_b));
-
- arr_points[1].y = y3_max;
-
- arr_points[1].slope = dc_fixpt_zero;
+ corner_points[1].red.y = rgb_resulted[hw_points - 1].red;
+ corner_points[1].green.y = rgb_resulted[hw_points - 1].green;
+ corner_points[1].blue.y = rgb_resulted[hw_points - 1].blue;
+ corner_points[1].red.slope = dc_fixpt_zero;
+ corner_points[1].green.slope = dc_fixpt_zero;
+ corner_points[1].blue.slope = dc_fixpt_zero;
if (output_tf->tf == TRANSFER_FUNCTION_PQ) {
/* for PQ, we want to have a straight line from last HW X point,
@@ -522,9 +591,15 @@ bool cm_helper_translate_curve_to_degamma_hw_format(
const struct fixed31_32 end_value =
dc_fixpt_from_int(125);
- arr_points[1].slope = dc_fixpt_div(
- dc_fixpt_sub(dc_fixpt_one, arr_points[1].y),
- dc_fixpt_sub(end_value, arr_points[1].x));
+ corner_points[1].red.slope = dc_fixpt_div(
+ dc_fixpt_sub(dc_fixpt_one, corner_points[1].red.y),
+ dc_fixpt_sub(end_value, corner_points[1].red.x));
+ corner_points[1].green.slope = dc_fixpt_div(
+ dc_fixpt_sub(dc_fixpt_one, corner_points[1].green.y),
+ dc_fixpt_sub(end_value, corner_points[1].green.x));
+ corner_points[1].blue.slope = dc_fixpt_div(
+ dc_fixpt_sub(dc_fixpt_one, corner_points[1].blue.y),
+ dc_fixpt_sub(end_value, corner_points[1].blue.x));
}
lut_params->hw_points_num = hw_points;
@@ -564,7 +639,7 @@ bool cm_helper_translate_curve_to_degamma_hw_format(
++i;
}
cm_helper_convert_to_custom_float(rgb_resulted,
- lut_params->arr_points,
+ lut_params->corner_points,
hw_points, false);
return true;
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.h
index 7a531b0..5ae4d69 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.h
@@ -98,7 +98,7 @@ void cm_helper_program_xfer_func(
bool cm_helper_convert_to_custom_float(
struct pwl_result_data *rgb_resulted,
- struct curve_points *arr_points,
+ struct curve_points3 *corner_points,
uint32_t hw_points_num,
bool fixpoint);
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
index 193184a..87495de 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c
@@ -45,6 +45,7 @@
#include "dcn10_hubbub.h"
#include "dcn10_cm_common.h"
#include "dc_link_dp.h"
+#include "dccg.h"
#define DC_LOGGER_INIT(logger)
@@ -786,7 +787,7 @@ static bool dcn10_hw_wa_force_recovery(struct dc *dc)
&dc->current_state->res_ctx.pipe_ctx[i];
if (pipe_ctx != NULL) {
hubp = pipe_ctx->plane_res.hubp;
- if (hubp != NULL) {
+ if (hubp != NULL && hubp->funcs->hubp_get_underflow_status) {
if (hubp->funcs->hubp_get_underflow_status(hubp) != 0) {
/* one pipe underflow, we will reset all the pipes*/
need_recover = true;
@@ -812,7 +813,7 @@ static bool dcn10_hw_wa_force_recovery(struct dc *dc)
if (pipe_ctx != NULL) {
hubp = pipe_ctx->plane_res.hubp;
/*DCHUBP_CNTL:HUBP_BLANK_EN=1*/
- if (hubp != NULL)
+ if (hubp != NULL && hubp->funcs->set_hubp_blank_en)
hubp->funcs->set_hubp_blank_en(hubp, true);
}
}
@@ -825,7 +826,7 @@ static bool dcn10_hw_wa_force_recovery(struct dc *dc)
if (pipe_ctx != NULL) {
hubp = pipe_ctx->plane_res.hubp;
/*DCHUBP_CNTL:HUBP_DISABLE=1*/
- if (hubp != NULL)
+ if (hubp != NULL && hubp->funcs->hubp_disable_control)
hubp->funcs->hubp_disable_control(hubp, true);
}
}
@@ -835,7 +836,7 @@ static bool dcn10_hw_wa_force_recovery(struct dc *dc)
if (pipe_ctx != NULL) {
hubp = pipe_ctx->plane_res.hubp;
/*DCHUBP_CNTL:HUBP_DISABLE=0*/
- if (hubp != NULL)
+ if (hubp != NULL && hubp->funcs->hubp_disable_control)
hubp->funcs->hubp_disable_control(hubp, true);
}
}
@@ -847,7 +848,7 @@ static bool dcn10_hw_wa_force_recovery(struct dc *dc)
if (pipe_ctx != NULL) {
hubp = pipe_ctx->plane_res.hubp;
/*DCHUBP_CNTL:HUBP_BLANK_EN=0*/
- if (hubp != NULL)
+ if (hubp != NULL && hubp->funcs->set_hubp_blank_en)
hubp->funcs->set_hubp_blank_en(hubp, true);
}
}
@@ -1126,7 +1127,7 @@ static void dcn10_init_hw(struct dc *dc)
enable_power_gating_plane(dc->hwseq, true);
- memset(&dc->res_pool->dccg->clks, 0, sizeof(dc->res_pool->dccg->clks));
+ memset(&dc->res_pool->clk_mgr->clks, 0, sizeof(dc->res_pool->clk_mgr->clks));
}
static void reset_hw_ctx_wrap(
@@ -1603,7 +1604,7 @@ static void mmhub_read_vm_context0_settings(struct dcn10_hubp *hubp1,
}
-static void dcn10_program_pte_vm(struct dce_hwseq *hws, struct hubp *hubp)
+void dcn10_program_pte_vm(struct dce_hwseq *hws, struct hubp *hubp)
{
struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp);
struct vm_system_aperture_param apt = { {{ 0 } } };
@@ -1703,33 +1704,22 @@ static void program_gamut_remap(struct pipe_ctx *pipe_ctx)
pipe_ctx->plane_res.dpp->funcs->dpp_set_gamut_remap(pipe_ctx->plane_res.dpp, &adjust);
}
-
-static void program_csc_matrix(struct pipe_ctx *pipe_ctx,
- enum dc_color_space colorspace,
- uint16_t *matrix)
-{
- if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) {
- if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment != NULL)
- pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment(pipe_ctx->plane_res.dpp, matrix);
- } else {
- if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default != NULL)
- pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default(pipe_ctx->plane_res.dpp, colorspace);
- }
-}
-
static void dcn10_program_output_csc(struct dc *dc,
struct pipe_ctx *pipe_ctx,
enum dc_color_space colorspace,
uint16_t *matrix,
int opp_id)
{
- if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment != NULL)
- program_csc_matrix(pipe_ctx,
- colorspace,
- matrix);
+ if (pipe_ctx->stream->csc_color_matrix.enable_adjustment == true) {
+ if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment != NULL)
+ pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_adjustment(pipe_ctx->plane_res.dpp, matrix);
+ } else {
+ if (pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default != NULL)
+ pipe_ctx->plane_res.dpp->funcs->dpp_set_csc_default(pipe_ctx->plane_res.dpp, colorspace);
+ }
}
-static bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx)
+bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx)
{
if (pipe_ctx->plane_state->visible)
return true;
@@ -1738,7 +1728,7 @@ static bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx)
return false;
}
-static bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx)
+bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx)
{
if (pipe_ctx->plane_state->visible)
return true;
@@ -1747,7 +1737,7 @@ static bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx)
return false;
}
-static bool is_pipe_tree_visible(struct pipe_ctx *pipe_ctx)
+bool is_pipe_tree_visible(struct pipe_ctx *pipe_ctx)
{
if (pipe_ctx->plane_state->visible)
return true;
@@ -1943,10 +1933,6 @@ static void dcn10_update_mpcc(struct dc *dc, struct pipe_ctx *pipe_ctx)
struct mpc *mpc = dc->res_pool->mpc;
struct mpc_tree *mpc_tree_params = &(pipe_ctx->stream_res.opp->mpc_tree_params);
-
-
- /* TODO: proper fix once fpga works */
-
if (dc->debug.visual_confirm == VISUAL_CONFIRM_HDR) {
dcn10_get_hdr_visual_confirm_color(
pipe_ctx, &blnd_cfg.black_color);
@@ -2026,8 +2012,6 @@ static void update_scaler(struct pipe_ctx *pipe_ctx)
bool per_pixel_alpha =
pipe_ctx->plane_state->per_pixel_alpha && pipe_ctx->bottom_pipe;
- /* TODO: proper fix once fpga works */
-
pipe_ctx->plane_res.scl_data.lb_params.alpha_en = per_pixel_alpha;
pipe_ctx->plane_res.scl_data.lb_params.depth = LB_PIXEL_DEPTH_30BPP;
/* scaler configuration */
@@ -2035,7 +2019,7 @@ static void update_scaler(struct pipe_ctx *pipe_ctx)
pipe_ctx->plane_res.dpp, &pipe_ctx->plane_res.scl_data);
}
-static void update_dchubp_dpp(
+void update_dchubp_dpp(
struct dc *dc,
struct pipe_ctx *pipe_ctx,
struct dc_state *context)
@@ -2052,16 +2036,22 @@ static void update_dchubp_dpp(
*/
if (plane_state->update_flags.bits.full_update) {
bool should_divided_by_2 = context->bw.dcn.clk.dppclk_khz <=
- dc->res_pool->dccg->clks.dispclk_khz / 2;
+ dc->res_pool->clk_mgr->clks.dispclk_khz / 2;
dpp->funcs->dpp_dppclk_control(
dpp,
should_divided_by_2,
true);
- dc->res_pool->dccg->clks.dppclk_khz = should_divided_by_2 ?
- dc->res_pool->dccg->clks.dispclk_khz / 2 :
- dc->res_pool->dccg->clks.dispclk_khz;
+ if (dc->res_pool->dccg)
+ dc->res_pool->dccg->funcs->update_dpp_dto(
+ dc->res_pool->dccg,
+ dpp->inst,
+ pipe_ctx->plane_res.bw.calc.dppclk_khz);
+ else
+ dc->res_pool->clk_mgr->clks.dppclk_khz = should_divided_by_2 ?
+ dc->res_pool->clk_mgr->clks.dispclk_khz / 2 :
+ dc->res_pool->clk_mgr->clks.dispclk_khz;
}
/* TODO: Need input parameter to tell current DCHUB pipe tie to which OTG
@@ -2182,7 +2172,7 @@ static void dcn10_blank_pixel_data(
}
}
-static void set_hdr_multiplier(struct pipe_ctx *pipe_ctx)
+void set_hdr_multiplier(struct pipe_ctx *pipe_ctx)
{
struct fixed31_32 multiplier = dc_fixpt_from_fraction(
pipe_ctx->plane_state->sdr_white_level, 80);
@@ -2257,47 +2247,7 @@ static void program_all_pipe_in_tree(
}
}
-static void dcn10_pplib_apply_display_requirements(
- struct dc *dc,
- struct dc_state *context)
-{
- struct dm_pp_display_configuration *pp_display_cfg = &context->pp_display_cfg;
-
- pp_display_cfg->min_engine_clock_khz = dc->res_pool->dccg->clks.dcfclk_khz;
- pp_display_cfg->min_memory_clock_khz = dc->res_pool->dccg->clks.fclk_khz;
- pp_display_cfg->min_engine_clock_deep_sleep_khz = dc->res_pool->dccg->clks.dcfclk_deep_sleep_khz;
- pp_display_cfg->min_dcfc_deep_sleep_clock_khz = dc->res_pool->dccg->clks.dcfclk_deep_sleep_khz;
- pp_display_cfg->min_dcfclock_khz = dc->res_pool->dccg->clks.dcfclk_khz;
- pp_display_cfg->disp_clk_khz = dc->res_pool->dccg->clks.dispclk_khz;
- dce110_fill_display_configs(context, pp_display_cfg);
-
- if (memcmp(&dc->prev_display_config, pp_display_cfg, sizeof(
- struct dm_pp_display_configuration)) != 0)
- dm_pp_apply_display_requirements(dc->ctx, pp_display_cfg);
-
- dc->prev_display_config = *pp_display_cfg;
-}
-
-static void optimize_shared_resources(struct dc *dc)
-{
- if (dc->current_state->stream_count == 0) {
- /* S0i2 message */
- dcn10_pplib_apply_display_requirements(dc, dc->current_state);
- }
-
- if (dc->debug.pplib_wm_report_mode == WM_REPORT_OVERRIDE)
- dcn_bw_notify_pplib_of_wm_ranges(dc);
-}
-
-static void ready_shared_resources(struct dc *dc, struct dc_state *context)
-{
- /* S0i2 message */
- if (dc->current_state->stream_count == 0 &&
- context->stream_count != 0)
- dcn10_pplib_apply_display_requirements(dc, context);
-}
-
-static struct pipe_ctx *find_top_pipe_for_stream(
+struct pipe_ctx *find_top_pipe_for_stream(
struct dc *dc,
struct dc_state *context,
const struct dc_stream_state *stream)
@@ -2398,10 +2348,9 @@ static void dcn10_apply_ctx_for_surface(
hubbub1_wm_change_req_wa(dc->res_pool->hubbub);
}
-static void dcn10_set_bandwidth(
+static void dcn10_prepare_bandwidth(
struct dc *dc,
- struct dc_state *context,
- bool safe_to_lower)
+ struct dc_state *context)
{
if (dc->debug.sanity_checks)
dcn10_verify_allow_pstate_change_high(dc);
@@ -2410,12 +2359,10 @@ static void dcn10_set_bandwidth(
if (context->stream_count == 0)
context->bw.dcn.clk.phyclk_khz = 0;
- dc->res_pool->dccg->funcs->update_clocks(
- dc->res_pool->dccg,
- &context->bw.dcn.clk,
- safe_to_lower);
-
- dcn10_pplib_apply_display_requirements(dc, context);
+ dc->res_pool->clk_mgr->funcs->update_clocks(
+ dc->res_pool->clk_mgr,
+ context,
+ false);
}
hubbub1_program_watermarks(dc->res_pool->hubbub,
@@ -2423,6 +2370,38 @@ static void dcn10_set_bandwidth(
dc->res_pool->ref_clock_inKhz / 1000,
true);
+ if (dc->debug.pplib_wm_report_mode == WM_REPORT_OVERRIDE)
+ dcn_bw_notify_pplib_of_wm_ranges(dc);
+
+ if (dc->debug.sanity_checks)
+ dcn10_verify_allow_pstate_change_high(dc);
+}
+
+static void dcn10_optimize_bandwidth(
+ struct dc *dc,
+ struct dc_state *context)
+{
+ if (dc->debug.sanity_checks)
+ dcn10_verify_allow_pstate_change_high(dc);
+
+ if (!IS_FPGA_MAXIMUS_DC(dc->ctx->dce_environment)) {
+ if (context->stream_count == 0)
+ context->bw.dcn.clk.phyclk_khz = 0;
+
+ dc->res_pool->clk_mgr->funcs->update_clocks(
+ dc->res_pool->clk_mgr,
+ context,
+ true);
+ }
+
+ hubbub1_program_watermarks(dc->res_pool->hubbub,
+ &context->bw.dcn.watermarks,
+ dc->res_pool->ref_clock_inKhz / 1000,
+ true);
+
+ if (dc->debug.pplib_wm_report_mode == WM_REPORT_OVERRIDE)
+ dcn_bw_notify_pplib_of_wm_ranges(dc);
+
if (dc->debug.sanity_checks)
dcn10_verify_allow_pstate_change_high(dc);
}
@@ -2694,7 +2673,6 @@ static void dcn10_set_cursor_sdr_white_level(struct pipe_ctx *pipe_ctx)
static const struct hw_sequencer_funcs dcn10_funcs = {
.program_gamut_remap = program_gamut_remap,
- .program_csc_matrix = program_csc_matrix,
.init_hw = dcn10_init_hw,
.apply_ctx_to_hw = dce110_apply_ctx_to_hw,
.apply_ctx_for_surface = dcn10_apply_ctx_for_surface,
@@ -2721,7 +2699,8 @@ static const struct hw_sequencer_funcs dcn10_funcs = {
.disable_plane = dcn10_disable_plane,
.blank_pixel_data = dcn10_blank_pixel_data,
.pipe_control_lock = dcn10_pipe_control_lock,
- .set_bandwidth = dcn10_set_bandwidth,
+ .prepare_bandwidth = dcn10_prepare_bandwidth,
+ .optimize_bandwidth = dcn10_optimize_bandwidth,
.reset_hw_ctx_wrap = reset_hw_ctx_wrap,
.enable_stream_timing = dcn10_enable_stream_timing,
.set_drr = set_drr,
@@ -2732,10 +2711,6 @@ static const struct hw_sequencer_funcs dcn10_funcs = {
.log_hw_state = dcn10_log_hw_state,
.get_hw_state = dcn10_get_hw_state,
.wait_for_mpcc_disconnect = dcn10_wait_for_mpcc_disconnect,
- .ready_shared_resources = ready_shared_resources,
- .optimize_shared_resources = optimize_shared_resources,
- .pplib_apply_display_requirements =
- dcn10_pplib_apply_display_requirements,
.edp_backlight_control = hwss_edp_backlight_control,
.edp_power_control = hwss_edp_power_control,
.edp_wait_for_hpd_ready = hwss_edp_wait_for_hpd_ready,
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
index 84d461e..5e5610c 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.h
@@ -51,4 +51,24 @@ void dcn10_get_hw_state(
char *pBuf, unsigned int bufSize,
unsigned int mask);
+bool is_lower_pipe_tree_visible(struct pipe_ctx *pipe_ctx);
+
+bool is_upper_pipe_tree_visible(struct pipe_ctx *pipe_ctx);
+
+bool is_pipe_tree_visible(struct pipe_ctx *pipe_ctx);
+
+void dcn10_program_pte_vm(struct dce_hwseq *hws, struct hubp *hubp);
+
+void set_hdr_multiplier(struct pipe_ctx *pipe_ctx);
+
+void update_dchubp_dpp(
+ struct dc *dc,
+ struct pipe_ctx *pipe_ctx,
+ struct dc_state *context);
+
+struct pipe_ctx *find_top_pipe_for_stream(
+ struct dc *dc,
+ struct dc_state *context,
+ const struct dc_stream_state *stream);
+
#endif /* __DC_HWSS_DCN10_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
index ba6a8686..477ab92 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c
@@ -589,7 +589,7 @@ static bool dcn10_link_encoder_validate_hdmi_output(
return false;
/* DCE11 HW does not support 420 */
- if (!enc10->base.features.ycbcr420_supported &&
+ if (!enc10->base.features.hdmi_ycbcr420_supported &&
crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
return false;
@@ -606,8 +606,10 @@ bool dcn10_link_encoder_validate_dp_output(
const struct dcn10_link_encoder *enc10,
const struct dc_crtc_timing *crtc_timing)
{
- if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420)
- return false;
+ if (crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) {
+ if (!enc10->base.features.dp_ycbcr420_supported)
+ return false;
+ }
return true;
}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
index 5462668..7d1f667 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c
@@ -87,9 +87,8 @@ static void optc1_disable_stereo(struct timing_generator *optc)
REG_SET(OTG_STEREO_CONTROL, 0,
OTG_STEREO_EN, 0);
- REG_SET_3(OTG_3D_STRUCTURE_CONTROL, 0,
+ REG_SET_2(OTG_3D_STRUCTURE_CONTROL, 0,
OTG_3D_STRUCTURE_EN, 0,
- OTG_3D_STRUCTURE_V_UPDATE_MODE, 0,
OTG_3D_STRUCTURE_STEREO_SEL_OVR, 0);
}
@@ -274,10 +273,12 @@ void optc1_program_timing(
* program the reg for interrupt postition.
*/
vertical_line_start = asic_blank_end - optc->dlg_otg_param.vstartup_start + 1;
- if (vertical_line_start < 0) {
- ASSERT(0);
+ v_fp2 = 0;
+ if (vertical_line_start < 0)
+ v_fp2 = -vertical_line_start;
+ if (vertical_line_start < 0)
vertical_line_start = 0;
- }
+
REG_SET(OTG_VERTICAL_INTERRUPT2_POSITION, 0,
OTG_VERTICAL_INTERRUPT2_LINE_START, vertical_line_start);
@@ -296,9 +297,6 @@ void optc1_program_timing(
if (patched_crtc_timing.flags.INTERLACE == 1)
field_num = 1;
}
- v_fp2 = 0;
- if (optc->dlg_otg_param.vstartup_start > asic_blank_end)
- v_fp2 = optc->dlg_otg_param.vstartup_start > asic_blank_end;
/* Interlace */
if (patched_crtc_timing.flags.INTERLACE == 1) {
@@ -1155,9 +1153,8 @@ static void optc1_enable_stereo(struct timing_generator *optc,
OTG_DISABLE_STEREOSYNC_OUTPUT_FOR_DP, 1);
if (flags->PROGRAM_STEREO)
- REG_UPDATE_3(OTG_3D_STRUCTURE_CONTROL,
+ REG_UPDATE_2(OTG_3D_STRUCTURE_CONTROL,
OTG_3D_STRUCTURE_EN, flags->FRAME_PACKED,
- OTG_3D_STRUCTURE_V_UPDATE_MODE, flags->FRAME_PACKED,
OTG_3D_STRUCTURE_STEREO_SEL_OVR, flags->FRAME_PACKED);
}
diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
index a71453a..47dbe4b 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c
@@ -28,23 +28,23 @@
#include "resource.h"
#include "include/irq_service_interface.h"
-#include "dcn10/dcn10_resource.h"
+#include "dcn10_resource.h"
-#include "dcn10/dcn10_ipp.h"
-#include "dcn10/dcn10_mpc.h"
+#include "dcn10_ipp.h"
+#include "dcn10_mpc.h"
#include "irq/dcn10/irq_service_dcn10.h"
-#include "dcn10/dcn10_dpp.h"
+#include "dcn10_dpp.h"
#include "dcn10_optc.h"
-#include "dcn10/dcn10_hw_sequencer.h"
+#include "dcn10_hw_sequencer.h"
#include "dce110/dce110_hw_sequencer.h"
-#include "dcn10/dcn10_opp.h"
-#include "dcn10/dcn10_link_encoder.h"
-#include "dcn10/dcn10_stream_encoder.h"
-#include "dce/dce_clocks.h"
+#include "dcn10_opp.h"
+#include "dcn10_link_encoder.h"
+#include "dcn10_stream_encoder.h"
+#include "dcn10_clk_mgr.h"
#include "dce/dce_clock_source.h"
#include "dce/dce_audio.h"
#include "dce/dce_hwseq.h"
-#include "../virtual/virtual_stream_encoder.h"
+#include "virtual/virtual_stream_encoder.h"
#include "dce110/dce110_resource.h"
#include "dce112/dce112_resource.h"
#include "dcn10_hubp.h"
@@ -438,6 +438,7 @@ static const struct dcn_optc_mask tg_mask = {
static const struct bios_registers bios_regs = {
+ NBIO_SR(BIOS_SCRATCH_0),
NBIO_SR(BIOS_SCRATCH_3),
NBIO_SR(BIOS_SCRATCH_6)
};
@@ -719,7 +720,8 @@ static struct timing_generator *dcn10_timing_generator_create(
static const struct encoder_feature_support link_enc_feature = {
.max_hdmi_deep_color = COLOR_DEPTH_121212,
.max_hdmi_pixel_clock = 600000,
- .ycbcr420_supported = true,
+ .hdmi_ycbcr420_supported = true,
+ .dp_ycbcr420_supported = false,
.flags.bits.IS_HBR2_CAPABLE = true,
.flags.bits.IS_HBR3_CAPABLE = true,
.flags.bits.IS_TPS3_CAPABLE = true,
@@ -949,8 +951,8 @@ static void destruct(struct dcn10_resource_pool *pool)
if (pool->base.dmcu != NULL)
dce_dmcu_destroy(&pool->base.dmcu);
- if (pool->base.dccg != NULL)
- dce_dccg_destroy(&pool->base.dccg);
+ if (pool->base.clk_mgr != NULL)
+ dce_clk_mgr_destroy(&pool->base.clk_mgr);
kfree(pool->base.pp_smu);
}
@@ -1276,8 +1278,8 @@ static bool construct(
}
}
- pool->base.dccg = dcn1_dccg_create(ctx);
- if (pool->base.dccg == NULL) {
+ pool->base.clk_mgr = dcn1_clk_mgr_create(ctx);
+ if (pool->base.clk_mgr == NULL) {
dm_error("DC: failed to create display clock!\n");
BREAK_TO_DEBUGGER();
goto fail;
diff --git a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
index f2ea845..beb08fd 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_pp_smu.h
@@ -55,10 +55,10 @@ struct pp_smu {
struct pp_smu_wm_set_range {
unsigned int wm_inst;
- uint32_t min_fill_clk_khz;
- uint32_t max_fill_clk_khz;
- uint32_t min_drain_clk_khz;
- uint32_t max_drain_clk_khz;
+ uint32_t min_fill_clk_mhz;
+ uint32_t max_fill_clk_mhz;
+ uint32_t min_drain_clk_mhz;
+ uint32_t max_drain_clk_mhz;
};
#define MAX_WATERMARK_SETS 4
@@ -77,15 +77,15 @@ struct pp_smu_display_requirement_rv {
*/
unsigned int display_count;
- /* PPSMC_MSG_SetHardMinFclkByFreq: khz
+ /* PPSMC_MSG_SetHardMinFclkByFreq: mhz
* FCLK will vary with DPM, but never below requested hard min
*/
- unsigned int hard_min_fclk_khz;
+ unsigned int hard_min_fclk_mhz;
- /* PPSMC_MSG_SetHardMinDcefclkByFreq: khz
+ /* PPSMC_MSG_SetHardMinDcefclkByFreq: mhz
* fixed clock at requested freq, either from FCH bypass or DFS
*/
- unsigned int hard_min_dcefclk_khz;
+ unsigned int hard_min_dcefclk_mhz;
/* PPSMC_MSG_SetMinDeepSleepDcefclk: mhz
* when DF is in cstate, dcf clock is further divided down
@@ -103,13 +103,19 @@ struct pp_smu_funcs_rv {
void (*set_display_count)(struct pp_smu *pp, int count);
/* which SMU message? are reader and writer WM separate SMU msg? */
+ /*
+ * PPSMC_MSG_SetDriverDramAddrHigh
+ * PPSMC_MSG_SetDriverDramAddrLow
+ * PPSMC_MSG_TransferTableDram2Smu
+ *
+ * */
void (*set_wm_ranges)(struct pp_smu *pp,
struct pp_smu_wm_range_sets *ranges);
/* PPSMC_MSG_SetHardMinDcfclkByFreq
* fixed clock at requested freq, either from FCH bypass or DFS
*/
- void (*set_hard_min_dcfclk_by_freq)(struct pp_smu *pp, int khz);
+ void (*set_hard_min_dcfclk_by_freq)(struct pp_smu *pp, int mhz);
/* PPSMC_MSG_SetMinDeepSleepDcfclk
* when DF is in cstate, dcf clock is further divided down
@@ -120,12 +126,12 @@ struct pp_smu_funcs_rv {
/* PPSMC_MSG_SetHardMinFclkByFreq
* FCLK will vary with DPM, but never below requested hard min
*/
- void (*set_hard_min_fclk_by_freq)(struct pp_smu *pp, int khz);
+ void (*set_hard_min_fclk_by_freq)(struct pp_smu *pp, int mhz);
/* PPSMC_MSG_SetHardMinSocclkByFreq
* Needed for DWB support
*/
- void (*set_hard_min_socclk_by_freq)(struct pp_smu *pp, int khz);
+ void (*set_hard_min_socclk_by_freq)(struct pp_smu *pp, int mhz);
/* PME w/a */
void (*set_pme_wa_enable)(struct pp_smu *pp);
diff --git a/drivers/gpu/drm/amd/display/dc/dm_services_types.h b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
index 2b83f92..1af8c77 100644
--- a/drivers/gpu/drm/amd/display/dc/dm_services_types.h
+++ b/drivers/gpu/drm/amd/display/dc/dm_services_types.h
@@ -208,22 +208,20 @@ struct dm_bl_data_point {
/* Brightness level as effective value in range 0-255,
* corresponding to above percentage
*/
- uint8_t signalLevel;
+ uint8_t signal_level;
};
/* Total size of the structure should not exceed 256 bytes */
struct dm_acpi_atif_backlight_caps {
-
-
uint16_t size; /* Bytes 0-1 (2 bytes) */
uint16_t flags; /* Byted 2-3 (2 bytes) */
- uint8_t errorCode; /* Byte 4 */
- uint8_t acLevelPercentage; /* Byte 5 */
- uint8_t dcLevelPercentage; /* Byte 6 */
- uint8_t minInputSignal; /* Byte 7 */
- uint8_t maxInputSignal; /* Byte 8 */
- uint8_t numOfDataPoints; /* Byte 9 */
- struct dm_bl_data_point dataPoints[99]; /* Bytes 10-207 (198 bytes)*/
+ uint8_t error_code; /* Byte 4 */
+ uint8_t ac_level_percentage; /* Byte 5 */
+ uint8_t dc_level_percentage; /* Byte 6 */
+ uint8_t min_input_signal; /* Byte 7 */
+ uint8_t max_input_signal; /* Byte 8 */
+ uint8_t num_data_points; /* Byte 9 */
+ struct dm_bl_data_point data_points[99]; /* Bytes 10-207 (198 bytes)*/
};
enum dm_acpi_display_type {
diff --git a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
index cbafce6..5dd0452 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/display_mode_structs.h
@@ -113,7 +113,8 @@ struct _vcs_dpi_soc_bounding_box_st {
int use_urgent_burst_bw;
double max_hscl_ratio;
double max_vscl_ratio;
- struct _vcs_dpi_voltage_scaling_st clock_limits[7];
+ unsigned int num_states;
+ struct _vcs_dpi_voltage_scaling_st clock_limits[8];
};
struct _vcs_dpi_ip_params_st {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/bw_fixed.h b/drivers/gpu/drm/amd/display/dc/inc/bw_fixed.h
index 39ee8eba3..d1656c9 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/bw_fixed.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/bw_fixed.h
@@ -126,7 +126,7 @@ static inline struct bw_fixed bw_div(const struct bw_fixed arg1, const struct bw
static inline struct bw_fixed bw_mod(const struct bw_fixed arg1, const struct bw_fixed arg2)
{
struct bw_fixed res;
- div64_u64_rem(arg1.value, arg2.value, &res.value);
+ div64_u64_rem(arg1.value, arg2.value, (uint64_t *)&res.value);
return res;
}
diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
index c1976c1..e3ee96a 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h
@@ -82,7 +82,7 @@ void core_link_disable_stream(struct pipe_ctx *pipe_ctx, int option);
void core_link_set_avmute(struct pipe_ctx *pipe_ctx, bool enable);
/********** DAL Core*********************/
-#include "display_clock.h"
+#include "hw/clk_mgr.h"
#include "transform.h"
#include "dpp.h"
@@ -169,6 +169,7 @@ struct resource_pool {
unsigned int audio_count;
struct audio_support audio_support;
+ struct clk_mgr *clk_mgr;
struct dccg *dccg;
struct irq_service *irqs;
@@ -287,7 +288,7 @@ struct dc_state {
struct dcn_bw_internal_vars dcn_bw_vars;
#endif
- struct dccg *dis_clk;
+ struct clk_mgr *dccg;
struct kref refcount;
};
diff --git a/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h b/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h
index e688eb9..ece954a 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/dcn_calcs.h
@@ -31,8 +31,8 @@
#define __DCN_CALCS_H__
#include "bw_fixed.h"
-#include "display_clock.h"
#include "../dml/display_mode_lib.h"
+#include "hw/clk_mgr.h"
struct dc;
struct dc_state;
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
index a83a484..abc961c 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/abm.h
@@ -47,12 +47,18 @@ struct abm_funcs {
bool (*set_abm_level)(struct abm *abm, unsigned int abm_level);
bool (*set_abm_immediate_disable)(struct abm *abm);
bool (*init_backlight)(struct abm *abm);
- bool (*set_backlight_level)(struct abm *abm,
- unsigned int backlight_level,
+
+ /* backlight_pwm_u16_16 is unsigned 32 bit,
+ * 16 bit integer + 16 fractional, where 1.0 is max backlight value.
+ */
+ bool (*set_backlight_level_pwm)(struct abm *abm,
+ unsigned int backlight_pwm_u16_16,
unsigned int frame_ramp,
unsigned int controller_id,
bool use_smooth_brightness);
- unsigned int (*get_current_backlight_8_bit)(struct abm *abm);
+
+ unsigned int (*get_current_backlight)(struct abm *abm);
+ unsigned int (*get_target_backlight)(struct abm *abm);
};
#endif
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/display_clock.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
similarity index 63%
rename from drivers/gpu/drm/amd/display/dc/inc/hw/display_clock.h
rename to drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
index 689faa1..23a4b18 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/display_clock.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h
@@ -23,41 +23,25 @@
*
*/
-#ifndef __DISPLAY_CLOCK_H__
-#define __DISPLAY_CLOCK_H__
+#ifndef __DAL_CLK_MGR_H__
+#define __DAL_CLK_MGR_H__
#include "dm_services_types.h"
#include "dc.h"
-/* Structure containing all state-dependent clocks
- * (dependent on "enum clocks_state") */
-struct state_dependent_clocks {
- int display_clk_khz;
- int pixel_clk_khz;
-};
-
-struct dccg {
+struct clk_mgr {
struct dc_context *ctx;
- const struct display_clock_funcs *funcs;
+ const struct clk_mgr_funcs *funcs;
- enum dm_pp_clocks_state max_clks_state;
- enum dm_pp_clocks_state cur_min_clks_state;
struct dc_clocks clks;
};
-struct display_clock_funcs {
- void (*update_clocks)(struct dccg *dccg,
- struct dc_clocks *new_clocks,
+struct clk_mgr_funcs {
+ void (*update_clocks)(struct clk_mgr *clk_mgr,
+ struct dc_state *context,
bool safe_to_lower);
- int (*set_dispclk)(struct dccg *dccg,
- int requested_clock_khz);
- int (*get_dp_ref_clk_frequency)(struct dccg *dccg);
-
- bool (*update_dfs_bypass)(struct dccg *dccg,
- struct dc *dc,
- struct dc_state *context,
- int requested_clock_khz);
+ int (*get_dp_ref_clk_frequency)(struct clk_mgr *clk_mgr);
};
-#endif /* __DISPLAY_CLOCK_H__ */
+#endif /* __DAL_CLK_MGR_H__ */
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
new file mode 100644
index 0000000..95a56d0
--- /dev/null
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/dccg.h
@@ -0,0 +1,44 @@
+/*
+ * Copyright 2018 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: AMD
+ *
+ */
+
+#ifndef __DAL_DCCG_H__
+#define __DAL_DCCG_H__
+
+#include "dc_types.h"
+
+struct dccg {
+ struct dc_context *ctx;
+ const struct dccg_funcs *funcs;
+
+ int ref_dppclk;
+};
+
+struct dccg_funcs {
+ void (*update_dpp_dto)(struct dccg *dccg,
+ int dpp_inst,
+ int req_dppclk);
+};
+
+#endif //__DAL_DCCG_H__
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h
index cf7433e..da85537 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hw_shared.h
@@ -53,6 +53,12 @@ struct curve_points {
uint32_t custom_float_slope;
};
+struct curve_points3 {
+ struct curve_points red;
+ struct curve_points green;
+ struct curve_points blue;
+};
+
struct pwl_result_data {
struct fixed31_32 red;
struct fixed31_32 green;
@@ -71,9 +77,17 @@ struct pwl_result_data {
uint32_t delta_blue_reg;
};
+/* arr_curve_points - regamma regions/segments specification
+ * arr_points - beginning and end point specified separately (only one on DCE)
+ * corner_points - beginning and end point for all 3 colors (DCN)
+ * rgb_resulted - final curve
+ */
struct pwl_params {
struct gamma_curve arr_curve_points[34];
- struct curve_points arr_points[2];
+ union {
+ struct curve_points arr_points[2];
+ struct curve_points3 corner_points[2];
+ };
struct pwl_result_data rgb_resulted[256 + 3];
uint32_t hw_points_num;
};
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
index e28e977..c20fdca 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/link_encoder.h
@@ -65,7 +65,8 @@ struct encoder_feature_support {
enum dc_color_depth max_hdmi_deep_color;
unsigned int max_hdmi_pixel_clock;
- bool ycbcr420_supported;
+ bool hdmi_ycbcr420_supported;
+ bool dp_ycbcr420_supported;
};
union dpcd_psr_configuration {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h b/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h
index da89c2e..06df02d 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw/mem_input.h
@@ -31,7 +31,7 @@
#include "dml/display_mode_structs.h"
struct dchub_init_data;
-struct cstate_pstate_watermarks_st {
+struct cstate_pstate_watermarks_st1 {
uint32_t cstate_exit_ns;
uint32_t cstate_enter_plus_exit_ns;
uint32_t pstate_change_ns;
@@ -40,7 +40,7 @@ struct cstate_pstate_watermarks_st {
struct dcn_watermarks {
uint32_t pte_meta_urgent_ns;
uint32_t urgent_ns;
- struct cstate_pstate_watermarks_st cstate_pstate;
+ struct cstate_pstate_watermarks_st1 cstate_pstate;
};
struct dcn_watermark_set {
diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
index 26f29d5..e9b702c 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h
@@ -32,8 +32,6 @@
#include "inc/hw/link_encoder.h"
#include "core_status.h"
-#define EDP_BACKLIGHT_RAMP_DISABLE_LEVEL 0xFFFFFFFF
-
enum pipe_gating_control {
PIPE_GATING_CONTROL_DISABLE = 0,
PIPE_GATING_CONTROL_ENABLE,
@@ -87,11 +85,6 @@ struct hw_sequencer_funcs {
void (*program_gamut_remap)(
struct pipe_ctx *pipe_ctx);
- void (*program_csc_matrix)(
- struct pipe_ctx *pipe_ctx,
- enum dc_color_space colorspace,
- uint16_t *matrix);
-
void (*program_output_csc)(struct dc *dc,
struct pipe_ctx *pipe_ctx,
enum dc_color_space colorspace,
@@ -177,10 +170,12 @@ struct hw_sequencer_funcs {
struct pipe_ctx *pipe_ctx,
bool blank);
- void (*set_bandwidth)(
+ void (*prepare_bandwidth)(
struct dc *dc,
- struct dc_state *context,
- bool safe_to_lower);
+ struct dc_state *context);
+ void (*optimize_bandwidth)(
+ struct dc *dc,
+ struct dc_state *context);
void (*set_drr)(struct pipe_ctx **pipe_ctx, int num_pipes,
int vmin, int vmax);
@@ -210,11 +205,6 @@ struct hw_sequencer_funcs {
struct resource_pool *res_pool,
struct pipe_ctx *pipe_ctx);
- void (*ready_shared_resources)(struct dc *dc, struct dc_state *context);
- void (*optimize_shared_resources)(struct dc *dc);
- void (*pplib_apply_display_requirements)(
- struct dc *dc,
- struct dc_state *context);
void (*edp_power_control)(
struct dc_link *link,
bool enable);
diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h
index 33b99e3..0086a2f 100644
--- a/drivers/gpu/drm/amd/display/dc/inc/resource.h
+++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h
@@ -30,9 +30,6 @@
#include "dal_asic_id.h"
#include "dm_pp_smu.h"
-/* TODO unhardcode, 4 for CZ*/
-#define MEMORY_TYPE_MULTIPLIER 4
-
enum dce_version resource_parse_asic_id(
struct hw_asic_id asic_id);
diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
index cdcefd0..7480f07 100644
--- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
+++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.c
@@ -306,6 +306,18 @@ static struct fixed31_32 translate_from_linear_space(
a1);
}
+static struct fixed31_32 calculate_gamma22(struct fixed31_32 arg)
+{
+ struct fixed31_32 gamma = dc_fixpt_from_fraction(22, 10);
+
+ return translate_from_linear_space(arg,
+ dc_fixpt_zero,
+ dc_fixpt_zero,
+ dc_fixpt_zero,
+ dc_fixpt_zero,
+ gamma);
+}
+
static struct fixed31_32 translate_to_linear_space(
struct fixed31_32 arg,
struct fixed31_32 a0,
@@ -709,6 +721,169 @@ static void build_regamma(struct pwl_float_data_ex *rgb_regamma,
}
}
+static void hermite_spline_eetf(struct fixed31_32 input_x,
+ struct fixed31_32 max_display,
+ struct fixed31_32 min_display,
+ struct fixed31_32 max_content,
+ struct fixed31_32 *out_x)
+{
+ struct fixed31_32 min_lum_pq;
+ struct fixed31_32 max_lum_pq;
+ struct fixed31_32 max_content_pq;
+ struct fixed31_32 ks;
+ struct fixed31_32 E1;
+ struct fixed31_32 E2;
+ struct fixed31_32 E3;
+ struct fixed31_32 t;
+ struct fixed31_32 t2;
+ struct fixed31_32 t3;
+ struct fixed31_32 two;
+ struct fixed31_32 three;
+ struct fixed31_32 temp1;
+ struct fixed31_32 temp2;
+ struct fixed31_32 a = dc_fixpt_from_fraction(15, 10);
+ struct fixed31_32 b = dc_fixpt_from_fraction(5, 10);
+ struct fixed31_32 epsilon = dc_fixpt_from_fraction(1, 1000000); // dc_fixpt_epsilon is a bit too small
+
+ if (dc_fixpt_eq(max_content, dc_fixpt_zero)) {
+ *out_x = dc_fixpt_zero;
+ return;
+ }
+
+ compute_pq(input_x, &E1);
+ compute_pq(dc_fixpt_div(min_display, max_content), &min_lum_pq);
+ compute_pq(dc_fixpt_div(max_display, max_content), &max_lum_pq);
+ compute_pq(dc_fixpt_one, &max_content_pq); // always 1? DAL2 code is weird
+ a = dc_fixpt_div(dc_fixpt_add(dc_fixpt_one, b), max_content_pq); // (1+b)/maxContent
+ ks = dc_fixpt_sub(dc_fixpt_mul(a, max_lum_pq), b); // a * max_lum_pq - b
+
+ if (dc_fixpt_lt(E1, ks))
+ E2 = E1;
+ else if (dc_fixpt_le(ks, E1) && dc_fixpt_le(E1, dc_fixpt_one)) {
+ if (dc_fixpt_lt(epsilon, dc_fixpt_sub(dc_fixpt_one, ks)))
+ // t = (E1 - ks) / (1 - ks)
+ t = dc_fixpt_div(dc_fixpt_sub(E1, ks),
+ dc_fixpt_sub(dc_fixpt_one, ks));
+ else
+ t = dc_fixpt_zero;
+
+ two = dc_fixpt_from_int(2);
+ three = dc_fixpt_from_int(3);
+
+ t2 = dc_fixpt_mul(t, t);
+ t3 = dc_fixpt_mul(t2, t);
+ temp1 = dc_fixpt_mul(two, t3);
+ temp2 = dc_fixpt_mul(three, t2);
+
+ // (2t^3 - 3t^2 + 1) * ks
+ E2 = dc_fixpt_mul(ks, dc_fixpt_add(dc_fixpt_one,
+ dc_fixpt_sub(temp1, temp2)));
+
+ // (-2t^3 + 3t^2) * max_lum_pq
+ E2 = dc_fixpt_add(E2, dc_fixpt_mul(max_lum_pq,
+ dc_fixpt_sub(temp2, temp1)));
+
+ temp1 = dc_fixpt_mul(two, t2);
+ temp2 = dc_fixpt_sub(dc_fixpt_one, ks);
+
+ // (t^3 - 2t^2 + t) * (1-ks)
+ E2 = dc_fixpt_add(E2, dc_fixpt_mul(temp2,
+ dc_fixpt_add(t, dc_fixpt_sub(t3, temp1))));
+ } else
+ E2 = dc_fixpt_one;
+
+ temp1 = dc_fixpt_sub(dc_fixpt_one, E2);
+ temp2 = dc_fixpt_mul(temp1, temp1);
+ temp2 = dc_fixpt_mul(temp2, temp2);
+ // temp2 = (1-E2)^4
+
+ E3 = dc_fixpt_add(E2, dc_fixpt_mul(min_lum_pq, temp2));
+ compute_de_pq(E3, out_x);
+
+ *out_x = dc_fixpt_div(*out_x, dc_fixpt_div(max_display, max_content));
+}
+
+static bool build_freesync_hdr(struct pwl_float_data_ex *rgb_regamma,
+ uint32_t hw_points_num,
+ const struct hw_x_point *coordinate_x,
+ const struct freesync_hdr_tf_params *fs_params)
+{
+ uint32_t i;
+ struct pwl_float_data_ex *rgb = rgb_regamma;
+ const struct hw_x_point *coord_x = coordinate_x;
+ struct fixed31_32 scaledX = dc_fixpt_zero;
+ struct fixed31_32 scaledX1 = dc_fixpt_zero;
+ struct fixed31_32 max_display = dc_fixpt_from_int(fs_params->max_display);
+ struct fixed31_32 min_display = dc_fixpt_from_fraction(fs_params->min_display, 10000);
+ struct fixed31_32 max_content = dc_fixpt_from_int(fs_params->max_content);
+ struct fixed31_32 min_content = dc_fixpt_from_fraction(fs_params->min_content, 10000);
+ struct fixed31_32 clip = dc_fixpt_one;
+ struct fixed31_32 output;
+ bool use_eetf = false;
+ bool is_clipped = false;
+ struct fixed31_32 sdr_white_level = dc_fixpt_from_int(fs_params->sdr_white_level);
+
+ if (fs_params == NULL || fs_params->max_content == 0 ||
+ fs_params->max_display == 0)
+ return false;
+
+ if (fs_params->min_display > 1000) // cap at 0.1 at the bottom
+ min_display = dc_fixpt_from_fraction(1, 10);
+ if (fs_params->max_display < 100) // cap at 100 at the top
+ max_display = dc_fixpt_from_int(100);
+
+ if (fs_params->min_content < fs_params->min_display)
+ use_eetf = true;
+ else
+ min_content = min_display;
+
+ if (fs_params->max_content > fs_params->max_display)
+ use_eetf = true;
+ else
+ max_content = max_display;
+
+ rgb += 32; // first 32 points have problems with fixed point, too small
+ coord_x += 32;
+ for (i = 32; i <= hw_points_num; i++) {
+ if (!is_clipped) {
+ if (use_eetf) {
+ /*max content is equal 1 */
+ scaledX1 = dc_fixpt_div(coord_x->x,
+ dc_fixpt_div(max_content, sdr_white_level));
+ hermite_spline_eetf(scaledX1, max_display, min_display,
+ max_content, &scaledX);
+ } else
+ scaledX = dc_fixpt_div(coord_x->x,
+ dc_fixpt_div(max_display, sdr_white_level));
+
+ if (dc_fixpt_lt(scaledX, clip)) {
+ if (dc_fixpt_lt(scaledX, dc_fixpt_zero))
+ output = dc_fixpt_zero;
+ else
+ output = calculate_gamma22(scaledX);
+
+ rgb->r = output;
+ rgb->g = output;
+ rgb->b = output;
+ } else {
+ is_clipped = true;
+ rgb->r = clip;
+ rgb->g = clip;
+ rgb->b = clip;
+ }
+ } else {
+ rgb->r = clip;
+ rgb->g = clip;
+ rgb->b = clip;
+ }
+
+ ++coord_x;
+ ++rgb;
+ }
+
+ return true;
+}
+
static void build_degamma(struct pwl_float_data_ex *curve,
uint32_t hw_points_num,
const struct hw_x_point *coordinate_x, bool is_2_4)
@@ -1356,7 +1531,8 @@ static bool map_regamma_hw_to_x_user(
#define _EXTRA_POINTS 3
bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf,
- const struct dc_gamma *ramp, bool mapUserRamp, bool canRomBeUsed)
+ const struct dc_gamma *ramp, bool mapUserRamp, bool canRomBeUsed,
+ const struct freesync_hdr_tf_params *fs_params)
{
struct dc_transfer_func_distributed_points *tf_pts = &output_tf->tf_pts;
struct dividers dividers;
@@ -1374,7 +1550,7 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf,
/* we can use hardcoded curve for plain SRGB TF */
if (output_tf->type == TF_TYPE_PREDEFINED && canRomBeUsed == true &&
output_tf->tf == TRANSFER_FUNCTION_SRGB &&
- (!mapUserRamp && ramp->type == GAMMA_RGB_256))
+ (ramp->is_identity || (!mapUserRamp && ramp->type == GAMMA_RGB_256)))
return true;
output_tf->type = TF_TYPE_DISTRIBUTED_POINTS;
@@ -1424,6 +1600,12 @@ bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf,
MAX_HW_POINTS,
coordinates_x,
output_tf->sdr_ref_white_level);
+ } else if (tf == TRANSFER_FUNCTION_GAMMA22 &&
+ fs_params != NULL) {
+ build_freesync_hdr(rgb_regamma,
+ MAX_HW_POINTS,
+ coordinates_x,
+ fs_params);
} else {
tf_pts->end_exponent = 0;
tf_pts->x_point_at_y1_red = 1;
diff --git a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h
index 63ccb9c..a6e164d 100644
--- a/drivers/gpu/drm/amd/display/modules/color/color_gamma.h
+++ b/drivers/gpu/drm/amd/display/modules/color/color_gamma.h
@@ -73,12 +73,21 @@ struct regamma_lut {
};
};
+struct freesync_hdr_tf_params {
+ unsigned int sdr_white_level;
+ unsigned int min_content; // luminance in 1/10000 nits
+ unsigned int max_content; // luminance in nits
+ unsigned int min_display; // luminance in 1/10000 nits
+ unsigned int max_display; // luminance in nits
+};
+
void setup_x_points_distribution(void);
void precompute_pq(void);
void precompute_de_pq(void);
bool mod_color_calculate_regamma_params(struct dc_transfer_func *output_tf,
- const struct dc_gamma *ramp, bool mapUserRamp, bool canRomBeUsed);
+ const struct dc_gamma *ramp, bool mapUserRamp, bool canRomBeUsed,
+ const struct freesync_hdr_tf_params *fs_params);
bool mod_color_calculate_degamma_params(struct dc_transfer_func *output_tf,
const struct dc_gamma *ramp, bool mapUserRamp);
diff --git a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
index 4018c71..620a171 100644
--- a/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
+++ b/drivers/gpu/drm/amd/display/modules/freesync/freesync.c
@@ -37,6 +37,8 @@
#define RENDER_TIMES_MAX_COUNT 10
/* Threshold to exit BTR (to avoid frequent enter-exits at the lower limit) */
#define BTR_EXIT_MARGIN 2000
+/*Threshold to exit fixed refresh rate*/
+#define FIXED_REFRESH_EXIT_MARGIN_IN_HZ 4
/* Number of consecutive frames to check before entering/exiting fixed refresh*/
#define FIXED_REFRESH_ENTER_FRAME_COUNT 5
#define FIXED_REFRESH_EXIT_FRAME_COUNT 5
@@ -257,40 +259,14 @@ static void apply_below_the_range(struct core_freesync *core_freesync,
if (in_out_vrr->btr.btr_active) {
in_out_vrr->btr.frame_counter = 0;
in_out_vrr->btr.btr_active = false;
-
- /* Exit Fixed Refresh mode */
- } else if (in_out_vrr->fixed.fixed_active) {
-
- in_out_vrr->fixed.frame_counter++;
-
- if (in_out_vrr->fixed.frame_counter >
- FIXED_REFRESH_EXIT_FRAME_COUNT) {
- in_out_vrr->fixed.frame_counter = 0;
- in_out_vrr->fixed.fixed_active = false;
- }
}
} else if (last_render_time_in_us > max_render_time_in_us) {
/* Enter Below the Range */
- if (!in_out_vrr->btr.btr_active &&
- in_out_vrr->btr.btr_enabled) {
- in_out_vrr->btr.btr_active = true;
-
- /* Enter Fixed Refresh mode */
- } else if (!in_out_vrr->fixed.fixed_active &&
- !in_out_vrr->btr.btr_enabled) {
- in_out_vrr->fixed.frame_counter++;
-
- if (in_out_vrr->fixed.frame_counter >
- FIXED_REFRESH_ENTER_FRAME_COUNT) {
- in_out_vrr->fixed.frame_counter = 0;
- in_out_vrr->fixed.fixed_active = true;
- }
- }
+ in_out_vrr->btr.btr_active = true;
}
/* BTR set to "not active" so disengage */
if (!in_out_vrr->btr.btr_active) {
- in_out_vrr->btr.btr_active = false;
in_out_vrr->btr.inserted_duration_in_us = 0;
in_out_vrr->btr.frames_to_insert = 0;
in_out_vrr->btr.frame_counter = 0;
@@ -375,7 +351,12 @@ static void apply_fixed_refresh(struct core_freesync *core_freesync,
bool update = false;
unsigned int max_render_time_in_us = in_out_vrr->max_duration_in_us;
- if (last_render_time_in_us + BTR_EXIT_MARGIN < max_render_time_in_us) {
+ //Compute the exit refresh rate and exit frame duration
+ unsigned int exit_refresh_rate_in_milli_hz = ((1000000000/max_render_time_in_us)
+ + (1000*FIXED_REFRESH_EXIT_MARGIN_IN_HZ));
+ unsigned int exit_frame_duration_in_us = 1000000000/exit_refresh_rate_in_milli_hz;
+
+ if (last_render_time_in_us < exit_frame_duration_in_us) {
/* Exit Fixed Refresh mode */
if (in_out_vrr->fixed.fixed_active) {
in_out_vrr->fixed.frame_counter++;
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index 2083c30..470d7b8 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -133,6 +133,10 @@ enum PP_FEATURE_MASK {
PP_AVFS_MASK = 0x40000,
};
+enum DC_FEATURE_MASK {
+ DC_FBC_MASK = 0x1,
+};
+
/**
* struct amd_ip_funcs - general hooks for managing amdgpu IP Blocks
*/
diff --git a/drivers/gpu/drm/amd/include/atomfirmware.h b/drivers/gpu/drm/amd/include/atomfirmware.h
index d2e7c0f..8eb0bb2 100644
--- a/drivers/gpu/drm/amd/include/atomfirmware.h
+++ b/drivers/gpu/drm/amd/include/atomfirmware.h
@@ -1325,7 +1325,7 @@ struct atom_smu_info_v3_3 {
struct atom_common_table_header table_header;
uint8_t smuip_min_ver;
uint8_t smuip_max_ver;
- uint8_t smu_rsd1;
+ uint8_t waflclk_ss_mode;
uint8_t gpuclk_ss_mode;
uint16_t sclk_ss_percentage;
uint16_t sclk_ss_rate_10hz;
@@ -1355,7 +1355,10 @@ struct atom_smu_info_v3_3 {
uint32_t syspll3_1_vco_freq_10khz;
uint32_t bootup_fclk_10khz;
uint32_t bootup_waflclk_10khz;
- uint32_t reserved[3];
+ uint32_t smu_info_caps;
+ uint16_t waflclk_ss_percentage; // in unit of 0.001%
+ uint16_t smuinitoffset;
+ uint32_t reserved;
};
/*
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 64ecffd..58ac0b9 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -205,20 +205,6 @@ struct tile_config {
/**
* struct kfd2kgd_calls
*
- * @init_gtt_mem_allocation: Allocate a buffer on the gart aperture.
- * The buffer can be used for mqds, hpds, kernel queue, fence and runlists
- *
- * @free_gtt_mem: Frees a buffer that was allocated on the gart aperture
- *
- * @get_local_mem_info: Retrieves information about GPU local memory
- *
- * @get_gpu_clock_counter: Retrieves GPU clock counter
- *
- * @get_max_engine_clock_in_mhz: Retrieves maximum GPU clock in MHz
- *
- * @alloc_pasid: Allocate a PASID
- * @free_pasid: Free a PASID
- *
* @program_sh_mem_settings: A function that should initiate the memory
* properties such as main aperture memory type (cache / non cached) and
* secondary aperture base address, size and memory type.
@@ -255,64 +241,16 @@ struct tile_config {
*
* @get_tile_config: Returns GPU-specific tiling mode information
*
- * @get_cu_info: Retrieves activated cu info
- *
- * @get_vram_usage: Returns current VRAM usage
- *
- * @create_process_vm: Create a VM address space for a given process and GPU
- *
- * @destroy_process_vm: Destroy a VM
- *
- * @get_process_page_dir: Get physical address of a VM page directory
- *
* @set_vm_context_page_table_base: Program page table base for a VMID
*
- * @alloc_memory_of_gpu: Allocate GPUVM memory
- *
- * @free_memory_of_gpu: Free GPUVM memory
- *
- * @map_memory_to_gpu: Map GPUVM memory into a specific VM address
- * space. Allocates and updates page tables and page directories as
- * needed. This function may return before all page table updates have
- * completed. This allows multiple map operations (on multiple GPUs)
- * to happen concurrently. Use sync_memory to synchronize with all
- * pending updates.
- *
- * @unmap_memor_to_gpu: Unmap GPUVM memory from a specific VM address space
- *
- * @sync_memory: Wait for pending page table updates to complete
- *
- * @map_gtt_bo_to_kernel: Map a GTT BO for kernel access
- * Pins the BO, maps it to kernel address space. Such BOs are never evicted.
- * The kernel virtual address remains valid until the BO is freed.
- *
- * @restore_process_bos: Restore all BOs that belong to the
- * process. This is intended for restoring memory mappings after a TTM
- * eviction.
- *
* @invalidate_tlbs: Invalidate TLBs for a specific PASID
*
* @invalidate_tlbs_vmid: Invalidate TLBs for a specific VMID
*
- * @submit_ib: Submits an IB to the engine specified by inserting the
- * IB to the corresponding ring (ring type). The IB is executed with the
- * specified VMID in a user mode context.
- *
- * @get_vm_fault_info: Return information about a recent VM fault on
- * GFXv7 and v8. If multiple VM faults occurred since the last call of
- * this function, it will return information about the first of those
- * faults. On GFXv9 VM fault information is fully contained in the IH
- * packet and this function is not needed.
- *
* @read_vmid_from_vmfault_reg: On Hawaii the VMID is not set in the
* IH ring entry. This function allows the KFD ISR to get the VMID
* from the fault status register as early as possible.
*
- * @gpu_recover: let kgd reset gpu after kfd detect CPC hang
- *
- * @set_compute_idle: Indicates that compute is idle on a device. This
- * can be used to change power profiles depending on compute activity.
- *
* @get_hive_id: Returns hive id of current device, 0 if xgmi is not enabled
*
* This structure contains function pointers to services that the kgd driver
@@ -320,21 +258,6 @@ struct tile_config {
*
*/
struct kfd2kgd_calls {
- int (*init_gtt_mem_allocation)(struct kgd_dev *kgd, size_t size,
- void **mem_obj, uint64_t *gpu_addr,
- void **cpu_ptr, bool mqd_gfx9);
-
- void (*free_gtt_mem)(struct kgd_dev *kgd, void *mem_obj);
-
- void (*get_local_mem_info)(struct kgd_dev *kgd,
- struct kfd_local_mem_info *mem_info);
- uint64_t (*get_gpu_clock_counter)(struct kgd_dev *kgd);
-
- uint32_t (*get_max_engine_clock_in_mhz)(struct kgd_dev *kgd);
-
- int (*alloc_pasid)(unsigned int bits);
- void (*free_pasid)(unsigned int pasid);
-
/* Register access functions */
void (*program_sh_mem_settings)(struct kgd_dev *kgd, uint32_t vmid,
uint32_t sh_mem_config, uint32_t sh_mem_ape1_base,
@@ -398,49 +321,11 @@ struct kfd2kgd_calls {
uint64_t va, uint32_t vmid);
int (*get_tile_config)(struct kgd_dev *kgd, struct tile_config *config);
- void (*get_cu_info)(struct kgd_dev *kgd,
- struct kfd_cu_info *cu_info);
- uint64_t (*get_vram_usage)(struct kgd_dev *kgd);
-
- int (*create_process_vm)(struct kgd_dev *kgd, unsigned int pasid, void **vm,
- void **process_info, struct dma_fence **ef);
- int (*acquire_process_vm)(struct kgd_dev *kgd, struct file *filp,
- unsigned int pasid, void **vm, void **process_info,
- struct dma_fence **ef);
- void (*destroy_process_vm)(struct kgd_dev *kgd, void *vm);
- void (*release_process_vm)(struct kgd_dev *kgd, void *vm);
- uint64_t (*get_process_page_dir)(void *vm);
void (*set_vm_context_page_table_base)(struct kgd_dev *kgd,
uint32_t vmid, uint64_t page_table_base);
- int (*alloc_memory_of_gpu)(struct kgd_dev *kgd, uint64_t va,
- uint64_t size, void *vm,
- struct kgd_mem **mem, uint64_t *offset,
- uint32_t flags);
- int (*free_memory_of_gpu)(struct kgd_dev *kgd, struct kgd_mem *mem);
- int (*map_memory_to_gpu)(struct kgd_dev *kgd, struct kgd_mem *mem,
- void *vm);
- int (*unmap_memory_to_gpu)(struct kgd_dev *kgd, struct kgd_mem *mem,
- void *vm);
- int (*sync_memory)(struct kgd_dev *kgd, struct kgd_mem *mem, bool intr);
- int (*map_gtt_bo_to_kernel)(struct kgd_dev *kgd, struct kgd_mem *mem,
- void **kptr, uint64_t *size);
- int (*restore_process_bos)(void *process_info, struct dma_fence **ef);
-
int (*invalidate_tlbs)(struct kgd_dev *kgd, uint16_t pasid);
int (*invalidate_tlbs_vmid)(struct kgd_dev *kgd, uint16_t vmid);
-
- int (*submit_ib)(struct kgd_dev *kgd, enum kgd_engine_type engine,
- uint32_t vmid, uint64_t gpu_addr,
- uint32_t *ib_cmd, uint32_t ib_len);
-
- int (*get_vm_fault_info)(struct kgd_dev *kgd,
- struct kfd_vm_fault_info *info);
uint32_t (*read_vmid_from_vmfault_reg)(struct kgd_dev *kgd);
-
- void (*gpu_recover)(struct kgd_dev *kgd);
-
- void (*set_compute_idle)(struct kgd_dev *kgd, bool idle);
-
uint64_t (*get_hive_id)(struct kgd_dev *kgd);
};
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c
index 5e19f59..d138ddae 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_powertune.c
@@ -967,7 +967,7 @@ int smu7_enable_didt_config(struct pp_hwmgr *hwmgr)
PP_CAP(PHM_PlatformCaps_TDRamping) ||
PP_CAP(PHM_PlatformCaps_TCPRamping)) {
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
mutex_lock(&adev->grbm_idx_mutex);
value = 0;
value2 = cgs_read_register(hwmgr->device, mmGRBM_GFX_INDEX);
@@ -1014,13 +1014,13 @@ int smu7_enable_didt_config(struct pp_hwmgr *hwmgr)
"Failed to enable DPM DIDT.", goto error);
}
mutex_unlock(&adev->grbm_idx_mutex);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
return 0;
error:
mutex_unlock(&adev->grbm_idx_mutex);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return result;
}
@@ -1034,7 +1034,7 @@ int smu7_disable_didt_config(struct pp_hwmgr *hwmgr)
PP_CAP(PHM_PlatformCaps_TDRamping) ||
PP_CAP(PHM_PlatformCaps_TCPRamping)) {
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
result = smu7_enable_didt(hwmgr, false);
PP_ASSERT_WITH_CODE((result == 0),
@@ -1046,12 +1046,12 @@ int smu7_disable_didt_config(struct pp_hwmgr *hwmgr)
PP_ASSERT_WITH_CODE((0 == result),
"Failed to disable DPM DIDT.", goto error);
}
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
}
return 0;
error:
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return result;
}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c
index 99a33c3..101c09b 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/smu_helper.c
@@ -713,20 +713,20 @@ int smu_set_watermarks_for_clocks_ranges(void *wt_table,
for (i = 0; i < wm_with_clock_ranges->num_wm_dmif_sets; i++) {
table->WatermarkRow[1][i].MinClock =
cpu_to_le16((uint16_t)
- (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz) /
- 1000);
+ (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_min_dcfclk_clk_in_khz /
+ 1000));
table->WatermarkRow[1][i].MaxClock =
cpu_to_le16((uint16_t)
- (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz) /
- 1000);
+ (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_max_dcfclk_clk_in_khz /
+ 1000));
table->WatermarkRow[1][i].MinUclk =
cpu_to_le16((uint16_t)
- (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz) /
- 1000);
+ (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_min_mem_clk_in_khz /
+ 1000));
table->WatermarkRow[1][i].MaxUclk =
cpu_to_le16((uint16_t)
- (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz) /
- 1000);
+ (wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_max_mem_clk_in_khz /
+ 1000));
table->WatermarkRow[1][i].WmSetting = (uint8_t)
wm_with_clock_ranges->wm_dmif_clocks_ranges[i].wm_set_id;
}
@@ -734,20 +734,20 @@ int smu_set_watermarks_for_clocks_ranges(void *wt_table,
for (i = 0; i < wm_with_clock_ranges->num_wm_mcif_sets; i++) {
table->WatermarkRow[0][i].MinClock =
cpu_to_le16((uint16_t)
- (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz) /
- 1000);
+ (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_min_socclk_clk_in_khz /
+ 1000));
table->WatermarkRow[0][i].MaxClock =
cpu_to_le16((uint16_t)
- (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz) /
- 1000);
+ (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_max_socclk_clk_in_khz /
+ 1000));
table->WatermarkRow[0][i].MinUclk =
cpu_to_le16((uint16_t)
- (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz) /
- 1000);
+ (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_min_mem_clk_in_khz /
+ 1000));
table->WatermarkRow[0][i].MaxUclk =
cpu_to_le16((uint16_t)
- (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz) /
- 1000);
+ (wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_max_mem_clk_in_khz /
+ 1000));
table->WatermarkRow[0][i].WmSetting = (uint8_t)
wm_with_clock_ranges->wm_mcif_clocks_ranges[i].wm_set_id;
}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c
index 2d88abf..6f26cb2 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega10_powertune.c
@@ -937,7 +937,7 @@ static int vega10_enable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr)
num_se = adev->gfx.config.max_shader_engines;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
mutex_lock(&adev->grbm_idx_mutex);
for (count = 0; count < num_se; count++) {
@@ -962,7 +962,7 @@ static int vega10_enable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr)
vega10_didt_set_mask(hwmgr, true);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return 0;
}
@@ -971,11 +971,11 @@ static int vega10_disable_cac_driving_se_didt_config(struct pp_hwmgr *hwmgr)
{
struct amdgpu_device *adev = hwmgr->adev;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
vega10_didt_set_mask(hwmgr, false);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return 0;
}
@@ -988,7 +988,7 @@ static int vega10_enable_psm_gc_didt_config(struct pp_hwmgr *hwmgr)
num_se = adev->gfx.config.max_shader_engines;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
mutex_lock(&adev->grbm_idx_mutex);
for (count = 0; count < num_se; count++) {
@@ -1007,7 +1007,7 @@ static int vega10_enable_psm_gc_didt_config(struct pp_hwmgr *hwmgr)
vega10_didt_set_mask(hwmgr, true);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
vega10_program_gc_didt_config_registers(hwmgr, GCDiDtDroopCtrlConfig_vega10);
if (PP_CAP(PHM_PlatformCaps_GCEDC))
@@ -1024,11 +1024,11 @@ static int vega10_disable_psm_gc_didt_config(struct pp_hwmgr *hwmgr)
struct amdgpu_device *adev = hwmgr->adev;
uint32_t data;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
vega10_didt_set_mask(hwmgr, false);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
if (PP_CAP(PHM_PlatformCaps_GCEDC)) {
data = 0x00000000;
@@ -1049,7 +1049,7 @@ static int vega10_enable_se_edc_config(struct pp_hwmgr *hwmgr)
num_se = adev->gfx.config.max_shader_engines;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
mutex_lock(&adev->grbm_idx_mutex);
for (count = 0; count < num_se; count++) {
@@ -1070,7 +1070,7 @@ static int vega10_enable_se_edc_config(struct pp_hwmgr *hwmgr)
vega10_didt_set_mask(hwmgr, true);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return 0;
}
@@ -1079,11 +1079,11 @@ static int vega10_disable_se_edc_config(struct pp_hwmgr *hwmgr)
{
struct amdgpu_device *adev = hwmgr->adev;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
vega10_didt_set_mask(hwmgr, false);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return 0;
}
@@ -1097,7 +1097,7 @@ static int vega10_enable_psm_gc_edc_config(struct pp_hwmgr *hwmgr)
num_se = adev->gfx.config.max_shader_engines;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
vega10_program_gc_didt_config_registers(hwmgr, AvfsPSMResetConfig_vega10);
@@ -1118,7 +1118,7 @@ static int vega10_enable_psm_gc_edc_config(struct pp_hwmgr *hwmgr)
vega10_didt_set_mask(hwmgr, true);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
vega10_program_gc_didt_config_registers(hwmgr, PSMGCEDCDroopCtrlConfig_vega10);
@@ -1138,11 +1138,11 @@ static int vega10_disable_psm_gc_edc_config(struct pp_hwmgr *hwmgr)
struct amdgpu_device *adev = hwmgr->adev;
uint32_t data;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
vega10_didt_set_mask(hwmgr, false);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
if (PP_CAP(PHM_PlatformCaps_GCEDC)) {
data = 0x00000000;
@@ -1160,7 +1160,7 @@ static int vega10_enable_se_edc_force_stall_config(struct pp_hwmgr *hwmgr)
struct amdgpu_device *adev = hwmgr->adev;
int result;
- adev->gfx.rlc.funcs->enter_safe_mode(adev);
+ amdgpu_gfx_rlc_enter_safe_mode(adev);
mutex_lock(&adev->grbm_idx_mutex);
WREG32_SOC15(GC, 0, mmGRBM_GFX_INDEX, 0xE0000000);
@@ -1173,7 +1173,7 @@ static int vega10_enable_se_edc_force_stall_config(struct pp_hwmgr *hwmgr)
vega10_didt_set_mask(hwmgr, false);
- adev->gfx.rlc.funcs->exit_safe_mode(adev);
+ amdgpu_gfx_rlc_exit_safe_mode(adev);
return 0;
}
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
index 57143d5..f2daf00 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.c
@@ -120,6 +120,7 @@ static void vega20_set_default_registry_data(struct pp_hwmgr *hwmgr)
data->registry_data.disable_auto_wattman = 1;
data->registry_data.auto_wattman_debug = 0;
data->registry_data.auto_wattman_sample_period = 100;
+ data->registry_data.fclk_gfxclk_ratio = 0x3F6CCCCD;
data->registry_data.auto_wattman_threshold = 50;
data->registry_data.gfxoff_controlled_by_driver = 1;
data->gfxoff_allowed = false;
@@ -829,6 +830,28 @@ static int vega20_enable_all_smu_features(struct pp_hwmgr *hwmgr)
return 0;
}
+static int vega20_notify_smc_display_change(struct pp_hwmgr *hwmgr)
+{
+ struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend);
+
+ if (data->smu_features[GNLD_DPM_UCLK].enabled)
+ return smum_send_msg_to_smc_with_parameter(hwmgr,
+ PPSMC_MSG_SetUclkFastSwitch,
+ 1);
+
+ return 0;
+}
+
+static int vega20_send_clock_ratio(struct pp_hwmgr *hwmgr)
+{
+ struct vega20_hwmgr *data =
+ (struct vega20_hwmgr *)(hwmgr->backend);
+
+ return smum_send_msg_to_smc_with_parameter(hwmgr,
+ PPSMC_MSG_SetFclkGfxClkRatio,
+ data->registry_data.fclk_gfxclk_ratio);
+}
+
static int vega20_disable_all_smu_features(struct pp_hwmgr *hwmgr)
{
struct vega20_hwmgr *data =
@@ -1532,6 +1555,16 @@ static int vega20_enable_dpm_tasks(struct pp_hwmgr *hwmgr)
"[EnableDPMTasks] Failed to enable all smu features!",
return result);
+ result = vega20_notify_smc_display_change(hwmgr);
+ PP_ASSERT_WITH_CODE(!result,
+ "[EnableDPMTasks] Failed to notify smc display change!",
+ return result);
+
+ result = vega20_send_clock_ratio(hwmgr);
+ PP_ASSERT_WITH_CODE(!result,
+ "[EnableDPMTasks] Failed to send clock ratio!",
+ return result);
+
/* Initialize UVD/VCE powergating state */
vega20_init_powergate_state(hwmgr);
@@ -1972,19 +2005,6 @@ static int vega20_read_sensor(struct pp_hwmgr *hwmgr, int idx,
return ret;
}
-static int vega20_notify_smc_display_change(struct pp_hwmgr *hwmgr,
- bool has_disp)
-{
- struct vega20_hwmgr *data = (struct vega20_hwmgr *)(hwmgr->backend);
-
- if (data->smu_features[GNLD_DPM_UCLK].enabled)
- return smum_send_msg_to_smc_with_parameter(hwmgr,
- PPSMC_MSG_SetUclkFastSwitch,
- has_disp ? 1 : 0);
-
- return 0;
-}
-
int vega20_display_clock_voltage_request(struct pp_hwmgr *hwmgr,
struct pp_display_clock_request *clock_req)
{
@@ -2044,13 +2064,6 @@ static int vega20_notify_smc_display_config_after_ps_adjustment(
struct pp_display_clock_request clock_req;
int ret = 0;
- if ((hwmgr->display_config->num_display > 1) &&
- !hwmgr->display_config->multi_monitor_in_sync &&
- !hwmgr->display_config->nb_pstate_switch_disable)
- vega20_notify_smc_display_change(hwmgr, false);
- else
- vega20_notify_smc_display_change(hwmgr, true);
-
min_clocks.dcefClock = hwmgr->display_config->min_dcef_set_clk;
min_clocks.dcefClockInSR = hwmgr->display_config->min_dcef_deep_sleep_set_clk;
min_clocks.memoryClock = hwmgr->display_config->min_mem_set_clock;
@@ -2742,7 +2755,7 @@ static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr,
for (i = 0; i < clocks.num_levels; i++)
size += sprintf(buf + size, "%d: %uMhz %s\n",
i, clocks.data[i].clocks_in_khz / 1000,
- (clocks.data[i].clocks_in_khz == now) ? "*" : "");
+ (clocks.data[i].clocks_in_khz == now * 10) ? "*" : "");
break;
case PP_MCLK:
@@ -2759,7 +2772,7 @@ static int vega20_print_clock_levels(struct pp_hwmgr *hwmgr,
for (i = 0; i < clocks.num_levels; i++)
size += sprintf(buf + size, "%d: %uMhz %s\n",
i, clocks.data[i].clocks_in_khz / 1000,
- (clocks.data[i].clocks_in_khz == now) ? "*" : "");
+ (clocks.data[i].clocks_in_khz == now * 10) ? "*" : "");
break;
case PP_PCIE:
@@ -3441,109 +3454,64 @@ static int vega20_get_thermal_temperature_range(struct pp_hwmgr *hwmgr,
static const struct pp_hwmgr_func vega20_hwmgr_funcs = {
/* init/fini related */
- .backend_init =
- vega20_hwmgr_backend_init,
- .backend_fini =
- vega20_hwmgr_backend_fini,
- .asic_setup =
- vega20_setup_asic_task,
- .power_off_asic =
- vega20_power_off_asic,
- .dynamic_state_management_enable =
- vega20_enable_dpm_tasks,
- .dynamic_state_management_disable =
- vega20_disable_dpm_tasks,
+ .backend_init = vega20_hwmgr_backend_init,
+ .backend_fini = vega20_hwmgr_backend_fini,
+ .asic_setup = vega20_setup_asic_task,
+ .power_off_asic = vega20_power_off_asic,
+ .dynamic_state_management_enable = vega20_enable_dpm_tasks,
+ .dynamic_state_management_disable = vega20_disable_dpm_tasks,
/* power state related */
- .apply_clocks_adjust_rules =
- vega20_apply_clocks_adjust_rules,
- .pre_display_config_changed =
- vega20_pre_display_configuration_changed_task,
- .display_config_changed =
- vega20_display_configuration_changed_task,
+ .apply_clocks_adjust_rules = vega20_apply_clocks_adjust_rules,
+ .pre_display_config_changed = vega20_pre_display_configuration_changed_task,
+ .display_config_changed = vega20_display_configuration_changed_task,
.check_smc_update_required_for_display_configuration =
vega20_check_smc_update_required_for_display_configuration,
.notify_smc_display_config_after_ps_adjustment =
vega20_notify_smc_display_config_after_ps_adjustment,
/* export to DAL */
- .get_sclk =
- vega20_dpm_get_sclk,
- .get_mclk =
- vega20_dpm_get_mclk,
- .get_dal_power_level =
- vega20_get_dal_power_level,
- .get_clock_by_type_with_latency =
- vega20_get_clock_by_type_with_latency,
- .get_clock_by_type_with_voltage =
- vega20_get_clock_by_type_with_voltage,
- .set_watermarks_for_clocks_ranges =
- vega20_set_watermarks_for_clocks_ranges,
- .display_clock_voltage_request =
- vega20_display_clock_voltage_request,
- .get_performance_level =
- vega20_get_performance_level,
+ .get_sclk = vega20_dpm_get_sclk,
+ .get_mclk = vega20_dpm_get_mclk,
+ .get_dal_power_level = vega20_get_dal_power_level,
+ .get_clock_by_type_with_latency = vega20_get_clock_by_type_with_latency,
+ .get_clock_by_type_with_voltage = vega20_get_clock_by_type_with_voltage,
+ .set_watermarks_for_clocks_ranges = vega20_set_watermarks_for_clocks_ranges,
+ .display_clock_voltage_request = vega20_display_clock_voltage_request,
+ .get_performance_level = vega20_get_performance_level,
/* UMD pstate, profile related */
- .force_dpm_level =
- vega20_dpm_force_dpm_level,
- .get_power_profile_mode =
- vega20_get_power_profile_mode,
- .set_power_profile_mode =
- vega20_set_power_profile_mode,
+ .force_dpm_level = vega20_dpm_force_dpm_level,
+ .get_power_profile_mode = vega20_get_power_profile_mode,
+ .set_power_profile_mode = vega20_set_power_profile_mode,
/* od related */
- .set_power_limit =
- vega20_set_power_limit,
- .get_sclk_od =
- vega20_get_sclk_od,
- .set_sclk_od =
- vega20_set_sclk_od,
- .get_mclk_od =
- vega20_get_mclk_od,
- .set_mclk_od =
- vega20_set_mclk_od,
- .odn_edit_dpm_table =
- vega20_odn_edit_dpm_table,
+ .set_power_limit = vega20_set_power_limit,
+ .get_sclk_od = vega20_get_sclk_od,
+ .set_sclk_od = vega20_set_sclk_od,
+ .get_mclk_od = vega20_get_mclk_od,
+ .set_mclk_od = vega20_set_mclk_od,
+ .odn_edit_dpm_table = vega20_odn_edit_dpm_table,
/* for sysfs to retrive/set gfxclk/memclk */
- .force_clock_level =
- vega20_force_clock_level,
- .print_clock_levels =
- vega20_print_clock_levels,
- .read_sensor =
- vega20_read_sensor,
+ .force_clock_level = vega20_force_clock_level,
+ .print_clock_levels = vega20_print_clock_levels,
+ .read_sensor = vega20_read_sensor,
/* powergate related */
- .powergate_uvd =
- vega20_power_gate_uvd,
- .powergate_vce =
- vega20_power_gate_vce,
+ .powergate_uvd = vega20_power_gate_uvd,
+ .powergate_vce = vega20_power_gate_vce,
/* thermal related */
- .start_thermal_controller =
- vega20_start_thermal_controller,
- .stop_thermal_controller =
- vega20_thermal_stop_thermal_controller,
- .get_thermal_temperature_range =
- vega20_get_thermal_temperature_range,
- .register_irq_handlers =
- smu9_register_irq_handlers,
- .disable_smc_firmware_ctf =
- vega20_thermal_disable_alert,
+ .start_thermal_controller = vega20_start_thermal_controller,
+ .stop_thermal_controller = vega20_thermal_stop_thermal_controller,
+ .get_thermal_temperature_range = vega20_get_thermal_temperature_range,
+ .register_irq_handlers = smu9_register_irq_handlers,
+ .disable_smc_firmware_ctf = vega20_thermal_disable_alert,
/* fan control related */
- .get_fan_speed_percent =
- vega20_fan_ctrl_get_fan_speed_percent,
- .set_fan_speed_percent =
- vega20_fan_ctrl_set_fan_speed_percent,
- .get_fan_speed_info =
- vega20_fan_ctrl_get_fan_speed_info,
- .get_fan_speed_rpm =
- vega20_fan_ctrl_get_fan_speed_rpm,
- .set_fan_speed_rpm =
- vega20_fan_ctrl_set_fan_speed_rpm,
- .get_fan_control_mode =
- vega20_get_fan_control_mode,
- .set_fan_control_mode =
- vega20_set_fan_control_mode,
+ .get_fan_speed_percent = vega20_fan_ctrl_get_fan_speed_percent,
+ .set_fan_speed_percent = vega20_fan_ctrl_set_fan_speed_percent,
+ .get_fan_speed_info = vega20_fan_ctrl_get_fan_speed_info,
+ .get_fan_speed_rpm = vega20_fan_ctrl_get_fan_speed_rpm,
+ .set_fan_speed_rpm = vega20_fan_ctrl_set_fan_speed_rpm,
+ .get_fan_control_mode = vega20_get_fan_control_mode,
+ .set_fan_control_mode = vega20_set_fan_control_mode,
/* smu memory related */
- .notify_cac_buffer_info =
- vega20_notify_cac_buffer_info,
- .enable_mgpu_fan_boost =
- vega20_enable_mgpu_fan_boost,
+ .notify_cac_buffer_info = vega20_notify_cac_buffer_info,
+ .enable_mgpu_fan_boost = vega20_enable_mgpu_fan_boost,
};
int vega20_hwmgr_init(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
index 56fe6a0..25faaa5 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/vega20_hwmgr.h
@@ -328,6 +328,7 @@ struct vega20_registry_data {
uint8_t disable_auto_wattman;
uint32_t auto_wattman_debug;
uint32_t auto_wattman_sample_period;
+ uint32_t fclk_gfxclk_ratio;
uint8_t auto_wattman_threshold;
uint8_t log_avfs_param;
uint8_t enable_enginess;
diff --git a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
index e5a60aa..07d180ce 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/hwmgr.h
@@ -28,7 +28,6 @@
#include "hardwaremanager.h"
#include "hwmgr_ppt.h"
#include "ppatomctrl.h"
-#include "hwmgr_ppt.h"
#include "power_state.h"
#include "smu_helper.h"
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smu7_common.h b/drivers/gpu/drm/amd/powerplay/inc/smu7_common.h
index 65eb630..94bf7b6 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smu7_common.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/smu7_common.h
@@ -40,10 +40,6 @@
#include "bif/bif_5_0_d.h"
#include "bif/bif_5_0_sh_mask.h"
-
-#include "bif/bif_5_0_d.h"
-#include "bif/bif_5_0_sh_mask.h"
-
#include "dce/dce_10_0_d.h"
#include "dce/dce_10_0_sh_mask.h"
diff --git a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h b/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
index 45d64a8..4f63a73 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/vega20_ppsmc.h
@@ -105,7 +105,8 @@
#define PPSMC_MSG_SetSystemVirtualDramAddrHigh 0x4B
#define PPSMC_MSG_SetSystemVirtualDramAddrLow 0x4C
#define PPSMC_MSG_WaflTest 0x4D
-// Unused ID 0x4E to 0x50
+#define PPSMC_MSG_SetFclkGfxClkRatio 0x4E
+// Unused ID 0x4F to 0x50
#define PPSMC_MSG_AllowGfxOff 0x51
#define PPSMC_MSG_DisallowGfxOff 0x52
#define PPSMC_MSG_GetPptLimit 0x53
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c
index 872d382..2b2c266 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/polaris10_smumgr.c
@@ -44,7 +44,6 @@
#include "smu7_hwmgr.h"
#include "hardwaremanager.h"
-#include "ppatomctrl.h"
#include "atombios.h"
#include "pppcielanes.h"
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c
index d0eb8ab..d111dd4 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu10_smumgr.c
@@ -29,7 +29,6 @@
#include "rv_ppsmc.h"
#include "smu10_driver_if.h"
#include "smu10.h"
-#include "ppatomctrl.h"
#include "pp_debug.h"
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c
index 09b844e..e2787e1 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/smu8_smumgr.c
@@ -24,6 +24,7 @@
#include <linux/delay.h>
#include <linux/gfp.h>
#include <linux/kernel.h>
+#include <linux/ktime.h>
#include <linux/slab.h>
#include <linux/types.h>
@@ -61,9 +62,13 @@ static uint32_t smu8_get_argument(struct pp_hwmgr *hwmgr)
mmSMU_MP1_SRBM2P_ARG_0);
}
-static int smu8_send_msg_to_smc_async(struct pp_hwmgr *hwmgr, uint16_t msg)
+/* Send a message to the SMC, and wait for its response.*/
+static int smu8_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr,
+ uint16_t msg, uint32_t parameter)
{
int result = 0;
+ ktime_t t_start;
+ s64 elapsed_us;
if (hwmgr == NULL || hwmgr->device == NULL)
return -EINVAL;
@@ -74,28 +79,31 @@ static int smu8_send_msg_to_smc_async(struct pp_hwmgr *hwmgr, uint16_t msg)
/* Read the last message to SMU, to report actual cause */
uint32_t val = cgs_read_register(hwmgr->device,
mmSMU_MP1_SRBM2P_MSG_0);
- pr_err("smu8_send_msg_to_smc_async (0x%04x) failed\n", msg);
- pr_err("SMU still servicing msg (0x%04x)\n", val);
+ pr_err("%s(0x%04x) aborted; SMU still servicing msg (0x%04x)\n",
+ __func__, msg, val);
return result;
}
+ t_start = ktime_get();
+
+ cgs_write_register(hwmgr->device, mmSMU_MP1_SRBM2P_ARG_0, parameter);
cgs_write_register(hwmgr->device, mmSMU_MP1_SRBM2P_RESP_0, 0);
cgs_write_register(hwmgr->device, mmSMU_MP1_SRBM2P_MSG_0, msg);
- return 0;
+ result = PHM_WAIT_FIELD_UNEQUAL(hwmgr,
+ SMU_MP1_SRBM2P_RESP_0, CONTENT, 0);
+
+ elapsed_us = ktime_us_delta(ktime_get(), t_start);
+
+ WARN(result, "%s(0x%04x, %#x) timed out after %lld us\n",
+ __func__, msg, parameter, elapsed_us);
+
+ return result;
}
-/* Send a message to the SMC, and wait for its response.*/
static int smu8_send_msg_to_smc(struct pp_hwmgr *hwmgr, uint16_t msg)
{
- int result = 0;
-
- result = smu8_send_msg_to_smc_async(hwmgr, msg);
- if (result != 0)
- return result;
-
- return PHM_WAIT_FIELD_UNEQUAL(hwmgr,
- SMU_MP1_SRBM2P_RESP_0, CONTENT, 0);
+ return smu8_send_msg_to_smc_with_parameter(hwmgr, msg, 0);
}
static int smu8_set_smc_sram_address(struct pp_hwmgr *hwmgr,
@@ -135,17 +143,6 @@ static int smu8_write_smc_sram_dword(struct pp_hwmgr *hwmgr,
return result;
}
-static int smu8_send_msg_to_smc_with_parameter(struct pp_hwmgr *hwmgr,
- uint16_t msg, uint32_t parameter)
-{
- if (hwmgr == NULL || hwmgr->device == NULL)
- return -EINVAL;
-
- cgs_write_register(hwmgr->device, mmSMU_MP1_SRBM2P_ARG_0, parameter);
-
- return smu8_send_msg_to_smc(hwmgr, msg);
-}
-
static int smu8_check_fw_load_finish(struct pp_hwmgr *hwmgr,
uint32_t firmware)
{
@@ -737,6 +734,10 @@ static int smu8_start_smu(struct pp_hwmgr *hwmgr)
cgs_write_register(hwmgr->device, mmMP0PUB_IND_INDEX, index);
hwmgr->smu_version = cgs_read_register(hwmgr->device, mmMP0PUB_IND_DATA);
+ pr_info("smu version %02d.%02d.%02d\n",
+ ((hwmgr->smu_version >> 16) & 0xFF),
+ ((hwmgr->smu_version >> 8) & 0xFF),
+ (hwmgr->smu_version & 0xFF));
adev->pm.fw_version = hwmgr->smu_version >> 8;
return smu8_request_smu_load_fw(hwmgr);
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c
index 99d5e4f..a6edd5d 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/smumgr.c
@@ -37,10 +37,13 @@ MODULE_FIRMWARE("amdgpu/fiji_smc.bin");
MODULE_FIRMWARE("amdgpu/polaris10_smc.bin");
MODULE_FIRMWARE("amdgpu/polaris10_smc_sk.bin");
MODULE_FIRMWARE("amdgpu/polaris10_k_smc.bin");
+MODULE_FIRMWARE("amdgpu/polaris10_k2_smc.bin");
MODULE_FIRMWARE("amdgpu/polaris11_smc.bin");
MODULE_FIRMWARE("amdgpu/polaris11_smc_sk.bin");
MODULE_FIRMWARE("amdgpu/polaris11_k_smc.bin");
+MODULE_FIRMWARE("amdgpu/polaris11_k2_smc.bin");
MODULE_FIRMWARE("amdgpu/polaris12_smc.bin");
+MODULE_FIRMWARE("amdgpu/polaris12_k_smc.bin");
MODULE_FIRMWARE("amdgpu/vegam_smc.bin");
MODULE_FIRMWARE("amdgpu/vega10_smc.bin");
MODULE_FIRMWARE("amdgpu/vega10_acg_smc.bin");
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c b/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c
index 9f71512..1e69300 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/vegam_smumgr.c
@@ -40,7 +40,6 @@
#include "smu7_hwmgr.h"
#include "hardwaremanager.h"
-#include "ppatomctrl.h"
#include "atombios.h"
#include "pppcielanes.h"
diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h
index e6c4cd3..bfc6504 100644
--- a/drivers/gpu/drm/ast/ast_drv.h
+++ b/drivers/gpu/drm/ast/ast_drv.h
@@ -104,8 +104,6 @@ struct ast_private {
int fb_mtrr;
struct {
- struct drm_global_reference mem_global_ref;
- struct ttm_bo_global_ref bo_global_ref;
struct ttm_bo_device bdev;
} ttm;
diff --git a/drivers/gpu/drm/ast/ast_ttm.c b/drivers/gpu/drm/ast/ast_ttm.c
index fe354eb..c168d62 100644
--- a/drivers/gpu/drm/ast/ast_ttm.c
+++ b/drivers/gpu/drm/ast/ast_ttm.c
@@ -36,63 +36,6 @@ ast_bdev(struct ttm_bo_device *bd)
return container_of(bd, struct ast_private, ttm.bdev);
}
-static int
-ast_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void
-ast_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-static int ast_ttm_global_init(struct ast_private *ast)
-{
- struct drm_global_reference *global_ref;
- int r;
-
- global_ref = &ast->ttm.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &ast_ttm_mem_global_init;
- global_ref->release = &ast_ttm_mem_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM memory accounting "
- "subsystem.\n");
- return r;
- }
-
- ast->ttm.bo_global_ref.mem_glob =
- ast->ttm.mem_global_ref.object;
- global_ref = &ast->ttm.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- drm_global_item_unref(&ast->ttm.mem_global_ref);
- return r;
- }
- return 0;
-}
-
-static void
-ast_ttm_global_release(struct ast_private *ast)
-{
- if (ast->ttm.mem_global_ref.release == NULL)
- return;
-
- drm_global_item_unref(&ast->ttm.bo_global_ref.ref);
- drm_global_item_unref(&ast->ttm.mem_global_ref);
- ast->ttm.mem_global_ref.release = NULL;
-}
-
-
static void ast_bo_ttm_destroy(struct ttm_buffer_object *tbo)
{
struct ast_bo *bo;
@@ -232,12 +175,7 @@ int ast_mm_init(struct ast_private *ast)
struct drm_device *dev = ast->dev;
struct ttm_bo_device *bdev = &ast->ttm.bdev;
- ret = ast_ttm_global_init(ast);
- if (ret)
- return ret;
-
ret = ttm_bo_device_init(&ast->ttm.bdev,
- ast->ttm.bo_global_ref.ref.object,
&ast_bo_driver,
dev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET,
@@ -268,8 +206,6 @@ void ast_mm_fini(struct ast_private *ast)
ttm_bo_device_release(&ast->ttm.bdev);
- ast_ttm_global_release(ast);
-
arch_phys_wc_del(ast->fb_mtrr);
arch_io_free_memtype_wc(pci_resource_start(dev->pdev, 0),
pci_resource_len(dev->pdev, 0));
diff --git a/drivers/gpu/drm/bochs/bochs.h b/drivers/gpu/drm/bochs/bochs.h
index 577a8b9..fb38c8b 100644
--- a/drivers/gpu/drm/bochs/bochs.h
+++ b/drivers/gpu/drm/bochs/bochs.h
@@ -77,8 +77,6 @@ struct bochs_device {
/* ttm */
struct {
- struct drm_global_reference mem_global_ref;
- struct ttm_bo_global_ref bo_global_ref;
struct ttm_bo_device bdev;
bool initialized;
} ttm;
diff --git a/drivers/gpu/drm/bochs/bochs_mm.c b/drivers/gpu/drm/bochs/bochs_mm.c
index e6ccf7f..0980411 100644
--- a/drivers/gpu/drm/bochs/bochs_mm.c
+++ b/drivers/gpu/drm/bochs/bochs_mm.c
@@ -16,61 +16,6 @@ static inline struct bochs_device *bochs_bdev(struct ttm_bo_device *bd)
return container_of(bd, struct bochs_device, ttm.bdev);
}
-static int bochs_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void bochs_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-static int bochs_ttm_global_init(struct bochs_device *bochs)
-{
- struct drm_global_reference *global_ref;
- int r;
-
- global_ref = &bochs->ttm.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &bochs_ttm_mem_global_init;
- global_ref->release = &bochs_ttm_mem_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM memory accounting "
- "subsystem.\n");
- return r;
- }
-
- bochs->ttm.bo_global_ref.mem_glob =
- bochs->ttm.mem_global_ref.object;
- global_ref = &bochs->ttm.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- drm_global_item_unref(&bochs->ttm.mem_global_ref);
- return r;
- }
-
- return 0;
-}
-
-static void bochs_ttm_global_release(struct bochs_device *bochs)
-{
- if (bochs->ttm.mem_global_ref.release == NULL)
- return;
-
- drm_global_item_unref(&bochs->ttm.bo_global_ref.ref);
- drm_global_item_unref(&bochs->ttm.mem_global_ref);
- bochs->ttm.mem_global_ref.release = NULL;
-}
-
-
static void bochs_bo_ttm_destroy(struct ttm_buffer_object *tbo)
{
struct bochs_bo *bo;
@@ -208,12 +153,7 @@ int bochs_mm_init(struct bochs_device *bochs)
struct ttm_bo_device *bdev = &bochs->ttm.bdev;
int ret;
- ret = bochs_ttm_global_init(bochs);
- if (ret)
- return ret;
-
ret = ttm_bo_device_init(&bochs->ttm.bdev,
- bochs->ttm.bo_global_ref.ref.object,
&bochs_bo_driver,
bochs->dev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET,
@@ -240,7 +180,6 @@ void bochs_mm_fini(struct bochs_device *bochs)
return;
ttm_bo_device_release(&bochs->ttm.bdev);
- bochs_ttm_global_release(bochs);
bochs->ttm.initialized = false;
}
diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.h b/drivers/gpu/drm/cirrus/cirrus_drv.h
index a29f87e..f2b2e0d 100644
--- a/drivers/gpu/drm/cirrus/cirrus_drv.h
+++ b/drivers/gpu/drm/cirrus/cirrus_drv.h
@@ -136,8 +136,6 @@ struct cirrus_device {
int fb_mtrr;
struct {
- struct drm_global_reference mem_global_ref;
- struct ttm_bo_global_ref bo_global_ref;
struct ttm_bo_device bdev;
} ttm;
bool mm_inited;
diff --git a/drivers/gpu/drm/cirrus/cirrus_ttm.c b/drivers/gpu/drm/cirrus/cirrus_ttm.c
index f219532..e075810 100644
--- a/drivers/gpu/drm/cirrus/cirrus_ttm.c
+++ b/drivers/gpu/drm/cirrus/cirrus_ttm.c
@@ -36,63 +36,6 @@ cirrus_bdev(struct ttm_bo_device *bd)
return container_of(bd, struct cirrus_device, ttm.bdev);
}
-static int
-cirrus_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void
-cirrus_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-static int cirrus_ttm_global_init(struct cirrus_device *cirrus)
-{
- struct drm_global_reference *global_ref;
- int r;
-
- global_ref = &cirrus->ttm.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &cirrus_ttm_mem_global_init;
- global_ref->release = &cirrus_ttm_mem_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM memory accounting "
- "subsystem.\n");
- return r;
- }
-
- cirrus->ttm.bo_global_ref.mem_glob =
- cirrus->ttm.mem_global_ref.object;
- global_ref = &cirrus->ttm.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- drm_global_item_unref(&cirrus->ttm.mem_global_ref);
- return r;
- }
- return 0;
-}
-
-static void
-cirrus_ttm_global_release(struct cirrus_device *cirrus)
-{
- if (cirrus->ttm.mem_global_ref.release == NULL)
- return;
-
- drm_global_item_unref(&cirrus->ttm.bo_global_ref.ref);
- drm_global_item_unref(&cirrus->ttm.mem_global_ref);
- cirrus->ttm.mem_global_ref.release = NULL;
-}
-
-
static void cirrus_bo_ttm_destroy(struct ttm_buffer_object *tbo)
{
struct cirrus_bo *bo;
@@ -232,12 +175,7 @@ int cirrus_mm_init(struct cirrus_device *cirrus)
struct drm_device *dev = cirrus->dev;
struct ttm_bo_device *bdev = &cirrus->ttm.bdev;
- ret = cirrus_ttm_global_init(cirrus);
- if (ret)
- return ret;
-
ret = ttm_bo_device_init(&cirrus->ttm.bdev,
- cirrus->ttm.bo_global_ref.ref.object,
&cirrus_bo_driver,
dev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET,
@@ -273,8 +211,6 @@ void cirrus_mm_fini(struct cirrus_device *cirrus)
ttm_bo_device_release(&cirrus->ttm.bdev);
- cirrus_ttm_global_release(cirrus);
-
arch_phys_wc_del(cirrus->fb_mtrr);
cirrus->fb_mtrr = 0;
arch_io_free_memtype_wc(pci_resource_start(dev->pdev, 0),
diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c
index 8c3cfac..5294145 100644
--- a/drivers/gpu/drm/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/drm_dp_mst_topology.c
@@ -1275,6 +1275,9 @@ static struct drm_dp_mst_branch *drm_dp_get_mst_branch_device(struct drm_dp_mst_
mutex_lock(&mgr->lock);
mstb = mgr->mst_primary;
+ if (!mstb)
+ goto out;
+
for (i = 0; i < lct - 1; i++) {
int shift = (i % 2) ? 0 : 4;
int port_num = (rad[i / 2] >> shift) & 0xf;
diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c
index aa1cef7..e2ffecd 100644
--- a/drivers/gpu/drm/drm_drv.c
+++ b/drivers/gpu/drm/drm_drv.c
@@ -973,14 +973,12 @@ static void drm_core_exit(void)
drm_sysfs_destroy();
idr_destroy(&drm_minors_idr);
drm_connector_ida_destroy();
- drm_global_release();
}
static int __init drm_core_init(void)
{
int ret;
- drm_global_init();
drm_connector_ida_init();
idr_init(&drm_minors_idr);
diff --git a/drivers/gpu/drm/drm_global.c b/drivers/gpu/drm/drm_global.c
deleted file mode 100644
index 5799e27..0000000
--- a/drivers/gpu/drm/drm_global.c
+++ /dev/null
@@ -1,137 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0 OR MIT
-/**************************************************************************
- *
- * Copyright 2008-2009 VMware, Inc., Palo Alto, CA., USA
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
- *
- **************************************************************************/
-/*
- * Authors: Thomas Hellstrom <thellstrom-at-vmware-dot-com>
- */
-
-#include <linux/mutex.h>
-#include <linux/slab.h>
-#include <linux/module.h>
-#include <drm/drm_global.h>
-
-struct drm_global_item {
- struct mutex mutex;
- void *object;
- int refcount;
-};
-
-static struct drm_global_item glob[DRM_GLOBAL_NUM];
-
-void drm_global_init(void)
-{
- int i;
-
- for (i = 0; i < DRM_GLOBAL_NUM; ++i) {
- struct drm_global_item *item = &glob[i];
- mutex_init(&item->mutex);
- item->object = NULL;
- item->refcount = 0;
- }
-}
-
-void drm_global_release(void)
-{
- int i;
- for (i = 0; i < DRM_GLOBAL_NUM; ++i) {
- struct drm_global_item *item = &glob[i];
- BUG_ON(item->object != NULL);
- BUG_ON(item->refcount != 0);
- }
-}
-
-/**
- * drm_global_item_ref - Initialize and acquire reference to memory
- * object
- * @ref: Object for initialization
- *
- * This initializes a memory object, allocating memory and calling the
- * .init() hook. Further calls will increase the reference count for
- * that item.
- *
- * Returns:
- * Zero on success, non-zero otherwise.
- */
-int drm_global_item_ref(struct drm_global_reference *ref)
-{
- int ret = 0;
- struct drm_global_item *item = &glob[ref->global_type];
-
- mutex_lock(&item->mutex);
- if (item->refcount == 0) {
- ref->object = kzalloc(ref->size, GFP_KERNEL);
- if (unlikely(ref->object == NULL)) {
- ret = -ENOMEM;
- goto error_unlock;
- }
- ret = ref->init(ref);
- if (unlikely(ret != 0))
- goto error_free;
-
- item->object = ref->object;
- } else {
- ref->object = item->object;
- }
-
- ++item->refcount;
- mutex_unlock(&item->mutex);
- return 0;
-
-error_free:
- kfree(ref->object);
- ref->object = NULL;
-error_unlock:
- mutex_unlock(&item->mutex);
- return ret;
-}
-EXPORT_SYMBOL(drm_global_item_ref);
-
-/**
- * drm_global_item_unref - Drop reference to memory
- * object
- * @ref: Object being removed
- *
- * Drop a reference to the memory object and eventually call the
- * release() hook. The allocated object should be dropped in the
- * release() hook or before calling this function
- *
- */
-
-void drm_global_item_unref(struct drm_global_reference *ref)
-{
- struct drm_global_item *item = &glob[ref->global_type];
-
- mutex_lock(&item->mutex);
- BUG_ON(item->refcount == 0);
- BUG_ON(ref->object != item->object);
- if (--item->refcount == 0) {
- ref->release(ref);
- item->object = NULL;
- }
- mutex_unlock(&item->mutex);
-}
-EXPORT_SYMBOL(drm_global_item_unref);
-
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_sched.c b/drivers/gpu/drm/etnaviv/etnaviv_sched.c
index e7c3ed6..49a6763 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_sched.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_sched.c
@@ -93,7 +93,7 @@ static void etnaviv_sched_timedout_job(struct drm_sched_job *sched_job)
* If the GPU managed to complete this jobs fence, the timout is
* spurious. Bail out.
*/
- if (fence_completed(gpu, submit->out_fence->seqno))
+ if (dma_fence_is_signaled(submit->out_fence))
return;
/*
@@ -105,8 +105,6 @@ static void etnaviv_sched_timedout_job(struct drm_sched_job *sched_job)
change = dma_addr - gpu->hangcheck_dma_addr;
if (change < 0 || change > 16) {
gpu->hangcheck_dma_addr = dma_addr;
- schedule_delayed_work(&sched_job->sched->work_tdr,
- sched_job->sched->timeout);
return;
}
@@ -127,6 +125,8 @@ static void etnaviv_sched_free_job(struct drm_sched_job *sched_job)
{
struct etnaviv_gem_submit *submit = to_etnaviv_submit(sched_job);
+ drm_sched_job_cleanup(sched_job);
+
etnaviv_submit_put(submit);
}
@@ -159,6 +159,7 @@ int etnaviv_sched_push_job(struct drm_sched_entity *sched_entity,
submit->out_fence, 0,
INT_MAX, GFP_KERNEL);
if (submit->out_fence_id < 0) {
+ drm_sched_job_cleanup(&submit->sched_job);
ret = -ENOMEM;
goto out_unlock;
}
diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 94529aa..aef487d 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -164,13 +164,6 @@ static u32 decon_get_frame_count(struct decon_context *ctx, bool end)
return frm;
}
-static u32 decon_get_vblank_counter(struct exynos_drm_crtc *crtc)
-{
- struct decon_context *ctx = crtc->ctx;
-
- return decon_get_frame_count(ctx, false);
-}
-
static void decon_setup_trigger(struct decon_context *ctx)
{
if (!ctx->crtc->i80_mode && !(ctx->out_type & I80_HW_TRG))
@@ -536,7 +529,6 @@ static const struct exynos_drm_crtc_ops decon_crtc_ops = {
.disable = decon_disable,
.enable_vblank = decon_enable_vblank,
.disable_vblank = decon_disable_vblank,
- .get_vblank_counter = decon_get_vblank_counter,
.atomic_begin = decon_atomic_begin,
.update_plane = decon_update_plane,
.disable_plane = decon_disable_plane,
@@ -554,7 +546,6 @@ static int decon_bind(struct device *dev, struct device *master, void *data)
int ret;
ctx->drm_dev = drm_dev;
- drm_dev->max_vblank_count = 0xffffffff;
for (win = ctx->first_win; win < WINDOWS_NR; win++) {
ctx->configs[win].pixel_formats = decon_formats;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_crtc.c b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
index eea9025..2696289 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_crtc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_crtc.c
@@ -162,16 +162,6 @@ static void exynos_drm_crtc_disable_vblank(struct drm_crtc *crtc)
exynos_crtc->ops->disable_vblank(exynos_crtc);
}
-static u32 exynos_drm_crtc_get_vblank_counter(struct drm_crtc *crtc)
-{
- struct exynos_drm_crtc *exynos_crtc = to_exynos_crtc(crtc);
-
- if (exynos_crtc->ops->get_vblank_counter)
- return exynos_crtc->ops->get_vblank_counter(exynos_crtc);
-
- return 0;
-}
-
static const struct drm_crtc_funcs exynos_crtc_funcs = {
.set_config = drm_atomic_helper_set_config,
.page_flip = drm_atomic_helper_page_flip,
@@ -181,7 +171,6 @@ static const struct drm_crtc_funcs exynos_crtc_funcs = {
.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
.enable_vblank = exynos_drm_crtc_enable_vblank,
.disable_vblank = exynos_drm_crtc_disable_vblank,
- .get_vblank_counter = exynos_drm_crtc_get_vblank_counter,
};
struct exynos_drm_crtc *exynos_drm_crtc_create(struct drm_device *drm_dev,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index ec9604f..5e61e70 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -135,7 +135,6 @@ struct exynos_drm_crtc_ops {
void (*disable)(struct exynos_drm_crtc *crtc);
int (*enable_vblank)(struct exynos_drm_crtc *crtc);
void (*disable_vblank)(struct exynos_drm_crtc *crtc);
- u32 (*get_vblank_counter)(struct exynos_drm_crtc *crtc);
enum drm_mode_status (*mode_valid)(struct exynos_drm_crtc *crtc,
const struct drm_display_mode *mode);
bool (*mode_fixup)(struct exynos_drm_crtc *crtc,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index 07af775..d81e62a 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -14,6 +14,7 @@
#include <drm/drmP.h>
#include <drm/drm_crtc_helper.h>
+#include <drm/drm_fb_helper.h>
#include <drm/drm_mipi_dsi.h>
#include <drm/drm_panel.h>
#include <drm/drm_atomic_helper.h>
@@ -1474,12 +1475,12 @@ static int exynos_dsi_create_connector(struct drm_encoder *encoder)
{
struct exynos_dsi *dsi = encoder_to_dsi(encoder);
struct drm_connector *connector = &dsi->connector;
+ struct drm_device *drm = encoder->dev;
int ret;
connector->polled = DRM_CONNECTOR_POLL_HPD;
- ret = drm_connector_init(encoder->dev, connector,
- &exynos_dsi_connector_funcs,
+ ret = drm_connector_init(drm, connector, &exynos_dsi_connector_funcs,
DRM_MODE_CONNECTOR_DSI);
if (ret) {
DRM_ERROR("Failed to initialize connector with drm\n");
@@ -1489,7 +1490,12 @@ static int exynos_dsi_create_connector(struct drm_encoder *encoder)
connector->status = connector_status_disconnected;
drm_connector_helper_add(connector, &exynos_dsi_connector_helper_funcs);
drm_connector_attach_encoder(connector, encoder);
+ if (!drm->registered)
+ return 0;
+ connector->funcs->reset(connector);
+ drm_fb_helper_add_one_connector(drm->fb_helper, connector);
+ drm_connector_register(connector);
return 0;
}
@@ -1527,7 +1533,9 @@ static int exynos_dsi_host_attach(struct mipi_dsi_host *host,
}
dsi->panel = of_drm_find_panel(device->dev.of_node);
- if (dsi->panel) {
+ if (IS_ERR(dsi->panel)) {
+ dsi->panel = NULL;
+ } else {
drm_panel_attach(dsi->panel, &dsi->connector);
dsi->connector.status = connector_status_connected;
}
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index 918dd2c..01d1822 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -192,7 +192,7 @@ int exynos_drm_fbdev_init(struct drm_device *dev)
struct drm_fb_helper *helper;
int ret;
- if (!dev->mode_config.num_crtc || !dev->mode_config.num_connector)
+ if (!dev->mode_config.num_crtc)
return 0;
fbdev = kzalloc(sizeof(*fbdev), GFP_KERNEL);
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
index 45c25a4..3c168ae 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_drm_drv.h
@@ -49,8 +49,6 @@ struct hibmc_drm_private {
bool mode_config_initialized;
/* ttm */
- struct drm_global_reference mem_global_ref;
- struct ttm_bo_global_ref bo_global_ref;
struct ttm_bo_device bdev;
bool initialized;
diff --git a/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c b/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
index 2e3e0bd..dd38326 100644
--- a/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
+++ b/drivers/gpu/drm/hisilicon/hibmc/hibmc_ttm.c
@@ -29,55 +29,6 @@ hibmc_bdev(struct ttm_bo_device *bd)
return container_of(bd, struct hibmc_drm_private, bdev);
}
-static int
-hibmc_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void
-hibmc_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-static int hibmc_ttm_global_init(struct hibmc_drm_private *hibmc)
-{
- int ret;
-
- hibmc->mem_global_ref.global_type = DRM_GLOBAL_TTM_MEM;
- hibmc->mem_global_ref.size = sizeof(struct ttm_mem_global);
- hibmc->mem_global_ref.init = &hibmc_ttm_mem_global_init;
- hibmc->mem_global_ref.release = &hibmc_ttm_mem_global_release;
- ret = drm_global_item_ref(&hibmc->mem_global_ref);
- if (ret) {
- DRM_ERROR("could not get ref on ttm global: %d\n", ret);
- return ret;
- }
-
- hibmc->bo_global_ref.mem_glob =
- hibmc->mem_global_ref.object;
- hibmc->bo_global_ref.ref.global_type = DRM_GLOBAL_TTM_BO;
- hibmc->bo_global_ref.ref.size = sizeof(struct ttm_bo_global);
- hibmc->bo_global_ref.ref.init = &ttm_bo_global_init;
- hibmc->bo_global_ref.ref.release = &ttm_bo_global_release;
- ret = drm_global_item_ref(&hibmc->bo_global_ref.ref);
- if (ret) {
- DRM_ERROR("failed setting up TTM BO subsystem: %d\n", ret);
- drm_global_item_unref(&hibmc->mem_global_ref);
- return ret;
- }
- return 0;
-}
-
-static void
-hibmc_ttm_global_release(struct hibmc_drm_private *hibmc)
-{
- drm_global_item_unref(&hibmc->bo_global_ref.ref);
- drm_global_item_unref(&hibmc->mem_global_ref);
- hibmc->mem_global_ref.release = NULL;
-}
-
static void hibmc_bo_ttm_destroy(struct ttm_buffer_object *tbo)
{
struct hibmc_bo *bo = container_of(tbo, struct hibmc_bo, bo);
@@ -237,18 +188,12 @@ int hibmc_mm_init(struct hibmc_drm_private *hibmc)
struct drm_device *dev = hibmc->dev;
struct ttm_bo_device *bdev = &hibmc->bdev;
- ret = hibmc_ttm_global_init(hibmc);
- if (ret)
- return ret;
-
ret = ttm_bo_device_init(&hibmc->bdev,
- hibmc->bo_global_ref.ref.object,
&hibmc_bo_driver,
dev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET,
true);
if (ret) {
- hibmc_ttm_global_release(hibmc);
DRM_ERROR("error initializing bo driver: %d\n", ret);
return ret;
}
@@ -256,7 +201,6 @@ int hibmc_mm_init(struct hibmc_drm_private *hibmc)
ret = ttm_bo_init_mm(bdev, TTM_PL_VRAM,
hibmc->fb_size >> PAGE_SHIFT);
if (ret) {
- hibmc_ttm_global_release(hibmc);
DRM_ERROR("failed ttm VRAM init: %d\n", ret);
return ret;
}
@@ -271,7 +215,6 @@ void hibmc_mm_fini(struct hibmc_drm_private *hibmc)
return;
ttm_bo_device_release(&hibmc->bdev);
- hibmc_ttm_global_release(hibmc);
hibmc->mm_inited = false;
}
diff --git a/drivers/gpu/drm/i915/gvt/gtt.c b/drivers/gpu/drm/i915/gvt/gtt.c
index 2402395..58e166e 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.c
+++ b/drivers/gpu/drm/i915/gvt/gtt.c
@@ -1905,7 +1905,6 @@ static struct intel_vgpu_mm *intel_vgpu_create_ggtt_mm(struct intel_vgpu *vgpu)
vgpu_free_mm(mm);
return ERR_PTR(-ENOMEM);
}
- mm->ggtt_mm.last_partial_off = -1UL;
return mm;
}
@@ -1930,7 +1929,6 @@ void _intel_vgpu_mm_release(struct kref *mm_ref)
invalidate_ppgtt_mm(mm);
} else {
vfree(mm->ggtt_mm.virtual_ggtt);
- mm->ggtt_mm.last_partial_off = -1UL;
}
vgpu_free_mm(mm);
@@ -2168,6 +2166,8 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off,
struct intel_gvt_gtt_entry e, m;
dma_addr_t dma_addr;
int ret;
+ struct intel_gvt_partial_pte *partial_pte, *pos, *n;
+ bool partial_update = false;
if (bytes != 4 && bytes != 8)
return -EINVAL;
@@ -2178,68 +2178,57 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off,
if (!vgpu_gmadr_is_valid(vgpu, gma))
return 0;
- ggtt_get_guest_entry(ggtt_mm, &e, g_gtt_index);
-
+ e.type = GTT_TYPE_GGTT_PTE;
memcpy((void *)&e.val64 + (off & (info->gtt_entry_size - 1)), p_data,
bytes);
/* If ggtt entry size is 8 bytes, and it's split into two 4 bytes
- * write, we assume the two 4 bytes writes are consecutive.
- * Otherwise, we abort and report error
+ * write, save the first 4 bytes in a list and update virtual
+ * PTE. Only update shadow PTE when the second 4 bytes comes.
*/
if (bytes < info->gtt_entry_size) {
- if (ggtt_mm->ggtt_mm.last_partial_off == -1UL) {
- /* the first partial part*/
- ggtt_mm->ggtt_mm.last_partial_off = off;
- ggtt_mm->ggtt_mm.last_partial_data = e.val64;
- return 0;
- } else if ((g_gtt_index ==
- (ggtt_mm->ggtt_mm.last_partial_off >>
- info->gtt_entry_size_shift)) &&
- (off != ggtt_mm->ggtt_mm.last_partial_off)) {
- /* the second partial part */
+ bool found = false;
- int last_off = ggtt_mm->ggtt_mm.last_partial_off &
- (info->gtt_entry_size - 1);
+ list_for_each_entry_safe(pos, n,
+ &ggtt_mm->ggtt_mm.partial_pte_list, list) {
+ if (g_gtt_index == pos->offset >>
+ info->gtt_entry_size_shift) {
+ if (off != pos->offset) {
+ /* the second partial part*/
+ int last_off = pos->offset &
+ (info->gtt_entry_size - 1);
- memcpy((void *)&e.val64 + last_off,
- (void *)&ggtt_mm->ggtt_mm.last_partial_data +
- last_off, bytes);
+ memcpy((void *)&e.val64 + last_off,
+ (void *)&pos->data + last_off,
+ bytes);
- ggtt_mm->ggtt_mm.last_partial_off = -1UL;
- } else {
- int last_offset;
+ list_del(&pos->list);
+ kfree(pos);
+ found = true;
+ break;
+ }
- gvt_vgpu_err("failed to populate guest ggtt entry: abnormal ggtt entry write sequence, last_partial_off=%lx, offset=%x, bytes=%d, ggtt entry size=%d\n",
- ggtt_mm->ggtt_mm.last_partial_off, off,
- bytes, info->gtt_entry_size);
+ /* update of the first partial part */
+ pos->data = e.val64;
+ ggtt_set_guest_entry(ggtt_mm, &e, g_gtt_index);
+ return 0;
+ }
+ }
- /* set host ggtt entry to scratch page and clear
- * virtual ggtt entry as not present for last
- * partially write offset
- */
- last_offset = ggtt_mm->ggtt_mm.last_partial_off &
- (~(info->gtt_entry_size - 1));
-
- ggtt_get_host_entry(ggtt_mm, &m, last_offset);
- ggtt_invalidate_pte(vgpu, &m);
- ops->set_pfn(&m, gvt->gtt.scratch_mfn);
- ops->clear_present(&m);
- ggtt_set_host_entry(ggtt_mm, &m, last_offset);
- ggtt_invalidate(gvt->dev_priv);
-
- ggtt_get_guest_entry(ggtt_mm, &e, last_offset);
- ops->clear_present(&e);
- ggtt_set_guest_entry(ggtt_mm, &e, last_offset);
-
- ggtt_mm->ggtt_mm.last_partial_off = off;
- ggtt_mm->ggtt_mm.last_partial_data = e.val64;
-
- return 0;
+ if (!found) {
+ /* the first partial part */
+ partial_pte = kzalloc(sizeof(*partial_pte), GFP_KERNEL);
+ if (!partial_pte)
+ return -ENOMEM;
+ partial_pte->offset = off;
+ partial_pte->data = e.val64;
+ list_add_tail(&partial_pte->list,
+ &ggtt_mm->ggtt_mm.partial_pte_list);
+ partial_update = true;
}
}
- if (ops->test_present(&e)) {
+ if (!partial_update && (ops->test_present(&e))) {
gfn = ops->get_pfn(&e);
m = e;
@@ -2263,16 +2252,18 @@ static int emulate_ggtt_mmio_write(struct intel_vgpu *vgpu, unsigned int off,
} else
ops->set_pfn(&m, dma_addr >> PAGE_SHIFT);
} else {
- ggtt_get_host_entry(ggtt_mm, &m, g_gtt_index);
- ggtt_invalidate_pte(vgpu, &m);
ops->set_pfn(&m, gvt->gtt.scratch_mfn);
ops->clear_present(&m);
}
out:
+ ggtt_set_guest_entry(ggtt_mm, &e, g_gtt_index);
+
+ ggtt_get_host_entry(ggtt_mm, &e, g_gtt_index);
+ ggtt_invalidate_pte(vgpu, &e);
+
ggtt_set_host_entry(ggtt_mm, &m, g_gtt_index);
ggtt_invalidate(gvt->dev_priv);
- ggtt_set_guest_entry(ggtt_mm, &e, g_gtt_index);
return 0;
}
@@ -2430,6 +2421,8 @@ int intel_vgpu_init_gtt(struct intel_vgpu *vgpu)
intel_vgpu_reset_ggtt(vgpu, false);
+ INIT_LIST_HEAD(>t->ggtt_mm->ggtt_mm.partial_pte_list);
+
return create_scratch_page_tree(vgpu);
}
@@ -2454,6 +2447,14 @@ static void intel_vgpu_destroy_all_ppgtt_mm(struct intel_vgpu *vgpu)
static void intel_vgpu_destroy_ggtt_mm(struct intel_vgpu *vgpu)
{
+ struct intel_gvt_partial_pte *pos;
+
+ list_for_each_entry(pos,
+ &vgpu->gtt.ggtt_mm->ggtt_mm.partial_pte_list, list) {
+ gvt_dbg_mm("partial PTE update on hold 0x%lx : 0x%llx\n",
+ pos->offset, pos->data);
+ kfree(pos);
+ }
intel_vgpu_destroy_mm(vgpu->gtt.ggtt_mm);
vgpu->gtt.ggtt_mm = NULL;
}
diff --git a/drivers/gpu/drm/i915/gvt/gtt.h b/drivers/gpu/drm/i915/gvt/gtt.h
index 7a9b361..d8cb04c 100644
--- a/drivers/gpu/drm/i915/gvt/gtt.h
+++ b/drivers/gpu/drm/i915/gvt/gtt.h
@@ -35,7 +35,6 @@
#define _GVT_GTT_H_
#define I915_GTT_PAGE_SHIFT 12
-#define I915_GTT_PAGE_MASK (~(I915_GTT_PAGE_SIZE - 1))
struct intel_vgpu_mm;
@@ -133,6 +132,12 @@ enum intel_gvt_mm_type {
#define GVT_RING_CTX_NR_PDPS GEN8_3LVL_PDPES
+struct intel_gvt_partial_pte {
+ unsigned long offset;
+ u64 data;
+ struct list_head list;
+};
+
struct intel_vgpu_mm {
enum intel_gvt_mm_type type;
struct intel_vgpu *vgpu;
@@ -157,8 +162,7 @@ struct intel_vgpu_mm {
} ppgtt_mm;
struct {
void *virtual_ggtt;
- unsigned long last_partial_off;
- u64 last_partial_data;
+ struct list_head partial_pte_list;
} ggtt_mm;
};
};
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 90f50f6..aa280bb 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -1609,7 +1609,7 @@ static int bxt_gt_disp_pwron_write(struct intel_vgpu *vgpu,
return 0;
}
-static int bxt_edp_psr_imr_iir_write(struct intel_vgpu *vgpu,
+static int edp_psr_imr_iir_write(struct intel_vgpu *vgpu,
unsigned int offset, void *p_data, unsigned int bytes)
{
vgpu_vreg(vgpu, offset) = 0;
@@ -2607,6 +2607,9 @@ static int init_generic_mmio_info(struct intel_gvt *gvt)
MMIO_DFH(_MMIO(0x1a178), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
MMIO_DFH(_MMIO(0x1a17c), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
MMIO_DFH(_MMIO(0x2217c), D_BDW_PLUS, F_CMD_ACCESS, NULL, NULL);
+
+ MMIO_DH(EDP_PSR_IMR, D_BDW_PLUS, NULL, edp_psr_imr_iir_write);
+ MMIO_DH(EDP_PSR_IIR, D_BDW_PLUS, NULL, edp_psr_imr_iir_write);
return 0;
}
@@ -3205,9 +3208,6 @@ static int init_bxt_mmio_info(struct intel_gvt *gvt)
MMIO_D(HSW_TVIDEO_DIP_GCP(TRANSCODER_B), D_BXT);
MMIO_D(HSW_TVIDEO_DIP_GCP(TRANSCODER_C), D_BXT);
- MMIO_DH(EDP_PSR_IMR, D_BXT, NULL, bxt_edp_psr_imr_iir_write);
- MMIO_DH(EDP_PSR_IIR, D_BXT, NULL, bxt_edp_psr_imr_iir_write);
-
MMIO_D(RC6_CTX_BASE, D_BXT);
MMIO_D(GEN8_PUSHBUS_CONTROL, D_BXT);
diff --git a/drivers/gpu/drm/i915/gvt/mmio_context.c b/drivers/gpu/drm/i915/gvt/mmio_context.c
index 10e63ee..36a5147 100644
--- a/drivers/gpu/drm/i915/gvt/mmio_context.c
+++ b/drivers/gpu/drm/i915/gvt/mmio_context.c
@@ -131,7 +131,7 @@ static struct engine_mmio gen9_engine_mmio_list[] __cacheline_aligned = {
{RCS, GAMT_CHKN_BIT_REG, 0x0, false}, /* 0x4ab8 */
{RCS, GEN9_GAMT_ECO_REG_RW_IA, 0x0, false}, /* 0x4ab0 */
- {RCS, GEN9_CSFE_CHICKEN1_RCS, 0x0, false}, /* 0x20d4 */
+ {RCS, GEN9_CSFE_CHICKEN1_RCS, 0xffff, false}, /* 0x20d4 */
{RCS, GEN8_GARBCNTL, 0x0, false}, /* 0xb004 */
{RCS, GEN7_FF_THREAD_MODE, 0x0, false}, /* 0x20a0 */
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 44e2c0f..ffdbbac 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -1175,8 +1175,6 @@ skl_dram_get_channels_info(struct drm_i915_private *dev_priv)
return -EINVAL;
}
- dram_info->valid_dimm = true;
-
/*
* If any of the channel is single rank channel, worst case output
* will be same as if single rank memory, so consider single rank
@@ -1193,8 +1191,7 @@ skl_dram_get_channels_info(struct drm_i915_private *dev_priv)
return -EINVAL;
}
- if (ch0.is_16gb_dimm || ch1.is_16gb_dimm)
- dram_info->is_16gb_dimm = true;
+ dram_info->is_16gb_dimm = ch0.is_16gb_dimm || ch1.is_16gb_dimm;
dev_priv->dram_info.symmetric_memory = intel_is_dram_symmetric(val_ch0,
val_ch1,
@@ -1314,7 +1311,6 @@ bxt_get_dram_info(struct drm_i915_private *dev_priv)
return -EINVAL;
}
- dram_info->valid_dimm = true;
dram_info->valid = true;
return 0;
}
@@ -1327,12 +1323,17 @@ intel_get_dram_info(struct drm_i915_private *dev_priv)
int ret;
dram_info->valid = false;
- dram_info->valid_dimm = false;
- dram_info->is_16gb_dimm = false;
dram_info->rank = I915_DRAM_RANK_INVALID;
dram_info->bandwidth_kbps = 0;
dram_info->num_channels = 0;
+ /*
+ * Assume 16Gb DIMMs are present until proven otherwise.
+ * This is only used for the level 0 watermark latency
+ * w/a which does not apply to bxt/glk.
+ */
+ dram_info->is_16gb_dimm = !IS_GEN9_LP(dev_priv);
+
if (INTEL_GEN(dev_priv) < 9 || IS_GEMINILAKE(dev_priv))
return;
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 8624b4b..9102571 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -1948,7 +1948,6 @@ struct drm_i915_private {
struct dram_info {
bool valid;
- bool valid_dimm;
bool is_16gb_dimm;
u8 num_channels;
enum dram_rank {
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 0918728..1aaccbe 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -460,7 +460,7 @@ eb_validate_vma(struct i915_execbuffer *eb,
* any non-page-aligned or non-canonical addresses.
*/
if (unlikely(entry->flags & EXEC_OBJECT_PINNED &&
- entry->offset != gen8_canonical_addr(entry->offset & PAGE_MASK)))
+ entry->offset != gen8_canonical_addr(entry->offset & I915_GTT_PAGE_MASK)))
return -EINVAL;
/* pad_to_size was once a reserved field, so sanitize it */
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 56c7f86..47c3025 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1757,7 +1757,7 @@ static void gen6_dump_ppgtt(struct i915_hw_ppgtt *base, struct seq_file *m)
if (i == 4)
continue;
- seq_printf(m, "\t\t(%03d, %04d) %08lx: ",
+ seq_printf(m, "\t\t(%03d, %04d) %08llx: ",
pde, pte,
(pde * GEN6_PTES + pte) * I915_GTT_PAGE_SIZE);
for (i = 0; i < 4; i++) {
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index 7e2af5f..2803929 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -42,13 +42,15 @@
#include "i915_selftest.h"
#include "i915_timeline.h"
-#define I915_GTT_PAGE_SIZE_4K BIT(12)
-#define I915_GTT_PAGE_SIZE_64K BIT(16)
-#define I915_GTT_PAGE_SIZE_2M BIT(21)
+#define I915_GTT_PAGE_SIZE_4K BIT_ULL(12)
+#define I915_GTT_PAGE_SIZE_64K BIT_ULL(16)
+#define I915_GTT_PAGE_SIZE_2M BIT_ULL(21)
#define I915_GTT_PAGE_SIZE I915_GTT_PAGE_SIZE_4K
#define I915_GTT_MAX_PAGE_SIZE I915_GTT_PAGE_SIZE_2M
+#define I915_GTT_PAGE_MASK -I915_GTT_PAGE_SIZE
+
#define I915_GTT_MIN_ALIGNMENT I915_GTT_PAGE_SIZE
#define I915_FENCE_REG_NONE -1
@@ -659,20 +661,20 @@ int i915_gem_gtt_insert(struct i915_address_space *vm,
u64 start, u64 end, unsigned int flags);
/* Flags used by pin/bind&friends. */
-#define PIN_NONBLOCK BIT(0)
-#define PIN_MAPPABLE BIT(1)
-#define PIN_ZONE_4G BIT(2)
-#define PIN_NONFAULT BIT(3)
-#define PIN_NOEVICT BIT(4)
+#define PIN_NONBLOCK BIT_ULL(0)
+#define PIN_MAPPABLE BIT_ULL(1)
+#define PIN_ZONE_4G BIT_ULL(2)
+#define PIN_NONFAULT BIT_ULL(3)
+#define PIN_NOEVICT BIT_ULL(4)
-#define PIN_MBZ BIT(5) /* I915_VMA_PIN_OVERFLOW */
-#define PIN_GLOBAL BIT(6) /* I915_VMA_GLOBAL_BIND */
-#define PIN_USER BIT(7) /* I915_VMA_LOCAL_BIND */
-#define PIN_UPDATE BIT(8)
+#define PIN_MBZ BIT_ULL(5) /* I915_VMA_PIN_OVERFLOW */
+#define PIN_GLOBAL BIT_ULL(6) /* I915_VMA_GLOBAL_BIND */
+#define PIN_USER BIT_ULL(7) /* I915_VMA_LOCAL_BIND */
+#define PIN_UPDATE BIT_ULL(8)
-#define PIN_HIGH BIT(9)
-#define PIN_OFFSET_BIAS BIT(10)
-#define PIN_OFFSET_FIXED BIT(11)
+#define PIN_HIGH BIT_ULL(9)
+#define PIN_OFFSET_BIAS BIT_ULL(10)
+#define PIN_OFFSET_FIXED BIT_ULL(11)
#define PIN_OFFSET_MASK (-I915_GTT_PAGE_SIZE)
#endif
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 7c491ea..e31c27e 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -2095,8 +2095,12 @@ enum i915_power_well_id {
/* ICL PHY DFLEX registers */
#define PORT_TX_DFLEXDPMLE1 _MMIO(0x1638C0)
-#define DFLEXDPMLE1_DPMLETC_MASK(n) (0xf << (4 * (n)))
-#define DFLEXDPMLE1_DPMLETC(n, x) ((x) << (4 * (n)))
+#define DFLEXDPMLE1_DPMLETC_MASK(tc_port) (0xf << (4 * (tc_port)))
+#define DFLEXDPMLE1_DPMLETC_ML0(tc_port) (1 << (4 * (tc_port)))
+#define DFLEXDPMLE1_DPMLETC_ML1_0(tc_port) (3 << (4 * (tc_port)))
+#define DFLEXDPMLE1_DPMLETC_ML3(tc_port) (8 << (4 * (tc_port)))
+#define DFLEXDPMLE1_DPMLETC_ML3_2(tc_port) (12 << (4 * (tc_port)))
+#define DFLEXDPMLE1_DPMLETC_ML3_0(tc_port) (15 << (4 * (tc_port)))
/* BXT PHY Ref registers */
#define _PORT_REF_DW3_A 0x16218C
@@ -4593,12 +4597,12 @@ enum {
#define DRM_DIP_ENABLE (1 << 28)
#define PSR_VSC_BIT_7_SET (1 << 27)
-#define VSC_SELECT_MASK (0x3 << 26)
-#define VSC_SELECT_SHIFT 26
-#define VSC_DIP_HW_HEA_DATA (0 << 26)
-#define VSC_DIP_HW_HEA_SW_DATA (1 << 26)
-#define VSC_DIP_HW_DATA_SW_HEA (2 << 26)
-#define VSC_DIP_SW_HEA_DATA (3 << 26)
+#define VSC_SELECT_MASK (0x3 << 25)
+#define VSC_SELECT_SHIFT 25
+#define VSC_DIP_HW_HEA_DATA (0 << 25)
+#define VSC_DIP_HW_HEA_SW_DATA (1 << 25)
+#define VSC_DIP_HW_DATA_SW_HEA (2 << 25)
+#define VSC_DIP_SW_HEA_DATA (3 << 25)
#define VDIP_ENABLE_PPS (1 << 24)
/* Panel power sequencing */
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index 769f3f5..ee3ca2d 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -144,6 +144,9 @@ static const struct {
/* HDMI N/CTS table */
#define TMDS_297M 297000
#define TMDS_296M 296703
+#define TMDS_594M 594000
+#define TMDS_593M 593407
+
static const struct {
int sample_rate;
int clock;
@@ -164,6 +167,20 @@ static const struct {
{ 176400, TMDS_297M, 18816, 247500 },
{ 192000, TMDS_296M, 23296, 281250 },
{ 192000, TMDS_297M, 20480, 247500 },
+ { 44100, TMDS_593M, 8918, 937500 },
+ { 44100, TMDS_594M, 9408, 990000 },
+ { 48000, TMDS_593M, 5824, 562500 },
+ { 48000, TMDS_594M, 6144, 594000 },
+ { 32000, TMDS_593M, 5824, 843750 },
+ { 32000, TMDS_594M, 3072, 445500 },
+ { 88200, TMDS_593M, 17836, 937500 },
+ { 88200, TMDS_594M, 18816, 990000 },
+ { 96000, TMDS_593M, 11648, 562500 },
+ { 96000, TMDS_594M, 12288, 594000 },
+ { 176400, TMDS_593M, 35672, 937500 },
+ { 176400, TMDS_594M, 37632, 990000 },
+ { 192000, TMDS_593M, 23296, 562500 },
+ { 192000, TMDS_594M, 24576, 594000 },
};
/* get AUD_CONFIG_PIXEL_CLOCK_HDMI_* value for mode */
diff --git a/drivers/gpu/drm/i915/intel_cdclk.c b/drivers/gpu/drm/i915/intel_cdclk.c
index 29075c7..8d74276 100644
--- a/drivers/gpu/drm/i915/intel_cdclk.c
+++ b/drivers/gpu/drm/i915/intel_cdclk.c
@@ -2138,16 +2138,8 @@ void intel_set_cdclk(struct drm_i915_private *dev_priv,
static int intel_pixel_rate_to_cdclk(struct drm_i915_private *dev_priv,
int pixel_rate)
{
- if (INTEL_GEN(dev_priv) >= 10)
+ if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
return DIV_ROUND_UP(pixel_rate, 2);
- else if (IS_GEMINILAKE(dev_priv))
- /*
- * FIXME: Avoid using a pixel clock that is more than 99% of the cdclk
- * as a temporary workaround. Use a higher cdclk instead. (Note that
- * intel_compute_max_dotclk() limits the max pixel clock to 99% of max
- * cdclk.)
- */
- return DIV_ROUND_UP(pixel_rate * 100, 2 * 99);
else if (IS_GEN9(dev_priv) ||
IS_BROADWELL(dev_priv) || IS_HASWELL(dev_priv))
return pixel_rate;
@@ -2543,14 +2535,8 @@ static int intel_compute_max_dotclk(struct drm_i915_private *dev_priv)
{
int max_cdclk_freq = dev_priv->max_cdclk_freq;
- if (INTEL_GEN(dev_priv) >= 10)
+ if (INTEL_GEN(dev_priv) >= 10 || IS_GEMINILAKE(dev_priv))
return 2 * max_cdclk_freq;
- else if (IS_GEMINILAKE(dev_priv))
- /*
- * FIXME: Limiting to 99% as a temporary workaround. See
- * intel_min_cdclk() for details.
- */
- return 2 * max_cdclk_freq * 99 / 100;
else if (IS_GEN9(dev_priv) ||
IS_BROADWELL(dev_priv) || IS_HASWELL(dev_priv))
return max_cdclk_freq;
diff --git a/drivers/gpu/drm/i915/intel_device_info.c b/drivers/gpu/drm/i915/intel_device_info.c
index 0ef0c64..01fa982 100644
--- a/drivers/gpu/drm/i915/intel_device_info.c
+++ b/drivers/gpu/drm/i915/intel_device_info.c
@@ -474,7 +474,7 @@ static void broadwell_sseu_info_init(struct drm_i915_private *dev_priv)
u8 eu_disabled_mask;
u32 n_disabled;
- if (!(sseu->subslice_mask[ss] & BIT(ss)))
+ if (!(sseu->subslice_mask[s] & BIT(ss)))
/* skip disabled subslice */
continue;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 9741cc4..a54843f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -4850,8 +4850,31 @@ static void cpt_verify_modeset(struct drm_device *dev, int pipe)
* chroma samples for both of the luma samples, and thus we don't
* actually get the expected MPEG2 chroma siting convention :(
* The same behaviour is observed on pre-SKL platforms as well.
+ *
+ * Theory behind the formula (note that we ignore sub-pixel
+ * source coordinates):
+ * s = source sample position
+ * d = destination sample position
+ *
+ * Downscaling 4:1:
+ * -0.5
+ * | 0.0
+ * | | 1.5 (initial phase)
+ * | | |
+ * v v v
+ * | s | s | s | s |
+ * | d |
+ *
+ * Upscaling 1:4:
+ * -0.5
+ * | -0.375 (initial phase)
+ * | | 0.0
+ * | | |
+ * v v v
+ * | s |
+ * | d | d | d | d |
*/
-u16 skl_scaler_calc_phase(int sub, bool chroma_cosited)
+u16 skl_scaler_calc_phase(int sub, int scale, bool chroma_cosited)
{
int phase = -0x8000;
u16 trip = 0;
@@ -4859,6 +4882,15 @@ u16 skl_scaler_calc_phase(int sub, bool chroma_cosited)
if (chroma_cosited)
phase += (sub - 1) * 0x8000 / sub;
+ phase += scale / (2 * sub);
+
+ /*
+ * Hardware initial phase limited to [-0.5:1.5].
+ * Since the max hardware scale factor is 3.0, we
+ * should never actually excdeed 1.0 here.
+ */
+ WARN_ON(phase < -0x8000 || phase > 0x18000);
+
if (phase < 0)
phase = 0x10000 + phase;
else
@@ -5067,13 +5099,20 @@ static void skylake_pfit_enable(struct intel_crtc *crtc)
if (crtc->config->pch_pfit.enabled) {
u16 uv_rgb_hphase, uv_rgb_vphase;
+ int pfit_w, pfit_h, hscale, vscale;
int id;
if (WARN_ON(crtc->config->scaler_state.scaler_id < 0))
return;
- uv_rgb_hphase = skl_scaler_calc_phase(1, false);
- uv_rgb_vphase = skl_scaler_calc_phase(1, false);
+ pfit_w = (crtc->config->pch_pfit.size >> 16) & 0xFFFF;
+ pfit_h = crtc->config->pch_pfit.size & 0xFFFF;
+
+ hscale = (crtc->config->pipe_src_w << 16) / pfit_w;
+ vscale = (crtc->config->pipe_src_h << 16) / pfit_h;
+
+ uv_rgb_hphase = skl_scaler_calc_phase(1, hscale, false);
+ uv_rgb_vphase = skl_scaler_calc_phase(1, vscale, false);
id = scaler_state->scaler_id;
I915_WRITE(SKL_PS_CTRL(pipe, id), PS_SCALER_EN |
@@ -12768,17 +12807,12 @@ static void intel_atomic_commit_tail(struct drm_atomic_state *state)
intel_check_cpu_fifo_underruns(dev_priv);
intel_check_pch_fifo_underruns(dev_priv);
- if (!new_crtc_state->active) {
- /*
- * Make sure we don't call initial_watermarks
- * for ILK-style watermark updates.
- *
- * No clue what this is supposed to achieve.
- */
- if (INTEL_GEN(dev_priv) >= 9)
- dev_priv->display.initial_watermarks(intel_state,
- to_intel_crtc_state(new_crtc_state));
- }
+ /* FIXME unify this for all platforms */
+ if (!new_crtc_state->active &&
+ !HAS_GMCH_DISPLAY(dev_priv) &&
+ dev_priv->display.initial_watermarks)
+ dev_priv->display.initial_watermarks(intel_state,
+ to_intel_crtc_state(new_crtc_state));
}
}
@@ -14646,7 +14680,7 @@ static int intel_framebuffer_init(struct intel_framebuffer *intel_fb,
fb->height < SKL_MIN_YUV_420_SRC_H ||
(fb->width % 4) != 0 || (fb->height % 4) != 0)) {
DRM_DEBUG_KMS("src dimensions not correct for NV12\n");
- return -EINVAL;
+ goto err;
}
for (i = 0; i < fb->format->num_planes; i++) {
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index 1b00f8e..a911691 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -452,6 +452,10 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo
if (!intel_connector)
return NULL;
+ intel_connector->get_hw_state = intel_dp_mst_get_hw_state;
+ intel_connector->mst_port = intel_dp;
+ intel_connector->port = port;
+
connector = &intel_connector->base;
ret = drm_connector_init(dev, connector, &intel_dp_mst_connector_funcs,
DRM_MODE_CONNECTOR_DisplayPort);
@@ -462,10 +466,6 @@ static struct drm_connector *intel_dp_add_mst_connector(struct drm_dp_mst_topolo
drm_connector_helper_add(connector, &intel_dp_mst_connector_helper_funcs);
- intel_connector->get_hw_state = intel_dp_mst_get_hw_state;
- intel_connector->mst_port = intel_dp;
- intel_connector->port = port;
-
for_each_pipe(dev_priv, pipe) {
struct drm_encoder *enc =
&intel_dp->mst_encoders[pipe]->base.base;
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index f8dc84b..8b298e5 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -1646,7 +1646,7 @@ void intel_mode_from_pipe_config(struct drm_display_mode *mode,
void intel_crtc_arm_fifo_underrun(struct intel_crtc *crtc,
struct intel_crtc_state *crtc_state);
-u16 skl_scaler_calc_phase(int sub, bool chroma_center);
+u16 skl_scaler_calc_phase(int sub, int scale, bool chroma_center);
int skl_update_scaler_crtc(struct intel_crtc_state *crtc_state);
int skl_max_scale(const struct intel_crtc_state *crtc_state,
u32 pixel_format);
diff --git a/drivers/gpu/drm/i915/intel_hotplug.c b/drivers/gpu/drm/i915/intel_hotplug.c
index 648a13c..9a80181 100644
--- a/drivers/gpu/drm/i915/intel_hotplug.c
+++ b/drivers/gpu/drm/i915/intel_hotplug.c
@@ -228,7 +228,9 @@ static void intel_hpd_irq_storm_reenable_work(struct work_struct *work)
drm_for_each_connector_iter(connector, &conn_iter) {
struct intel_connector *intel_connector = to_intel_connector(connector);
- if (intel_connector->encoder->hpd_pin == pin) {
+ /* Don't check MST ports, they don't have pins */
+ if (!intel_connector->mst_port &&
+ intel_connector->encoder->hpd_pin == pin) {
if (connector->polled != intel_connector->polled)
DRM_DEBUG_DRIVER("Reenabling HPD on connector %s\n",
connector->name);
@@ -395,37 +397,54 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
struct intel_encoder *encoder;
bool storm_detected = false;
bool queue_dig = false, queue_hp = false;
+ u32 long_hpd_pulse_mask = 0;
+ u32 short_hpd_pulse_mask = 0;
+ enum hpd_pin pin;
if (!pin_mask)
return;
spin_lock(&dev_priv->irq_lock);
- for_each_intel_encoder(&dev_priv->drm, encoder) {
- enum hpd_pin pin = encoder->hpd_pin;
- bool has_hpd_pulse = intel_encoder_has_hpd_pulse(encoder);
+ /*
+ * Determine whether ->hpd_pulse() exists for each pin, and
+ * whether we have a short or a long pulse. This is needed
+ * as each pin may have up to two encoders (HDMI and DP) and
+ * only the one of them (DP) will have ->hpd_pulse().
+ */
+ for_each_intel_encoder(&dev_priv->drm, encoder) {
+ bool has_hpd_pulse = intel_encoder_has_hpd_pulse(encoder);
+ enum port port = encoder->port;
+ bool long_hpd;
+
+ pin = encoder->hpd_pin;
if (!(BIT(pin) & pin_mask))
continue;
- if (has_hpd_pulse) {
- bool long_hpd = long_mask & BIT(pin);
- enum port port = encoder->port;
+ if (!has_hpd_pulse)
+ continue;
- DRM_DEBUG_DRIVER("digital hpd port %c - %s\n", port_name(port),
- long_hpd ? "long" : "short");
- /*
- * For long HPD pulses we want to have the digital queue happen,
- * but we still want HPD storm detection to function.
- */
- queue_dig = true;
- if (long_hpd) {
- dev_priv->hotplug.long_port_mask |= (1 << port);
- } else {
- /* for short HPD just trigger the digital queue */
- dev_priv->hotplug.short_port_mask |= (1 << port);
- continue;
- }
+ long_hpd = long_mask & BIT(pin);
+
+ DRM_DEBUG_DRIVER("digital hpd port %c - %s\n", port_name(port),
+ long_hpd ? "long" : "short");
+ queue_dig = true;
+
+ if (long_hpd) {
+ long_hpd_pulse_mask |= BIT(pin);
+ dev_priv->hotplug.long_port_mask |= BIT(port);
+ } else {
+ short_hpd_pulse_mask |= BIT(pin);
+ dev_priv->hotplug.short_port_mask |= BIT(port);
}
+ }
+
+ /* Now process each pin just once */
+ for_each_hpd_pin(pin) {
+ bool long_hpd;
+
+ if (!(BIT(pin) & pin_mask))
+ continue;
if (dev_priv->hotplug.stats[pin].state == HPD_DISABLED) {
/*
@@ -442,11 +461,22 @@ void intel_hpd_irq_handler(struct drm_i915_private *dev_priv,
if (dev_priv->hotplug.stats[pin].state != HPD_ENABLED)
continue;
- if (!has_hpd_pulse) {
+ /*
+ * Delegate to ->hpd_pulse() if one of the encoders for this
+ * pin has it, otherwise let the hotplug_work deal with this
+ * pin directly.
+ */
+ if (((short_hpd_pulse_mask | long_hpd_pulse_mask) & BIT(pin))) {
+ long_hpd = long_hpd_pulse_mask & BIT(pin);
+ } else {
dev_priv->hotplug.event_bits |= BIT(pin);
+ long_hpd = true;
queue_hp = true;
}
+ if (!long_hpd)
+ continue;
+
if (intel_hpd_irq_storm_detect(dev_priv, pin)) {
dev_priv->hotplug.event_bits &= ~BIT(pin);
storm_detected = true;
diff --git a/drivers/gpu/drm/i915/intel_lpe_audio.c b/drivers/gpu/drm/i915/intel_lpe_audio.c
index cdf1955..5d5336f 100644
--- a/drivers/gpu/drm/i915/intel_lpe_audio.c
+++ b/drivers/gpu/drm/i915/intel_lpe_audio.c
@@ -297,8 +297,10 @@ void intel_lpe_audio_teardown(struct drm_i915_private *dev_priv)
lpe_audio_platdev_destroy(dev_priv);
irq_free_desc(dev_priv->lpe_audio.irq);
-}
+ dev_priv->lpe_audio.irq = -1;
+ dev_priv->lpe_audio.platdev = NULL;
+}
/**
* intel_lpe_audio_notify() - notify lpe audio event
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 43957bb..37c94a5 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -424,7 +424,8 @@ static u64 execlists_update_context(struct i915_request *rq)
reg_state[CTX_RING_TAIL+1] = intel_ring_set_tail(rq->ring, rq->tail);
- /* True 32b PPGTT with dynamic page allocation: update PDP
+ /*
+ * True 32b PPGTT with dynamic page allocation: update PDP
* registers and point the unallocated PDPs to scratch page.
* PML4 is allocated during ppgtt init, so this is not needed
* in 48-bit mode.
@@ -432,6 +433,17 @@ static u64 execlists_update_context(struct i915_request *rq)
if (ppgtt && !i915_vm_is_48bit(&ppgtt->vm))
execlists_update_context_pdps(ppgtt, reg_state);
+ /*
+ * Make sure the context image is complete before we submit it to HW.
+ *
+ * Ostensibly, writes (including the WCB) should be flushed prior to
+ * an uncached write such as our mmio register access, the empirical
+ * evidence (esp. on Braswell) suggests that the WC write into memory
+ * may not be visible to the HW prior to the completion of the UC
+ * register write and that we may begin execution from the context
+ * before its image is complete leading to invalid PD chasing.
+ */
+ wmb();
return ce->lrc_desc;
}
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 1db9b832..245f002 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -2881,8 +2881,7 @@ static void intel_read_wm_latency(struct drm_i915_private *dev_priv,
* any underrun. If not able to get Dimm info assume 16GB dimm
* to avoid any underrun.
*/
- if (!dev_priv->dram_info.valid_dimm ||
- dev_priv->dram_info.is_16gb_dimm)
+ if (dev_priv->dram_info.is_16gb_dimm)
wm[0] += 1;
} else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index d0ef50b..187bb0c 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -91,6 +91,7 @@ static int
gen4_render_ring_flush(struct i915_request *rq, u32 mode)
{
u32 cmd, *cs;
+ int i;
/*
* read/write caches:
@@ -127,12 +128,45 @@ gen4_render_ring_flush(struct i915_request *rq, u32 mode)
cmd |= MI_INVALIDATE_ISP;
}
- cs = intel_ring_begin(rq, 2);
+ i = 2;
+ if (mode & EMIT_INVALIDATE)
+ i += 20;
+
+ cs = intel_ring_begin(rq, i);
if (IS_ERR(cs))
return PTR_ERR(cs);
*cs++ = cmd;
- *cs++ = MI_NOOP;
+
+ /*
+ * A random delay to let the CS invalidate take effect? Without this
+ * delay, the GPU relocation path fails as the CS does not see
+ * the updated contents. Just as important, if we apply the flushes
+ * to the EMIT_FLUSH branch (i.e. immediately after the relocation
+ * write and before the invalidate on the next batch), the relocations
+ * still fail. This implies that is a delay following invalidation
+ * that is required to reset the caches as opposed to a delay to
+ * ensure the memory is written.
+ */
+ if (mode & EMIT_INVALIDATE) {
+ *cs++ = GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE;
+ *cs++ = i915_ggtt_offset(rq->engine->scratch) |
+ PIPE_CONTROL_GLOBAL_GTT;
+ *cs++ = 0;
+ *cs++ = 0;
+
+ for (i = 0; i < 12; i++)
+ *cs++ = MI_FLUSH;
+
+ *cs++ = GFX_OP_PIPE_CONTROL(4) | PIPE_CONTROL_QW_WRITE;
+ *cs++ = i915_ggtt_offset(rq->engine->scratch) |
+ PIPE_CONTROL_GLOBAL_GTT;
+ *cs++ = 0;
+ *cs++ = 0;
+ }
+
+ *cs++ = cmd;
+
intel_ring_advance(rq, cs);
return 0;
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index 0fdabce..44e4491 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -2749,6 +2749,12 @@ static const struct i915_power_well_desc icl_power_wells[] = {
},
},
{
+ .name = "DC off",
+ .domains = ICL_DISPLAY_DC_OFF_POWER_DOMAINS,
+ .ops = &gen9_dc_off_power_well_ops,
+ .id = DISP_PW_ID_NONE,
+ },
+ {
.name = "power well 2",
.domains = ICL_PW_2_POWER_DOMAINS,
.ops = &hsw_power_well_ops,
@@ -2760,12 +2766,6 @@ static const struct i915_power_well_desc icl_power_wells[] = {
},
},
{
- .name = "DC off",
- .domains = ICL_DISPLAY_DC_OFF_POWER_DOMAINS,
- .ops = &gen9_dc_off_power_well_ops,
- .id = DISP_PW_ID_NONE,
- },
- {
.name = "power well 3",
.domains = ICL_PW_3_POWER_DOMAINS,
.ops = &hsw_power_well_ops,
@@ -3176,8 +3176,7 @@ static u8 intel_dbuf_max_slices(struct drm_i915_private *dev_priv)
void icl_dbuf_slices_update(struct drm_i915_private *dev_priv,
u8 req_slices)
{
- u8 hw_enabled_slices = dev_priv->wm.skl_hw.ddb.enabled_slices;
- u32 val;
+ const u8 hw_enabled_slices = dev_priv->wm.skl_hw.ddb.enabled_slices;
bool ret;
if (req_slices > intel_dbuf_max_slices(dev_priv)) {
@@ -3188,7 +3187,6 @@ void icl_dbuf_slices_update(struct drm_i915_private *dev_priv,
if (req_slices == hw_enabled_slices || req_slices == 0)
return;
- val = I915_READ(DBUF_CTL_S2);
if (req_slices > hw_enabled_slices)
ret = intel_dbuf_slice_set(dev_priv, DBUF_CTL_S2, true);
else
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 5fd2f7b..d3090a7 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -302,13 +302,65 @@ skl_plane_max_stride(struct intel_plane *plane,
return min(8192 * cpp, 32768);
}
+static void
+skl_program_scaler(struct intel_plane *plane,
+ const struct intel_crtc_state *crtc_state,
+ const struct intel_plane_state *plane_state)
+{
+ struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
+ enum pipe pipe = plane->pipe;
+ int scaler_id = plane_state->scaler_id;
+ const struct intel_scaler *scaler =
+ &crtc_state->scaler_state.scalers[scaler_id];
+ int crtc_x = plane_state->base.dst.x1;
+ int crtc_y = plane_state->base.dst.y1;
+ uint32_t crtc_w = drm_rect_width(&plane_state->base.dst);
+ uint32_t crtc_h = drm_rect_height(&plane_state->base.dst);
+ u16 y_hphase, uv_rgb_hphase;
+ u16 y_vphase, uv_rgb_vphase;
+ int hscale, vscale;
+
+ hscale = drm_rect_calc_hscale(&plane_state->base.src,
+ &plane_state->base.dst,
+ 0, INT_MAX);
+ vscale = drm_rect_calc_vscale(&plane_state->base.src,
+ &plane_state->base.dst,
+ 0, INT_MAX);
+
+ /* TODO: handle sub-pixel coordinates */
+ if (plane_state->base.fb->format->format == DRM_FORMAT_NV12) {
+ y_hphase = skl_scaler_calc_phase(1, hscale, false);
+ y_vphase = skl_scaler_calc_phase(1, vscale, false);
+
+ /* MPEG2 chroma siting convention */
+ uv_rgb_hphase = skl_scaler_calc_phase(2, hscale, true);
+ uv_rgb_vphase = skl_scaler_calc_phase(2, vscale, false);
+ } else {
+ /* not used */
+ y_hphase = 0;
+ y_vphase = 0;
+
+ uv_rgb_hphase = skl_scaler_calc_phase(1, hscale, false);
+ uv_rgb_vphase = skl_scaler_calc_phase(1, vscale, false);
+ }
+
+ I915_WRITE_FW(SKL_PS_CTRL(pipe, scaler_id),
+ PS_SCALER_EN | PS_PLANE_SEL(plane->id) | scaler->mode);
+ I915_WRITE_FW(SKL_PS_PWR_GATE(pipe, scaler_id), 0);
+ I915_WRITE_FW(SKL_PS_VPHASE(pipe, scaler_id),
+ PS_Y_PHASE(y_vphase) | PS_UV_RGB_PHASE(uv_rgb_vphase));
+ I915_WRITE_FW(SKL_PS_HPHASE(pipe, scaler_id),
+ PS_Y_PHASE(y_hphase) | PS_UV_RGB_PHASE(uv_rgb_hphase));
+ I915_WRITE_FW(SKL_PS_WIN_POS(pipe, scaler_id), (crtc_x << 16) | crtc_y);
+ I915_WRITE_FW(SKL_PS_WIN_SZ(pipe, scaler_id), (crtc_w << 16) | crtc_h);
+}
+
void
skl_update_plane(struct intel_plane *plane,
const struct intel_crtc_state *crtc_state,
const struct intel_plane_state *plane_state)
{
struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
- const struct drm_framebuffer *fb = plane_state->base.fb;
enum plane_id plane_id = plane->id;
enum pipe pipe = plane->pipe;
u32 plane_ctl = plane_state->ctl;
@@ -318,8 +370,6 @@ skl_update_plane(struct intel_plane *plane,
u32 aux_stride = skl_plane_stride(plane_state, 1);
int crtc_x = plane_state->base.dst.x1;
int crtc_y = plane_state->base.dst.y1;
- uint32_t crtc_w = drm_rect_width(&plane_state->base.dst);
- uint32_t crtc_h = drm_rect_height(&plane_state->base.dst);
uint32_t x = plane_state->color_plane[0].x;
uint32_t y = plane_state->color_plane[0].y;
uint32_t src_w = drm_rect_width(&plane_state->base.src) >> 16;
@@ -329,8 +379,6 @@ skl_update_plane(struct intel_plane *plane,
/* Sizes are 0 based */
src_w--;
src_h--;
- crtc_w--;
- crtc_h--;
spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
@@ -353,41 +401,8 @@ skl_update_plane(struct intel_plane *plane,
(plane_state->color_plane[1].y << 16) |
plane_state->color_plane[1].x);
- /* program plane scaler */
if (plane_state->scaler_id >= 0) {
- int scaler_id = plane_state->scaler_id;
- const struct intel_scaler *scaler =
- &crtc_state->scaler_state.scalers[scaler_id];
- u16 y_hphase, uv_rgb_hphase;
- u16 y_vphase, uv_rgb_vphase;
-
- /* TODO: handle sub-pixel coordinates */
- if (fb->format->format == DRM_FORMAT_NV12) {
- y_hphase = skl_scaler_calc_phase(1, false);
- y_vphase = skl_scaler_calc_phase(1, false);
-
- /* MPEG2 chroma siting convention */
- uv_rgb_hphase = skl_scaler_calc_phase(2, true);
- uv_rgb_vphase = skl_scaler_calc_phase(2, false);
- } else {
- /* not used */
- y_hphase = 0;
- y_vphase = 0;
-
- uv_rgb_hphase = skl_scaler_calc_phase(1, false);
- uv_rgb_vphase = skl_scaler_calc_phase(1, false);
- }
-
- I915_WRITE_FW(SKL_PS_CTRL(pipe, scaler_id),
- PS_SCALER_EN | PS_PLANE_SEL(plane_id) | scaler->mode);
- I915_WRITE_FW(SKL_PS_PWR_GATE(pipe, scaler_id), 0);
- I915_WRITE_FW(SKL_PS_VPHASE(pipe, scaler_id),
- PS_Y_PHASE(y_vphase) | PS_UV_RGB_PHASE(uv_rgb_vphase));
- I915_WRITE_FW(SKL_PS_HPHASE(pipe, scaler_id),
- PS_Y_PHASE(y_hphase) | PS_UV_RGB_PHASE(uv_rgb_hphase));
- I915_WRITE_FW(SKL_PS_WIN_POS(pipe, scaler_id), (crtc_x << 16) | crtc_y);
- I915_WRITE_FW(SKL_PS_WIN_SZ(pipe, scaler_id),
- ((crtc_w + 1) << 16)|(crtc_h + 1));
+ skl_program_scaler(plane, crtc_state, plane_state);
I915_WRITE_FW(PLANE_POS(pipe, plane_id), 0);
} else {
diff --git a/drivers/gpu/drm/i915/selftests/huge_pages.c b/drivers/gpu/drm/i915/selftests/huge_pages.c
index 8d03f64..5c22f2c 100644
--- a/drivers/gpu/drm/i915/selftests/huge_pages.c
+++ b/drivers/gpu/drm/i915/selftests/huge_pages.c
@@ -551,7 +551,7 @@ static int igt_mock_ppgtt_misaligned_dma(void *arg)
err = igt_check_page_sizes(vma);
if (vma->page_sizes.gtt != I915_GTT_PAGE_SIZE_4K) {
- pr_err("page_sizes.gtt=%u, expected %lu\n",
+ pr_err("page_sizes.gtt=%u, expected %llu\n",
vma->page_sizes.gtt, I915_GTT_PAGE_SIZE_4K);
err = -EINVAL;
}
diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index 8e2e269..127d815 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -1337,7 +1337,7 @@ static int igt_gtt_reserve(void *arg)
GEM_BUG_ON(!drm_mm_node_allocated(&vma->node));
if (vma->node.start != total ||
vma->node.size != 2*I915_GTT_PAGE_SIZE) {
- pr_err("i915_gem_gtt_reserve (pass 1) placement failed, found (%llx + %llx), expected (%llx + %lx)\n",
+ pr_err("i915_gem_gtt_reserve (pass 1) placement failed, found (%llx + %llx), expected (%llx + %llx)\n",
vma->node.start, vma->node.size,
total, 2*I915_GTT_PAGE_SIZE);
err = -EINVAL;
@@ -1386,7 +1386,7 @@ static int igt_gtt_reserve(void *arg)
GEM_BUG_ON(!drm_mm_node_allocated(&vma->node));
if (vma->node.start != total ||
vma->node.size != 2*I915_GTT_PAGE_SIZE) {
- pr_err("i915_gem_gtt_reserve (pass 2) placement failed, found (%llx + %llx), expected (%llx + %lx)\n",
+ pr_err("i915_gem_gtt_reserve (pass 2) placement failed, found (%llx + %llx), expected (%llx + %llx)\n",
vma->node.start, vma->node.size,
total, 2*I915_GTT_PAGE_SIZE);
err = -EINVAL;
@@ -1430,7 +1430,7 @@ static int igt_gtt_reserve(void *arg)
GEM_BUG_ON(!drm_mm_node_allocated(&vma->node));
if (vma->node.start != offset ||
vma->node.size != 2*I915_GTT_PAGE_SIZE) {
- pr_err("i915_gem_gtt_reserve (pass 3) placement failed, found (%llx + %llx), expected (%llx + %lx)\n",
+ pr_err("i915_gem_gtt_reserve (pass 3) placement failed, found (%llx + %llx), expected (%llx + %llx)\n",
vma->node.start, vma->node.size,
offset, 2*I915_GTT_PAGE_SIZE);
err = -EINVAL;
diff --git a/drivers/gpu/drm/meson/meson_venc.c b/drivers/gpu/drm/meson/meson_venc.c
index 514245e..acbbad3 100644
--- a/drivers/gpu/drm/meson/meson_venc.c
+++ b/drivers/gpu/drm/meson/meson_venc.c
@@ -854,6 +854,13 @@ void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic,
unsigned int sof_lines;
unsigned int vsync_lines;
+ /* Use VENCI for 480i and 576i and double HDMI pixels */
+ if (mode->flags & DRM_MODE_FLAG_DBLCLK) {
+ hdmi_repeat = true;
+ use_enci = true;
+ venc_hdmi_latency = 1;
+ }
+
if (meson_venc_hdmi_supported_vic(vic)) {
vmode = meson_venc_hdmi_get_vic_vmode(vic);
if (!vmode) {
@@ -865,13 +872,7 @@ void meson_venc_hdmi_mode_set(struct meson_drm *priv, int vic,
} else {
meson_venc_hdmi_get_dmt_vmode(mode, &vmode_dmt);
vmode = &vmode_dmt;
- }
-
- /* Use VENCI for 480i and 576i and double HDMI pixels */
- if (mode->flags & DRM_MODE_FLAG_DBLCLK) {
- hdmi_repeat = true;
- use_enci = true;
- venc_hdmi_latency = 1;
+ use_enci = false;
}
/* Repeat VENC pixels for 480/576i/p, 720p50/60 and 1080p50/60 */
diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h
index 04f1dfb..0aaedc5 100644
--- a/drivers/gpu/drm/mgag200/mgag200_drv.h
+++ b/drivers/gpu/drm/mgag200/mgag200_drv.h
@@ -212,8 +212,6 @@ struct mga_device {
int fb_mtrr;
struct {
- struct drm_global_reference mem_global_ref;
- struct ttm_bo_global_ref bo_global_ref;
struct ttm_bo_device bdev;
} ttm;
diff --git a/drivers/gpu/drm/mgag200/mgag200_ttm.c b/drivers/gpu/drm/mgag200/mgag200_ttm.c
index 05570f0..d96a9b3 100644
--- a/drivers/gpu/drm/mgag200/mgag200_ttm.c
+++ b/drivers/gpu/drm/mgag200/mgag200_ttm.c
@@ -36,63 +36,6 @@ mgag200_bdev(struct ttm_bo_device *bd)
return container_of(bd, struct mga_device, ttm.bdev);
}
-static int
-mgag200_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void
-mgag200_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-static int mgag200_ttm_global_init(struct mga_device *ast)
-{
- struct drm_global_reference *global_ref;
- int r;
-
- global_ref = &ast->ttm.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &mgag200_ttm_mem_global_init;
- global_ref->release = &mgag200_ttm_mem_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM memory accounting "
- "subsystem.\n");
- return r;
- }
-
- ast->ttm.bo_global_ref.mem_glob =
- ast->ttm.mem_global_ref.object;
- global_ref = &ast->ttm.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- drm_global_item_unref(&ast->ttm.mem_global_ref);
- return r;
- }
- return 0;
-}
-
-static void
-mgag200_ttm_global_release(struct mga_device *ast)
-{
- if (ast->ttm.mem_global_ref.release == NULL)
- return;
-
- drm_global_item_unref(&ast->ttm.bo_global_ref.ref);
- drm_global_item_unref(&ast->ttm.mem_global_ref);
- ast->ttm.mem_global_ref.release = NULL;
-}
-
-
static void mgag200_bo_ttm_destroy(struct ttm_buffer_object *tbo)
{
struct mgag200_bo *bo;
@@ -232,12 +175,7 @@ int mgag200_mm_init(struct mga_device *mdev)
struct drm_device *dev = mdev->dev;
struct ttm_bo_device *bdev = &mdev->ttm.bdev;
- ret = mgag200_ttm_global_init(mdev);
- if (ret)
- return ret;
-
ret = ttm_bo_device_init(&mdev->ttm.bdev,
- mdev->ttm.bo_global_ref.ref.object,
&mgag200_bo_driver,
dev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET,
@@ -268,8 +206,6 @@ void mgag200_mm_fini(struct mga_device *mdev)
ttm_bo_device_release(&mdev->ttm.bdev);
- mgag200_ttm_global_release(mdev);
-
arch_io_free_memtype_wc(pci_resource_start(dev->pdev, 0),
pci_resource_len(dev->pdev, 0));
arch_phys_wc_del(mdev->fb_mtrr);
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index 0b2191f..d20b9ba 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -146,8 +146,6 @@ struct nouveau_drm {
/* TTM interface support */
struct {
- struct drm_global_reference mem_global_ref;
- struct ttm_bo_global_ref bo_global_ref;
struct ttm_bo_device bdev;
atomic_t validate_sequence;
int (*move)(struct nouveau_channel *,
diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c
index 8edb9f2..1543c2f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c
@@ -175,66 +175,6 @@ nouveau_ttm_mmap(struct file *filp, struct vm_area_struct *vma)
}
static int
-nouveau_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void
-nouveau_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-int
-nouveau_ttm_global_init(struct nouveau_drm *drm)
-{
- struct drm_global_reference *global_ref;
- int ret;
-
- global_ref = &drm->ttm.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &nouveau_ttm_mem_global_init;
- global_ref->release = &nouveau_ttm_mem_global_release;
-
- ret = drm_global_item_ref(global_ref);
- if (unlikely(ret != 0)) {
- DRM_ERROR("Failed setting up TTM memory accounting\n");
- drm->ttm.mem_global_ref.release = NULL;
- return ret;
- }
-
- drm->ttm.bo_global_ref.mem_glob = global_ref->object;
- global_ref = &drm->ttm.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
-
- ret = drm_global_item_ref(global_ref);
- if (unlikely(ret != 0)) {
- DRM_ERROR("Failed setting up TTM BO subsystem\n");
- drm_global_item_unref(&drm->ttm.mem_global_ref);
- drm->ttm.mem_global_ref.release = NULL;
- return ret;
- }
-
- return 0;
-}
-
-void
-nouveau_ttm_global_release(struct nouveau_drm *drm)
-{
- if (drm->ttm.mem_global_ref.release == NULL)
- return;
-
- drm_global_item_unref(&drm->ttm.bo_global_ref.ref);
- drm_global_item_unref(&drm->ttm.mem_global_ref);
- drm->ttm.mem_global_ref.release = NULL;
-}
-
-static int
nouveau_ttm_init_host(struct nouveau_drm *drm, u8 kind)
{
struct nvif_mmu *mmu = &drm->client.mmu;
@@ -296,12 +236,7 @@ nouveau_ttm_init(struct nouveau_drm *drm)
drm->agp.cma = pci->agp.cma;
}
- ret = nouveau_ttm_global_init(drm);
- if (ret)
- return ret;
-
ret = ttm_bo_device_init(&drm->ttm.bdev,
- drm->ttm.bo_global_ref.ref.object,
&nouveau_bo_driver,
dev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET,
@@ -356,8 +291,6 @@ nouveau_ttm_fini(struct nouveau_drm *drm)
ttm_bo_device_release(&drm->ttm.bdev);
- nouveau_ttm_global_release(drm);
-
arch_phys_wc_del(drm->ttm.mtrr);
drm->ttm.mtrr = 0;
arch_io_free_memtype_wc(device->func->resource_addr(device, 1),
diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c
index 394c129..0a485c5 100644
--- a/drivers/gpu/drm/omapdrm/dss/dsi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dsi.c
@@ -5409,11 +5409,14 @@ static int dsi_probe(struct platform_device *pdev)
/* DSI on OMAP3 doesn't have register DSI_GNQ, set number
* of data to 3 by default */
- if (dsi->data->quirks & DSI_QUIRK_GNQ)
+ if (dsi->data->quirks & DSI_QUIRK_GNQ) {
+ dsi_runtime_get(dsi);
/* NB_DATA_LANES */
dsi->num_lanes_supported = 1 + REG_GET(dsi, DSI_GNQ, 11, 9);
- else
+ dsi_runtime_put(dsi);
+ } else {
dsi->num_lanes_supported = 3;
+ }
r = dsi_init_output(dsi);
if (r)
@@ -5426,15 +5429,19 @@ static int dsi_probe(struct platform_device *pdev)
}
r = of_platform_populate(dev->of_node, NULL, NULL, dev);
- if (r)
+ if (r) {
DSSERR("Failed to populate DSI child devices: %d\n", r);
+ goto err_uninit_output;
+ }
r = component_add(&pdev->dev, &dsi_component_ops);
if (r)
- goto err_uninit_output;
+ goto err_of_depopulate;
return 0;
+err_of_depopulate:
+ of_platform_depopulate(dev);
err_uninit_output:
dsi_uninit_output(dsi);
err_pm_disable:
@@ -5470,19 +5477,12 @@ static int dsi_runtime_suspend(struct device *dev)
/* wait for current handler to finish before turning the DSI off */
synchronize_irq(dsi->irq);
- dispc_runtime_put(dsi->dss->dispc);
-
return 0;
}
static int dsi_runtime_resume(struct device *dev)
{
struct dsi_data *dsi = dev_get_drvdata(dev);
- int r;
-
- r = dispc_runtime_get(dsi->dss->dispc);
- if (r)
- return r;
dsi->is_enabled = true;
/* ensure the irq handler sees the is_enabled value */
diff --git a/drivers/gpu/drm/omapdrm/dss/dss.c b/drivers/gpu/drm/omapdrm/dss/dss.c
index 1aaf260..7553c7f 100644
--- a/drivers/gpu/drm/omapdrm/dss/dss.c
+++ b/drivers/gpu/drm/omapdrm/dss/dss.c
@@ -1484,16 +1484,23 @@ static int dss_probe(struct platform_device *pdev)
dss);
/* Add all the child devices as components. */
+ r = of_platform_populate(pdev->dev.of_node, NULL, NULL, &pdev->dev);
+ if (r)
+ goto err_uninit_debugfs;
+
omapdss_gather_components(&pdev->dev);
device_for_each_child(&pdev->dev, &match, dss_add_child_component);
r = component_master_add_with_match(&pdev->dev, &dss_component_ops, match);
if (r)
- goto err_uninit_debugfs;
+ goto err_of_depopulate;
return 0;
+err_of_depopulate:
+ of_platform_depopulate(&pdev->dev);
+
err_uninit_debugfs:
dss_debugfs_remove_file(dss->debugfs.clk);
dss_debugfs_remove_file(dss->debugfs.dss);
@@ -1522,6 +1529,8 @@ static int dss_remove(struct platform_device *pdev)
{
struct dss_device *dss = platform_get_drvdata(pdev);
+ of_platform_depopulate(&pdev->dev);
+
component_master_del(&pdev->dev, &dss_component_ops);
dss_debugfs_remove_file(dss->debugfs.clk);
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
index cf6230e..aabdda3 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
@@ -635,10 +635,14 @@ static int hdmi4_bind(struct device *dev, struct device *master, void *data)
hdmi->dss = dss;
- r = hdmi_pll_init(dss, hdmi->pdev, &hdmi->pll, &hdmi->wp);
+ r = hdmi_runtime_get(hdmi);
if (r)
return r;
+ r = hdmi_pll_init(dss, hdmi->pdev, &hdmi->pll, &hdmi->wp);
+ if (r)
+ goto err_runtime_put;
+
r = hdmi4_cec_init(hdmi->pdev, &hdmi->core, &hdmi->wp);
if (r)
goto err_pll_uninit;
@@ -652,12 +656,16 @@ static int hdmi4_bind(struct device *dev, struct device *master, void *data)
hdmi->debugfs = dss_debugfs_create_file(dss, "hdmi", hdmi_dump_regs,
hdmi);
+ hdmi_runtime_put(hdmi);
+
return 0;
err_cec_uninit:
hdmi4_cec_uninit(&hdmi->core);
err_pll_uninit:
hdmi_pll_uninit(&hdmi->pll);
+err_runtime_put:
+ hdmi_runtime_put(hdmi);
return r;
}
@@ -833,32 +841,6 @@ static int hdmi4_remove(struct platform_device *pdev)
return 0;
}
-static int hdmi_runtime_suspend(struct device *dev)
-{
- struct omap_hdmi *hdmi = dev_get_drvdata(dev);
-
- dispc_runtime_put(hdmi->dss->dispc);
-
- return 0;
-}
-
-static int hdmi_runtime_resume(struct device *dev)
-{
- struct omap_hdmi *hdmi = dev_get_drvdata(dev);
- int r;
-
- r = dispc_runtime_get(hdmi->dss->dispc);
- if (r < 0)
- return r;
-
- return 0;
-}
-
-static const struct dev_pm_ops hdmi_pm_ops = {
- .runtime_suspend = hdmi_runtime_suspend,
- .runtime_resume = hdmi_runtime_resume,
-};
-
static const struct of_device_id hdmi_of_match[] = {
{ .compatible = "ti,omap4-hdmi", },
{},
@@ -869,7 +851,6 @@ struct platform_driver omapdss_hdmi4hw_driver = {
.remove = hdmi4_remove,
.driver = {
.name = "omapdss_hdmi",
- .pm = &hdmi_pm_ops,
.of_match_table = hdmi_of_match,
.suppress_bind_attrs = true,
},
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
index b0e4a74..9e8556f 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
@@ -825,32 +825,6 @@ static int hdmi5_remove(struct platform_device *pdev)
return 0;
}
-static int hdmi_runtime_suspend(struct device *dev)
-{
- struct omap_hdmi *hdmi = dev_get_drvdata(dev);
-
- dispc_runtime_put(hdmi->dss->dispc);
-
- return 0;
-}
-
-static int hdmi_runtime_resume(struct device *dev)
-{
- struct omap_hdmi *hdmi = dev_get_drvdata(dev);
- int r;
-
- r = dispc_runtime_get(hdmi->dss->dispc);
- if (r < 0)
- return r;
-
- return 0;
-}
-
-static const struct dev_pm_ops hdmi_pm_ops = {
- .runtime_suspend = hdmi_runtime_suspend,
- .runtime_resume = hdmi_runtime_resume,
-};
-
static const struct of_device_id hdmi_of_match[] = {
{ .compatible = "ti,omap5-hdmi", },
{ .compatible = "ti,dra7-hdmi", },
@@ -862,7 +836,6 @@ struct platform_driver omapdss_hdmi5hw_driver = {
.remove = hdmi5_remove,
.driver = {
.name = "omapdss_hdmi5",
- .pm = &hdmi_pm_ops,
.of_match_table = hdmi_of_match,
.suppress_bind_attrs = true,
},
diff --git a/drivers/gpu/drm/omapdrm/dss/venc.c b/drivers/gpu/drm/omapdrm/dss/venc.c
index ff0b18c..b5f5272 100644
--- a/drivers/gpu/drm/omapdrm/dss/venc.c
+++ b/drivers/gpu/drm/omapdrm/dss/venc.c
@@ -946,19 +946,12 @@ static int venc_runtime_suspend(struct device *dev)
if (venc->tv_dac_clk)
clk_disable_unprepare(venc->tv_dac_clk);
- dispc_runtime_put(venc->dss->dispc);
-
return 0;
}
static int venc_runtime_resume(struct device *dev)
{
struct venc_device *venc = dev_get_drvdata(dev);
- int r;
-
- r = dispc_runtime_get(venc->dss->dispc);
- if (r < 0)
- return r;
if (venc->tv_dac_clk)
clk_prepare_enable(venc->tv_dac_clk);
diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index 62928ec..caffc54 100644
--- a/drivers/gpu/drm/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -350,11 +350,14 @@ static void omap_crtc_arm_event(struct drm_crtc *crtc)
static void omap_crtc_atomic_enable(struct drm_crtc *crtc,
struct drm_crtc_state *old_state)
{
+ struct omap_drm_private *priv = crtc->dev->dev_private;
struct omap_crtc *omap_crtc = to_omap_crtc(crtc);
int ret;
DBG("%s", omap_crtc->name);
+ priv->dispc_ops->runtime_get(priv->dispc);
+
spin_lock_irq(&crtc->dev->event_lock);
drm_crtc_vblank_on(crtc);
ret = drm_crtc_vblank_get(crtc);
@@ -367,6 +370,7 @@ static void omap_crtc_atomic_enable(struct drm_crtc *crtc,
static void omap_crtc_atomic_disable(struct drm_crtc *crtc,
struct drm_crtc_state *old_state)
{
+ struct omap_drm_private *priv = crtc->dev->dev_private;
struct omap_crtc *omap_crtc = to_omap_crtc(crtc);
DBG("%s", omap_crtc->name);
@@ -379,6 +383,8 @@ static void omap_crtc_atomic_disable(struct drm_crtc *crtc,
spin_unlock_irq(&crtc->dev->event_lock);
drm_crtc_vblank_off(crtc);
+
+ priv->dispc_ops->runtime_put(priv->dispc);
}
static enum drm_mode_status omap_crtc_mode_valid(struct drm_crtc *crtc,
diff --git a/drivers/gpu/drm/qxl/qxl_drv.h b/drivers/gpu/drm/qxl/qxl_drv.h
index 14d3fa8..13a0254 100644
--- a/drivers/gpu/drm/qxl/qxl_drv.h
+++ b/drivers/gpu/drm/qxl/qxl_drv.h
@@ -126,9 +126,6 @@ struct qxl_output {
#define drm_encoder_to_qxl_output(x) container_of(x, struct qxl_output, enc)
struct qxl_mman {
- struct ttm_bo_global_ref bo_global_ref;
- struct drm_global_reference mem_global_ref;
- unsigned int mem_global_referenced:1;
struct ttm_bo_device bdev;
};
diff --git a/drivers/gpu/drm/qxl/qxl_ttm.c b/drivers/gpu/drm/qxl/qxl_ttm.c
index 559a101..886f61e 100644
--- a/drivers/gpu/drm/qxl/qxl_ttm.c
+++ b/drivers/gpu/drm/qxl/qxl_ttm.c
@@ -46,62 +46,6 @@ static struct qxl_device *qxl_get_qdev(struct ttm_bo_device *bdev)
return qdev;
}
-static int qxl_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void qxl_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-static int qxl_ttm_global_init(struct qxl_device *qdev)
-{
- struct drm_global_reference *global_ref;
- int r;
-
- qdev->mman.mem_global_referenced = false;
- global_ref = &qdev->mman.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &qxl_ttm_mem_global_init;
- global_ref->release = &qxl_ttm_mem_global_release;
-
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM memory accounting "
- "subsystem.\n");
- return r;
- }
-
- qdev->mman.bo_global_ref.mem_glob =
- qdev->mman.mem_global_ref.object;
- global_ref = &qdev->mman.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- drm_global_item_unref(&qdev->mman.mem_global_ref);
- return r;
- }
-
- qdev->mman.mem_global_referenced = true;
- return 0;
-}
-
-static void qxl_ttm_global_fini(struct qxl_device *qdev)
-{
- if (qdev->mman.mem_global_referenced) {
- drm_global_item_unref(&qdev->mman.bo_global_ref.ref);
- drm_global_item_unref(&qdev->mman.mem_global_ref);
- qdev->mman.mem_global_referenced = false;
- }
-}
-
static struct vm_operations_struct qxl_ttm_vm_ops;
static const struct vm_operations_struct *ttm_vm_ops;
@@ -372,12 +316,8 @@ int qxl_ttm_init(struct qxl_device *qdev)
int r;
int num_io_pages; /* != rom->num_io_pages, we include surface0 */
- r = qxl_ttm_global_init(qdev);
- if (r)
- return r;
/* No others user of address space so set it to 0 */
r = ttm_bo_device_init(&qdev->mman.bdev,
- qdev->mman.bo_global_ref.ref.object,
&qxl_bo_driver,
qdev->ddev.anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET, 0);
@@ -413,7 +353,6 @@ void qxl_ttm_fini(struct qxl_device *qdev)
ttm_bo_clean_mm(&qdev->mman.bdev, TTM_PL_VRAM);
ttm_bo_clean_mm(&qdev->mman.bdev, TTM_PL_PRIV);
ttm_bo_device_release(&qdev->mman.bdev);
- qxl_ttm_global_fini(qdev);
DRM_INFO("qxl: ttm finalized\n");
}
diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c
index 21161aa..652126f 100644
--- a/drivers/gpu/drm/radeon/r300.c
+++ b/drivers/gpu/drm/radeon/r300.c
@@ -814,7 +814,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
((idx_value >> 21) & 0xF));
return -EINVAL;
}
- /* Pass through. */
+ /* Fall through. */
case 6:
track->cb[i].cpp = 4;
break;
@@ -965,7 +965,7 @@ static int r300_packet0_check(struct radeon_cs_parser *p,
return -EINVAL;
}
/* The same rules apply as for DXT3/5. */
- /* Pass through. */
+ /* Fall through. */
case R300_TX_FORMAT_DXT3:
case R300_TX_FORMAT_DXT5:
track->textures[i].cpp = 1;
diff --git a/drivers/gpu/drm/radeon/r420.c b/drivers/gpu/drm/radeon/r420.c
index 45e1d4e..2318d9e3 100644
--- a/drivers/gpu/drm/radeon/r420.c
+++ b/drivers/gpu/drm/radeon/r420.c
@@ -109,6 +109,7 @@ void r420_pipes_init(struct radeon_device *rdev)
default:
/* force to 1 pipe */
num_pipes = 1;
+ /* fall through */
case 1:
tmp = (0 << 1);
break;
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 1a6f6ed..32808e5 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -448,10 +448,7 @@ struct radeon_surface_reg {
* TTM.
*/
struct radeon_mman {
- struct ttm_bo_global_ref bo_global_ref;
- struct drm_global_reference mem_global_ref;
struct ttm_bo_device bdev;
- bool mem_global_referenced;
bool initialized;
#if defined(CONFIG_DEBUG_FS)
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_tv.c b/drivers/gpu/drm/radeon/radeon_legacy_tv.c
index 4278272..3dae2c4 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_tv.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_tv.c
@@ -421,24 +421,14 @@ static void radeon_legacy_write_tv_restarts(struct radeon_encoder *radeon_encode
static bool radeon_legacy_tv_init_restarts(struct drm_encoder *encoder)
{
- struct drm_device *dev = encoder->dev;
- struct radeon_device *rdev = dev->dev_private;
struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
struct radeon_encoder_tv_dac *tv_dac = radeon_encoder->enc_priv;
- struct radeon_crtc *radeon_crtc;
int restart;
unsigned int h_total, v_total, f_total;
int v_offset, h_offset;
u16 p1, p2, h_inc;
bool h_changed;
const struct radeon_tv_mode_constants *const_ptr;
- struct radeon_pll *pll;
-
- radeon_crtc = to_radeon_crtc(radeon_encoder->base.crtc);
- if (radeon_crtc->crtc_id == 1)
- pll = &rdev->clock.p2pll;
- else
- pll = &rdev->clock.p1pll;
const_ptr = radeon_legacy_tv_get_std_mode(radeon_encoder, NULL);
if (!const_ptr)
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 92f6d40..833e909 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -314,11 +314,9 @@ struct radeon_bo *radeon_bo_ref(struct radeon_bo *bo)
void radeon_bo_unref(struct radeon_bo **bo)
{
struct ttm_buffer_object *tbo;
- struct radeon_device *rdev;
if ((*bo) == NULL)
return;
- rdev = (*bo)->rdev;
tbo = &((*bo)->tbo);
ttm_bo_put(tbo);
*bo = NULL;
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index cbb67e9..9920a6f 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -60,65 +60,6 @@ static struct radeon_device *radeon_get_rdev(struct ttm_bo_device *bdev)
return rdev;
}
-
-/*
- * Global memory.
- */
-static int radeon_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void radeon_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-static int radeon_ttm_global_init(struct radeon_device *rdev)
-{
- struct drm_global_reference *global_ref;
- int r;
-
- rdev->mman.mem_global_referenced = false;
- global_ref = &rdev->mman.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &radeon_ttm_mem_global_init;
- global_ref->release = &radeon_ttm_mem_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM memory accounting "
- "subsystem.\n");
- return r;
- }
-
- rdev->mman.bo_global_ref.mem_glob =
- rdev->mman.mem_global_ref.object;
- global_ref = &rdev->mman.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- drm_global_item_unref(&rdev->mman.mem_global_ref);
- return r;
- }
-
- rdev->mman.mem_global_referenced = true;
- return 0;
-}
-
-static void radeon_ttm_global_fini(struct radeon_device *rdev)
-{
- if (rdev->mman.mem_global_referenced) {
- drm_global_item_unref(&rdev->mman.bo_global_ref.ref);
- drm_global_item_unref(&rdev->mman.mem_global_ref);
- rdev->mman.mem_global_referenced = false;
- }
-}
-
static int radeon_invalidate_caches(struct ttm_bo_device *bdev, uint32_t flags)
{
return 0;
@@ -847,13 +788,8 @@ int radeon_ttm_init(struct radeon_device *rdev)
{
int r;
- r = radeon_ttm_global_init(rdev);
- if (r) {
- return r;
- }
/* No others user of address space so set it to 0 */
r = ttm_bo_device_init(&rdev->mman.bdev,
- rdev->mman.bo_global_ref.ref.object,
&radeon_bo_driver,
rdev->ddev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET,
@@ -925,7 +861,6 @@ void radeon_ttm_fini(struct radeon_device *rdev)
ttm_bo_clean_mm(&rdev->mman.bdev, TTM_PL_TT);
ttm_bo_device_release(&rdev->mman.bdev);
radeon_gart_fini(rdev);
- radeon_ttm_global_fini(rdev);
rdev->mman.initialized = false;
DRM_INFO("radeon: ttm finalized\n");
}
diff --git a/drivers/gpu/drm/scheduler/sched_entity.c b/drivers/gpu/drm/scheduler/sched_entity.c
index 3e22a54..4463d38 100644
--- a/drivers/gpu/drm/scheduler/sched_entity.c
+++ b/drivers/gpu/drm/scheduler/sched_entity.c
@@ -130,7 +130,14 @@ drm_sched_entity_get_free_sched(struct drm_sched_entity *entity)
int i;
for (i = 0; i < entity->num_rq_list; ++i) {
- num_jobs = atomic_read(&entity->rq_list[i]->sched->num_jobs);
+ struct drm_gpu_scheduler *sched = entity->rq_list[i]->sched;
+
+ if (!entity->rq_list[i]->sched->ready) {
+ DRM_WARN("sched%s is not ready, skipping", sched->name);
+ continue;
+ }
+
+ num_jobs = atomic_read(&sched->num_jobs);
if (num_jobs < min_jobs) {
min_jobs = num_jobs;
rq = entity->rq_list[i];
@@ -204,7 +211,6 @@ static void drm_sched_entity_kill_jobs_cb(struct dma_fence *f,
drm_sched_fence_finished(job->s_fence);
WARN_ON(job->s_fence->parent);
- dma_fence_put(&job->s_fence->finished);
job->sched->ops->free_job(job);
}
diff --git a/drivers/gpu/drm/scheduler/sched_main.c b/drivers/gpu/drm/scheduler/sched_main.c
index 44fe587..18ebbb0 100644
--- a/drivers/gpu/drm/scheduler/sched_main.c
+++ b/drivers/gpu/drm/scheduler/sched_main.c
@@ -196,6 +196,19 @@ static void drm_sched_start_timeout(struct drm_gpu_scheduler *sched)
schedule_delayed_work(&sched->work_tdr, sched->timeout);
}
+/**
+ * drm_sched_fault - immediately start timeout handler
+ *
+ * @sched: scheduler where the timeout handling should be started.
+ *
+ * Start timeout handling immediately when the driver detects a hardware fault.
+ */
+void drm_sched_fault(struct drm_gpu_scheduler *sched)
+{
+ mod_delayed_work(system_wq, &sched->work_tdr, 0);
+}
+EXPORT_SYMBOL(drm_sched_fault);
+
/* job_finish is called after hw fence signaled
*/
static void drm_sched_job_finish(struct work_struct *work)
@@ -220,7 +233,6 @@ static void drm_sched_job_finish(struct work_struct *work)
drm_sched_start_timeout(sched);
spin_unlock(&sched->job_list_lock);
- dma_fence_put(&s_job->s_fence->finished);
sched->ops->free_job(s_job);
}
@@ -283,6 +295,7 @@ static void drm_sched_job_timedout(struct work_struct *work)
already_signaled:
;
}
+ drm_sched_start_timeout(sched);
spin_unlock(&sched->job_list_lock);
}
@@ -406,6 +419,9 @@ int drm_sched_job_init(struct drm_sched_job *job,
struct drm_gpu_scheduler *sched;
drm_sched_entity_select_rq(entity);
+ if (!entity->rq)
+ return -ENOENT;
+
sched = entity->rq->sched;
job->sched = sched;
@@ -424,6 +440,18 @@ int drm_sched_job_init(struct drm_sched_job *job,
EXPORT_SYMBOL(drm_sched_job_init);
/**
+ * drm_sched_job_cleanup - clean up scheduler job resources
+ *
+ * @job: scheduler job to clean up
+ */
+void drm_sched_job_cleanup(struct drm_sched_job *job)
+{
+ dma_fence_put(&job->s_fence->finished);
+ job->s_fence = NULL;
+}
+EXPORT_SYMBOL(drm_sched_job_cleanup);
+
+/**
* drm_sched_ready - is the scheduler ready
*
* @sched: scheduler instance
@@ -619,6 +647,7 @@ int drm_sched_init(struct drm_gpu_scheduler *sched,
return PTR_ERR(sched->thread);
}
+ sched->ready = true;
return 0;
}
EXPORT_SYMBOL(drm_sched_init);
@@ -634,5 +663,7 @@ void drm_sched_fini(struct drm_gpu_scheduler *sched)
{
if (sched->thread)
kthread_stop(sched->thread);
+
+ sched->ready = false;
}
EXPORT_SYMBOL(drm_sched_fini);
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 83b4657..d87935b 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -45,6 +45,14 @@
static void ttm_bo_global_kobj_release(struct kobject *kobj);
+/**
+ * ttm_global_mutex - protecting the global BO state
+ */
+DEFINE_MUTEX(ttm_global_mutex);
+struct ttm_bo_global ttm_bo_glob = {
+ .use_count = 0
+};
+
static struct attribute ttm_bo_count = {
.name = "bo_count",
.mode = S_IRUGO
@@ -1519,35 +1527,45 @@ static void ttm_bo_global_kobj_release(struct kobject *kobj)
container_of(kobj, struct ttm_bo_global, kobj);
__free_page(glob->dummy_read_page);
- kfree(glob);
}
-void ttm_bo_global_release(struct drm_global_reference *ref)
+static void ttm_bo_global_release(void)
{
- struct ttm_bo_global *glob = ref->object;
+ struct ttm_bo_global *glob = &ttm_bo_glob;
+
+ mutex_lock(&ttm_global_mutex);
+ if (--glob->use_count > 0)
+ goto out;
kobject_del(&glob->kobj);
kobject_put(&glob->kobj);
+ ttm_mem_global_release(&ttm_mem_glob);
+out:
+ mutex_unlock(&ttm_global_mutex);
}
-EXPORT_SYMBOL(ttm_bo_global_release);
-int ttm_bo_global_init(struct drm_global_reference *ref)
+static int ttm_bo_global_init(void)
{
- struct ttm_bo_global_ref *bo_ref =
- container_of(ref, struct ttm_bo_global_ref, ref);
- struct ttm_bo_global *glob = ref->object;
- int ret;
+ struct ttm_bo_global *glob = &ttm_bo_glob;
+ int ret = 0;
unsigned i;
- mutex_init(&glob->device_list_mutex);
+ mutex_lock(&ttm_global_mutex);
+ if (++glob->use_count > 1)
+ goto out;
+
+ ret = ttm_mem_global_init(&ttm_mem_glob);
+ if (ret)
+ goto out;
+
spin_lock_init(&glob->lru_lock);
- glob->mem_glob = bo_ref->mem_glob;
+ glob->mem_glob = &ttm_mem_glob;
glob->mem_glob->bo_glob = glob;
glob->dummy_read_page = alloc_page(__GFP_ZERO | GFP_DMA32);
if (unlikely(glob->dummy_read_page == NULL)) {
ret = -ENOMEM;
- goto out_no_drp;
+ goto out;
}
for (i = 0; i < TTM_MAX_BO_PRIORITY; ++i)
@@ -1559,13 +1577,10 @@ int ttm_bo_global_init(struct drm_global_reference *ref)
&glob->kobj, &ttm_bo_glob_kobj_type, ttm_get_kobj(), "buffer_objects");
if (unlikely(ret != 0))
kobject_put(&glob->kobj);
- return ret;
-out_no_drp:
- kfree(glob);
+out:
+ mutex_unlock(&ttm_global_mutex);
return ret;
}
-EXPORT_SYMBOL(ttm_bo_global_init);
-
int ttm_bo_device_release(struct ttm_bo_device *bdev)
{
@@ -1587,9 +1602,9 @@ int ttm_bo_device_release(struct ttm_bo_device *bdev)
}
}
- mutex_lock(&glob->device_list_mutex);
+ mutex_lock(&ttm_global_mutex);
list_del(&bdev->device_list);
- mutex_unlock(&glob->device_list_mutex);
+ mutex_unlock(&ttm_global_mutex);
cancel_delayed_work_sync(&bdev->wq);
@@ -1604,18 +1619,25 @@ int ttm_bo_device_release(struct ttm_bo_device *bdev)
drm_vma_offset_manager_destroy(&bdev->vma_manager);
+ if (!ret)
+ ttm_bo_global_release();
+
return ret;
}
EXPORT_SYMBOL(ttm_bo_device_release);
int ttm_bo_device_init(struct ttm_bo_device *bdev,
- struct ttm_bo_global *glob,
struct ttm_bo_driver *driver,
struct address_space *mapping,
uint64_t file_page_offset,
bool need_dma32)
{
- int ret = -EINVAL;
+ struct ttm_bo_global *glob = &ttm_bo_glob;
+ int ret;
+
+ ret = ttm_bo_global_init();
+ if (ret)
+ return ret;
bdev->driver = driver;
@@ -1636,12 +1658,13 @@ int ttm_bo_device_init(struct ttm_bo_device *bdev,
bdev->dev_mapping = mapping;
bdev->glob = glob;
bdev->need_dma32 = need_dma32;
- mutex_lock(&glob->device_list_mutex);
+ mutex_lock(&ttm_global_mutex);
list_add_tail(&bdev->device_list, &glob->device_list);
- mutex_unlock(&glob->device_list_mutex);
+ mutex_unlock(&ttm_global_mutex);
return 0;
out_no_sys:
+ ttm_bo_global_release();
return ret;
}
EXPORT_SYMBOL(ttm_bo_device_init);
diff --git a/drivers/gpu/drm/ttm/ttm_execbuf_util.c b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
index e493edb..efa005a 100644
--- a/drivers/gpu/drm/ttm/ttm_execbuf_util.c
+++ b/drivers/gpu/drm/ttm/ttm_execbuf_util.c
@@ -187,14 +187,12 @@ void ttm_eu_fence_buffer_objects(struct ww_acquire_ctx *ticket,
struct ttm_buffer_object *bo;
struct ttm_bo_global *glob;
struct ttm_bo_device *bdev;
- struct ttm_bo_driver *driver;
if (list_empty(list))
return;
bo = list_first_entry(list, struct ttm_validate_buffer, head)->bo;
bdev = bo->bdev;
- driver = bdev->driver;
glob = bo->bdev->glob;
spin_lock(&glob->lru_lock);
diff --git a/drivers/gpu/drm/ttm/ttm_memory.c b/drivers/gpu/drm/ttm/ttm_memory.c
index 450387c..f1567c3 100644
--- a/drivers/gpu/drm/ttm/ttm_memory.c
+++ b/drivers/gpu/drm/ttm/ttm_memory.c
@@ -41,6 +41,9 @@
#define TTM_MEMORY_ALLOC_RETRIES 4
+struct ttm_mem_global ttm_mem_glob;
+EXPORT_SYMBOL(ttm_mem_glob);
+
struct ttm_mem_zone {
struct kobject kobj;
struct ttm_mem_global *glob;
@@ -216,14 +219,6 @@ static ssize_t ttm_mem_global_store(struct kobject *kobj,
return size;
}
-static void ttm_mem_global_kobj_release(struct kobject *kobj)
-{
- struct ttm_mem_global *glob =
- container_of(kobj, struct ttm_mem_global, kobj);
-
- kfree(glob);
-}
-
static struct attribute *ttm_mem_global_attrs[] = {
&ttm_mem_global_lower_mem_limit,
NULL
@@ -235,7 +230,6 @@ static const struct sysfs_ops ttm_mem_global_ops = {
};
static struct kobj_type ttm_mem_glob_kobj_type = {
- .release = &ttm_mem_global_kobj_release,
.sysfs_ops = &ttm_mem_global_ops,
.default_attrs = ttm_mem_global_attrs,
};
@@ -464,7 +458,6 @@ int ttm_mem_global_init(struct ttm_mem_global *glob)
ttm_mem_global_release(glob);
return ret;
}
-EXPORT_SYMBOL(ttm_mem_global_init);
void ttm_mem_global_release(struct ttm_mem_global *glob)
{
@@ -486,7 +479,6 @@ void ttm_mem_global_release(struct ttm_mem_global *glob)
kobject_del(&glob->kobj);
kobject_put(&glob->kobj);
}
-EXPORT_SYMBOL(ttm_mem_global_release);
static void ttm_check_swapping(struct ttm_mem_global *glob)
{
diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c
index 9243dea..445b2ef 100644
--- a/drivers/gpu/drm/v3d/v3d_sched.c
+++ b/drivers/gpu/drm/v3d/v3d_sched.c
@@ -35,6 +35,8 @@ v3d_job_free(struct drm_sched_job *sched_job)
{
struct v3d_job *job = to_v3d_job(sched_job);
+ drm_sched_job_cleanup(sched_job);
+
v3d_exec_put(job->exec);
}
@@ -167,9 +169,6 @@ v3d_job_timedout(struct drm_sched_job *sched_job)
if (job->timedout_ctca != ctca || job->timedout_ctra != ctra) {
job->timedout_ctca = ctca;
job->timedout_ctra = ctra;
-
- schedule_delayed_work(&job->base.sched->work_tdr,
- job->base.sched->timeout);
return;
}
diff --git a/drivers/gpu/drm/virtio/virtgpu_drv.h b/drivers/gpu/drm/virtio/virtgpu_drv.h
index 4632bd7..9db5680 100644
--- a/drivers/gpu/drm/virtio/virtgpu_drv.h
+++ b/drivers/gpu/drm/virtio/virtgpu_drv.h
@@ -144,9 +144,6 @@ struct virtio_gpu_fbdev {
};
struct virtio_gpu_mman {
- struct ttm_bo_global_ref bo_global_ref;
- struct drm_global_reference mem_global_ref;
- bool mem_global_referenced;
struct ttm_bo_device bdev;
};
diff --git a/drivers/gpu/drm/virtio/virtgpu_ttm.c b/drivers/gpu/drm/virtio/virtgpu_ttm.c
index cd63dff..4bfbf25fa 100644
--- a/drivers/gpu/drm/virtio/virtgpu_ttm.c
+++ b/drivers/gpu/drm/virtio/virtgpu_ttm.c
@@ -50,62 +50,6 @@ virtio_gpu_device *virtio_gpu_get_vgdev(struct ttm_bo_device *bdev)
return vgdev;
}
-static int virtio_gpu_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void virtio_gpu_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-static int virtio_gpu_ttm_global_init(struct virtio_gpu_device *vgdev)
-{
- struct drm_global_reference *global_ref;
- int r;
-
- vgdev->mman.mem_global_referenced = false;
- global_ref = &vgdev->mman.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &virtio_gpu_ttm_mem_global_init;
- global_ref->release = &virtio_gpu_ttm_mem_global_release;
-
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM memory accounting "
- "subsystem.\n");
- return r;
- }
-
- vgdev->mman.bo_global_ref.mem_glob =
- vgdev->mman.mem_global_ref.object;
- global_ref = &vgdev->mman.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- r = drm_global_item_ref(global_ref);
- if (r != 0) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- drm_global_item_unref(&vgdev->mman.mem_global_ref);
- return r;
- }
-
- vgdev->mman.mem_global_referenced = true;
- return 0;
-}
-
-static void virtio_gpu_ttm_global_fini(struct virtio_gpu_device *vgdev)
-{
- if (vgdev->mman.mem_global_referenced) {
- drm_global_item_unref(&vgdev->mman.bo_global_ref.ref);
- drm_global_item_unref(&vgdev->mman.mem_global_ref);
- vgdev->mman.mem_global_referenced = false;
- }
-}
-
int virtio_gpu_mmap(struct file *filp, struct vm_area_struct *vma)
{
struct drm_file *file_priv;
@@ -382,12 +326,8 @@ int virtio_gpu_ttm_init(struct virtio_gpu_device *vgdev)
{
int r;
- r = virtio_gpu_ttm_global_init(vgdev);
- if (r)
- return r;
/* No others user of address space so set it to 0 */
r = ttm_bo_device_init(&vgdev->mman.bdev,
- vgdev->mman.bo_global_ref.ref.object,
&virtio_gpu_bo_driver,
vgdev->ddev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET, 0);
@@ -406,13 +346,11 @@ int virtio_gpu_ttm_init(struct virtio_gpu_device *vgdev)
err_mm_init:
ttm_bo_device_release(&vgdev->mman.bdev);
err_dev_init:
- virtio_gpu_ttm_global_fini(vgdev);
return r;
}
void virtio_gpu_ttm_fini(struct virtio_gpu_device *vgdev)
{
ttm_bo_device_release(&vgdev->mman.bdev);
- virtio_gpu_ttm_global_fini(vgdev);
DRM_INFO("virtio_gpu: ttm finalized\n");
}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 61a84b9..b9c0788 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -801,11 +801,6 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
DRM_INFO("MMIO at 0x%08x size is %u kiB\n",
dev_priv->mmio_start, dev_priv->mmio_size / 1024);
- ret = vmw_ttm_global_init(dev_priv);
- if (unlikely(ret != 0))
- goto out_err0;
-
-
vmw_master_init(&dev_priv->fbdev_master);
ttm_lock_set_kill(&dev_priv->fbdev_master.lock, false, SIGTERM);
dev_priv->active_master = &dev_priv->fbdev_master;
@@ -816,7 +811,7 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
if (unlikely(dev_priv->mmio_virt == NULL)) {
ret = -ENOMEM;
DRM_ERROR("Failed mapping MMIO.\n");
- goto out_err3;
+ goto out_err0;
}
/* Need mmio memory to check for fifo pitchlock cap. */
@@ -828,8 +823,8 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
goto out_err4;
}
- dev_priv->tdev = ttm_object_device_init
- (dev_priv->mem_global_ref.object, 12, &vmw_prime_dmabuf_ops);
+ dev_priv->tdev = ttm_object_device_init(&ttm_mem_glob, 12,
+ &vmw_prime_dmabuf_ops);
if (unlikely(dev_priv->tdev == NULL)) {
DRM_ERROR("Unable to initialize TTM object management.\n");
@@ -870,7 +865,6 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
}
ret = ttm_bo_device_init(&dev_priv->bdev,
- dev_priv->bo_global_ref.ref.object,
&vmw_bo_driver,
dev->anon_inode->i_mapping,
VMWGFX_FILE_PAGE_OFFSET,
@@ -992,8 +986,6 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset)
ttm_object_device_release(&dev_priv->tdev);
out_err4:
memunmap(dev_priv->mmio_virt);
-out_err3:
- vmw_ttm_global_release(dev_priv);
out_err0:
for (i = vmw_res_context; i < vmw_res_max; ++i)
idr_destroy(&dev_priv->res_idr[i]);
@@ -1045,7 +1037,6 @@ static void vmw_driver_unload(struct drm_device *dev)
memunmap(dev_priv->mmio_virt);
if (dev_priv->ctx.staged_bindings)
vmw_binding_state_free(dev_priv->ctx.staged_bindings);
- vmw_ttm_global_release(dev_priv);
for (i = vmw_res_context; i < vmw_res_max; ++i)
idr_destroy(&dev_priv->res_idr[i]);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 59f6142..28df788 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -417,8 +417,6 @@ enum {
struct vmw_private {
struct ttm_bo_device bdev;
- struct ttm_bo_global_ref bo_global_ref;
- struct drm_global_reference mem_global_ref;
struct vmw_fifo_state fifo;
@@ -842,8 +840,6 @@ extern int vmw_fifo_flush(struct vmw_private *dev_priv,
* TTM glue - vmwgfx_ttm_glue.c
*/
-extern int vmw_ttm_global_init(struct vmw_private *dev_priv);
-extern void vmw_ttm_global_release(struct vmw_private *dev_priv);
extern int vmw_mmap(struct file *filp, struct vm_area_struct *vma);
/**
@@ -1363,7 +1359,7 @@ vmw_bo_reference(struct vmw_buffer_object *buf)
static inline struct ttm_mem_global *vmw_mem_glob(struct vmw_private *dev_priv)
{
- return (struct ttm_mem_global *) dev_priv->mem_global_ref.object;
+ return &ttm_mem_glob;
}
static inline void vmw_fifo_resource_inc(struct vmw_private *dev_priv)
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c
index 7b1e5a5..154eb09 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ttm_glue.c
@@ -42,57 +42,3 @@ int vmw_mmap(struct file *filp, struct vm_area_struct *vma)
dev_priv = vmw_priv(file_priv->minor->dev);
return ttm_bo_mmap(filp, vma, &dev_priv->bdev);
}
-
-static int vmw_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- DRM_INFO("global init.\n");
- return ttm_mem_global_init(ref->object);
-}
-
-static void vmw_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-int vmw_ttm_global_init(struct vmw_private *dev_priv)
-{
- struct drm_global_reference *global_ref;
- int ret;
-
- global_ref = &dev_priv->mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &vmw_ttm_mem_global_init;
- global_ref->release = &vmw_ttm_mem_global_release;
-
- ret = drm_global_item_ref(global_ref);
- if (unlikely(ret != 0)) {
- DRM_ERROR("Failed setting up TTM memory accounting.\n");
- return ret;
- }
-
- dev_priv->bo_global_ref.mem_glob =
- dev_priv->mem_global_ref.object;
- global_ref = &dev_priv->bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
- ret = drm_global_item_ref(global_ref);
-
- if (unlikely(ret != 0)) {
- DRM_ERROR("Failed setting up TTM buffer objects.\n");
- goto out_no_bo;
- }
-
- return 0;
-out_no_bo:
- drm_global_item_unref(&dev_priv->mem_global_ref);
- return ret;
-}
-
-void vmw_ttm_global_release(struct vmw_private *dev_priv)
-{
- drm_global_item_unref(&dev_priv->bo_global_ref.ref);
- drm_global_item_unref(&dev_priv->mem_global_ref);
-}
diff --git a/drivers/gpu/vga/vga_switcheroo.c b/drivers/gpu/vga/vga_switcheroo.c
index cf2a185..a132c37 100644
--- a/drivers/gpu/vga/vga_switcheroo.c
+++ b/drivers/gpu/vga/vga_switcheroo.c
@@ -380,6 +380,9 @@ int vga_switcheroo_register_audio_client(struct pci_dev *pdev,
mutex_unlock(&vgasr_mutex);
return -EINVAL;
}
+ /* notify if GPU has been already bound */
+ if (ops->gpu_bound)
+ ops->gpu_bound(pdev, id);
}
mutex_unlock(&vgasr_mutex);
diff --git a/drivers/hid/hid-alps.c b/drivers/hid/hid-alps.c
index aec253b..3cd7229 100644
--- a/drivers/hid/hid-alps.c
+++ b/drivers/hid/hid-alps.c
@@ -660,6 +660,20 @@ static int T4_init(struct hid_device *hdev, struct alps_dev *pri_data)
return ret;
}
+static int alps_sp_open(struct input_dev *dev)
+{
+ struct hid_device *hid = input_get_drvdata(dev);
+
+ return hid_hw_open(hid);
+}
+
+static void alps_sp_close(struct input_dev *dev)
+{
+ struct hid_device *hid = input_get_drvdata(dev);
+
+ hid_hw_close(hid);
+}
+
static int alps_input_configured(struct hid_device *hdev, struct hid_input *hi)
{
struct alps_dev *data = hid_get_drvdata(hdev);
@@ -733,6 +747,10 @@ static int alps_input_configured(struct hid_device *hdev, struct hid_input *hi)
input2->id.version = input->id.version;
input2->dev.parent = input->dev.parent;
+ input_set_drvdata(input2, hdev);
+ input2->open = alps_sp_open;
+ input2->close = alps_sp_close;
+
__set_bit(EV_KEY, input2->evbit);
data->sp_btn_cnt = (data->sp_btn_info & 0x0F);
for (i = 0; i < data->sp_btn_cnt; i++)
diff --git a/drivers/hid/hid-asus.c b/drivers/hid/hid-asus.c
index dc6d647..a1fa2fc 100644
--- a/drivers/hid/hid-asus.c
+++ b/drivers/hid/hid-asus.c
@@ -359,6 +359,9 @@ static bool asus_kbd_wmi_led_control_present(struct hid_device *hdev)
u32 value;
int ret;
+ if (!IS_ENABLED(CONFIG_ASUS_WMI))
+ return false;
+
ret = asus_wmi_evaluate_method(ASUS_WMI_METHODID_DSTS2,
ASUS_WMI_DEVID_KBD_BACKLIGHT, 0, &value);
hid_dbg(hdev, "WMI backlight check: rc %d value %x", ret, value);
diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index f63489c..c0d6689 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h
@@ -927,6 +927,9 @@
#define USB_DEVICE_ID_QUANTA_OPTICAL_TOUCH_3003 0x3003
#define USB_DEVICE_ID_QUANTA_OPTICAL_TOUCH_3008 0x3008
+#define I2C_VENDOR_ID_RAYDIUM 0x2386
+#define I2C_PRODUCT_ID_RAYDIUM_4B33 0x4b33
+
#define USB_VENDOR_ID_RAZER 0x1532
#define USB_DEVICE_ID_RAZER_BLADE_14 0x011D
diff --git a/drivers/hid/hid-quirks.c b/drivers/hid/hid-quirks.c
index 52c3b01..8237dd8 100644
--- a/drivers/hid/hid-quirks.c
+++ b/drivers/hid/hid-quirks.c
@@ -107,7 +107,6 @@ static const struct hid_device_id hid_quirks[] = {
{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_MOUSE_C05A), HID_QUIRK_ALWAYS_POLL },
{ HID_USB_DEVICE(USB_VENDOR_ID_LOGITECH, USB_DEVICE_ID_LOGITECH_MOUSE_C06A), HID_QUIRK_ALWAYS_POLL },
{ HID_USB_DEVICE(USB_VENDOR_ID_MCS, USB_DEVICE_ID_MCS_GAMEPADBLOCK), HID_QUIRK_MULTI_INPUT },
- { HID_USB_DEVICE(USB_VENDOR_ID_MGE, USB_DEVICE_ID_MGE_UPS), HID_QUIRK_NOGET },
{ HID_USB_DEVICE(USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_POWER_COVER), HID_QUIRK_NO_INIT_REPORTS },
{ HID_USB_DEVICE(USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_SURFACE_PRO_2), HID_QUIRK_NO_INIT_REPORTS },
{ HID_USB_DEVICE(USB_VENDOR_ID_MICROSOFT, USB_DEVICE_ID_MS_TOUCH_COVER_2), HID_QUIRK_NO_INIT_REPORTS },
diff --git a/drivers/hid/i2c-hid/i2c-hid-core.c b/drivers/hid/i2c-hid/i2c-hid-core.c
index 4aab96c..3cde7c1 100644
--- a/drivers/hid/i2c-hid/i2c-hid-core.c
+++ b/drivers/hid/i2c-hid/i2c-hid-core.c
@@ -49,6 +49,7 @@
#define I2C_HID_QUIRK_SET_PWR_WAKEUP_DEV BIT(0)
#define I2C_HID_QUIRK_NO_IRQ_AFTER_RESET BIT(1)
#define I2C_HID_QUIRK_NO_RUNTIME_PM BIT(2)
+#define I2C_HID_QUIRK_DELAY_AFTER_SLEEP BIT(3)
/* flags */
#define I2C_HID_STARTED 0
@@ -158,6 +159,8 @@ struct i2c_hid {
bool irq_wake_enabled;
struct mutex reset_lock;
+
+ unsigned long sleep_delay;
};
static const struct i2c_hid_quirks {
@@ -172,6 +175,8 @@ static const struct i2c_hid_quirks {
{ I2C_VENDOR_ID_HANTICK, I2C_PRODUCT_ID_HANTICK_5288,
I2C_HID_QUIRK_NO_IRQ_AFTER_RESET |
I2C_HID_QUIRK_NO_RUNTIME_PM },
+ { I2C_VENDOR_ID_RAYDIUM, I2C_PRODUCT_ID_RAYDIUM_4B33,
+ I2C_HID_QUIRK_DELAY_AFTER_SLEEP },
{ 0, 0 }
};
@@ -387,6 +392,7 @@ static int i2c_hid_set_power(struct i2c_client *client, int power_state)
{
struct i2c_hid *ihid = i2c_get_clientdata(client);
int ret;
+ unsigned long now, delay;
i2c_hid_dbg(ihid, "%s\n", __func__);
@@ -404,9 +410,22 @@ static int i2c_hid_set_power(struct i2c_client *client, int power_state)
goto set_pwr_exit;
}
+ if (ihid->quirks & I2C_HID_QUIRK_DELAY_AFTER_SLEEP &&
+ power_state == I2C_HID_PWR_ON) {
+ now = jiffies;
+ if (time_after(ihid->sleep_delay, now)) {
+ delay = jiffies_to_usecs(ihid->sleep_delay - now);
+ usleep_range(delay, delay + 1);
+ }
+ }
+
ret = __i2c_hid_command(client, &hid_set_power_cmd, power_state,
0, NULL, 0, NULL, 0);
+ if (ihid->quirks & I2C_HID_QUIRK_DELAY_AFTER_SLEEP &&
+ power_state == I2C_HID_PWR_SLEEP)
+ ihid->sleep_delay = jiffies + msecs_to_jiffies(20);
+
if (ret)
dev_err(&client->dev, "failed to change power setting.\n");
diff --git a/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c b/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c
index cac262a..89f2976 100644
--- a/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c
+++ b/drivers/hid/i2c-hid/i2c-hid-dmi-quirks.c
@@ -331,6 +331,14 @@ static const struct dmi_system_id i2c_hid_dmi_desc_override_table[] = {
.driver_data = (void *)&sipodev_desc
},
{
+ .ident = "Direkt-Tek DTLAPY133-1",
+ .matches = {
+ DMI_EXACT_MATCH(DMI_SYS_VENDOR, "Direkt-Tek"),
+ DMI_EXACT_MATCH(DMI_PRODUCT_NAME, "DTLAPY133-1"),
+ },
+ .driver_data = (void *)&sipodev_desc
+ },
+ {
.ident = "Mediacom Flexbook Edge 11",
.matches = {
DMI_EXACT_MATCH(DMI_SYS_VENDOR, "MEDIACOM"),
diff --git a/drivers/hid/usbhid/hiddev.c b/drivers/hid/usbhid/hiddev.c
index 23872d0..a746017 100644
--- a/drivers/hid/usbhid/hiddev.c
+++ b/drivers/hid/usbhid/hiddev.c
@@ -512,14 +512,24 @@ static noinline int hiddev_ioctl_usage(struct hiddev *hiddev, unsigned int cmd,
if (cmd == HIDIOCGCOLLECTIONINDEX) {
if (uref->usage_index >= field->maxusage)
goto inval;
+ uref->usage_index =
+ array_index_nospec(uref->usage_index,
+ field->maxusage);
} else if (uref->usage_index >= field->report_count)
goto inval;
}
- if ((cmd == HIDIOCGUSAGES || cmd == HIDIOCSUSAGES) &&
- (uref_multi->num_values > HID_MAX_MULTI_USAGES ||
- uref->usage_index + uref_multi->num_values > field->report_count))
- goto inval;
+ if (cmd == HIDIOCGUSAGES || cmd == HIDIOCSUSAGES) {
+ if (uref_multi->num_values > HID_MAX_MULTI_USAGES ||
+ uref->usage_index + uref_multi->num_values >
+ field->report_count)
+ goto inval;
+
+ uref->usage_index =
+ array_index_nospec(uref->usage_index,
+ field->report_count -
+ uref_multi->num_values);
+ }
switch (cmd) {
case HIDIOCGUSAGE:
diff --git a/drivers/hwmon/hwmon.c b/drivers/hwmon/hwmon.c
index 975c951..84f61ce 100644
--- a/drivers/hwmon/hwmon.c
+++ b/drivers/hwmon/hwmon.c
@@ -649,8 +649,10 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,
if (info[i]->config[j] & HWMON_T_INPUT) {
err = hwmon_thermal_add_sensor(dev,
hwdev, j);
- if (err)
- goto free_device;
+ if (err) {
+ device_unregister(hdev);
+ goto ida_remove;
+ }
}
}
}
@@ -658,8 +660,6 @@ __hwmon_device_register(struct device *dev, const char *name, void *drvdata,
return hdev;
-free_device:
- device_unregister(hdev);
free_hwmon:
kfree(hwdev);
ida_remove:
diff --git a/drivers/hwmon/ibmpowernv.c b/drivers/hwmon/ibmpowernv.c
index 0ccca87..293dd1c 100644
--- a/drivers/hwmon/ibmpowernv.c
+++ b/drivers/hwmon/ibmpowernv.c
@@ -181,7 +181,7 @@ static ssize_t show_label(struct device *dev, struct device_attribute *devattr,
return sprintf(buf, "%s\n", sdata->label);
}
-static int __init get_logical_cpu(int hwcpu)
+static int get_logical_cpu(int hwcpu)
{
int cpu;
@@ -192,9 +192,8 @@ static int __init get_logical_cpu(int hwcpu)
return -ENOENT;
}
-static void __init make_sensor_label(struct device_node *np,
- struct sensor_data *sdata,
- const char *label)
+static void make_sensor_label(struct device_node *np,
+ struct sensor_data *sdata, const char *label)
{
u32 id;
size_t n;
diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig
index 56ccb1e..f2c6819 100644
--- a/drivers/i2c/busses/Kconfig
+++ b/drivers/i2c/busses/Kconfig
@@ -224,6 +224,15 @@
This driver can also be built as a module. If so, the module
will be called i2c-nforce2-s4985.
+config I2C_NVIDIA_GPU
+ tristate "NVIDIA GPU I2C controller"
+ depends on PCI
+ help
+ If you say yes to this option, support will be included for the
+ NVIDIA GPU I2C controller which is used to communicate with the GPU's
+ Type-C controller. This driver can also be built as a module called
+ i2c-nvidia-gpu.
+
config I2C_SIS5595
tristate "SiS 5595"
depends on PCI
@@ -752,7 +761,7 @@
config I2C_OMAP
tristate "OMAP I2C adapter"
- depends on ARCH_OMAP
+ depends on ARCH_OMAP || ARCH_K3
default y if MACH_OMAP_H3 || MACH_OMAP_OSK
help
If you say yes to this option, support will be included for the
diff --git a/drivers/i2c/busses/Makefile b/drivers/i2c/busses/Makefile
index 18b26af..5f0cb69 100644
--- a/drivers/i2c/busses/Makefile
+++ b/drivers/i2c/busses/Makefile
@@ -19,6 +19,7 @@
obj-$(CONFIG_I2C_ISMT) += i2c-ismt.o
obj-$(CONFIG_I2C_NFORCE2) += i2c-nforce2.o
obj-$(CONFIG_I2C_NFORCE2_S4985) += i2c-nforce2-s4985.o
+obj-$(CONFIG_I2C_NVIDIA_GPU) += i2c-nvidia-gpu.o
obj-$(CONFIG_I2C_PIIX4) += i2c-piix4.o
obj-$(CONFIG_I2C_SIS5595) += i2c-sis5595.o
obj-$(CONFIG_I2C_SIS630) += i2c-sis630.o
diff --git a/drivers/i2c/busses/i2c-nvidia-gpu.c b/drivers/i2c/busses/i2c-nvidia-gpu.c
new file mode 100644
index 0000000..8822357
--- /dev/null
+++ b/drivers/i2c/busses/i2c-nvidia-gpu.c
@@ -0,0 +1,368 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Nvidia GPU I2C controller Driver
+ *
+ * Copyright (C) 2018 NVIDIA Corporation. All rights reserved.
+ * Author: Ajay Gupta <ajayg@nvidia.com>
+ */
+#include <linux/delay.h>
+#include <linux/i2c.h>
+#include <linux/interrupt.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+#include <linux/pm.h>
+#include <linux/pm_runtime.h>
+
+#include <asm/unaligned.h>
+
+/* I2C definitions */
+#define I2C_MST_CNTL 0x00
+#define I2C_MST_CNTL_GEN_START BIT(0)
+#define I2C_MST_CNTL_GEN_STOP BIT(1)
+#define I2C_MST_CNTL_CMD_READ (1 << 2)
+#define I2C_MST_CNTL_CMD_WRITE (2 << 2)
+#define I2C_MST_CNTL_BURST_SIZE_SHIFT 6
+#define I2C_MST_CNTL_GEN_NACK BIT(28)
+#define I2C_MST_CNTL_STATUS GENMASK(30, 29)
+#define I2C_MST_CNTL_STATUS_OKAY (0 << 29)
+#define I2C_MST_CNTL_STATUS_NO_ACK (1 << 29)
+#define I2C_MST_CNTL_STATUS_TIMEOUT (2 << 29)
+#define I2C_MST_CNTL_STATUS_BUS_BUSY (3 << 29)
+#define I2C_MST_CNTL_CYCLE_TRIGGER BIT(31)
+
+#define I2C_MST_ADDR 0x04
+
+#define I2C_MST_I2C0_TIMING 0x08
+#define I2C_MST_I2C0_TIMING_SCL_PERIOD_100KHZ 0x10e
+#define I2C_MST_I2C0_TIMING_TIMEOUT_CLK_CNT 16
+#define I2C_MST_I2C0_TIMING_TIMEOUT_CLK_CNT_MAX 255
+#define I2C_MST_I2C0_TIMING_TIMEOUT_CHECK BIT(24)
+
+#define I2C_MST_DATA 0x0c
+
+#define I2C_MST_HYBRID_PADCTL 0x20
+#define I2C_MST_HYBRID_PADCTL_MODE_I2C BIT(0)
+#define I2C_MST_HYBRID_PADCTL_I2C_SCL_INPUT_RCV BIT(14)
+#define I2C_MST_HYBRID_PADCTL_I2C_SDA_INPUT_RCV BIT(15)
+
+struct gpu_i2c_dev {
+ struct device *dev;
+ void __iomem *regs;
+ struct i2c_adapter adapter;
+ struct i2c_board_info *gpu_ccgx_ucsi;
+};
+
+static void gpu_enable_i2c_bus(struct gpu_i2c_dev *i2cd)
+{
+ u32 val;
+
+ /* enable I2C */
+ val = readl(i2cd->regs + I2C_MST_HYBRID_PADCTL);
+ val |= I2C_MST_HYBRID_PADCTL_MODE_I2C |
+ I2C_MST_HYBRID_PADCTL_I2C_SCL_INPUT_RCV |
+ I2C_MST_HYBRID_PADCTL_I2C_SDA_INPUT_RCV;
+ writel(val, i2cd->regs + I2C_MST_HYBRID_PADCTL);
+
+ /* enable 100KHZ mode */
+ val = I2C_MST_I2C0_TIMING_SCL_PERIOD_100KHZ;
+ val |= (I2C_MST_I2C0_TIMING_TIMEOUT_CLK_CNT_MAX
+ << I2C_MST_I2C0_TIMING_TIMEOUT_CLK_CNT);
+ val |= I2C_MST_I2C0_TIMING_TIMEOUT_CHECK;
+ writel(val, i2cd->regs + I2C_MST_I2C0_TIMING);
+}
+
+static int gpu_i2c_check_status(struct gpu_i2c_dev *i2cd)
+{
+ unsigned long target = jiffies + msecs_to_jiffies(1000);
+ u32 val;
+
+ do {
+ val = readl(i2cd->regs + I2C_MST_CNTL);
+ if (!(val & I2C_MST_CNTL_CYCLE_TRIGGER))
+ break;
+ if ((val & I2C_MST_CNTL_STATUS) !=
+ I2C_MST_CNTL_STATUS_BUS_BUSY)
+ break;
+ usleep_range(500, 600);
+ } while (time_is_after_jiffies(target));
+
+ if (time_is_before_jiffies(target)) {
+ dev_err(i2cd->dev, "i2c timeout error %x\n", val);
+ return -ETIME;
+ }
+
+ val = readl(i2cd->regs + I2C_MST_CNTL);
+ switch (val & I2C_MST_CNTL_STATUS) {
+ case I2C_MST_CNTL_STATUS_OKAY:
+ return 0;
+ case I2C_MST_CNTL_STATUS_NO_ACK:
+ return -EIO;
+ case I2C_MST_CNTL_STATUS_TIMEOUT:
+ return -ETIME;
+ default:
+ return 0;
+ }
+}
+
+static int gpu_i2c_read(struct gpu_i2c_dev *i2cd, u8 *data, u16 len)
+{
+ int status;
+ u32 val;
+
+ val = I2C_MST_CNTL_GEN_START | I2C_MST_CNTL_CMD_READ |
+ (len << I2C_MST_CNTL_BURST_SIZE_SHIFT) |
+ I2C_MST_CNTL_CYCLE_TRIGGER | I2C_MST_CNTL_GEN_NACK;
+ writel(val, i2cd->regs + I2C_MST_CNTL);
+
+ status = gpu_i2c_check_status(i2cd);
+ if (status < 0)
+ return status;
+
+ val = readl(i2cd->regs + I2C_MST_DATA);
+ switch (len) {
+ case 1:
+ data[0] = val;
+ break;
+ case 2:
+ put_unaligned_be16(val, data);
+ break;
+ case 3:
+ put_unaligned_be16(val >> 8, data);
+ data[2] = val;
+ break;
+ case 4:
+ put_unaligned_be32(val, data);
+ break;
+ default:
+ break;
+ }
+ return status;
+}
+
+static int gpu_i2c_start(struct gpu_i2c_dev *i2cd)
+{
+ writel(I2C_MST_CNTL_GEN_START, i2cd->regs + I2C_MST_CNTL);
+ return gpu_i2c_check_status(i2cd);
+}
+
+static int gpu_i2c_stop(struct gpu_i2c_dev *i2cd)
+{
+ writel(I2C_MST_CNTL_GEN_STOP, i2cd->regs + I2C_MST_CNTL);
+ return gpu_i2c_check_status(i2cd);
+}
+
+static int gpu_i2c_write(struct gpu_i2c_dev *i2cd, u8 data)
+{
+ u32 val;
+
+ writel(data, i2cd->regs + I2C_MST_DATA);
+
+ val = I2C_MST_CNTL_CMD_WRITE | (1 << I2C_MST_CNTL_BURST_SIZE_SHIFT);
+ writel(val, i2cd->regs + I2C_MST_CNTL);
+
+ return gpu_i2c_check_status(i2cd);
+}
+
+static int gpu_i2c_master_xfer(struct i2c_adapter *adap,
+ struct i2c_msg *msgs, int num)
+{
+ struct gpu_i2c_dev *i2cd = i2c_get_adapdata(adap);
+ int status, status2;
+ int i, j;
+
+ /*
+ * The controller supports maximum 4 byte read due to known
+ * limitation of sending STOP after every read.
+ */
+ for (i = 0; i < num; i++) {
+ if (msgs[i].flags & I2C_M_RD) {
+ /* program client address before starting read */
+ writel(msgs[i].addr, i2cd->regs + I2C_MST_ADDR);
+ /* gpu_i2c_read has implicit start */
+ status = gpu_i2c_read(i2cd, msgs[i].buf, msgs[i].len);
+ if (status < 0)
+ goto stop;
+ } else {
+ u8 addr = i2c_8bit_addr_from_msg(msgs + i);
+
+ status = gpu_i2c_start(i2cd);
+ if (status < 0) {
+ if (i == 0)
+ return status;
+ goto stop;
+ }
+
+ status = gpu_i2c_write(i2cd, addr);
+ if (status < 0)
+ goto stop;
+
+ for (j = 0; j < msgs[i].len; j++) {
+ status = gpu_i2c_write(i2cd, msgs[i].buf[j]);
+ if (status < 0)
+ goto stop;
+ }
+ }
+ }
+ status = gpu_i2c_stop(i2cd);
+ if (status < 0)
+ return status;
+
+ return i;
+stop:
+ status2 = gpu_i2c_stop(i2cd);
+ if (status2 < 0)
+ dev_err(i2cd->dev, "i2c stop failed %d\n", status2);
+ return status;
+}
+
+static const struct i2c_adapter_quirks gpu_i2c_quirks = {
+ .max_read_len = 4,
+ .flags = I2C_AQ_COMB_WRITE_THEN_READ,
+};
+
+static u32 gpu_i2c_functionality(struct i2c_adapter *adap)
+{
+ return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
+}
+
+static const struct i2c_algorithm gpu_i2c_algorithm = {
+ .master_xfer = gpu_i2c_master_xfer,
+ .functionality = gpu_i2c_functionality,
+};
+
+/*
+ * This driver is for Nvidia GPU cards with USB Type-C interface.
+ * We want to identify the cards using vendor ID and class code only
+ * to avoid dependency of adding product id for any new card which
+ * requires this driver.
+ * Currently there is no class code defined for UCSI device over PCI
+ * so using UNKNOWN class for now and it will be updated when UCSI
+ * over PCI gets a class code.
+ * There is no other NVIDIA cards with UNKNOWN class code. Even if the
+ * driver gets loaded for an undesired card then eventually i2c_read()
+ * (initiated from UCSI i2c_client) will timeout or UCSI commands will
+ * timeout.
+ */
+#define PCI_CLASS_SERIAL_UNKNOWN 0x0c80
+static const struct pci_device_id gpu_i2c_ids[] = {
+ { PCI_VENDOR_ID_NVIDIA, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
+ PCI_CLASS_SERIAL_UNKNOWN << 8, 0xffffff00},
+ { }
+};
+MODULE_DEVICE_TABLE(pci, gpu_i2c_ids);
+
+static int gpu_populate_client(struct gpu_i2c_dev *i2cd, int irq)
+{
+ struct i2c_client *ccgx_client;
+
+ i2cd->gpu_ccgx_ucsi = devm_kzalloc(i2cd->dev,
+ sizeof(*i2cd->gpu_ccgx_ucsi),
+ GFP_KERNEL);
+ if (!i2cd->gpu_ccgx_ucsi)
+ return -ENOMEM;
+
+ strlcpy(i2cd->gpu_ccgx_ucsi->type, "ccgx-ucsi",
+ sizeof(i2cd->gpu_ccgx_ucsi->type));
+ i2cd->gpu_ccgx_ucsi->addr = 0x8;
+ i2cd->gpu_ccgx_ucsi->irq = irq;
+ ccgx_client = i2c_new_device(&i2cd->adapter, i2cd->gpu_ccgx_ucsi);
+ if (!ccgx_client)
+ return -ENODEV;
+
+ return 0;
+}
+
+static int gpu_i2c_probe(struct pci_dev *pdev, const struct pci_device_id *id)
+{
+ struct gpu_i2c_dev *i2cd;
+ int status;
+
+ i2cd = devm_kzalloc(&pdev->dev, sizeof(*i2cd), GFP_KERNEL);
+ if (!i2cd)
+ return -ENOMEM;
+
+ i2cd->dev = &pdev->dev;
+ dev_set_drvdata(&pdev->dev, i2cd);
+
+ status = pcim_enable_device(pdev);
+ if (status < 0) {
+ dev_err(&pdev->dev, "pcim_enable_device failed %d\n", status);
+ return status;
+ }
+
+ pci_set_master(pdev);
+
+ i2cd->regs = pcim_iomap(pdev, 0, 0);
+ if (!i2cd->regs) {
+ dev_err(&pdev->dev, "pcim_iomap failed\n");
+ return -ENOMEM;
+ }
+
+ status = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_MSI);
+ if (status < 0) {
+ dev_err(&pdev->dev, "pci_alloc_irq_vectors err %d\n", status);
+ return status;
+ }
+
+ gpu_enable_i2c_bus(i2cd);
+
+ i2c_set_adapdata(&i2cd->adapter, i2cd);
+ i2cd->adapter.owner = THIS_MODULE;
+ strlcpy(i2cd->adapter.name, "NVIDIA GPU I2C adapter",
+ sizeof(i2cd->adapter.name));
+ i2cd->adapter.algo = &gpu_i2c_algorithm;
+ i2cd->adapter.quirks = &gpu_i2c_quirks;
+ i2cd->adapter.dev.parent = &pdev->dev;
+ status = i2c_add_adapter(&i2cd->adapter);
+ if (status < 0)
+ goto free_irq_vectors;
+
+ status = gpu_populate_client(i2cd, pdev->irq);
+ if (status < 0) {
+ dev_err(&pdev->dev, "gpu_populate_client failed %d\n", status);
+ goto del_adapter;
+ }
+
+ return 0;
+
+del_adapter:
+ i2c_del_adapter(&i2cd->adapter);
+free_irq_vectors:
+ pci_free_irq_vectors(pdev);
+ return status;
+}
+
+static void gpu_i2c_remove(struct pci_dev *pdev)
+{
+ struct gpu_i2c_dev *i2cd = dev_get_drvdata(&pdev->dev);
+
+ i2c_del_adapter(&i2cd->adapter);
+ pci_free_irq_vectors(pdev);
+}
+
+static int gpu_i2c_resume(struct device *dev)
+{
+ struct gpu_i2c_dev *i2cd = dev_get_drvdata(dev);
+
+ gpu_enable_i2c_bus(i2cd);
+ return 0;
+}
+
+static UNIVERSAL_DEV_PM_OPS(gpu_i2c_driver_pm, NULL, gpu_i2c_resume, NULL);
+
+static struct pci_driver gpu_i2c_driver = {
+ .name = "nvidia-gpu",
+ .id_table = gpu_i2c_ids,
+ .probe = gpu_i2c_probe,
+ .remove = gpu_i2c_remove,
+ .driver = {
+ .pm = &gpu_i2c_driver_pm,
+ },
+};
+
+module_pci_driver(gpu_i2c_driver);
+
+MODULE_AUTHOR("Ajay Gupta <ajayg@nvidia.com>");
+MODULE_DESCRIPTION("Nvidia GPU I2C controller Driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/i2c/busses/i2c-qcom-geni.c b/drivers/i2c/busses/i2c-qcom-geni.c
index 527f55c..db075bc0 100644
--- a/drivers/i2c/busses/i2c-qcom-geni.c
+++ b/drivers/i2c/busses/i2c-qcom-geni.c
@@ -571,18 +571,19 @@ static int geni_i2c_probe(struct platform_device *pdev)
dev_dbg(&pdev->dev, "i2c fifo/se-dma mode. fifo depth:%d\n", tx_depth);
- ret = i2c_add_adapter(&gi2c->adap);
- if (ret) {
- dev_err(&pdev->dev, "Error adding i2c adapter %d\n", ret);
- return ret;
- }
-
gi2c->suspended = 1;
pm_runtime_set_suspended(gi2c->se.dev);
pm_runtime_set_autosuspend_delay(gi2c->se.dev, I2C_AUTO_SUSPEND_DELAY);
pm_runtime_use_autosuspend(gi2c->se.dev);
pm_runtime_enable(gi2c->se.dev);
+ ret = i2c_add_adapter(&gi2c->adap);
+ if (ret) {
+ dev_err(&pdev->dev, "Error adding i2c adapter %d\n", ret);
+ pm_runtime_disable(gi2c->se.dev);
+ return ret;
+ }
+
return 0;
}
@@ -590,8 +591,8 @@ static int geni_i2c_remove(struct platform_device *pdev)
{
struct geni_i2c_dev *gi2c = platform_get_drvdata(pdev);
- pm_runtime_disable(gi2c->se.dev);
i2c_del_adapter(&gi2c->adap);
+ pm_runtime_disable(gi2c->se.dev);
return 0;
}
diff --git a/drivers/leds/trigger/ledtrig-pattern.c b/drivers/leds/trigger/ledtrig-pattern.c
index ce7acd1..1870cf8 100644
--- a/drivers/leds/trigger/ledtrig-pattern.c
+++ b/drivers/leds/trigger/ledtrig-pattern.c
@@ -75,8 +75,6 @@ static void pattern_trig_timer_function(struct timer_list *t)
{
struct pattern_trig_data *data = from_timer(data, t, timer);
- mutex_lock(&data->lock);
-
for (;;) {
if (!data->is_indefinite && !data->repeat)
break;
@@ -87,9 +85,10 @@ static void pattern_trig_timer_function(struct timer_list *t)
data->curr->brightness);
mod_timer(&data->timer,
jiffies + msecs_to_jiffies(data->curr->delta_t));
-
- /* Skip the tuple with zero duration */
- pattern_trig_update_patterns(data);
+ if (!data->next->delta_t) {
+ /* Skip the tuple with zero duration */
+ pattern_trig_update_patterns(data);
+ }
/* Select next tuple */
pattern_trig_update_patterns(data);
} else {
@@ -116,8 +115,6 @@ static void pattern_trig_timer_function(struct timer_list *t)
break;
}
-
- mutex_unlock(&data->lock);
}
static int pattern_trig_start_pattern(struct led_classdev *led_cdev)
@@ -176,14 +173,10 @@ static ssize_t repeat_store(struct device *dev, struct device_attribute *attr,
if (res < -1 || res == 0)
return -EINVAL;
- /*
- * Clear previous patterns' performence firstly, and remove the timer
- * without mutex lock to avoid dead lock.
- */
- del_timer_sync(&data->timer);
-
mutex_lock(&data->lock);
+ del_timer_sync(&data->timer);
+
if (data->is_hw_pattern)
led_cdev->pattern_clear(led_cdev);
@@ -234,14 +227,10 @@ static ssize_t pattern_trig_store_patterns(struct led_classdev *led_cdev,
struct pattern_trig_data *data = led_cdev->trigger_data;
int ccount, cr, offset = 0, err = 0;
- /*
- * Clear previous patterns' performence firstly, and remove the timer
- * without mutex lock to avoid dead lock.
- */
- del_timer_sync(&data->timer);
-
mutex_lock(&data->lock);
+ del_timer_sync(&data->timer);
+
if (data->is_hw_pattern)
led_cdev->pattern_clear(led_cdev);
diff --git a/drivers/mtd/devices/Kconfig b/drivers/mtd/devices/Kconfig
index e514d57..aa98342 100644
--- a/drivers/mtd/devices/Kconfig
+++ b/drivers/mtd/devices/Kconfig
@@ -207,7 +207,7 @@
config MTD_DOCG3
tristate "M-Systems Disk-On-Chip G3"
select BCH
- select BCH_CONST_PARAMS
+ select BCH_CONST_PARAMS if !MTD_NAND_BCH
select BITREVERSE
help
This provides an MTD device driver for the M-Systems DiskOnChip
diff --git a/drivers/mtd/maps/sa1100-flash.c b/drivers/mtd/maps/sa1100-flash.c
index 784c6e1..fd5fe12 100644
--- a/drivers/mtd/maps/sa1100-flash.c
+++ b/drivers/mtd/maps/sa1100-flash.c
@@ -221,7 +221,14 @@ static struct sa_info *sa1100_setup_mtd(struct platform_device *pdev,
info->mtd = info->subdev[0].mtd;
ret = 0;
} else if (info->num_subdev > 1) {
- struct mtd_info *cdev[nr];
+ struct mtd_info **cdev;
+
+ cdev = kmalloc_array(nr, sizeof(*cdev), GFP_KERNEL);
+ if (!cdev) {
+ ret = -ENOMEM;
+ goto err;
+ }
+
/*
* We detected multiple devices. Concatenate them together.
*/
@@ -230,6 +237,7 @@ static struct sa_info *sa1100_setup_mtd(struct platform_device *pdev,
info->mtd = mtd_concat_create(cdev, info->num_subdev,
plat->name);
+ kfree(cdev);
if (info->mtd == NULL) {
ret = -ENXIO;
goto err;
diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c
index 05bd077..71050a0 100644
--- a/drivers/mtd/nand/raw/nand_base.c
+++ b/drivers/mtd/nand/raw/nand_base.c
@@ -590,7 +590,6 @@ nand_get_device(struct mtd_info *mtd, int new_state)
/**
* panic_nand_wait - [GENERIC] wait until the command is done
- * @mtd: MTD device structure
* @chip: NAND chip structure
* @timeo: timeout
*
diff --git a/drivers/mtd/spi-nor/cadence-quadspi.c b/drivers/mtd/spi-nor/cadence-quadspi.c
index e24db81..d846428 100644
--- a/drivers/mtd/spi-nor/cadence-quadspi.c
+++ b/drivers/mtd/spi-nor/cadence-quadspi.c
@@ -996,7 +996,7 @@ static int cqspi_direct_read_execute(struct spi_nor *nor, u_char *buf,
err_unmap:
dma_unmap_single(nor->dev, dma_dst, len, DMA_FROM_DEVICE);
- return 0;
+ return ret;
}
static ssize_t cqspi_read(struct spi_nor *nor, loff_t from,
diff --git a/drivers/mtd/spi-nor/spi-nor.c b/drivers/mtd/spi-nor/spi-nor.c
index 9407ca5..3e54e31 100644
--- a/drivers/mtd/spi-nor/spi-nor.c
+++ b/drivers/mtd/spi-nor/spi-nor.c
@@ -3250,12 +3250,14 @@ static int spi_nor_init_params(struct spi_nor *nor,
memcpy(&sfdp_params, params, sizeof(sfdp_params));
memcpy(&prev_map, &nor->erase_map, sizeof(prev_map));
- if (spi_nor_parse_sfdp(nor, &sfdp_params))
+ if (spi_nor_parse_sfdp(nor, &sfdp_params)) {
+ nor->addr_width = 0;
/* restore previous erase map */
memcpy(&nor->erase_map, &prev_map,
sizeof(nor->erase_map));
- else
+ } else {
memcpy(params, &sfdp_params, sizeof(*params));
+ }
}
return 0;
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index ffa37ad..333387f 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -3112,13 +3112,13 @@ static int bond_slave_netdev_event(unsigned long event,
case NETDEV_CHANGE:
/* For 802.3ad mode only:
* Getting invalid Speed/Duplex values here will put slave
- * in weird state. So mark it as link-down for the time
+ * in weird state. So mark it as link-fail for the time
* being and let link-monitoring (miimon) set it right when
* correct speeds/duplex are available.
*/
if (bond_update_speed_duplex(slave) &&
BOND_MODE(bond) == BOND_MODE_8023AD)
- slave->link = BOND_LINK_DOWN;
+ slave->link = BOND_LINK_FAIL;
if (BOND_MODE(bond) == BOND_MODE_8023AD)
bond_3ad_adapter_speed_duplex_changed(slave);
diff --git a/drivers/net/dsa/microchip/ksz_common.c b/drivers/net/dsa/microchip/ksz_common.c
index 54e0ca6..86b6464 100644
--- a/drivers/net/dsa/microchip/ksz_common.c
+++ b/drivers/net/dsa/microchip/ksz_common.c
@@ -1117,11 +1117,6 @@ static int ksz_switch_init(struct ksz_device *dev)
{
int i;
- mutex_init(&dev->reg_mutex);
- mutex_init(&dev->stats_mutex);
- mutex_init(&dev->alu_mutex);
- mutex_init(&dev->vlan_mutex);
-
dev->ds->ops = &ksz_switch_ops;
for (i = 0; i < ARRAY_SIZE(ksz_switch_chips); i++) {
@@ -1206,6 +1201,11 @@ int ksz_switch_register(struct ksz_device *dev)
if (dev->pdata)
dev->chip_id = dev->pdata->chip_id;
+ mutex_init(&dev->reg_mutex);
+ mutex_init(&dev->stats_mutex);
+ mutex_init(&dev->alu_mutex);
+ mutex_init(&dev->vlan_mutex);
+
if (ksz_switch_detect(dev))
return -EINVAL;
diff --git a/drivers/net/dsa/mv88e6xxx/global1.c b/drivers/net/dsa/mv88e6xxx/global1.c
index d721ccf..38e399e 100644
--- a/drivers/net/dsa/mv88e6xxx/global1.c
+++ b/drivers/net/dsa/mv88e6xxx/global1.c
@@ -567,6 +567,8 @@ int mv88e6xxx_g1_stats_clear(struct mv88e6xxx_chip *chip)
if (err)
return err;
+ /* Keep the histogram mode bits */
+ val &= MV88E6XXX_G1_STATS_OP_HIST_RX_TX;
val |= MV88E6XXX_G1_STATS_OP_BUSY | MV88E6XXX_G1_STATS_OP_FLUSH_ALL;
err = mv88e6xxx_g1_write(chip, MV88E6XXX_G1_STATS_OP, val);
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
index 6a633c7..99ef1da 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
@@ -407,13 +407,13 @@ static void aq_ethtool_get_pauseparam(struct net_device *ndev,
struct ethtool_pauseparam *pause)
{
struct aq_nic_s *aq_nic = netdev_priv(ndev);
+ u32 fc = aq_nic->aq_nic_cfg.flow_control;
pause->autoneg = 0;
- if (aq_nic->aq_hw->aq_nic_cfg->flow_control & AQ_NIC_FC_RX)
- pause->rx_pause = 1;
- if (aq_nic->aq_hw->aq_nic_cfg->flow_control & AQ_NIC_FC_TX)
- pause->tx_pause = 1;
+ pause->rx_pause = !!(fc & AQ_NIC_FC_RX);
+ pause->tx_pause = !!(fc & AQ_NIC_FC_TX);
+
}
static int aq_ethtool_set_pauseparam(struct net_device *ndev,
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h
index e868924..a1e70da 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_hw.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_hw.h
@@ -204,6 +204,10 @@ struct aq_hw_ops {
int (*hw_get_fw_version)(struct aq_hw_s *self, u32 *fw_version);
+ int (*hw_set_offload)(struct aq_hw_s *self,
+ struct aq_nic_cfg_s *aq_nic_cfg);
+
+ int (*hw_set_fc)(struct aq_hw_s *self, u32 fc, u32 tc);
};
struct aq_fw_ops {
@@ -226,6 +230,8 @@ struct aq_fw_ops {
int (*update_stats)(struct aq_hw_s *self);
+ u32 (*get_flow_control)(struct aq_hw_s *self, u32 *fcmode);
+
int (*set_flow_control)(struct aq_hw_s *self);
int (*set_power)(struct aq_hw_s *self, unsigned int power_state,
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_main.c b/drivers/net/ethernet/aquantia/atlantic/aq_main.c
index e3ae29e..7c07eef 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_main.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_main.c
@@ -99,8 +99,11 @@ static int aq_ndev_set_features(struct net_device *ndev,
struct aq_nic_s *aq_nic = netdev_priv(ndev);
struct aq_nic_cfg_s *aq_cfg = aq_nic_get_cfg(aq_nic);
bool is_lro = false;
+ int err = 0;
- if (aq_cfg->hw_features & NETIF_F_LRO) {
+ aq_cfg->features = features;
+
+ if (aq_cfg->aq_hw_caps->hw_features & NETIF_F_LRO) {
is_lro = features & NETIF_F_LRO;
if (aq_cfg->is_lro != is_lro) {
@@ -112,8 +115,11 @@ static int aq_ndev_set_features(struct net_device *ndev,
}
}
}
+ if ((aq_nic->ndev->features ^ features) & NETIF_F_RXCSUM)
+ err = aq_nic->aq_hw_ops->hw_set_offload(aq_nic->aq_hw,
+ aq_cfg);
- return 0;
+ return err;
}
static int aq_ndev_set_mac_address(struct net_device *ndev, void *addr)
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
index 5fed244..7abdc09 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
@@ -118,12 +118,13 @@ void aq_nic_cfg_start(struct aq_nic_s *self)
}
cfg->link_speed_msk &= cfg->aq_hw_caps->link_speed_msk;
- cfg->hw_features = cfg->aq_hw_caps->hw_features;
+ cfg->features = cfg->aq_hw_caps->hw_features;
}
static int aq_nic_update_link_status(struct aq_nic_s *self)
{
int err = self->aq_fw_ops->update_link_status(self->aq_hw);
+ u32 fc = 0;
if (err)
return err;
@@ -133,6 +134,15 @@ static int aq_nic_update_link_status(struct aq_nic_s *self)
AQ_CFG_DRV_NAME, self->link_status.mbps,
self->aq_hw->aq_link_status.mbps);
aq_nic_update_interrupt_moderation_settings(self);
+
+ /* Driver has to update flow control settings on RX block
+ * on any link event.
+ * We should query FW whether it negotiated FC.
+ */
+ if (self->aq_fw_ops->get_flow_control)
+ self->aq_fw_ops->get_flow_control(self->aq_hw, &fc);
+ if (self->aq_hw_ops->hw_set_fc)
+ self->aq_hw_ops->hw_set_fc(self->aq_hw, fc, 0);
}
self->link_status = self->aq_hw->aq_link_status;
@@ -590,7 +600,7 @@ int aq_nic_set_multicast_list(struct aq_nic_s *self, struct net_device *ndev)
}
}
- if (i > 0 && i < AQ_HW_MULTICAST_ADDRESS_MAX) {
+ if (i > 0 && i <= AQ_HW_MULTICAST_ADDRESS_MAX) {
packet_filter |= IFF_MULTICAST;
self->mc_list.count = i;
self->aq_hw_ops->hw_multicast_list_set(self->aq_hw,
@@ -772,7 +782,9 @@ void aq_nic_get_link_ksettings(struct aq_nic_s *self,
ethtool_link_ksettings_add_link_mode(cmd, advertising,
Pause);
- if (self->aq_nic_cfg.flow_control & AQ_NIC_FC_TX)
+ /* Asym is when either RX or TX, but not both */
+ if (!!(self->aq_nic_cfg.flow_control & AQ_NIC_FC_TX) ^
+ !!(self->aq_nic_cfg.flow_control & AQ_NIC_FC_RX))
ethtool_link_ksettings_add_link_mode(cmd, advertising,
Asym_Pause);
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h
index c1582f4..44ec47a 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.h
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.h
@@ -23,7 +23,7 @@ struct aq_vec_s;
struct aq_nic_cfg_s {
const struct aq_hw_caps_s *aq_hw_caps;
- u64 hw_features;
+ u64 features;
u32 rxds; /* rx ring size, descriptors # */
u32 txds; /* tx ring size, descriptors # */
u32 vecs; /* vecs==allocated irqs */
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c
index 3db9144..74550cc 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_ring.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ring.c
@@ -172,6 +172,27 @@ bool aq_ring_tx_clean(struct aq_ring_s *self)
return !!budget;
}
+static void aq_rx_checksum(struct aq_ring_s *self,
+ struct aq_ring_buff_s *buff,
+ struct sk_buff *skb)
+{
+ if (!(self->aq_nic->ndev->features & NETIF_F_RXCSUM))
+ return;
+
+ if (unlikely(buff->is_cso_err)) {
+ ++self->stats.rx.errors;
+ skb->ip_summed = CHECKSUM_NONE;
+ return;
+ }
+ if (buff->is_ip_cso) {
+ __skb_incr_checksum_unnecessary(skb);
+ if (buff->is_udp_cso || buff->is_tcp_cso)
+ __skb_incr_checksum_unnecessary(skb);
+ } else {
+ skb->ip_summed = CHECKSUM_NONE;
+ }
+}
+
#define AQ_SKB_ALIGN SKB_DATA_ALIGN(sizeof(struct skb_shared_info))
int aq_ring_rx_clean(struct aq_ring_s *self,
struct napi_struct *napi,
@@ -267,18 +288,8 @@ int aq_ring_rx_clean(struct aq_ring_s *self,
}
skb->protocol = eth_type_trans(skb, ndev);
- if (unlikely(buff->is_cso_err)) {
- ++self->stats.rx.errors;
- skb->ip_summed = CHECKSUM_NONE;
- } else {
- if (buff->is_ip_cso) {
- __skb_incr_checksum_unnecessary(skb);
- if (buff->is_udp_cso || buff->is_tcp_cso)
- __skb_incr_checksum_unnecessary(skb);
- } else {
- skb->ip_summed = CHECKSUM_NONE;
- }
- }
+
+ aq_rx_checksum(self, buff, skb);
skb_set_hash(skb, buff->rss_hash,
buff->is_hash_l4 ? PKT_HASH_TYPE_L4 :
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
index 76d25d5..f02592f 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_b0.c
@@ -100,12 +100,17 @@ static int hw_atl_b0_hw_reset(struct aq_hw_s *self)
return err;
}
+static int hw_atl_b0_set_fc(struct aq_hw_s *self, u32 fc, u32 tc)
+{
+ hw_atl_rpb_rx_xoff_en_per_tc_set(self, !!(fc & AQ_NIC_FC_RX), tc);
+ return 0;
+}
+
static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self)
{
u32 tc = 0U;
u32 buff_size = 0U;
unsigned int i_priority = 0U;
- bool is_rx_flow_control = false;
/* TPS Descriptor rate init */
hw_atl_tps_tx_pkt_shed_desc_rate_curr_time_res_set(self, 0x0U);
@@ -138,7 +143,6 @@ static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self)
/* QoS Rx buf size per TC */
tc = 0;
- is_rx_flow_control = (AQ_NIC_FC_RX & self->aq_nic_cfg->flow_control);
buff_size = HW_ATL_B0_RXBUF_MAX;
hw_atl_rpb_rx_pkt_buff_size_per_tc_set(self, buff_size, tc);
@@ -150,7 +154,8 @@ static int hw_atl_b0_hw_qos_set(struct aq_hw_s *self)
(buff_size *
(1024U / 32U) * 50U) /
100U, tc);
- hw_atl_rpb_rx_xoff_en_per_tc_set(self, is_rx_flow_control ? 1U : 0U, tc);
+
+ hw_atl_b0_set_fc(self, self->aq_nic_cfg->flow_control, tc);
/* QoS 802.1p priority -> TC mapping */
for (i_priority = 8U; i_priority--;)
@@ -229,8 +234,10 @@ static int hw_atl_b0_hw_offload_set(struct aq_hw_s *self,
hw_atl_tpo_tcp_udp_crc_offload_en_set(self, 1);
/* RX checksums offloads*/
- hw_atl_rpo_ipv4header_crc_offload_en_set(self, 1);
- hw_atl_rpo_tcp_udp_crc_offload_en_set(self, 1);
+ hw_atl_rpo_ipv4header_crc_offload_en_set(self, !!(aq_nic_cfg->features &
+ NETIF_F_RXCSUM));
+ hw_atl_rpo_tcp_udp_crc_offload_en_set(self, !!(aq_nic_cfg->features &
+ NETIF_F_RXCSUM));
/* LSO offloads*/
hw_atl_tdm_large_send_offload_en_set(self, 0xFFFFFFFFU);
@@ -655,9 +662,9 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self,
struct hw_atl_rxd_wb_s *rxd_wb = (struct hw_atl_rxd_wb_s *)
&ring->dx_ring[ring->hw_head * HW_ATL_B0_RXD_SIZE];
- unsigned int is_err = 1U;
unsigned int is_rx_check_sum_enabled = 0U;
unsigned int pkt_type = 0U;
+ u8 rx_stat = 0U;
if (!(rxd_wb->status & 0x1U)) { /* RxD is not done */
break;
@@ -665,35 +672,35 @@ static int hw_atl_b0_hw_ring_rx_receive(struct aq_hw_s *self,
buff = &ring->buff_ring[ring->hw_head];
- is_err = (0x0000003CU & rxd_wb->status);
+ rx_stat = (0x0000003CU & rxd_wb->status) >> 2;
is_rx_check_sum_enabled = (rxd_wb->type) & (0x3U << 19);
- is_err &= ~0x20U; /* exclude validity bit */
pkt_type = 0xFFU & (rxd_wb->type >> 4);
- if (is_rx_check_sum_enabled) {
- if (0x0U == (pkt_type & 0x3U))
- buff->is_ip_cso = (is_err & 0x08U) ? 0U : 1U;
+ if (is_rx_check_sum_enabled & BIT(0) &&
+ (0x0U == (pkt_type & 0x3U)))
+ buff->is_ip_cso = (rx_stat & BIT(1)) ? 0U : 1U;
+ if (is_rx_check_sum_enabled & BIT(1)) {
if (0x4U == (pkt_type & 0x1CU))
- buff->is_udp_cso = buff->is_cso_err ? 0U : 1U;
+ buff->is_udp_cso = (rx_stat & BIT(2)) ? 0U :
+ !!(rx_stat & BIT(3));
else if (0x0U == (pkt_type & 0x1CU))
- buff->is_tcp_cso = buff->is_cso_err ? 0U : 1U;
-
- /* Checksum offload workaround for small packets */
- if (rxd_wb->pkt_len <= 60) {
- buff->is_ip_cso = 0U;
- buff->is_cso_err = 0U;
- }
+ buff->is_tcp_cso = (rx_stat & BIT(2)) ? 0U :
+ !!(rx_stat & BIT(3));
}
-
- is_err &= ~0x18U;
+ buff->is_cso_err = !!(rx_stat & 0x6);
+ /* Checksum offload workaround for small packets */
+ if (unlikely(rxd_wb->pkt_len <= 60)) {
+ buff->is_ip_cso = 0U;
+ buff->is_cso_err = 0U;
+ }
dma_unmap_page(ndev, buff->pa, buff->len, DMA_FROM_DEVICE);
- if (is_err || rxd_wb->type & 0x1000U) {
- /* status error or DMA error */
+ if ((rx_stat & BIT(0)) || rxd_wb->type & 0x1000U) {
+ /* MAC error or DMA error */
buff->is_error = 1U;
} else {
if (self->aq_nic_cfg->is_rss) {
@@ -915,6 +922,12 @@ static int hw_atl_b0_hw_interrupt_moderation_set(struct aq_hw_s *self)
static int hw_atl_b0_hw_stop(struct aq_hw_s *self)
{
hw_atl_b0_hw_irq_disable(self, HW_ATL_B0_INT_MASK);
+
+ /* Invalidate Descriptor Cache to prevent writing to the cached
+ * descriptors and to the data pointer of those descriptors
+ */
+ hw_atl_rdm_rx_dma_desc_cache_init_set(self, 1);
+
return aq_hw_err_from_flags(self);
}
@@ -963,4 +976,6 @@ const struct aq_hw_ops hw_atl_ops_b0 = {
.hw_get_regs = hw_atl_utils_hw_get_regs,
.hw_get_hw_stats = hw_atl_utils_get_hw_stats,
.hw_get_fw_version = hw_atl_utils_get_fw_version,
+ .hw_set_offload = hw_atl_b0_hw_offload_set,
+ .hw_set_fc = hw_atl_b0_set_fc,
};
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c
index be0a3a9..5502ec5f 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.c
@@ -619,6 +619,14 @@ void hw_atl_rpb_rx_flow_ctl_mode_set(struct aq_hw_s *aq_hw, u32 rx_flow_ctl_mode
HW_ATL_RPB_RX_FC_MODE_SHIFT, rx_flow_ctl_mode);
}
+void hw_atl_rdm_rx_dma_desc_cache_init_set(struct aq_hw_s *aq_hw, u32 init)
+{
+ aq_hw_write_reg_bit(aq_hw, HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR,
+ HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK,
+ HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT,
+ init);
+}
+
void hw_atl_rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
u32 rx_pkt_buff_size_per_tc, u32 buffer)
{
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h
index 7056c73..41f2399 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh.h
@@ -325,6 +325,9 @@ void hw_atl_rpb_rx_pkt_buff_size_per_tc_set(struct aq_hw_s *aq_hw,
u32 rx_pkt_buff_size_per_tc,
u32 buffer);
+/* set rdm rx dma descriptor cache init */
+void hw_atl_rdm_rx_dma_desc_cache_init_set(struct aq_hw_s *aq_hw, u32 init);
+
/* set rx xoff enable (per tc) */
void hw_atl_rpb_rx_xoff_en_per_tc_set(struct aq_hw_s *aq_hw, u32 rx_xoff_en_per_tc,
u32 buffer);
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h
index 716674a..a715fa3 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_llh_internal.h
@@ -293,6 +293,24 @@
/* default value of bitfield desc{d}_reset */
#define HW_ATL_RDM_DESCDRESET_DEFAULT 0x0
+/* rdm_desc_init_i bitfield definitions
+ * preprocessor definitions for the bitfield rdm_desc_init_i.
+ * port="pif_rdm_desc_init_i"
+ */
+
+/* register address for bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_ADR 0x00005a00
+/* bitmask for bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSK 0xffffffff
+/* inverted bitmask for bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_MSKN 0x00000000
+/* lower bit position of bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_SHIFT 0
+/* width of bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_WIDTH 32
+/* default value of bitfield rdm_desc_init_i */
+#define HW_ATL_RDM_RX_DMA_DESC_CACHE_INIT_DEFAULT 0x0
+
/* rx int_desc_wrb_en bitfield definitions
* preprocessor definitions for the bitfield "int_desc_wrb_en".
* port="pif_rdm_int_desc_wrb_en_i"
diff --git a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c
index 096ca57..7de3220 100644
--- a/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c
+++ b/drivers/net/ethernet/aquantia/atlantic/hw_atl/hw_atl_utils_fw2x.c
@@ -30,6 +30,8 @@
#define HW_ATL_FW2X_MPI_STATE_ADDR 0x370
#define HW_ATL_FW2X_MPI_STATE2_ADDR 0x374
+#define HW_ATL_FW2X_CAP_PAUSE BIT(CAPS_HI_PAUSE)
+#define HW_ATL_FW2X_CAP_ASYM_PAUSE BIT(CAPS_HI_ASYMMETRIC_PAUSE)
#define HW_ATL_FW2X_CAP_SLEEP_PROXY BIT(CAPS_HI_SLEEP_PROXY)
#define HW_ATL_FW2X_CAP_WOL BIT(CAPS_HI_WOL)
@@ -451,6 +453,24 @@ static int aq_fw2x_set_flow_control(struct aq_hw_s *self)
return 0;
}
+static u32 aq_fw2x_get_flow_control(struct aq_hw_s *self, u32 *fcmode)
+{
+ u32 mpi_state = aq_hw_read_reg(self, HW_ATL_FW2X_MPI_STATE2_ADDR);
+
+ if (mpi_state & HW_ATL_FW2X_CAP_PAUSE)
+ if (mpi_state & HW_ATL_FW2X_CAP_ASYM_PAUSE)
+ *fcmode = AQ_NIC_FC_RX;
+ else
+ *fcmode = AQ_NIC_FC_RX | AQ_NIC_FC_TX;
+ else
+ if (mpi_state & HW_ATL_FW2X_CAP_ASYM_PAUSE)
+ *fcmode = AQ_NIC_FC_TX;
+ else
+ *fcmode = 0;
+
+ return 0;
+}
+
const struct aq_fw_ops aq_fw_2x_ops = {
.init = aq_fw2x_init,
.deinit = aq_fw2x_deinit,
@@ -465,4 +485,5 @@ const struct aq_fw_ops aq_fw_2x_ops = {
.set_eee_rate = aq_fw2x_set_eee_rate,
.get_eee_rate = aq_fw2x_get_eee_rate,
.set_flow_control = aq_fw2x_set_flow_control,
+ .get_flow_control = aq_fw2x_get_flow_control
};
diff --git a/drivers/net/ethernet/atheros/alx/alx.h b/drivers/net/ethernet/atheros/alx/alx.h
index 78c5de4..9d0e74f 100644
--- a/drivers/net/ethernet/atheros/alx/alx.h
+++ b/drivers/net/ethernet/atheros/alx/alx.h
@@ -140,6 +140,5 @@ struct alx_priv {
};
extern const struct ethtool_ops alx_ethtool_ops;
-extern const char alx_drv_name[];
#endif
diff --git a/drivers/net/ethernet/atheros/alx/main.c b/drivers/net/ethernet/atheros/alx/main.c
index 7968c64..c131cfc1 100644
--- a/drivers/net/ethernet/atheros/alx/main.c
+++ b/drivers/net/ethernet/atheros/alx/main.c
@@ -49,7 +49,7 @@
#include "hw.h"
#include "reg.h"
-const char alx_drv_name[] = "alx";
+static const char alx_drv_name[] = "alx";
static void alx_free_txbuf(struct alx_tx_queue *txq, int entry)
{
diff --git a/drivers/net/ethernet/broadcom/bcmsysport.c b/drivers/net/ethernet/broadcom/bcmsysport.c
index 4122553..0e2d99c 100644
--- a/drivers/net/ethernet/broadcom/bcmsysport.c
+++ b/drivers/net/ethernet/broadcom/bcmsysport.c
@@ -1902,9 +1902,6 @@ static void bcm_sysport_netif_start(struct net_device *dev)
intrl2_1_mask_clear(priv, 0xffffffff);
else
intrl2_0_mask_clear(priv, INTRL2_0_TDMA_MBDONE_MASK);
-
- /* Last call before we start the real business */
- netif_tx_start_all_queues(dev);
}
static void rbuf_init(struct bcm_sysport_priv *priv)
@@ -2048,6 +2045,8 @@ static int bcm_sysport_open(struct net_device *dev)
bcm_sysport_netif_start(dev);
+ netif_tx_start_all_queues(dev);
+
return 0;
out_clear_rx_int:
@@ -2071,7 +2070,7 @@ static void bcm_sysport_netif_stop(struct net_device *dev)
struct bcm_sysport_priv *priv = netdev_priv(dev);
/* stop all software from updating hardware */
- netif_tx_stop_all_queues(dev);
+ netif_tx_disable(dev);
napi_disable(&priv->napi);
cancel_work_sync(&priv->dim.dim.work);
phy_stop(dev->phydev);
@@ -2658,12 +2657,12 @@ static int __maybe_unused bcm_sysport_suspend(struct device *d)
if (!netif_running(dev))
return 0;
+ netif_device_detach(dev);
+
bcm_sysport_netif_stop(dev);
phy_suspend(dev->phydev);
- netif_device_detach(dev);
-
/* Disable UniMAC RX */
umac_enable_set(priv, CMD_RX_EN, 0);
@@ -2746,8 +2745,6 @@ static int __maybe_unused bcm_sysport_resume(struct device *d)
goto out_free_rx_ring;
}
- netif_device_attach(dev);
-
/* RX pipe enable */
topctrl_writel(priv, 0, RX_FLUSH_CNTL);
@@ -2788,6 +2785,8 @@ static int __maybe_unused bcm_sysport_resume(struct device *d)
bcm_sysport_netif_start(dev);
+ netif_device_attach(dev);
+
return 0;
out_free_rx_ring:
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index 20c1681..2d6f090 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -2855,7 +2855,6 @@ static void bcmgenet_netif_start(struct net_device *dev)
umac_enable_set(priv, CMD_TX_EN | CMD_RX_EN, true);
- netif_tx_start_all_queues(dev);
bcmgenet_enable_tx_napi(priv);
/* Monitor link interrupts now */
@@ -2937,6 +2936,8 @@ static int bcmgenet_open(struct net_device *dev)
bcmgenet_netif_start(dev);
+ netif_tx_start_all_queues(dev);
+
return 0;
err_irq1:
@@ -2958,7 +2959,7 @@ static void bcmgenet_netif_stop(struct net_device *dev)
struct bcmgenet_priv *priv = netdev_priv(dev);
bcmgenet_disable_tx_napi(priv);
- netif_tx_stop_all_queues(dev);
+ netif_tx_disable(dev);
/* Disable MAC receive */
umac_enable_set(priv, CMD_RX_EN, false);
@@ -3620,13 +3621,13 @@ static int bcmgenet_suspend(struct device *d)
if (!netif_running(dev))
return 0;
+ netif_device_detach(dev);
+
bcmgenet_netif_stop(dev);
if (!device_may_wakeup(d))
phy_suspend(dev->phydev);
- netif_device_detach(dev);
-
/* Prepare the device for Wake-on-LAN and switch to the slow clock */
if (device_may_wakeup(d) && priv->wolopts) {
ret = bcmgenet_power_down(priv, GENET_POWER_WOL_MAGIC);
@@ -3700,8 +3701,6 @@ static int bcmgenet_resume(struct device *d)
/* Always enable ring 16 - descriptor ring */
bcmgenet_enable_dma(priv, dma_ctrl);
- netif_device_attach(dev);
-
if (!device_may_wakeup(d))
phy_resume(dev->phydev);
@@ -3710,6 +3709,8 @@ static int bcmgenet_resume(struct device *d)
bcmgenet_netif_start(dev);
+ netif_device_attach(dev);
+
return 0;
out_clk_disable:
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
index 3f96aa3..20fcf0d 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_enet.c
@@ -3760,7 +3760,8 @@ static int hns3_reset_notify_init_enet(struct hnae3_handle *handle)
/* Hardware table is only clear when pf resets */
if (!(handle->flags & HNAE3_SUPPORT_VF)) {
ret = hns3_restore_vlan(netdev);
- return ret;
+ if (ret)
+ return ret;
}
ret = hns3_restore_fd_rules(netdev);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c
index aa5cb98..494e562 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_tm.c
@@ -1168,14 +1168,14 @@ static int hclge_pfc_setup_hw(struct hclge_dev *hdev)
*/
static int hclge_bp_setup_hw(struct hclge_dev *hdev, u8 tc)
{
- struct hclge_vport *vport = hdev->vport;
- u32 i, k, qs_bitmap;
- int ret;
+ int i;
for (i = 0; i < HCLGE_BP_GRP_NUM; i++) {
- qs_bitmap = 0;
+ u32 qs_bitmap = 0;
+ int k, ret;
for (k = 0; k < hdev->num_alloc_vport; k++) {
+ struct hclge_vport *vport = &hdev->vport[k];
u16 qs_id = vport->qs_offset + tc;
u8 grp, sub_grp;
@@ -1185,8 +1185,6 @@ static int hclge_bp_setup_hw(struct hclge_dev *hdev, u8 tc)
HCLGE_BP_SUB_GRP_ID_S);
if (i == grp)
qs_bitmap |= (1 << sub_grp);
-
- vport++;
}
ret = hclge_tm_qs_bp_cfg(hdev, tc, i, qs_bitmap);
diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index 7893bef..c9d5d0a 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -1545,7 +1545,7 @@ static netdev_tx_t ibmvnic_xmit(struct sk_buff *skb, struct net_device *netdev)
tx_crq.v1.sge_len = cpu_to_be32(skb->len);
tx_crq.v1.ioba = cpu_to_be64(data_dma_addr);
- if (adapter->vlan_header_insertion) {
+ if (adapter->vlan_header_insertion && skb_vlan_tag_present(skb)) {
tx_crq.v1.flags2 |= IBMVNIC_TX_VLAN_INSERT;
tx_crq.v1.vlan_id = cpu_to_be16(skb->vlan_tci);
}
diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c
index bc71a21..21c2688 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_main.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_main.c
@@ -12249,6 +12249,8 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
NETIF_F_GSO_GRE |
NETIF_F_GSO_GRE_CSUM |
NETIF_F_GSO_PARTIAL |
+ NETIF_F_GSO_IPXIP4 |
+ NETIF_F_GSO_IPXIP6 |
NETIF_F_GSO_UDP_TUNNEL |
NETIF_F_GSO_UDP_TUNNEL_CSUM |
NETIF_F_SCTP_CRC |
@@ -12266,13 +12268,13 @@ static int i40e_config_netdev(struct i40e_vsi *vsi)
/* record features VLANs can make use of */
netdev->vlan_features |= hw_enc_features | NETIF_F_TSO_MANGLEID;
- if (!(pf->flags & I40E_FLAG_MFP_ENABLED))
- netdev->hw_features |= NETIF_F_NTUPLE | NETIF_F_HW_TC;
-
hw_features = hw_enc_features |
NETIF_F_HW_VLAN_CTAG_TX |
NETIF_F_HW_VLAN_CTAG_RX;
+ if (!(pf->flags & I40E_FLAG_MFP_ENABLED))
+ hw_features |= NETIF_F_NTUPLE | NETIF_F_HW_TC;
+
netdev->hw_features |= hw_features;
netdev->features |= hw_features | NETIF_F_HW_VLAN_CTAG_FILTER;
diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index 4c4b571..b854837 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -76,6 +76,8 @@ extern const char ice_drv_ver[];
#define ICE_MIN_INTR_PER_VF (ICE_MIN_QS_PER_VF + 1)
#define ICE_DFLT_INTR_PER_VF (ICE_DFLT_QS_PER_VF + 1)
+#define ICE_MAX_RESET_WAIT 20
+
#define ICE_VSIQF_HKEY_ARRAY_SIZE ((VSIQF_HKEY_MAX_INDEX + 1) * 4)
#define ICE_DFLT_NETIF_M (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK)
@@ -189,7 +191,6 @@ struct ice_vsi {
u64 tx_linearize;
DECLARE_BITMAP(state, __ICE_STATE_NBITS);
DECLARE_BITMAP(flags, ICE_VSI_FLAG_NBITS);
- unsigned long active_vlans[BITS_TO_LONGS(VLAN_N_VID)];
unsigned int current_netdev_flags;
u32 tx_restart;
u32 tx_busy;
@@ -369,5 +370,6 @@ int ice_set_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size);
int ice_get_rss(struct ice_vsi *vsi, u8 *seed, u8 *lut, u16 lut_size);
void ice_fill_rss_lut(u8 *lut, u16 rss_table_size, u16 rss_size);
void ice_print_link_msg(struct ice_vsi *vsi, bool isup);
+void ice_napi_del(struct ice_vsi *vsi);
#endif /* _ICE_H_ */
diff --git a/drivers/net/ethernet/intel/ice/ice_common.c b/drivers/net/ethernet/intel/ice/ice_common.c
index 8cd6a24..554fd70 100644
--- a/drivers/net/ethernet/intel/ice/ice_common.c
+++ b/drivers/net/ethernet/intel/ice/ice_common.c
@@ -811,6 +811,9 @@ void ice_deinit_hw(struct ice_hw *hw)
/* Attempt to disable FW logging before shutting down control queues */
ice_cfg_fw_log(hw, false);
ice_shutdown_all_ctrlq(hw);
+
+ /* Clear VSI contexts if not already cleared */
+ ice_clear_all_vsi_ctx(hw);
}
/**
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 9692358..648acdb 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -1517,10 +1517,15 @@ ice_set_pauseparam(struct net_device *netdev, struct ethtool_pauseparam *pause)
}
if (!test_bit(__ICE_DOWN, pf->state)) {
- /* Give it a little more time to try to come back */
+ /* Give it a little more time to try to come back. If still
+ * down, restart autoneg link or reinitialize the interface.
+ */
msleep(75);
if (!test_bit(__ICE_DOWN, pf->state))
return ice_nway_reset(netdev);
+
+ ice_down(vsi);
+ ice_up(vsi);
}
return err;
diff --git a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
index 5fdea6e..596b9fb 100644
--- a/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
+++ b/drivers/net/ethernet/intel/ice/ice_hw_autogen.h
@@ -242,6 +242,8 @@
#define GLNVM_ULD 0x000B6008
#define GLNVM_ULD_CORER_DONE_M BIT(3)
#define GLNVM_ULD_GLOBR_DONE_M BIT(4)
+#define GLPCI_CNF2 0x000BE004
+#define GLPCI_CNF2_CACHELINE_SIZE_M BIT(1)
#define PF_FUNC_RID 0x0009E880
#define PF_FUNC_RID_FUNC_NUM_S 0
#define PF_FUNC_RID_FUNC_NUM_M ICE_M(0x7, 0)
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 5bacad0..1041fa2 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -1997,7 +1997,7 @@ int ice_cfg_vlan_pruning(struct ice_vsi *vsi, bool ena)
status = ice_update_vsi(&vsi->back->hw, vsi->idx, ctxt, NULL);
if (status) {
netdev_err(vsi->netdev, "%sabling VLAN pruning on VSI handle: %d, VSI HW ID: %d failed, err = %d, aq_err = %d\n",
- ena ? "Ena" : "Dis", vsi->idx, vsi->vsi_num, status,
+ ena ? "En" : "Dis", vsi->idx, vsi->vsi_num, status,
vsi->back->hw.adminq.sq_last_status);
goto err_out;
}
@@ -2458,6 +2458,7 @@ int ice_vsi_release(struct ice_vsi *vsi)
* on this wq
*/
if (vsi->netdev && !ice_is_reset_in_progress(pf->state)) {
+ ice_napi_del(vsi);
unregister_netdev(vsi->netdev);
free_netdev(vsi->netdev);
vsi->netdev = NULL;
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 0599345..333312a 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -1465,7 +1465,7 @@ static int ice_req_irq_msix_misc(struct ice_pf *pf)
* ice_napi_del - Remove NAPI handler for the VSI
* @vsi: VSI for which NAPI handler is to be removed
*/
-static void ice_napi_del(struct ice_vsi *vsi)
+void ice_napi_del(struct ice_vsi *vsi)
{
int v_idx;
@@ -1622,7 +1622,6 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
{
struct ice_netdev_priv *np = netdev_priv(netdev);
struct ice_vsi *vsi = np->vsi;
- int ret;
if (vid >= VLAN_N_VID) {
netdev_err(netdev, "VLAN id requested %d is out of range %d\n",
@@ -1635,7 +1634,8 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
/* Enable VLAN pruning when VLAN 0 is added */
if (unlikely(!vid)) {
- ret = ice_cfg_vlan_pruning(vsi, true);
+ int ret = ice_cfg_vlan_pruning(vsi, true);
+
if (ret)
return ret;
}
@@ -1644,12 +1644,7 @@ static int ice_vlan_rx_add_vid(struct net_device *netdev,
* needed to continue allowing all untagged packets since VLAN prune
* list is applied to all packets by the switch
*/
- ret = ice_vsi_add_vlan(vsi, vid);
-
- if (!ret)
- set_bit(vid, vsi->active_vlans);
-
- return ret;
+ return ice_vsi_add_vlan(vsi, vid);
}
/**
@@ -1677,8 +1672,6 @@ static int ice_vlan_rx_kill_vid(struct net_device *netdev,
if (status)
return status;
- clear_bit(vid, vsi->active_vlans);
-
/* Disable VLAN pruning when VLAN 0 is removed */
if (unlikely(!vid))
status = ice_cfg_vlan_pruning(vsi, false);
@@ -2002,6 +1995,22 @@ static int ice_init_interrupt_scheme(struct ice_pf *pf)
}
/**
+ * ice_verify_cacheline_size - verify driver's assumption of 64 Byte cache lines
+ * @pf: pointer to the PF structure
+ *
+ * There is no error returned here because the driver should be able to handle
+ * 128 Byte cache lines, so we only print a warning in case issues are seen,
+ * specifically with Tx.
+ */
+static void ice_verify_cacheline_size(struct ice_pf *pf)
+{
+ if (rd32(&pf->hw, GLPCI_CNF2) & GLPCI_CNF2_CACHELINE_SIZE_M)
+ dev_warn(&pf->pdev->dev,
+ "%d Byte cache line assumption is invalid, driver may have Tx timeouts!\n",
+ ICE_CACHE_LINE_BYTES);
+}
+
+/**
* ice_probe - Device initialization routine
* @pdev: PCI device information struct
* @ent: entry in ice_pci_tbl
@@ -2151,6 +2160,8 @@ static int ice_probe(struct pci_dev *pdev,
/* since everything is good, start the service timer */
mod_timer(&pf->serv_tmr, round_jiffies(jiffies + pf->serv_tmr_period));
+ ice_verify_cacheline_size(pf);
+
return 0;
err_alloc_sw_unroll:
@@ -2182,6 +2193,12 @@ static void ice_remove(struct pci_dev *pdev)
if (!pf)
return;
+ for (i = 0; i < ICE_MAX_RESET_WAIT; i++) {
+ if (!ice_is_reset_in_progress(pf->state))
+ break;
+ msleep(100);
+ }
+
set_bit(__ICE_DOWN, pf->state);
ice_service_task_stop(pf);
@@ -2510,31 +2527,6 @@ static int ice_vsi_vlan_setup(struct ice_vsi *vsi)
}
/**
- * ice_restore_vlan - Reinstate VLANs when vsi/netdev comes back up
- * @vsi: the VSI being brought back up
- */
-static int ice_restore_vlan(struct ice_vsi *vsi)
-{
- int err;
- u16 vid;
-
- if (!vsi->netdev)
- return -EINVAL;
-
- err = ice_vsi_vlan_setup(vsi);
- if (err)
- return err;
-
- for_each_set_bit(vid, vsi->active_vlans, VLAN_N_VID) {
- err = ice_vlan_rx_add_vid(vsi->netdev, htons(ETH_P_8021Q), vid);
- if (err)
- break;
- }
-
- return err;
-}
-
-/**
* ice_vsi_cfg - Setup the VSI
* @vsi: the VSI being configured
*
@@ -2546,7 +2538,9 @@ static int ice_vsi_cfg(struct ice_vsi *vsi)
if (vsi->netdev) {
ice_set_rx_mode(vsi->netdev);
- err = ice_restore_vlan(vsi);
+
+ err = ice_vsi_vlan_setup(vsi);
+
if (err)
return err;
}
@@ -3296,7 +3290,7 @@ static void ice_rebuild(struct ice_pf *pf)
struct device *dev = &pf->pdev->dev;
struct ice_hw *hw = &pf->hw;
enum ice_status ret;
- int err;
+ int err, i;
if (test_bit(__ICE_DOWN, pf->state))
goto clear_recovery;
@@ -3370,6 +3364,22 @@ static void ice_rebuild(struct ice_pf *pf)
}
ice_reset_all_vfs(pf, true);
+
+ for (i = 0; i < pf->num_alloc_vsi; i++) {
+ bool link_up;
+
+ if (!pf->vsi[i] || pf->vsi[i]->type != ICE_VSI_PF)
+ continue;
+ ice_get_link_status(pf->vsi[i]->port_info, &link_up);
+ if (link_up) {
+ netif_carrier_on(pf->vsi[i]->netdev);
+ netif_tx_wake_all_queues(pf->vsi[i]->netdev);
+ } else {
+ netif_carrier_off(pf->vsi[i]->netdev);
+ netif_tx_stop_all_queues(pf->vsi[i]->netdev);
+ }
+ }
+
/* if we get here, reset flow is successful */
clear_bit(__ICE_RESET_FAILED, pf->state);
return;
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.c b/drivers/net/ethernet/intel/ice/ice_switch.c
index 33403f3..40c9c65 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.c
+++ b/drivers/net/ethernet/intel/ice/ice_switch.c
@@ -348,6 +348,18 @@ static void ice_clear_vsi_ctx(struct ice_hw *hw, u16 vsi_handle)
}
/**
+ * ice_clear_all_vsi_ctx - clear all the VSI context entries
+ * @hw: pointer to the hw struct
+ */
+void ice_clear_all_vsi_ctx(struct ice_hw *hw)
+{
+ u16 i;
+
+ for (i = 0; i < ICE_MAX_VSI; i++)
+ ice_clear_vsi_ctx(hw, i);
+}
+
+/**
* ice_add_vsi - add VSI context to the hardware and VSI handle list
* @hw: pointer to the hw struct
* @vsi_handle: unique VSI handle provided by drivers
diff --git a/drivers/net/ethernet/intel/ice/ice_switch.h b/drivers/net/ethernet/intel/ice/ice_switch.h
index b88d96a..d5ef0bd 100644
--- a/drivers/net/ethernet/intel/ice/ice_switch.h
+++ b/drivers/net/ethernet/intel/ice/ice_switch.h
@@ -190,6 +190,8 @@ ice_update_vsi(struct ice_hw *hw, u16 vsi_handle, struct ice_vsi_ctx *vsi_ctx,
struct ice_sq_cd *cd);
bool ice_is_vsi_valid(struct ice_hw *hw, u16 vsi_handle);
struct ice_vsi_ctx *ice_get_vsi_ctx(struct ice_hw *hw, u16 vsi_handle);
+void ice_clear_all_vsi_ctx(struct ice_hw *hw);
+/* Switch config */
enum ice_status ice_get_initial_sw_cfg(struct ice_hw *hw);
/* Switch/bridge related commands */
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.c b/drivers/net/ethernet/intel/ice/ice_txrx.c
index 5dae968..fe5bbab 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.c
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.c
@@ -1520,7 +1520,7 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)
/* update gso_segs and bytecount */
first->gso_segs = skb_shinfo(skb)->gso_segs;
- first->bytecount = (first->gso_segs - 1) * off->header_len;
+ first->bytecount += (first->gso_segs - 1) * off->header_len;
cd_tso_len = skb->len - off->header_len;
cd_mss = skb_shinfo(skb)->gso_size;
@@ -1556,15 +1556,15 @@ int ice_tso(struct ice_tx_buf *first, struct ice_tx_offload_params *off)
* magnitude greater than our largest possible GSO size.
*
* This would then be implemented as:
- * return (((size >> 12) * 85) >> 8) + 1;
+ * return (((size >> 12) * 85) >> 8) + ICE_DESCS_FOR_SKB_DATA_PTR;
*
* Since multiplication and division are commutative, we can reorder
* operations into:
- * return ((size * 85) >> 20) + 1;
+ * return ((size * 85) >> 20) + ICE_DESCS_FOR_SKB_DATA_PTR;
*/
static unsigned int ice_txd_use_count(unsigned int size)
{
- return ((size * 85) >> 20) + 1;
+ return ((size * 85) >> 20) + ICE_DESCS_FOR_SKB_DATA_PTR;
}
/**
@@ -1706,7 +1706,8 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_ring *tx_ring)
* + 1 desc for context descriptor,
* otherwise try next time
*/
- if (ice_maybe_stop_tx(tx_ring, count + 4 + 1)) {
+ if (ice_maybe_stop_tx(tx_ring, count + ICE_DESCS_PER_CACHE_LINE +
+ ICE_DESCS_FOR_CTX_DESC)) {
tx_ring->tx_stats.tx_busy++;
return NETDEV_TX_BUSY;
}
diff --git a/drivers/net/ethernet/intel/ice/ice_txrx.h b/drivers/net/ethernet/intel/ice/ice_txrx.h
index 1d0f58b..75d0eaf 100644
--- a/drivers/net/ethernet/intel/ice/ice_txrx.h
+++ b/drivers/net/ethernet/intel/ice/ice_txrx.h
@@ -22,8 +22,21 @@
#define ICE_RX_BUF_WRITE 16 /* Must be power of 2 */
#define ICE_MAX_TXQ_PER_TXQG 128
-/* Tx Descriptors needed, worst case */
-#define DESC_NEEDED (MAX_SKB_FRAGS + 4)
+/* We are assuming that the cache line is always 64 Bytes here for ice.
+ * In order to make sure that is a correct assumption there is a check in probe
+ * to print a warning if the read from GLPCI_CNF2 tells us that the cache line
+ * size is 128 bytes. We do it this way because we do not want to read the
+ * GLPCI_CNF2 register or a variable containing the value on every pass through
+ * the Tx path.
+ */
+#define ICE_CACHE_LINE_BYTES 64
+#define ICE_DESCS_PER_CACHE_LINE (ICE_CACHE_LINE_BYTES / \
+ sizeof(struct ice_tx_desc))
+#define ICE_DESCS_FOR_CTX_DESC 1
+#define ICE_DESCS_FOR_SKB_DATA_PTR 1
+/* Tx descriptors needed, worst case */
+#define DESC_NEEDED (MAX_SKB_FRAGS + ICE_DESCS_FOR_CTX_DESC + \
+ ICE_DESCS_PER_CACHE_LINE + ICE_DESCS_FOR_SKB_DATA_PTR)
#define ICE_DESC_UNUSED(R) \
((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \
(R)->next_to_clean - (R)->next_to_use - 1)
diff --git a/drivers/net/ethernet/intel/ice/ice_type.h b/drivers/net/ethernet/intel/ice/ice_type.h
index 12f9432..f4dbc81 100644
--- a/drivers/net/ethernet/intel/ice/ice_type.h
+++ b/drivers/net/ethernet/intel/ice/ice_type.h
@@ -92,12 +92,12 @@ struct ice_link_status {
u64 phy_type_low;
u16 max_frame_size;
u16 link_speed;
+ u16 req_speeds;
u8 lse_ena; /* Link Status Event notification */
u8 link_info;
u8 an_info;
u8 ext_info;
u8 pacing;
- u8 req_speeds;
/* Refer to #define from module_type[ICE_MODULE_TYPE_TOTAL_BYTE] of
* ice_aqc_get_phy_caps structure
*/
diff --git a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
index 45f10f8..e71065f 100644
--- a/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
+++ b/drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c
@@ -348,7 +348,7 @@ static int ice_vsi_set_pvid(struct ice_vsi *vsi, u16 vid)
struct ice_vsi_ctx ctxt = { 0 };
enum ice_status status;
- ctxt.info.vlan_flags = ICE_AQ_VSI_VLAN_MODE_TAGGED |
+ ctxt.info.vlan_flags = ICE_AQ_VSI_VLAN_MODE_UNTAGGED |
ICE_AQ_VSI_PVLAN_INSERT_PVID |
ICE_AQ_VSI_VLAN_EMOD_STR;
ctxt.info.pvid = cpu_to_le16(vid);
@@ -2171,7 +2171,6 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
if (!ice_vsi_add_vlan(vsi, vid)) {
vf->num_vlan++;
- set_bit(vid, vsi->active_vlans);
/* Enable VLAN pruning when VLAN 0 is added */
if (unlikely(!vid))
@@ -2190,7 +2189,6 @@ static int ice_vc_process_vlan_msg(struct ice_vf *vf, u8 *msg, bool add_v)
*/
if (!ice_vsi_kill_vlan(vsi, vid)) {
vf->num_vlan--;
- clear_bit(vid, vsi->active_vlans);
/* Disable VLAN pruning when removing VLAN 0 */
if (unlikely(!vid))
diff --git a/drivers/net/ethernet/intel/igb/igb_ptp.c b/drivers/net/ethernet/intel/igb/igb_ptp.c
index 29ced6b..2b95dc9 100644
--- a/drivers/net/ethernet/intel/igb/igb_ptp.c
+++ b/drivers/net/ethernet/intel/igb/igb_ptp.c
@@ -53,13 +53,15 @@
* 2^40 * 10^-9 / 60 = 18.3 minutes.
*
* SYSTIM is converted to real time using a timecounter. As
- * timecounter_cyc2time() allows old timestamps, the timecounter
- * needs to be updated at least once per half of the SYSTIM interval.
- * Scheduling of delayed work is not very accurate, so we aim for 8
- * minutes to be sure the actual interval is shorter than 9.16 minutes.
+ * timecounter_cyc2time() allows old timestamps, the timecounter needs
+ * to be updated at least once per half of the SYSTIM interval.
+ * Scheduling of delayed work is not very accurate, and also the NIC
+ * clock can be adjusted to run up to 6% faster and the system clock
+ * up to 10% slower, so we aim for 6 minutes to be sure the actual
+ * interval in the NIC time is shorter than 9.16 minutes.
*/
-#define IGB_SYSTIM_OVERFLOW_PERIOD (HZ * 60 * 8)
+#define IGB_SYSTIM_OVERFLOW_PERIOD (HZ * 60 * 6)
#define IGB_PTP_TX_TIMEOUT (HZ * 15)
#define INCPERIOD_82576 BIT(E1000_TIMINCA_16NS_SHIFT)
#define INCVALUE_82576_MASK GENMASK(E1000_TIMINCA_16NS_SHIFT - 1, 0)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 5bfd349..3ba672e 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -494,7 +494,7 @@ struct mvneta_port {
#if defined(__LITTLE_ENDIAN)
struct mvneta_tx_desc {
u32 command; /* Options used by HW for packet transmitting.*/
- u16 reserverd1; /* csum_l4 (for future use) */
+ u16 reserved1; /* csum_l4 (for future use) */
u16 data_size; /* Data size of transmitted packet in bytes */
u32 buf_phys_addr; /* Physical addr of transmitted buffer */
u32 reserved2; /* hw_cmd - (for future use, PMT) */
@@ -519,7 +519,7 @@ struct mvneta_rx_desc {
#else
struct mvneta_tx_desc {
u16 data_size; /* Data size of transmitted packet in bytes */
- u16 reserverd1; /* csum_l4 (for future use) */
+ u16 reserved1; /* csum_l4 (for future use) */
u32 command; /* Options used by HW for packet transmitting.*/
u32 reserved2; /* hw_cmd - (for future use, PMT) */
u32 buf_phys_addr; /* Physical addr of transmitted buffer */
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 1857ee0..6f5153a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -1006,7 +1006,6 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
ring->packets++;
}
ring->bytes += tx_info->nr_bytes;
- netdev_tx_sent_queue(ring->tx_queue, tx_info->nr_bytes);
AVG_PERF_COUNTER(priv->pstats.tx_pktsz_avg, skb->len);
if (tx_info->inl)
@@ -1044,7 +1043,10 @@ netdev_tx_t mlx4_en_xmit(struct sk_buff *skb, struct net_device *dev)
netif_tx_stop_queue(ring->tx_queue);
ring->queue_stopped++;
}
- send_doorbell = !skb->xmit_more || netif_xmit_stopped(ring->tx_queue);
+
+ send_doorbell = __netdev_tx_sent_queue(ring->tx_queue,
+ tx_info->nr_bytes,
+ skb->xmit_more);
real_size = (real_size / 16) & 0x3f;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index a2df12b..9bec9403 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -3568,7 +3568,6 @@ static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core)
burst_size = 7;
break;
case MLXSW_REG_HTGT_TRAP_GROUP_SP_IP2ME:
- is_bytes = true;
rate = 4 * 1024;
burst_size = 4;
break;
diff --git a/drivers/net/ethernet/qlogic/qed/qed_fcoe.c b/drivers/net/ethernet/qlogic/qed/qed_fcoe.c
index cc1b373..46dc93d 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_fcoe.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_fcoe.c
@@ -147,7 +147,8 @@ qed_sp_fcoe_func_start(struct qed_hwfn *p_hwfn,
"Cannot satisfy CQ amount. CQs requested %d, CQs available %d. Aborting function start\n",
fcoe_pf_params->num_cqs,
p_hwfn->hw_info.feat_num[QED_FCOE_CQ]);
- return -EINVAL;
+ rc = -EINVAL;
+ goto err;
}
p_data->mtu = cpu_to_le16(fcoe_pf_params->mtu);
@@ -156,14 +157,14 @@ qed_sp_fcoe_func_start(struct qed_hwfn *p_hwfn,
rc = qed_cxt_acquire_cid(p_hwfn, PROTOCOLID_FCOE, &dummy_cid);
if (rc)
- return rc;
+ goto err;
cxt_info.iid = dummy_cid;
rc = qed_cxt_get_cid_info(p_hwfn, &cxt_info);
if (rc) {
DP_NOTICE(p_hwfn, "Cannot find context info for dummy cid=%d\n",
dummy_cid);
- return rc;
+ goto err;
}
p_cxt = cxt_info.p_cxt;
SET_FIELD(p_cxt->tstorm_ag_context.flags3,
@@ -240,6 +241,10 @@ qed_sp_fcoe_func_start(struct qed_hwfn *p_hwfn,
rc = qed_spq_post(p_hwfn, p_ent, NULL);
return rc;
+
+err:
+ qed_sp_destroy_request(p_hwfn, p_ent);
+ return rc;
}
static int
diff --git a/drivers/net/ethernet/qlogic/qed/qed_iscsi.c b/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
index 1135387..4f8a685 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_iscsi.c
@@ -200,6 +200,7 @@ qed_sp_iscsi_func_start(struct qed_hwfn *p_hwfn,
"Cannot satisfy CQ amount. Queues requested %d, CQs available %d. Aborting function start\n",
p_params->num_queues,
p_hwfn->hw_info.feat_num[QED_ISCSI_CQ]);
+ qed_sp_destroy_request(p_hwfn, p_ent);
return -EINVAL;
}
diff --git a/drivers/net/ethernet/qlogic/qed/qed_l2.c b/drivers/net/ethernet/qlogic/qed/qed_l2.c
index 82a1bd1..67c02ea93 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_l2.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_l2.c
@@ -740,8 +740,7 @@ int qed_sp_vport_update(struct qed_hwfn *p_hwfn,
rc = qed_sp_vport_update_rss(p_hwfn, p_ramrod, p_rss_params);
if (rc) {
- /* Return spq entry which is taken in qed_sp_init_request()*/
- qed_spq_return_entry(p_hwfn, p_ent);
+ qed_sp_destroy_request(p_hwfn, p_ent);
return rc;
}
@@ -1355,6 +1354,7 @@ qed_filter_ucast_common(struct qed_hwfn *p_hwfn,
DP_NOTICE(p_hwfn,
"%d is not supported yet\n",
p_filter_cmd->opcode);
+ qed_sp_destroy_request(p_hwfn, *pp_ent);
return -EINVAL;
}
@@ -2056,13 +2056,13 @@ qed_configure_rfs_ntuple_filter(struct qed_hwfn *p_hwfn,
} else {
rc = qed_fw_vport(p_hwfn, p_params->vport_id, &abs_vport_id);
if (rc)
- return rc;
+ goto err;
if (p_params->qid != QED_RFS_NTUPLE_QID_RSS) {
rc = qed_fw_l2_queue(p_hwfn, p_params->qid,
&abs_rx_q_id);
if (rc)
- return rc;
+ goto err;
p_ramrod->rx_qid_valid = 1;
p_ramrod->rx_qid = cpu_to_le16(abs_rx_q_id);
@@ -2083,6 +2083,10 @@ qed_configure_rfs_ntuple_filter(struct qed_hwfn *p_hwfn,
(u64)p_params->addr, p_params->length);
return qed_spq_post(p_hwfn, p_ent, NULL);
+
+err:
+ qed_sp_destroy_request(p_hwfn, p_ent);
+ return rc;
}
int qed_get_rxq_coalesce(struct qed_hwfn *p_hwfn,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_mcp.c b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
index f40f654..a96364d 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_mcp.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_mcp.c
@@ -1944,9 +1944,12 @@ int qed_mcp_trans_speed_mask(struct qed_hwfn *p_hwfn,
struct qed_ptt *p_ptt, u32 *p_speed_mask)
{
u32 transceiver_type, transceiver_state;
+ int ret;
- qed_mcp_get_transceiver_data(p_hwfn, p_ptt, &transceiver_state,
- &transceiver_type);
+ ret = qed_mcp_get_transceiver_data(p_hwfn, p_ptt, &transceiver_state,
+ &transceiver_type);
+ if (ret)
+ return ret;
if (qed_is_transceiver_ready(transceiver_state, transceiver_type) ==
false)
diff --git a/drivers/net/ethernet/qlogic/qed/qed_rdma.c b/drivers/net/ethernet/qlogic/qed/qed_rdma.c
index c71391b..6211343 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_rdma.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_rdma.c
@@ -1514,6 +1514,7 @@ qed_rdma_register_tid(void *rdma_cxt,
default:
rc = -EINVAL;
DP_VERBOSE(p_hwfn, QED_MSG_RDMA, "rc = %d\n", rc);
+ qed_sp_destroy_request(p_hwfn, p_ent);
return rc;
}
SET_FIELD(p_ramrod->flags1,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_roce.c b/drivers/net/ethernet/qlogic/qed/qed_roce.c
index f9167d1..e49fada 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_roce.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_roce.c
@@ -745,6 +745,7 @@ static int qed_roce_sp_destroy_qp_responder(struct qed_hwfn *p_hwfn,
DP_NOTICE(p_hwfn,
"qed destroy responder failed: cannot allocate memory (ramrod). rc = %d\n",
rc);
+ qed_sp_destroy_request(p_hwfn, p_ent);
return rc;
}
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp.h b/drivers/net/ethernet/qlogic/qed/qed_sp.h
index e95431f..3157c0d 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp.h
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp.h
@@ -167,6 +167,9 @@ struct qed_spq_entry {
enum spq_mode comp_mode;
struct qed_spq_comp_cb comp_cb;
struct qed_spq_comp_done comp_done; /* SPQ_MODE_EBLOCK */
+
+ /* Posted entry for unlimited list entry in EBLOCK mode */
+ struct qed_spq_entry *post_ent;
};
struct qed_eq {
@@ -396,6 +399,17 @@ struct qed_sp_init_data {
struct qed_spq_comp_cb *p_comp_data;
};
+/**
+ * @brief Returns a SPQ entry to the pool / frees the entry if allocated.
+ * Should be called on in error flows after initializing the SPQ entry
+ * and before posting it.
+ *
+ * @param p_hwfn
+ * @param p_ent
+ */
+void qed_sp_destroy_request(struct qed_hwfn *p_hwfn,
+ struct qed_spq_entry *p_ent);
+
int qed_sp_init_request(struct qed_hwfn *p_hwfn,
struct qed_spq_entry **pp_ent,
u8 cmd,
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
index 77b6248..888274f 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sp_commands.c
@@ -47,6 +47,19 @@
#include "qed_sp.h"
#include "qed_sriov.h"
+void qed_sp_destroy_request(struct qed_hwfn *p_hwfn,
+ struct qed_spq_entry *p_ent)
+{
+ /* qed_spq_get_entry() can either get an entry from the free_pool,
+ * or, if no entries are left, allocate a new entry and add it to
+ * the unlimited_pending list.
+ */
+ if (p_ent->queue == &p_hwfn->p_spq->unlimited_pending)
+ kfree(p_ent);
+ else
+ qed_spq_return_entry(p_hwfn, p_ent);
+}
+
int qed_sp_init_request(struct qed_hwfn *p_hwfn,
struct qed_spq_entry **pp_ent,
u8 cmd, u8 protocol, struct qed_sp_init_data *p_data)
@@ -80,7 +93,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
case QED_SPQ_MODE_BLOCK:
if (!p_data->p_comp_data)
- return -EINVAL;
+ goto err;
p_ent->comp_cb.cookie = p_data->p_comp_data->cookie;
break;
@@ -95,7 +108,7 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
default:
DP_NOTICE(p_hwfn, "Unknown SPQE completion mode %d\n",
p_ent->comp_mode);
- return -EINVAL;
+ goto err;
}
DP_VERBOSE(p_hwfn, QED_MSG_SPQ,
@@ -109,6 +122,11 @@ int qed_sp_init_request(struct qed_hwfn *p_hwfn,
memset(&p_ent->ramrod, 0, sizeof(p_ent->ramrod));
return 0;
+
+err:
+ qed_sp_destroy_request(p_hwfn, p_ent);
+
+ return -EINVAL;
}
static enum tunnel_clss qed_tunn_clss_to_fw_clss(u8 type)
diff --git a/drivers/net/ethernet/qlogic/qed/qed_spq.c b/drivers/net/ethernet/qlogic/qed/qed_spq.c
index c4a6274..0a9c5bb 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_spq.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_spq.c
@@ -142,6 +142,7 @@ static int qed_spq_block(struct qed_hwfn *p_hwfn,
DP_INFO(p_hwfn, "Ramrod is stuck, requesting MCP drain\n");
rc = qed_mcp_drain(p_hwfn, p_ptt);
+ qed_ptt_release(p_hwfn, p_ptt);
if (rc) {
DP_NOTICE(p_hwfn, "MCP drain failed\n");
goto err;
@@ -150,18 +151,15 @@ static int qed_spq_block(struct qed_hwfn *p_hwfn,
/* Retry after drain */
rc = __qed_spq_block(p_hwfn, p_ent, p_fw_ret, true);
if (!rc)
- goto out;
+ return 0;
comp_done = (struct qed_spq_comp_done *)p_ent->comp_cb.cookie;
- if (comp_done->done == 1)
+ if (comp_done->done == 1) {
if (p_fw_ret)
*p_fw_ret = comp_done->fw_return_code;
-out:
- qed_ptt_release(p_hwfn, p_ptt);
- return 0;
-
+ return 0;
+ }
err:
- qed_ptt_release(p_hwfn, p_ptt);
DP_NOTICE(p_hwfn,
"Ramrod is stuck [CID %08x cmd %02x protocol %02x echo %04x]\n",
le32_to_cpu(p_ent->elem.hdr.cid),
@@ -685,6 +683,8 @@ static int qed_spq_add_entry(struct qed_hwfn *p_hwfn,
/* EBLOCK responsible to free the allocated p_ent */
if (p_ent->comp_mode != QED_SPQ_MODE_EBLOCK)
kfree(p_ent);
+ else
+ p_ent->post_ent = p_en2;
p_ent = p_en2;
}
@@ -767,6 +767,25 @@ static int qed_spq_pend_post(struct qed_hwfn *p_hwfn)
SPQ_HIGH_PRI_RESERVE_DEFAULT);
}
+/* Avoid overriding of SPQ entries when getting out-of-order completions, by
+ * marking the completions in a bitmap and increasing the chain consumer only
+ * for the first successive completed entries.
+ */
+static void qed_spq_comp_bmap_update(struct qed_hwfn *p_hwfn, __le16 echo)
+{
+ u16 pos = le16_to_cpu(echo) % SPQ_RING_SIZE;
+ struct qed_spq *p_spq = p_hwfn->p_spq;
+
+ __set_bit(pos, p_spq->p_comp_bitmap);
+ while (test_bit(p_spq->comp_bitmap_idx,
+ p_spq->p_comp_bitmap)) {
+ __clear_bit(p_spq->comp_bitmap_idx,
+ p_spq->p_comp_bitmap);
+ p_spq->comp_bitmap_idx++;
+ qed_chain_return_produced(&p_spq->chain);
+ }
+}
+
int qed_spq_post(struct qed_hwfn *p_hwfn,
struct qed_spq_entry *p_ent, u8 *fw_return_code)
{
@@ -824,11 +843,12 @@ int qed_spq_post(struct qed_hwfn *p_hwfn,
p_ent->queue == &p_spq->unlimited_pending);
if (p_ent->queue == &p_spq->unlimited_pending) {
- /* This is an allocated p_ent which does not need to
- * return to pool.
- */
+ struct qed_spq_entry *p_post_ent = p_ent->post_ent;
+
kfree(p_ent);
- return rc;
+
+ /* Return the entry which was actually posted */
+ p_ent = p_post_ent;
}
if (rc)
@@ -842,7 +862,7 @@ int qed_spq_post(struct qed_hwfn *p_hwfn,
spq_post_fail2:
spin_lock_bh(&p_spq->lock);
list_del(&p_ent->list);
- qed_chain_return_produced(&p_spq->chain);
+ qed_spq_comp_bmap_update(p_hwfn, p_ent->elem.hdr.echo);
spq_post_fail:
/* return to the free pool */
@@ -874,25 +894,8 @@ int qed_spq_completion(struct qed_hwfn *p_hwfn,
spin_lock_bh(&p_spq->lock);
list_for_each_entry_safe(p_ent, tmp, &p_spq->completion_pending, list) {
if (p_ent->elem.hdr.echo == echo) {
- u16 pos = le16_to_cpu(echo) % SPQ_RING_SIZE;
-
list_del(&p_ent->list);
-
- /* Avoid overriding of SPQ entries when getting
- * out-of-order completions, by marking the completions
- * in a bitmap and increasing the chain consumer only
- * for the first successive completed entries.
- */
- __set_bit(pos, p_spq->p_comp_bitmap);
-
- while (test_bit(p_spq->comp_bitmap_idx,
- p_spq->p_comp_bitmap)) {
- __clear_bit(p_spq->comp_bitmap_idx,
- p_spq->p_comp_bitmap);
- p_spq->comp_bitmap_idx++;
- qed_chain_return_produced(&p_spq->chain);
- }
-
+ qed_spq_comp_bmap_update(p_hwfn, echo);
p_spq->comp_count++;
found = p_ent;
break;
@@ -931,11 +934,9 @@ int qed_spq_completion(struct qed_hwfn *p_hwfn,
QED_MSG_SPQ,
"Got a completion without a callback function\n");
- if ((found->comp_mode != QED_SPQ_MODE_EBLOCK) ||
- (found->queue == &p_spq->unlimited_pending))
+ if (found->comp_mode != QED_SPQ_MODE_EBLOCK)
/* EBLOCK is responsible for returning its own entry into the
- * free list, unless it originally added the entry into the
- * unlimited pending list.
+ * free list.
*/
qed_spq_return_entry(p_hwfn, found);
diff --git a/drivers/net/ethernet/qlogic/qed/qed_sriov.c b/drivers/net/ethernet/qlogic/qed/qed_sriov.c
index 9b08a9d..ca6290f 100644
--- a/drivers/net/ethernet/qlogic/qed/qed_sriov.c
+++ b/drivers/net/ethernet/qlogic/qed/qed_sriov.c
@@ -101,6 +101,7 @@ static int qed_sp_vf_start(struct qed_hwfn *p_hwfn, struct qed_vf_info *p_vf)
default:
DP_NOTICE(p_hwfn, "Unknown VF personality %d\n",
p_hwfn->hw_info.personality);
+ qed_sp_destroy_request(p_hwfn, p_ent);
return -EINVAL;
}
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
index 9647578..14f26bf 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_io.c
@@ -459,7 +459,7 @@ static int qlcnic_tx_pkt(struct qlcnic_adapter *adapter,
struct cmd_desc_type0 *first_desc, struct sk_buff *skb,
struct qlcnic_host_tx_ring *tx_ring)
{
- u8 l4proto, opcode = 0, hdr_len = 0;
+ u8 l4proto, opcode = 0, hdr_len = 0, tag_vlan = 0;
u16 flags = 0, vlan_tci = 0;
int copied, offset, copy_len, size;
struct cmd_desc_type0 *hwdesc;
@@ -472,14 +472,16 @@ static int qlcnic_tx_pkt(struct qlcnic_adapter *adapter,
flags = QLCNIC_FLAGS_VLAN_TAGGED;
vlan_tci = ntohs(vh->h_vlan_TCI);
protocol = ntohs(vh->h_vlan_encapsulated_proto);
+ tag_vlan = 1;
} else if (skb_vlan_tag_present(skb)) {
flags = QLCNIC_FLAGS_VLAN_OOB;
vlan_tci = skb_vlan_tag_get(skb);
+ tag_vlan = 1;
}
if (unlikely(adapter->tx_pvid)) {
- if (vlan_tci && !(adapter->flags & QLCNIC_TAGGING_ENABLED))
+ if (tag_vlan && !(adapter->flags & QLCNIC_TAGGING_ENABLED))
return -EIO;
- if (vlan_tci && (adapter->flags & QLCNIC_TAGGING_ENABLED))
+ if (tag_vlan && (adapter->flags & QLCNIC_TAGGING_ENABLED))
goto set_flags;
flags = QLCNIC_FLAGS_VLAN_OOB;
diff --git a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c
index 0afc3d3..d11c16a 100644
--- a/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c
+++ b/drivers/net/ethernet/qualcomm/rmnet/rmnet_vnd.c
@@ -234,7 +234,7 @@ int rmnet_vnd_newlink(u8 id, struct net_device *rmnet_dev,
struct net_device *real_dev,
struct rmnet_endpoint *ep)
{
- struct rmnet_priv *priv;
+ struct rmnet_priv *priv = netdev_priv(rmnet_dev);
int rc;
if (ep->egress_dev)
@@ -247,6 +247,8 @@ int rmnet_vnd_newlink(u8 id, struct net_device *rmnet_dev,
rmnet_dev->hw_features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
rmnet_dev->hw_features |= NETIF_F_SG;
+ priv->real_dev = real_dev;
+
rc = register_netdevice(rmnet_dev);
if (!rc) {
ep->egress_dev = rmnet_dev;
@@ -255,9 +257,7 @@ int rmnet_vnd_newlink(u8 id, struct net_device *rmnet_dev,
rmnet_dev->rtnl_link_ops = &rmnet_link_ops;
- priv = netdev_priv(rmnet_dev);
priv->mux_id = id;
- priv->real_dev = real_dev;
netdev_dbg(rmnet_dev, "rmnet dev created\n");
}
diff --git a/drivers/net/ethernet/stmicro/stmmac/common.h b/drivers/net/ethernet/stmicro/stmmac/common.h
index b1b305f..272b9ca6 100644
--- a/drivers/net/ethernet/stmicro/stmmac/common.h
+++ b/drivers/net/ethernet/stmicro/stmmac/common.h
@@ -365,7 +365,8 @@ struct dma_features {
/* GMAC TX FIFO is 8K, Rx FIFO is 16K */
#define BUF_SIZE_16KiB 16384
-#define BUF_SIZE_8KiB 8192
+/* RX Buffer size must be < 8191 and multiple of 4/8/16 bytes */
+#define BUF_SIZE_8KiB 8188
#define BUF_SIZE_4KiB 4096
#define BUF_SIZE_2KiB 2048
diff --git a/drivers/net/ethernet/stmicro/stmmac/descs_com.h b/drivers/net/ethernet/stmicro/stmmac/descs_com.h
index ca9d7e4..40d6356 100644
--- a/drivers/net/ethernet/stmicro/stmmac/descs_com.h
+++ b/drivers/net/ethernet/stmicro/stmmac/descs_com.h
@@ -31,7 +31,7 @@
/* Enhanced descriptors */
static inline void ehn_desc_rx_set_on_ring(struct dma_desc *p, int end)
{
- p->des1 |= cpu_to_le32(((BUF_SIZE_8KiB - 1)
+ p->des1 |= cpu_to_le32((BUF_SIZE_8KiB
<< ERDES1_BUFFER2_SIZE_SHIFT)
& ERDES1_BUFFER2_SIZE_MASK);
diff --git a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
index 77914c8..5ef91a7 100644
--- a/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
+++ b/drivers/net/ethernet/stmicro/stmmac/enh_desc.c
@@ -262,7 +262,7 @@ static void enh_desc_init_rx_desc(struct dma_desc *p, int disable_rx_ic,
int mode, int end)
{
p->des0 |= cpu_to_le32(RDES0_OWN);
- p->des1 |= cpu_to_le32((BUF_SIZE_8KiB - 1) & ERDES1_BUFFER1_SIZE_MASK);
+ p->des1 |= cpu_to_le32(BUF_SIZE_8KiB & ERDES1_BUFFER1_SIZE_MASK);
if (mode == STMMAC_CHAIN_MODE)
ehn_desc_rx_set_on_chain(p);
diff --git a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
index abc3f85..d8c5bc4 100644
--- a/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
+++ b/drivers/net/ethernet/stmicro/stmmac/ring_mode.c
@@ -140,7 +140,7 @@ static void clean_desc3(void *priv_ptr, struct dma_desc *p)
static int set_16kib_bfsize(int mtu)
{
int ret = 0;
- if (unlikely(mtu >= BUF_SIZE_8KiB))
+ if (unlikely(mtu > BUF_SIZE_8KiB))
ret = BUF_SIZE_16KiB;
return ret;
}
diff --git a/drivers/net/fddi/defza.c b/drivers/net/fddi/defza.c
index 3b7f10a..c5cae8e 100644
--- a/drivers/net/fddi/defza.c
+++ b/drivers/net/fddi/defza.c
@@ -1,4 +1,4 @@
-// SPDX-License-Identifier: GPL-2.0
+// SPDX-License-Identifier: GPL-2.0+
/* FDDI network adapter driver for DEC FDDIcontroller 700/700-C devices.
*
* Copyright (c) 2018 Maciej W. Rozycki
@@ -56,7 +56,7 @@
#define DRV_VERSION "v.1.1.4"
#define DRV_RELDATE "Oct 6 2018"
-static char version[] =
+static const char version[] =
DRV_NAME ": " DRV_VERSION " " DRV_RELDATE " Maciej W. Rozycki\n";
MODULE_AUTHOR("Maciej W. Rozycki <macro@linux-mips.org>");
@@ -784,7 +784,7 @@ static void fza_rx(struct net_device *dev)
static void fza_tx_smt(struct net_device *dev)
{
struct fza_private *fp = netdev_priv(dev);
- struct fza_buffer_tx __iomem *smt_tx_ptr, *skb_data_ptr;
+ struct fza_buffer_tx __iomem *smt_tx_ptr;
int i, len;
u32 own;
@@ -799,6 +799,7 @@ static void fza_tx_smt(struct net_device *dev)
if (!netif_queue_stopped(dev)) {
if (dev_nit_active(dev)) {
+ struct fza_buffer_tx *skb_data_ptr;
struct sk_buff *skb;
/* Length must be a multiple of 4 as only word
diff --git a/drivers/net/fddi/defza.h b/drivers/net/fddi/defza.h
index b06acf3..93bda61 100644
--- a/drivers/net/fddi/defza.h
+++ b/drivers/net/fddi/defza.h
@@ -1,4 +1,4 @@
-/* SPDX-License-Identifier: GPL-2.0 */
+/* SPDX-License-Identifier: GPL-2.0+ */
/* FDDI network adapter driver for DEC FDDIcontroller 700/700-C devices.
*
* Copyright (c) 2018 Maciej W. Rozycki
@@ -235,6 +235,7 @@ struct fza_ring_cmd {
#define FZA_RING_CMD 0x200400 /* command ring address */
#define FZA_RING_CMD_SIZE 0x40 /* command descriptor ring
* size
+ */
/* Command constants. */
#define FZA_RING_CMD_MASK 0x7fffffff
#define FZA_RING_CMD_NOP 0x00000000 /* nop */
diff --git a/drivers/net/phy/broadcom.c b/drivers/net/phy/broadcom.c
index e86ea10..7045370 100644
--- a/drivers/net/phy/broadcom.c
+++ b/drivers/net/phy/broadcom.c
@@ -92,7 +92,7 @@ static int bcm54612e_config_init(struct phy_device *phydev)
return 0;
}
-static int bcm5481x_config(struct phy_device *phydev)
+static int bcm54xx_config_clock_delay(struct phy_device *phydev)
{
int rc, val;
@@ -429,7 +429,7 @@ static int bcm5481_config_aneg(struct phy_device *phydev)
ret = genphy_config_aneg(phydev);
/* Then we can set up the delay. */
- bcm5481x_config(phydev);
+ bcm54xx_config_clock_delay(phydev);
if (of_property_read_bool(np, "enet-phy-lane-swap")) {
/* Lane Swap - Undocumented register...magic! */
@@ -442,6 +442,19 @@ static int bcm5481_config_aneg(struct phy_device *phydev)
return ret;
}
+static int bcm54616s_config_aneg(struct phy_device *phydev)
+{
+ int ret;
+
+ /* Aneg firsly. */
+ ret = genphy_config_aneg(phydev);
+
+ /* Then we can set up the delay. */
+ bcm54xx_config_clock_delay(phydev);
+
+ return ret;
+}
+
static int brcm_phy_setbits(struct phy_device *phydev, int reg, int set)
{
int val;
@@ -636,6 +649,7 @@ static struct phy_driver broadcom_drivers[] = {
.features = PHY_GBIT_FEATURES,
.flags = PHY_HAS_INTERRUPT,
.config_init = bcm54xx_config_init,
+ .config_aneg = bcm54616s_config_aneg,
.ack_interrupt = bcm_phy_ack_intr,
.config_intr = bcm_phy_config_intr,
}, {
diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
index 7fc8508..271e8ad 100644
--- a/drivers/net/phy/realtek.c
+++ b/drivers/net/phy/realtek.c
@@ -220,7 +220,7 @@ static struct phy_driver realtek_drvs[] = {
.flags = PHY_HAS_INTERRUPT,
}, {
.phy_id = 0x001cc816,
- .name = "RTL8201F 10/100Mbps Ethernet",
+ .name = "RTL8201F Fast Ethernet",
.phy_id_mask = 0x001fffff,
.features = PHY_BASIC_FEATURES,
.flags = PHY_HAS_INTERRUPT,
diff --git a/drivers/net/usb/smsc95xx.c b/drivers/net/usb/smsc95xx.c
index 262e7a3..f2d01cb 100644
--- a/drivers/net/usb/smsc95xx.c
+++ b/drivers/net/usb/smsc95xx.c
@@ -1321,6 +1321,8 @@ static int smsc95xx_bind(struct usbnet *dev, struct usb_interface *intf)
dev->net->ethtool_ops = &smsc95xx_ethtool_ops;
dev->net->flags |= IFF_MULTICAST;
dev->net->hard_header_len += SMSC95XX_TX_OVERHEAD_CSUM;
+ dev->net->min_mtu = ETH_MIN_MTU;
+ dev->net->max_mtu = ETH_DATA_LEN;
dev->hard_mtu = dev->net->mtu + dev->net->hard_header_len;
pdata->dev = dev;
@@ -1598,6 +1600,8 @@ static int smsc95xx_suspend(struct usb_interface *intf, pm_message_t message)
return ret;
}
+ cancel_delayed_work_sync(&pdata->carrier_check);
+
if (pdata->suspend_flags) {
netdev_warn(dev->net, "error during last resume\n");
pdata->suspend_flags = 0;
@@ -1840,6 +1844,11 @@ static int smsc95xx_suspend(struct usb_interface *intf, pm_message_t message)
*/
if (ret && PMSG_IS_AUTO(message))
usbnet_resume(intf);
+
+ if (ret)
+ schedule_delayed_work(&pdata->carrier_check,
+ CARRIER_CHECK_DELAY);
+
return ret;
}
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index 2e65be8..559d567 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -1519,8 +1519,10 @@ static void __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
if (ns->ndev)
nvme_nvm_update_nvm_info(ns);
#ifdef CONFIG_NVME_MULTIPATH
- if (ns->head->disk)
+ if (ns->head->disk) {
nvme_update_disk_info(ns->head->disk, ns, id);
+ blk_queue_stack_limits(ns->head->disk->queue, ns->queue);
+ }
#endif
}
diff --git a/drivers/nvme/host/multipath.c b/drivers/nvme/host/multipath.c
index 5e3cc8c..9901afd 100644
--- a/drivers/nvme/host/multipath.c
+++ b/drivers/nvme/host/multipath.c
@@ -285,6 +285,7 @@ int nvme_mpath_alloc_disk(struct nvme_ctrl *ctrl, struct nvme_ns_head *head)
blk_queue_flag_set(QUEUE_FLAG_NONROT, q);
/* set to a default value for 512 until disk is validated */
blk_queue_logical_block_size(q, 512);
+ blk_set_stacking_limits(&q->limits);
/* we need to propagate up the VMC settings */
if (ctrl->vwc & NVME_CTRL_VWC_PRESENT)
diff --git a/drivers/nvme/target/core.c b/drivers/nvme/target/core.c
index f4efe28..a5f9bbc 100644
--- a/drivers/nvme/target/core.c
+++ b/drivers/nvme/target/core.c
@@ -420,7 +420,7 @@ static void nvmet_p2pmem_ns_add_p2p(struct nvmet_ctrl *ctrl,
struct pci_dev *p2p_dev;
int ret;
- if (!ctrl->p2p_client)
+ if (!ctrl->p2p_client || !ns->use_p2pmem)
return;
if (ns->p2p_dev) {
diff --git a/drivers/nvme/target/rdma.c b/drivers/nvme/target/rdma.c
index ddce100..3f7971d3 100644
--- a/drivers/nvme/target/rdma.c
+++ b/drivers/nvme/target/rdma.c
@@ -122,7 +122,6 @@ struct nvmet_rdma_device {
int inline_page_count;
};
-static struct workqueue_struct *nvmet_rdma_delete_wq;
static bool nvmet_rdma_use_srq;
module_param_named(use_srq, nvmet_rdma_use_srq, bool, 0444);
MODULE_PARM_DESC(use_srq, "Use shared receive queue.");
@@ -1274,12 +1273,12 @@ static int nvmet_rdma_queue_connect(struct rdma_cm_id *cm_id,
if (queue->host_qid == 0) {
/* Let inflight controller teardown complete */
- flush_workqueue(nvmet_rdma_delete_wq);
+ flush_scheduled_work();
}
ret = nvmet_rdma_cm_accept(cm_id, queue, &event->param.conn);
if (ret) {
- queue_work(nvmet_rdma_delete_wq, &queue->release_work);
+ schedule_work(&queue->release_work);
/* Destroying rdma_cm id is not needed here */
return 0;
}
@@ -1344,7 +1343,7 @@ static void __nvmet_rdma_queue_disconnect(struct nvmet_rdma_queue *queue)
if (disconnect) {
rdma_disconnect(queue->cm_id);
- queue_work(nvmet_rdma_delete_wq, &queue->release_work);
+ schedule_work(&queue->release_work);
}
}
@@ -1374,7 +1373,7 @@ static void nvmet_rdma_queue_connect_fail(struct rdma_cm_id *cm_id,
mutex_unlock(&nvmet_rdma_queue_mutex);
pr_err("failed to connect queue %d\n", queue->idx);
- queue_work(nvmet_rdma_delete_wq, &queue->release_work);
+ schedule_work(&queue->release_work);
}
/**
@@ -1656,17 +1655,8 @@ static int __init nvmet_rdma_init(void)
if (ret)
goto err_ib_client;
- nvmet_rdma_delete_wq = alloc_workqueue("nvmet-rdma-delete-wq",
- WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS, 0);
- if (!nvmet_rdma_delete_wq) {
- ret = -ENOMEM;
- goto err_unreg_transport;
- }
-
return 0;
-err_unreg_transport:
- nvmet_unregister_transport(&nvmet_rdma_ops);
err_ib_client:
ib_unregister_client(&nvmet_rdma_ib_client);
return ret;
@@ -1674,7 +1664,6 @@ static int __init nvmet_rdma_init(void)
static void __exit nvmet_rdma_exit(void)
{
- destroy_workqueue(nvmet_rdma_delete_wq);
nvmet_unregister_transport(&nvmet_rdma_ops);
ib_unregister_client(&nvmet_rdma_ib_client);
WARN_ON_ONCE(!list_empty(&nvmet_rdma_queue_list));
diff --git a/drivers/of/device.c b/drivers/of/device.c
index 0f27fad..5592437 100644
--- a/drivers/of/device.c
+++ b/drivers/of/device.c
@@ -149,9 +149,11 @@ int of_dma_configure(struct device *dev, struct device_node *np, bool force_dma)
* set by the driver.
*/
mask = DMA_BIT_MASK(ilog2(dma_addr + size - 1) + 1);
- dev->bus_dma_mask = mask;
dev->coherent_dma_mask &= mask;
*dev->dma_mask &= mask;
+ /* ...but only set bus mask if we found valid dma-ranges earlier */
+ if (!ret)
+ dev->bus_dma_mask = mask;
coherent = of_dma_is_coherent(np);
dev_dbg(dev, "device is%sdma coherent\n",
diff --git a/drivers/of/of_numa.c b/drivers/of/of_numa.c
index 35c64a4..fe6b136 100644
--- a/drivers/of/of_numa.c
+++ b/drivers/of/of_numa.c
@@ -104,9 +104,14 @@ static int __init of_numa_parse_distance_map_v1(struct device_node *map)
distance = of_read_number(matrix, 1);
matrix++;
+ if ((nodea == nodeb && distance != LOCAL_DISTANCE) ||
+ (nodea != nodeb && distance <= LOCAL_DISTANCE)) {
+ pr_err("Invalid distance[node%d -> node%d] = %d\n",
+ nodea, nodeb, distance);
+ return -EINVAL;
+ }
+
numa_set_distance(nodea, nodeb, distance);
- pr_debug("distance[node%d -> node%d] = %d\n",
- nodea, nodeb, distance);
/* Set default distance of node B->A same as A->B */
if (nodeb > nodea)
diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index 2a4aa64..921db6f 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -793,15 +793,10 @@ static void pci_acpi_setup(struct device *dev)
{
struct pci_dev *pci_dev = to_pci_dev(dev);
struct acpi_device *adev = ACPI_COMPANION(dev);
- int node;
if (!adev)
return;
- node = acpi_get_node(adev->handle);
- if (node != NUMA_NO_NODE)
- set_dev_node(dev, node);
-
pci_acpi_optimize_delay(pci_dev, adev->handle);
pci_acpi_add_pm_notifier(adev, pci_dev);
diff --git a/drivers/pinctrl/meson/pinctrl-meson-gxbb.c b/drivers/pinctrl/meson/pinctrl-meson-gxbb.c
index 4ceb06f..4edeb4c 100644
--- a/drivers/pinctrl/meson/pinctrl-meson-gxbb.c
+++ b/drivers/pinctrl/meson/pinctrl-meson-gxbb.c
@@ -830,7 +830,7 @@ static struct meson_bank meson_gxbb_periphs_banks[] = {
static struct meson_bank meson_gxbb_aobus_banks[] = {
/* name first last irq pullen pull dir out in */
- BANK("AO", GPIOAO_0, GPIOAO_13, 0, 13, 0, 0, 0, 16, 0, 0, 0, 16, 1, 0),
+ BANK("AO", GPIOAO_0, GPIOAO_13, 0, 13, 0, 16, 0, 0, 0, 0, 0, 16, 1, 0),
};
static struct meson_pinctrl_data meson_gxbb_periphs_pinctrl_data = {
diff --git a/drivers/pinctrl/meson/pinctrl-meson-gxl.c b/drivers/pinctrl/meson/pinctrl-meson-gxl.c
index 7dae1d7..158f618 100644
--- a/drivers/pinctrl/meson/pinctrl-meson-gxl.c
+++ b/drivers/pinctrl/meson/pinctrl-meson-gxl.c
@@ -807,7 +807,7 @@ static struct meson_bank meson_gxl_periphs_banks[] = {
static struct meson_bank meson_gxl_aobus_banks[] = {
/* name first last irq pullen pull dir out in */
- BANK("AO", GPIOAO_0, GPIOAO_9, 0, 9, 0, 0, 0, 16, 0, 0, 0, 16, 1, 0),
+ BANK("AO", GPIOAO_0, GPIOAO_9, 0, 9, 0, 16, 0, 0, 0, 0, 0, 16, 1, 0),
};
static struct meson_pinctrl_data meson_gxl_periphs_pinctrl_data = {
diff --git a/drivers/pinctrl/meson/pinctrl-meson.c b/drivers/pinctrl/meson/pinctrl-meson.c
index f8b778a..53d4490 100644
--- a/drivers/pinctrl/meson/pinctrl-meson.c
+++ b/drivers/pinctrl/meson/pinctrl-meson.c
@@ -192,7 +192,7 @@ static int meson_pinconf_set(struct pinctrl_dev *pcdev, unsigned int pin,
dev_dbg(pc->dev, "pin %u: disable bias\n", pin);
meson_calc_reg_and_bit(bank, pin, REG_PULL, ®, &bit);
- ret = regmap_update_bits(pc->reg_pull, reg,
+ ret = regmap_update_bits(pc->reg_pullen, reg,
BIT(bit), 0);
if (ret)
return ret;
diff --git a/drivers/pinctrl/meson/pinctrl-meson8.c b/drivers/pinctrl/meson/pinctrl-meson8.c
index c6d7931..8646617 100644
--- a/drivers/pinctrl/meson/pinctrl-meson8.c
+++ b/drivers/pinctrl/meson/pinctrl-meson8.c
@@ -1053,7 +1053,7 @@ static struct meson_bank meson8_cbus_banks[] = {
static struct meson_bank meson8_aobus_banks[] = {
/* name first last irq pullen pull dir out in */
- BANK("AO", GPIOAO_0, GPIO_TEST_N, 0, 13, 0, 0, 0, 16, 0, 0, 0, 16, 1, 0),
+ BANK("AO", GPIOAO_0, GPIO_TEST_N, 0, 13, 0, 16, 0, 0, 0, 0, 0, 16, 1, 0),
};
static struct meson_pinctrl_data meson8_cbus_pinctrl_data = {
diff --git a/drivers/pinctrl/meson/pinctrl-meson8b.c b/drivers/pinctrl/meson/pinctrl-meson8b.c
index bb2a309..647ad15 100644
--- a/drivers/pinctrl/meson/pinctrl-meson8b.c
+++ b/drivers/pinctrl/meson/pinctrl-meson8b.c
@@ -906,7 +906,7 @@ static struct meson_bank meson8b_cbus_banks[] = {
static struct meson_bank meson8b_aobus_banks[] = {
/* name first lastc irq pullen pull dir out in */
- BANK("AO", GPIOAO_0, GPIO_TEST_N, 0, 13, 0, 0, 0, 16, 0, 0, 0, 16, 1, 0),
+ BANK("AO", GPIOAO_0, GPIO_TEST_N, 0, 13, 0, 16, 0, 0, 0, 0, 0, 16, 1, 0),
};
static struct meson_pinctrl_data meson8b_cbus_pinctrl_data = {
diff --git a/drivers/rtc/hctosys.c b/drivers/rtc/hctosys.c
index e79f2a1..b9ec4a1 100644
--- a/drivers/rtc/hctosys.c
+++ b/drivers/rtc/hctosys.c
@@ -50,8 +50,10 @@ static int __init rtc_hctosys(void)
tv64.tv_sec = rtc_tm_to_time64(&tm);
#if BITS_PER_LONG == 32
- if (tv64.tv_sec > INT_MAX)
+ if (tv64.tv_sec > INT_MAX) {
+ err = -ERANGE;
goto err_read;
+ }
#endif
err = do_settimeofday64(&tv64);
diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c
index df0c577..a5a19ff 100644
--- a/drivers/rtc/rtc-cmos.c
+++ b/drivers/rtc/rtc-cmos.c
@@ -257,6 +257,7 @@ static int cmos_read_alarm(struct device *dev, struct rtc_wkalrm *t)
struct cmos_rtc *cmos = dev_get_drvdata(dev);
unsigned char rtc_control;
+ /* This not only a rtc_op, but also called directly */
if (!is_valid_irq(cmos->irq))
return -EIO;
@@ -452,6 +453,7 @@ static int cmos_set_alarm(struct device *dev, struct rtc_wkalrm *t)
unsigned char mon, mday, hrs, min, sec, rtc_control;
int ret;
+ /* This not only a rtc_op, but also called directly */
if (!is_valid_irq(cmos->irq))
return -EIO;
@@ -516,9 +518,6 @@ static int cmos_alarm_irq_enable(struct device *dev, unsigned int enabled)
struct cmos_rtc *cmos = dev_get_drvdata(dev);
unsigned long flags;
- if (!is_valid_irq(cmos->irq))
- return -EINVAL;
-
spin_lock_irqsave(&rtc_lock, flags);
if (enabled)
@@ -579,6 +578,12 @@ static const struct rtc_class_ops cmos_rtc_ops = {
.alarm_irq_enable = cmos_alarm_irq_enable,
};
+static const struct rtc_class_ops cmos_rtc_ops_no_alarm = {
+ .read_time = cmos_read_time,
+ .set_time = cmos_set_time,
+ .proc = cmos_procfs,
+};
+
/*----------------------------------------------------------------*/
/*
@@ -855,9 +860,12 @@ cmos_do_probe(struct device *dev, struct resource *ports, int rtc_irq)
dev_dbg(dev, "IRQ %d is already in use\n", rtc_irq);
goto cleanup1;
}
+
+ cmos_rtc.rtc->ops = &cmos_rtc_ops;
+ } else {
+ cmos_rtc.rtc->ops = &cmos_rtc_ops_no_alarm;
}
- cmos_rtc.rtc->ops = &cmos_rtc_ops;
cmos_rtc.rtc->nvram_old_abi = true;
retval = rtc_register_device(cmos_rtc.rtc);
if (retval)
diff --git a/drivers/rtc/rtc-pcf2127.c b/drivers/rtc/rtc-pcf2127.c
index 9f99a09..7cb786d 100644
--- a/drivers/rtc/rtc-pcf2127.c
+++ b/drivers/rtc/rtc-pcf2127.c
@@ -303,6 +303,9 @@ static int pcf2127_i2c_gather_write(void *context,
memcpy(buf + 1, val, val_size);
ret = i2c_master_send(client, buf, val_size + 1);
+
+ kfree(buf);
+
if (ret != val_size + 1)
return ret < 0 ? ret : -EIO;
diff --git a/drivers/s390/net/qeth_core.h b/drivers/s390/net/qeth_core.h
index 6843bc7..04e294d 100644
--- a/drivers/s390/net/qeth_core.h
+++ b/drivers/s390/net/qeth_core.h
@@ -87,6 +87,18 @@ struct qeth_dbf_info {
#define SENSE_RESETTING_EVENT_BYTE 1
#define SENSE_RESETTING_EVENT_FLAG 0x80
+static inline u32 qeth_get_device_id(struct ccw_device *cdev)
+{
+ struct ccw_dev_id dev_id;
+ u32 id;
+
+ ccw_device_get_id(cdev, &dev_id);
+ id = dev_id.devno;
+ id |= (u32) (dev_id.ssid << 16);
+
+ return id;
+}
+
/*
* Common IO related definitions
*/
@@ -97,7 +109,8 @@ struct qeth_dbf_info {
#define CARD_RDEV_ID(card) dev_name(&card->read.ccwdev->dev)
#define CARD_WDEV_ID(card) dev_name(&card->write.ccwdev->dev)
#define CARD_DDEV_ID(card) dev_name(&card->data.ccwdev->dev)
-#define CHANNEL_ID(channel) dev_name(&channel->ccwdev->dev)
+#define CCW_DEVID(cdev) (qeth_get_device_id(cdev))
+#define CARD_DEVID(card) (CCW_DEVID(CARD_RDEV(card)))
/**
* card stuff
@@ -830,6 +843,11 @@ struct qeth_trap_id {
/*some helper functions*/
#define QETH_CARD_IFNAME(card) (((card)->dev)? (card)->dev->name : "")
+static inline bool qeth_netdev_is_registered(struct net_device *dev)
+{
+ return dev->netdev_ops != NULL;
+}
+
static inline void qeth_scrub_qdio_buffer(struct qdio_buffer *buf,
unsigned int elements)
{
@@ -973,7 +991,7 @@ int qeth_wait_for_threads(struct qeth_card *, unsigned long);
int qeth_do_run_thread(struct qeth_card *, unsigned long);
void qeth_clear_thread_start_bit(struct qeth_card *, unsigned long);
void qeth_clear_thread_running_bit(struct qeth_card *, unsigned long);
-int qeth_core_hardsetup_card(struct qeth_card *);
+int qeth_core_hardsetup_card(struct qeth_card *card, bool *carrier_ok);
void qeth_print_status_message(struct qeth_card *);
int qeth_init_qdio_queues(struct qeth_card *);
int qeth_send_ipa_cmd(struct qeth_card *, struct qeth_cmd_buffer *,
@@ -1028,11 +1046,6 @@ int qeth_configure_cq(struct qeth_card *, enum qeth_cq);
int qeth_hw_trap(struct qeth_card *, enum qeth_diags_trap_action);
void qeth_trace_features(struct qeth_card *);
void qeth_close_dev(struct qeth_card *);
-int qeth_send_setassparms(struct qeth_card *, struct qeth_cmd_buffer *, __u16,
- long,
- int (*reply_cb)(struct qeth_card *,
- struct qeth_reply *, unsigned long),
- void *);
int qeth_setassparms_cb(struct qeth_card *, struct qeth_reply *, unsigned long);
struct qeth_cmd_buffer *qeth_get_setassparms_cmd(struct qeth_card *,
enum qeth_ipa_funcs,
diff --git a/drivers/s390/net/qeth_core_main.c b/drivers/s390/net/qeth_core_main.c
index 3274f13..4bce5ae 100644
--- a/drivers/s390/net/qeth_core_main.c
+++ b/drivers/s390/net/qeth_core_main.c
@@ -167,6 +167,8 @@ const char *qeth_get_cardname_short(struct qeth_card *card)
return "OSD_1000";
case QETH_LINK_TYPE_10GBIT_ETH:
return "OSD_10GIG";
+ case QETH_LINK_TYPE_25GBIT_ETH:
+ return "OSD_25GIG";
case QETH_LINK_TYPE_LANE_ETH100:
return "OSD_FE_LANE";
case QETH_LINK_TYPE_LANE_TR:
@@ -554,8 +556,8 @@ static int __qeth_issue_next_read(struct qeth_card *card)
if (!iob) {
dev_warn(&card->gdev->dev, "The qeth device driver "
"failed to recover an error on the device\n");
- QETH_DBF_MESSAGE(2, "%s issue_next_read failed: no iob "
- "available\n", dev_name(&card->gdev->dev));
+ QETH_DBF_MESSAGE(2, "issue_next_read on device %x failed: no iob available\n",
+ CARD_DEVID(card));
return -ENOMEM;
}
qeth_setup_ccw(channel->ccw, CCW_CMD_READ, QETH_BUFSIZE, iob->data);
@@ -563,8 +565,8 @@ static int __qeth_issue_next_read(struct qeth_card *card)
rc = ccw_device_start(channel->ccwdev, channel->ccw,
(addr_t) iob, 0, 0);
if (rc) {
- QETH_DBF_MESSAGE(2, "%s error in starting next read ccw! "
- "rc=%i\n", dev_name(&card->gdev->dev), rc);
+ QETH_DBF_MESSAGE(2, "error %i on device %x when starting next read ccw!\n",
+ rc, CARD_DEVID(card));
atomic_set(&channel->irq_pending, 0);
card->read_or_write_problem = 1;
qeth_schedule_recovery(card);
@@ -613,16 +615,14 @@ static void qeth_issue_ipa_msg(struct qeth_ipa_cmd *cmd, int rc,
const char *ipa_name;
int com = cmd->hdr.command;
ipa_name = qeth_get_ipa_cmd_name(com);
+
if (rc)
- QETH_DBF_MESSAGE(2, "IPA: %s(x%X) for %s/%s returned "
- "x%X \"%s\"\n",
- ipa_name, com, dev_name(&card->gdev->dev),
- QETH_CARD_IFNAME(card), rc,
- qeth_get_ipa_msg(rc));
+ QETH_DBF_MESSAGE(2, "IPA: %s(%#x) for device %x returned %#x \"%s\"\n",
+ ipa_name, com, CARD_DEVID(card), rc,
+ qeth_get_ipa_msg(rc));
else
- QETH_DBF_MESSAGE(5, "IPA: %s(x%X) for %s/%s succeeded\n",
- ipa_name, com, dev_name(&card->gdev->dev),
- QETH_CARD_IFNAME(card));
+ QETH_DBF_MESSAGE(5, "IPA: %s(%#x) for device %x succeeded\n",
+ ipa_name, com, CARD_DEVID(card));
}
static struct qeth_ipa_cmd *qeth_check_ipa_data(struct qeth_card *card,
@@ -711,7 +711,7 @@ static int qeth_check_idx_response(struct qeth_card *card,
QETH_DBF_HEX(CTRL, 2, buffer, QETH_DBF_CTRL_LEN);
if ((buffer[2] & 0xc0) == 0xc0) {
- QETH_DBF_MESSAGE(2, "received an IDX TERMINATE with cause code %#02x\n",
+ QETH_DBF_MESSAGE(2, "received an IDX TERMINATE with cause code %#04x\n",
buffer[4]);
QETH_CARD_TEXT(card, 2, "ckidxres");
QETH_CARD_TEXT(card, 2, " idxterm");
@@ -972,8 +972,8 @@ static int qeth_get_problem(struct qeth_card *card, struct ccw_device *cdev,
QETH_CARD_TEXT(card, 2, "CGENCHK");
dev_warn(&cdev->dev, "The qeth device driver "
"failed to recover an error on the device\n");
- QETH_DBF_MESSAGE(2, "%s check on device dstat=x%x, cstat=x%x\n",
- dev_name(&cdev->dev), dstat, cstat);
+ QETH_DBF_MESSAGE(2, "check on channel %x with dstat=%#x, cstat=%#x\n",
+ CCW_DEVID(cdev), dstat, cstat);
print_hex_dump(KERN_WARNING, "qeth: irb ", DUMP_PREFIX_OFFSET,
16, 1, irb, 64, 1);
return 1;
@@ -1013,8 +1013,8 @@ static long qeth_check_irb_error(struct qeth_card *card,
switch (PTR_ERR(irb)) {
case -EIO:
- QETH_DBF_MESSAGE(2, "%s i/o-error on device\n",
- dev_name(&cdev->dev));
+ QETH_DBF_MESSAGE(2, "i/o-error on channel %x\n",
+ CCW_DEVID(cdev));
QETH_CARD_TEXT(card, 2, "ckirberr");
QETH_CARD_TEXT_(card, 2, " rc%d", -EIO);
break;
@@ -1031,8 +1031,8 @@ static long qeth_check_irb_error(struct qeth_card *card,
}
break;
default:
- QETH_DBF_MESSAGE(2, "%s unknown error %ld on device\n",
- dev_name(&cdev->dev), PTR_ERR(irb));
+ QETH_DBF_MESSAGE(2, "unknown error %ld on channel %x\n",
+ PTR_ERR(irb), CCW_DEVID(cdev));
QETH_CARD_TEXT(card, 2, "ckirberr");
QETH_CARD_TEXT(card, 2, " rc???");
}
@@ -1114,9 +1114,9 @@ static void qeth_irq(struct ccw_device *cdev, unsigned long intparm,
dev_warn(&channel->ccwdev->dev,
"The qeth device driver failed to recover "
"an error on the device\n");
- QETH_DBF_MESSAGE(2, "%s sense data available. cstat "
- "0x%X dstat 0x%X\n",
- dev_name(&channel->ccwdev->dev), cstat, dstat);
+ QETH_DBF_MESSAGE(2, "sense data available on channel %x: cstat %#X dstat %#X\n",
+ CCW_DEVID(channel->ccwdev), cstat,
+ dstat);
print_hex_dump(KERN_WARNING, "qeth: irb ",
DUMP_PREFIX_OFFSET, 16, 1, irb, 32, 1);
print_hex_dump(KERN_WARNING, "qeth: sense data ",
@@ -1890,8 +1890,8 @@ static int qeth_idx_activate_channel(struct qeth_card *card,
if (channel->state != CH_STATE_ACTIVATING) {
dev_warn(&channel->ccwdev->dev, "The qeth device driver"
" failed to recover an error on the device\n");
- QETH_DBF_MESSAGE(2, "%s IDX activate timed out\n",
- dev_name(&channel->ccwdev->dev));
+ QETH_DBF_MESSAGE(2, "IDX activate timed out on channel %x\n",
+ CCW_DEVID(channel->ccwdev));
QETH_DBF_TEXT_(SETUP, 2, "2err%d", -ETIME);
return -ETIME;
}
@@ -1926,17 +1926,15 @@ static void qeth_idx_write_cb(struct qeth_card *card,
"The adapter is used exclusively by another "
"host\n");
else
- QETH_DBF_MESSAGE(2, "%s IDX_ACTIVATE on write channel:"
- " negative reply\n",
- dev_name(&channel->ccwdev->dev));
+ QETH_DBF_MESSAGE(2, "IDX_ACTIVATE on channel %x: negative reply\n",
+ CCW_DEVID(channel->ccwdev));
goto out;
}
memcpy(&temp, QETH_IDX_ACT_FUNC_LEVEL(iob->data), 2);
if ((temp & ~0x0100) != qeth_peer_func_level(card->info.func_level)) {
- QETH_DBF_MESSAGE(2, "%s IDX_ACTIVATE on write channel: "
- "function level mismatch (sent: 0x%x, received: "
- "0x%x)\n", dev_name(&channel->ccwdev->dev),
- card->info.func_level, temp);
+ QETH_DBF_MESSAGE(2, "IDX_ACTIVATE on channel %x: function level mismatch (sent: %#x, received: %#x)\n",
+ CCW_DEVID(channel->ccwdev),
+ card->info.func_level, temp);
goto out;
}
channel->state = CH_STATE_UP;
@@ -1973,9 +1971,8 @@ static void qeth_idx_read_cb(struct qeth_card *card,
"insufficient authorization\n");
break;
default:
- QETH_DBF_MESSAGE(2, "%s IDX_ACTIVATE on read channel:"
- " negative reply\n",
- dev_name(&channel->ccwdev->dev));
+ QETH_DBF_MESSAGE(2, "IDX_ACTIVATE on channel %x: negative reply\n",
+ CCW_DEVID(channel->ccwdev));
}
QETH_CARD_TEXT_(card, 2, "idxread%c",
QETH_IDX_ACT_CAUSE_CODE(iob->data));
@@ -1984,10 +1981,9 @@ static void qeth_idx_read_cb(struct qeth_card *card,
memcpy(&temp, QETH_IDX_ACT_FUNC_LEVEL(iob->data), 2);
if (temp != qeth_peer_func_level(card->info.func_level)) {
- QETH_DBF_MESSAGE(2, "%s IDX_ACTIVATE on read channel: function "
- "level mismatch (sent: 0x%x, received: 0x%x)\n",
- dev_name(&channel->ccwdev->dev),
- card->info.func_level, temp);
+ QETH_DBF_MESSAGE(2, "IDX_ACTIVATE on channel %x: function level mismatch (sent: %#x, received: %#x)\n",
+ CCW_DEVID(channel->ccwdev),
+ card->info.func_level, temp);
goto out;
}
memcpy(&card->token.issuer_rm_r,
@@ -2096,9 +2092,8 @@ int qeth_send_control_data(struct qeth_card *card, int len,
(addr_t) iob, 0, 0, event_timeout);
spin_unlock_irq(get_ccwdev_lock(channel->ccwdev));
if (rc) {
- QETH_DBF_MESSAGE(2, "%s qeth_send_control_data: "
- "ccw_device_start rc = %i\n",
- dev_name(&channel->ccwdev->dev), rc);
+ QETH_DBF_MESSAGE(2, "qeth_send_control_data on device %x: ccw_device_start rc = %i\n",
+ CARD_DEVID(card), rc);
QETH_CARD_TEXT_(card, 2, " err%d", rc);
spin_lock_irq(&card->lock);
list_del_init(&reply->list);
@@ -2853,8 +2848,8 @@ struct qeth_cmd_buffer *qeth_get_ipacmd_buffer(struct qeth_card *card,
} else {
dev_warn(&card->gdev->dev,
"The qeth driver ran out of channel command buffers\n");
- QETH_DBF_MESSAGE(1, "%s The qeth driver ran out of channel command buffers",
- dev_name(&card->gdev->dev));
+ QETH_DBF_MESSAGE(1, "device %x ran out of channel command buffers",
+ CARD_DEVID(card));
}
return iob;
@@ -2989,10 +2984,9 @@ static int qeth_query_ipassists_cb(struct qeth_card *card,
return 0;
default:
if (cmd->hdr.return_code) {
- QETH_DBF_MESSAGE(1, "%s IPA_CMD_QIPASSIST: Unhandled "
- "rc=%d\n",
- dev_name(&card->gdev->dev),
- cmd->hdr.return_code);
+ QETH_DBF_MESSAGE(1, "IPA_CMD_QIPASSIST on device %x: Unhandled rc=%#x\n",
+ CARD_DEVID(card),
+ cmd->hdr.return_code);
return 0;
}
}
@@ -3004,8 +2998,8 @@ static int qeth_query_ipassists_cb(struct qeth_card *card,
card->options.ipa6.supported_funcs = cmd->hdr.ipa_supported;
card->options.ipa6.enabled_funcs = cmd->hdr.ipa_enabled;
} else
- QETH_DBF_MESSAGE(1, "%s IPA_CMD_QIPASSIST: Flawed LIC detected"
- "\n", dev_name(&card->gdev->dev));
+ QETH_DBF_MESSAGE(1, "IPA_CMD_QIPASSIST on device %x: Flawed LIC detected\n",
+ CARD_DEVID(card));
return 0;
}
@@ -4297,10 +4291,9 @@ static int qeth_setadpparms_set_access_ctrl_cb(struct qeth_card *card,
cmd->data.setadapterparms.hdr.return_code);
if (cmd->data.setadapterparms.hdr.return_code !=
SET_ACCESS_CTRL_RC_SUCCESS)
- QETH_DBF_MESSAGE(3, "ERR:SET_ACCESS_CTRL(%s,%d)==%d\n",
- card->gdev->dev.kobj.name,
- access_ctrl_req->subcmd_code,
- cmd->data.setadapterparms.hdr.return_code);
+ QETH_DBF_MESSAGE(3, "ERR:SET_ACCESS_CTRL(%#x) on device %x: %#x\n",
+ access_ctrl_req->subcmd_code, CARD_DEVID(card),
+ cmd->data.setadapterparms.hdr.return_code);
switch (cmd->data.setadapterparms.hdr.return_code) {
case SET_ACCESS_CTRL_RC_SUCCESS:
if (card->options.isolation == ISOLATION_MODE_NONE) {
@@ -4312,14 +4305,14 @@ static int qeth_setadpparms_set_access_ctrl_cb(struct qeth_card *card,
}
break;
case SET_ACCESS_CTRL_RC_ALREADY_NOT_ISOLATED:
- QETH_DBF_MESSAGE(2, "%s QDIO data connection isolation already "
- "deactivated\n", dev_name(&card->gdev->dev));
+ QETH_DBF_MESSAGE(2, "QDIO data connection isolation on device %x already deactivated\n",
+ CARD_DEVID(card));
if (fallback)
card->options.isolation = card->options.prev_isolation;
break;
case SET_ACCESS_CTRL_RC_ALREADY_ISOLATED:
- QETH_DBF_MESSAGE(2, "%s QDIO data connection isolation already"
- " activated\n", dev_name(&card->gdev->dev));
+ QETH_DBF_MESSAGE(2, "QDIO data connection isolation on device %x already activated\n",
+ CARD_DEVID(card));
if (fallback)
card->options.isolation = card->options.prev_isolation;
break;
@@ -4405,10 +4398,8 @@ int qeth_set_access_ctrl_online(struct qeth_card *card, int fallback)
rc = qeth_setadpparms_set_access_ctrl(card,
card->options.isolation, fallback);
if (rc) {
- QETH_DBF_MESSAGE(3,
- "IPA(SET_ACCESS_CTRL,%s,%d) sent failed\n",
- card->gdev->dev.kobj.name,
- rc);
+ QETH_DBF_MESSAGE(3, "IPA(SET_ACCESS_CTRL(%d) on device %x: sent failed\n",
+ rc, CARD_DEVID(card));
rc = -EOPNOTSUPP;
}
} else if (card->options.isolation != ISOLATION_MODE_NONE) {
@@ -4443,7 +4434,8 @@ static int qeth_mdio_read(struct net_device *dev, int phy_id, int regnum)
rc = BMCR_FULLDPLX;
if ((card->info.link_type != QETH_LINK_TYPE_GBIT_ETH) &&
(card->info.link_type != QETH_LINK_TYPE_OSN) &&
- (card->info.link_type != QETH_LINK_TYPE_10GBIT_ETH))
+ (card->info.link_type != QETH_LINK_TYPE_10GBIT_ETH) &&
+ (card->info.link_type != QETH_LINK_TYPE_25GBIT_ETH))
rc |= BMCR_SPEED100;
break;
case MII_BMSR: /* Basic mode status register */
@@ -4634,8 +4626,8 @@ static int qeth_snmp_command(struct qeth_card *card, char __user *udata)
rc = qeth_send_ipa_snmp_cmd(card, iob, QETH_SETADP_BASE_LEN + req_len,
qeth_snmp_command_cb, (void *)&qinfo);
if (rc)
- QETH_DBF_MESSAGE(2, "SNMP command failed on %s: (0x%x)\n",
- QETH_CARD_IFNAME(card), rc);
+ QETH_DBF_MESSAGE(2, "SNMP command failed on device %x: (%#x)\n",
+ CARD_DEVID(card), rc);
else {
if (copy_to_user(udata, qinfo.udata, qinfo.udata_len))
rc = -EFAULT;
@@ -4869,8 +4861,8 @@ static void qeth_determine_capabilities(struct qeth_card *card)
rc = qeth_read_conf_data(card, (void **) &prcd, &length);
if (rc) {
- QETH_DBF_MESSAGE(2, "%s qeth_read_conf_data returned %i\n",
- dev_name(&card->gdev->dev), rc);
+ QETH_DBF_MESSAGE(2, "qeth_read_conf_data on device %x returned %i\n",
+ CARD_DEVID(card), rc);
QETH_DBF_TEXT_(SETUP, 2, "5err%d", rc);
goto out_offline;
}
@@ -5086,7 +5078,7 @@ static struct ccw_driver qeth_ccw_driver = {
.remove = ccwgroup_remove_ccwdev,
};
-int qeth_core_hardsetup_card(struct qeth_card *card)
+int qeth_core_hardsetup_card(struct qeth_card *card, bool *carrier_ok)
{
int retries = 3;
int rc;
@@ -5096,8 +5088,8 @@ int qeth_core_hardsetup_card(struct qeth_card *card)
qeth_update_from_chp_desc(card);
retry:
if (retries < 3)
- QETH_DBF_MESSAGE(2, "%s Retrying to do IDX activates.\n",
- dev_name(&card->gdev->dev));
+ QETH_DBF_MESSAGE(2, "Retrying to do IDX activates on device %x.\n",
+ CARD_DEVID(card));
rc = qeth_qdio_clear_card(card, card->info.type != QETH_CARD_TYPE_IQD);
ccw_device_set_offline(CARD_DDEV(card));
ccw_device_set_offline(CARD_WDEV(card));
@@ -5161,13 +5153,20 @@ int qeth_core_hardsetup_card(struct qeth_card *card)
if (rc == IPA_RC_LAN_OFFLINE) {
dev_warn(&card->gdev->dev,
"The LAN is offline\n");
- netif_carrier_off(card->dev);
+ *carrier_ok = false;
} else {
rc = -ENODEV;
goto out;
}
} else {
- netif_carrier_on(card->dev);
+ *carrier_ok = true;
+ }
+
+ if (qeth_netdev_is_registered(card->dev)) {
+ if (*carrier_ok)
+ netif_carrier_on(card->dev);
+ else
+ netif_carrier_off(card->dev);
}
card->options.ipa4.supported_funcs = 0;
@@ -5201,8 +5200,8 @@ int qeth_core_hardsetup_card(struct qeth_card *card)
out:
dev_warn(&card->gdev->dev, "The qeth device driver failed to recover "
"an error on the device\n");
- QETH_DBF_MESSAGE(2, "%s Initialization in hardsetup failed! rc=%d\n",
- dev_name(&card->gdev->dev), rc);
+ QETH_DBF_MESSAGE(2, "Initialization for device %x failed in hardsetup! rc=%d\n",
+ CARD_DEVID(card), rc);
return rc;
}
EXPORT_SYMBOL_GPL(qeth_core_hardsetup_card);
@@ -5481,11 +5480,12 @@ struct qeth_cmd_buffer *qeth_get_setassparms_cmd(struct qeth_card *card,
}
EXPORT_SYMBOL_GPL(qeth_get_setassparms_cmd);
-int qeth_send_setassparms(struct qeth_card *card,
- struct qeth_cmd_buffer *iob, __u16 len, long data,
- int (*reply_cb)(struct qeth_card *,
- struct qeth_reply *, unsigned long),
- void *reply_param)
+static int qeth_send_setassparms(struct qeth_card *card,
+ struct qeth_cmd_buffer *iob, u16 len,
+ long data, int (*reply_cb)(struct qeth_card *,
+ struct qeth_reply *,
+ unsigned long),
+ void *reply_param)
{
int rc;
struct qeth_ipa_cmd *cmd;
@@ -5501,7 +5501,6 @@ int qeth_send_setassparms(struct qeth_card *card,
rc = qeth_send_ipa_cmd(card, iob, reply_cb, reply_param);
return rc;
}
-EXPORT_SYMBOL_GPL(qeth_send_setassparms);
int qeth_send_simple_setassparms_prot(struct qeth_card *card,
enum qeth_ipa_funcs ipa_func,
@@ -6170,8 +6169,14 @@ static void qeth_set_cmd_adv_sup(struct ethtool_link_ksettings *cmd,
WARN_ON_ONCE(1);
}
- /* fallthrough from high to low, to select all legal speeds: */
+ /* partially does fall through, to also select lower speeds */
switch (maxspeed) {
+ case SPEED_25000:
+ ethtool_link_ksettings_add_link_mode(cmd, supported,
+ 25000baseSR_Full);
+ ethtool_link_ksettings_add_link_mode(cmd, advertising,
+ 25000baseSR_Full);
+ break;
case SPEED_10000:
ethtool_link_ksettings_add_link_mode(cmd, supported,
10000baseT_Full);
@@ -6254,6 +6259,10 @@ int qeth_core_ethtool_get_link_ksettings(struct net_device *netdev,
cmd->base.speed = SPEED_10000;
cmd->base.port = PORT_FIBRE;
break;
+ case QETH_LINK_TYPE_25GBIT_ETH:
+ cmd->base.speed = SPEED_25000;
+ cmd->base.port = PORT_FIBRE;
+ break;
default:
cmd->base.speed = SPEED_10;
cmd->base.port = PORT_TP;
@@ -6320,6 +6329,9 @@ int qeth_core_ethtool_get_link_ksettings(struct net_device *netdev,
case CARD_INFO_PORTS_10G:
cmd->base.speed = SPEED_10000;
break;
+ case CARD_INFO_PORTS_25G:
+ cmd->base.speed = SPEED_25000;
+ break;
}
return 0;
diff --git a/drivers/s390/net/qeth_core_mpc.h b/drivers/s390/net/qeth_core_mpc.h
index e850904..3e54be2 100644
--- a/drivers/s390/net/qeth_core_mpc.h
+++ b/drivers/s390/net/qeth_core_mpc.h
@@ -90,6 +90,7 @@ enum qeth_link_types {
QETH_LINK_TYPE_GBIT_ETH = 0x03,
QETH_LINK_TYPE_OSN = 0x04,
QETH_LINK_TYPE_10GBIT_ETH = 0x10,
+ QETH_LINK_TYPE_25GBIT_ETH = 0x12,
QETH_LINK_TYPE_LANE_ETH100 = 0x81,
QETH_LINK_TYPE_LANE_TR = 0x82,
QETH_LINK_TYPE_LANE_ETH1000 = 0x83,
@@ -347,6 +348,7 @@ enum qeth_card_info_port_speed {
CARD_INFO_PORTS_100M = 0x00000006,
CARD_INFO_PORTS_1G = 0x00000007,
CARD_INFO_PORTS_10G = 0x00000008,
+ CARD_INFO_PORTS_25G = 0x0000000A,
};
/* (SET)DELIP(M) IPA stuff ***************************************************/
@@ -436,7 +438,7 @@ struct qeth_ipacmd_setassparms {
__u32 flags_32bit;
struct qeth_ipa_caps caps;
struct qeth_checksum_cmd chksum;
- struct qeth_arp_cache_entry add_arp_entry;
+ struct qeth_arp_cache_entry arp_entry;
struct qeth_arp_query_data query_arp;
struct qeth_tso_start_data tso;
__u8 ip[16];
diff --git a/drivers/s390/net/qeth_l2_main.c b/drivers/s390/net/qeth_l2_main.c
index 23aaf37..2914a1a 100644
--- a/drivers/s390/net/qeth_l2_main.c
+++ b/drivers/s390/net/qeth_l2_main.c
@@ -146,11 +146,11 @@ static int qeth_l2_write_mac(struct qeth_card *card, u8 *mac)
QETH_CARD_TEXT(card, 2, "L2Wmac");
rc = qeth_l2_send_setdelmac(card, mac, cmd);
if (rc == -EEXIST)
- QETH_DBF_MESSAGE(2, "MAC %pM already registered on %s\n",
- mac, QETH_CARD_IFNAME(card));
+ QETH_DBF_MESSAGE(2, "MAC already registered on device %x\n",
+ CARD_DEVID(card));
else if (rc)
- QETH_DBF_MESSAGE(2, "Failed to register MAC %pM on %s: %d\n",
- mac, QETH_CARD_IFNAME(card), rc);
+ QETH_DBF_MESSAGE(2, "Failed to register MAC on device %x: %d\n",
+ CARD_DEVID(card), rc);
return rc;
}
@@ -163,8 +163,8 @@ static int qeth_l2_remove_mac(struct qeth_card *card, u8 *mac)
QETH_CARD_TEXT(card, 2, "L2Rmac");
rc = qeth_l2_send_setdelmac(card, mac, cmd);
if (rc)
- QETH_DBF_MESSAGE(2, "Failed to delete MAC %pM on %s: %d\n",
- mac, QETH_CARD_IFNAME(card), rc);
+ QETH_DBF_MESSAGE(2, "Failed to delete MAC on device %u: %d\n",
+ CARD_DEVID(card), rc);
return rc;
}
@@ -260,9 +260,9 @@ static int qeth_l2_send_setdelvlan_cb(struct qeth_card *card,
QETH_CARD_TEXT(card, 2, "L2sdvcb");
if (cmd->hdr.return_code) {
- QETH_DBF_MESSAGE(2, "Error in processing VLAN %i on %s: 0x%x.\n",
+ QETH_DBF_MESSAGE(2, "Error in processing VLAN %u on device %x: %#x.\n",
cmd->data.setdelvlan.vlan_id,
- QETH_CARD_IFNAME(card), cmd->hdr.return_code);
+ CARD_DEVID(card), cmd->hdr.return_code);
QETH_CARD_TEXT_(card, 2, "L2VL%4x", cmd->hdr.command);
QETH_CARD_TEXT_(card, 2, "err%d", cmd->hdr.return_code);
}
@@ -455,8 +455,8 @@ static int qeth_l2_request_initial_mac(struct qeth_card *card)
rc = qeth_vm_request_mac(card);
if (!rc)
goto out;
- QETH_DBF_MESSAGE(2, "z/VM MAC Service failed on device %s: x%x\n",
- CARD_BUS_ID(card), rc);
+ QETH_DBF_MESSAGE(2, "z/VM MAC Service failed on device %x: %#x\n",
+ CARD_DEVID(card), rc);
QETH_DBF_TEXT_(SETUP, 2, "err%04x", rc);
/* fall back to alternative mechanism: */
}
@@ -468,8 +468,8 @@ static int qeth_l2_request_initial_mac(struct qeth_card *card)
rc = qeth_setadpparms_change_macaddr(card);
if (!rc)
goto out;
- QETH_DBF_MESSAGE(2, "READ_MAC Assist failed on device %s: x%x\n",
- CARD_BUS_ID(card), rc);
+ QETH_DBF_MESSAGE(2, "READ_MAC Assist failed on device %x: %#x\n",
+ CARD_DEVID(card), rc);
QETH_DBF_TEXT_(SETUP, 2, "1err%04x", rc);
/* fall back once more: */
}
@@ -826,7 +826,8 @@ static void qeth_l2_remove_device(struct ccwgroup_device *cgdev)
if (cgdev->state == CCWGROUP_ONLINE)
qeth_l2_set_offline(cgdev);
- unregister_netdev(card->dev);
+ if (qeth_netdev_is_registered(card->dev))
+ unregister_netdev(card->dev);
}
static const struct ethtool_ops qeth_l2_ethtool_ops = {
@@ -862,11 +863,11 @@ static const struct net_device_ops qeth_l2_netdev_ops = {
.ndo_set_features = qeth_set_features
};
-static int qeth_l2_setup_netdev(struct qeth_card *card)
+static int qeth_l2_setup_netdev(struct qeth_card *card, bool carrier_ok)
{
int rc;
- if (card->dev->netdev_ops)
+ if (qeth_netdev_is_registered(card->dev))
return 0;
card->dev->priv_flags |= IFF_UNICAST_FLT;
@@ -919,6 +920,9 @@ static int qeth_l2_setup_netdev(struct qeth_card *card)
qeth_l2_request_initial_mac(card);
netif_napi_add(card->dev, &card->napi, qeth_poll, QETH_NAPI_WEIGHT);
rc = register_netdev(card->dev);
+ if (!rc && carrier_ok)
+ netif_carrier_on(card->dev);
+
if (rc)
card->dev->netdev_ops = NULL;
return rc;
@@ -949,6 +953,7 @@ static int __qeth_l2_set_online(struct ccwgroup_device *gdev, int recovery_mode)
struct qeth_card *card = dev_get_drvdata(&gdev->dev);
int rc = 0;
enum qeth_card_states recover_flag;
+ bool carrier_ok;
mutex_lock(&card->discipline_mutex);
mutex_lock(&card->conf_mutex);
@@ -956,7 +961,7 @@ static int __qeth_l2_set_online(struct ccwgroup_device *gdev, int recovery_mode)
QETH_DBF_HEX(SETUP, 2, &card, sizeof(void *));
recover_flag = card->state;
- rc = qeth_core_hardsetup_card(card);
+ rc = qeth_core_hardsetup_card(card, &carrier_ok);
if (rc) {
QETH_DBF_TEXT_(SETUP, 2, "2err%04x", rc);
rc = -ENODEV;
@@ -967,7 +972,7 @@ static int __qeth_l2_set_online(struct ccwgroup_device *gdev, int recovery_mode)
dev_info(&card->gdev->dev,
"The device represents a Bridge Capable Port\n");
- rc = qeth_l2_setup_netdev(card);
+ rc = qeth_l2_setup_netdev(card, carrier_ok);
if (rc)
goto out_remove;
diff --git a/drivers/s390/net/qeth_l3_main.c b/drivers/s390/net/qeth_l3_main.c
index 0b161cc..f08b745 100644
--- a/drivers/s390/net/qeth_l3_main.c
+++ b/drivers/s390/net/qeth_l3_main.c
@@ -278,9 +278,6 @@ static void qeth_l3_clear_ip_htable(struct qeth_card *card, int recover)
QETH_CARD_TEXT(card, 4, "clearip");
- if (recover && card->options.sniffer)
- return;
-
spin_lock_bh(&card->ip_lock);
hash_for_each_safe(card->ip_htable, i, tmp, addr, hnode) {
@@ -494,9 +491,8 @@ int qeth_l3_setrouting_v4(struct qeth_card *card)
QETH_PROT_IPV4);
if (rc) {
card->options.route4.type = NO_ROUTER;
- QETH_DBF_MESSAGE(2, "Error (0x%04x) while setting routing type"
- " on %s. Type set to 'no router'.\n", rc,
- QETH_CARD_IFNAME(card));
+ QETH_DBF_MESSAGE(2, "Error (%#06x) while setting routing type on device %x. Type set to 'no router'.\n",
+ rc, CARD_DEVID(card));
}
return rc;
}
@@ -518,9 +514,8 @@ int qeth_l3_setrouting_v6(struct qeth_card *card)
QETH_PROT_IPV6);
if (rc) {
card->options.route6.type = NO_ROUTER;
- QETH_DBF_MESSAGE(2, "Error (0x%04x) while setting routing type"
- " on %s. Type set to 'no router'.\n", rc,
- QETH_CARD_IFNAME(card));
+ QETH_DBF_MESSAGE(2, "Error (%#06x) while setting routing type on device %x. Type set to 'no router'.\n",
+ rc, CARD_DEVID(card));
}
return rc;
}
@@ -663,6 +658,8 @@ static int qeth_l3_register_addr_entry(struct qeth_card *card,
int rc = 0;
int cnt = 3;
+ if (card->options.sniffer)
+ return 0;
if (addr->proto == QETH_PROT_IPV4) {
QETH_CARD_TEXT(card, 2, "setaddr4");
@@ -697,6 +694,9 @@ static int qeth_l3_deregister_addr_entry(struct qeth_card *card,
{
int rc = 0;
+ if (card->options.sniffer)
+ return 0;
+
if (addr->proto == QETH_PROT_IPV4) {
QETH_CARD_TEXT(card, 2, "deladdr4");
QETH_CARD_HEX(card, 3, &addr->u.a4.addr, sizeof(int));
@@ -1070,8 +1070,8 @@ qeth_diags_trace_cb(struct qeth_card *card, struct qeth_reply *reply,
}
break;
default:
- QETH_DBF_MESSAGE(2, "Unknown sniffer action (0x%04x) on %s\n",
- cmd->data.diagass.action, QETH_CARD_IFNAME(card));
+ QETH_DBF_MESSAGE(2, "Unknown sniffer action (%#06x) on device %x\n",
+ cmd->data.diagass.action, CARD_DEVID(card));
}
return 0;
@@ -1517,32 +1517,25 @@ static void qeth_l3_set_rx_mode(struct net_device *dev)
qeth_l3_handle_promisc_mode(card);
}
-static const char *qeth_l3_arp_get_error_cause(int *rc)
+static int qeth_l3_arp_makerc(int rc)
{
- switch (*rc) {
- case QETH_IPA_ARP_RC_FAILED:
- *rc = -EIO;
- return "operation failed";
+ switch (rc) {
+ case IPA_RC_SUCCESS:
+ return 0;
case QETH_IPA_ARP_RC_NOTSUPP:
- *rc = -EOPNOTSUPP;
- return "operation not supported";
- case QETH_IPA_ARP_RC_OUT_OF_RANGE:
- *rc = -EINVAL;
- return "argument out of range";
case QETH_IPA_ARP_RC_Q_NOTSUPP:
- *rc = -EOPNOTSUPP;
- return "query operation not supported";
+ return -EOPNOTSUPP;
+ case QETH_IPA_ARP_RC_OUT_OF_RANGE:
+ return -EINVAL;
case QETH_IPA_ARP_RC_Q_NO_DATA:
- *rc = -ENOENT;
- return "no query data available";
+ return -ENOENT;
default:
- return "unknown error";
+ return -EIO;
}
}
static int qeth_l3_arp_set_no_entries(struct qeth_card *card, int no_entries)
{
- int tmp;
int rc;
QETH_CARD_TEXT(card, 3, "arpstnoe");
@@ -1560,13 +1553,10 @@ static int qeth_l3_arp_set_no_entries(struct qeth_card *card, int no_entries)
rc = qeth_send_simple_setassparms(card, IPA_ARP_PROCESSING,
IPA_CMD_ASS_ARP_SET_NO_ENTRIES,
no_entries);
- if (rc) {
- tmp = rc;
- QETH_DBF_MESSAGE(2, "Could not set number of ARP entries on "
- "%s: %s (0x%x/%d)\n", QETH_CARD_IFNAME(card),
- qeth_l3_arp_get_error_cause(&rc), tmp, tmp);
- }
- return rc;
+ if (rc)
+ QETH_DBF_MESSAGE(2, "Could not set number of ARP entries on device %x: %#x\n",
+ CARD_DEVID(card), rc);
+ return qeth_l3_arp_makerc(rc);
}
static __u32 get_arp_entry_size(struct qeth_card *card,
@@ -1716,7 +1706,6 @@ static int qeth_l3_query_arp_cache_info(struct qeth_card *card,
{
struct qeth_cmd_buffer *iob;
struct qeth_ipa_cmd *cmd;
- int tmp;
int rc;
QETH_CARD_TEXT_(card, 3, "qarpipv%i", prot);
@@ -1735,15 +1724,10 @@ static int qeth_l3_query_arp_cache_info(struct qeth_card *card,
rc = qeth_l3_send_ipa_arp_cmd(card, iob,
QETH_SETASS_BASE_LEN+QETH_ARP_CMD_LEN,
qeth_l3_arp_query_cb, (void *)qinfo);
- if (rc) {
- tmp = rc;
- QETH_DBF_MESSAGE(2,
- "Error while querying ARP cache on %s: %s "
- "(0x%x/%d)\n", QETH_CARD_IFNAME(card),
- qeth_l3_arp_get_error_cause(&rc), tmp, tmp);
- }
-
- return rc;
+ if (rc)
+ QETH_DBF_MESSAGE(2, "Error while querying ARP cache on device %x: %#x\n",
+ CARD_DEVID(card), rc);
+ return qeth_l3_arp_makerc(rc);
}
static int qeth_l3_arp_query(struct qeth_card *card, char __user *udata)
@@ -1793,15 +1777,18 @@ static int qeth_l3_arp_query(struct qeth_card *card, char __user *udata)
return rc;
}
-static int qeth_l3_arp_add_entry(struct qeth_card *card,
- struct qeth_arp_cache_entry *entry)
+static int qeth_l3_arp_modify_entry(struct qeth_card *card,
+ struct qeth_arp_cache_entry *entry,
+ enum qeth_arp_process_subcmds arp_cmd)
{
+ struct qeth_arp_cache_entry *cmd_entry;
struct qeth_cmd_buffer *iob;
- char buf[16];
- int tmp;
int rc;
- QETH_CARD_TEXT(card, 3, "arpadent");
+ if (arp_cmd == IPA_CMD_ASS_ARP_ADD_ENTRY)
+ QETH_CARD_TEXT(card, 3, "arpadd");
+ else
+ QETH_CARD_TEXT(card, 3, "arpdel");
/*
* currently GuestLAN only supports the ARP assist function
@@ -1814,71 +1801,25 @@ static int qeth_l3_arp_add_entry(struct qeth_card *card,
return -EOPNOTSUPP;
}
- iob = qeth_get_setassparms_cmd(card, IPA_ARP_PROCESSING,
- IPA_CMD_ASS_ARP_ADD_ENTRY,
- sizeof(struct qeth_arp_cache_entry),
- QETH_PROT_IPV4);
+ iob = qeth_get_setassparms_cmd(card, IPA_ARP_PROCESSING, arp_cmd,
+ sizeof(*cmd_entry), QETH_PROT_IPV4);
if (!iob)
return -ENOMEM;
- rc = qeth_send_setassparms(card, iob,
- sizeof(struct qeth_arp_cache_entry),
- (unsigned long) entry,
- qeth_setassparms_cb, NULL);
- if (rc) {
- tmp = rc;
- qeth_l3_ipaddr4_to_string((u8 *)entry->ipaddr, buf);
- QETH_DBF_MESSAGE(2, "Could not add ARP entry for address %s "
- "on %s: %s (0x%x/%d)\n", buf, QETH_CARD_IFNAME(card),
- qeth_l3_arp_get_error_cause(&rc), tmp, tmp);
- }
- return rc;
-}
-static int qeth_l3_arp_remove_entry(struct qeth_card *card,
- struct qeth_arp_cache_entry *entry)
-{
- struct qeth_cmd_buffer *iob;
- char buf[16] = {0, };
- int tmp;
- int rc;
+ cmd_entry = &__ipa_cmd(iob)->data.setassparms.data.arp_entry;
+ ether_addr_copy(cmd_entry->macaddr, entry->macaddr);
+ memcpy(cmd_entry->ipaddr, entry->ipaddr, 4);
+ rc = qeth_send_ipa_cmd(card, iob, qeth_setassparms_cb, NULL);
+ if (rc)
+ QETH_DBF_MESSAGE(2, "Could not modify (cmd: %#x) ARP entry on device %x: %#x\n",
+ arp_cmd, CARD_DEVID(card), rc);
- QETH_CARD_TEXT(card, 3, "arprment");
-
- /*
- * currently GuestLAN only supports the ARP assist function
- * IPA_CMD_ASS_ARP_QUERY_INFO, but not IPA_CMD_ASS_ARP_REMOVE_ENTRY;
- * thus we say EOPNOTSUPP for this ARP function
- */
- if (card->info.guestlan)
- return -EOPNOTSUPP;
- if (!qeth_is_supported(card, IPA_ARP_PROCESSING)) {
- return -EOPNOTSUPP;
- }
- memcpy(buf, entry, 12);
- iob = qeth_get_setassparms_cmd(card, IPA_ARP_PROCESSING,
- IPA_CMD_ASS_ARP_REMOVE_ENTRY,
- 12,
- QETH_PROT_IPV4);
- if (!iob)
- return -ENOMEM;
- rc = qeth_send_setassparms(card, iob,
- 12, (unsigned long)buf,
- qeth_setassparms_cb, NULL);
- if (rc) {
- tmp = rc;
- memset(buf, 0, 16);
- qeth_l3_ipaddr4_to_string((u8 *)entry->ipaddr, buf);
- QETH_DBF_MESSAGE(2, "Could not delete ARP entry for address %s"
- " on %s: %s (0x%x/%d)\n", buf, QETH_CARD_IFNAME(card),
- qeth_l3_arp_get_error_cause(&rc), tmp, tmp);
- }
- return rc;
+ return qeth_l3_arp_makerc(rc);
}
static int qeth_l3_arp_flush_cache(struct qeth_card *card)
{
int rc;
- int tmp;
QETH_CARD_TEXT(card, 3, "arpflush");
@@ -1894,19 +1835,17 @@ static int qeth_l3_arp_flush_cache(struct qeth_card *card)
}
rc = qeth_send_simple_setassparms(card, IPA_ARP_PROCESSING,
IPA_CMD_ASS_ARP_FLUSH_CACHE, 0);
- if (rc) {
- tmp = rc;
- QETH_DBF_MESSAGE(2, "Could not flush ARP cache on %s: %s "
- "(0x%x/%d)\n", QETH_CARD_IFNAME(card),
- qeth_l3_arp_get_error_cause(&rc), tmp, tmp);
- }
- return rc;
+ if (rc)
+ QETH_DBF_MESSAGE(2, "Could not flush ARP cache on device %x: %#x\n",
+ CARD_DEVID(card), rc);
+ return qeth_l3_arp_makerc(rc);
}
static int qeth_l3_do_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
{
struct qeth_card *card = dev->ml_priv;
struct qeth_arp_cache_entry arp_entry;
+ enum qeth_arp_process_subcmds arp_cmd;
int rc = 0;
switch (cmd) {
@@ -1925,27 +1864,16 @@ static int qeth_l3_do_ioctl(struct net_device *dev, struct ifreq *rq, int cmd)
rc = qeth_l3_arp_query(card, rq->ifr_ifru.ifru_data);
break;
case SIOC_QETH_ARP_ADD_ENTRY:
- if (!capable(CAP_NET_ADMIN)) {
- rc = -EPERM;
- break;
- }
- if (copy_from_user(&arp_entry, rq->ifr_ifru.ifru_data,
- sizeof(struct qeth_arp_cache_entry)))
- rc = -EFAULT;
- else
- rc = qeth_l3_arp_add_entry(card, &arp_entry);
- break;
case SIOC_QETH_ARP_REMOVE_ENTRY:
- if (!capable(CAP_NET_ADMIN)) {
- rc = -EPERM;
- break;
- }
- if (copy_from_user(&arp_entry, rq->ifr_ifru.ifru_data,
- sizeof(struct qeth_arp_cache_entry)))
- rc = -EFAULT;
- else
- rc = qeth_l3_arp_remove_entry(card, &arp_entry);
- break;
+ if (!capable(CAP_NET_ADMIN))
+ return -EPERM;
+ if (copy_from_user(&arp_entry, rq->ifr_data, sizeof(arp_entry)))
+ return -EFAULT;
+
+ arp_cmd = (cmd == SIOC_QETH_ARP_ADD_ENTRY) ?
+ IPA_CMD_ASS_ARP_ADD_ENTRY :
+ IPA_CMD_ASS_ARP_REMOVE_ENTRY;
+ return qeth_l3_arp_modify_entry(card, &arp_entry, arp_cmd);
case SIOC_QETH_ARP_FLUSH_CACHE:
if (!capable(CAP_NET_ADMIN)) {
rc = -EPERM;
@@ -2383,12 +2311,12 @@ static const struct net_device_ops qeth_l3_osa_netdev_ops = {
.ndo_neigh_setup = qeth_l3_neigh_setup,
};
-static int qeth_l3_setup_netdev(struct qeth_card *card)
+static int qeth_l3_setup_netdev(struct qeth_card *card, bool carrier_ok)
{
unsigned int headroom;
int rc;
- if (card->dev->netdev_ops)
+ if (qeth_netdev_is_registered(card->dev))
return 0;
if (card->info.type == QETH_CARD_TYPE_OSD ||
@@ -2457,6 +2385,9 @@ static int qeth_l3_setup_netdev(struct qeth_card *card)
netif_napi_add(card->dev, &card->napi, qeth_poll, QETH_NAPI_WEIGHT);
rc = register_netdev(card->dev);
+ if (!rc && carrier_ok)
+ netif_carrier_on(card->dev);
+
out:
if (rc)
card->dev->netdev_ops = NULL;
@@ -2497,7 +2428,8 @@ static void qeth_l3_remove_device(struct ccwgroup_device *cgdev)
if (cgdev->state == CCWGROUP_ONLINE)
qeth_l3_set_offline(cgdev);
- unregister_netdev(card->dev);
+ if (qeth_netdev_is_registered(card->dev))
+ unregister_netdev(card->dev);
qeth_l3_clear_ip_htable(card, 0);
qeth_l3_clear_ipato_list(card);
}
@@ -2507,6 +2439,7 @@ static int __qeth_l3_set_online(struct ccwgroup_device *gdev, int recovery_mode)
struct qeth_card *card = dev_get_drvdata(&gdev->dev);
int rc = 0;
enum qeth_card_states recover_flag;
+ bool carrier_ok;
mutex_lock(&card->discipline_mutex);
mutex_lock(&card->conf_mutex);
@@ -2514,14 +2447,14 @@ static int __qeth_l3_set_online(struct ccwgroup_device *gdev, int recovery_mode)
QETH_DBF_HEX(SETUP, 2, &card, sizeof(void *));
recover_flag = card->state;
- rc = qeth_core_hardsetup_card(card);
+ rc = qeth_core_hardsetup_card(card, &carrier_ok);
if (rc) {
QETH_DBF_TEXT_(SETUP, 2, "2err%04x", rc);
rc = -ENODEV;
goto out_remove;
}
- rc = qeth_l3_setup_netdev(card);
+ rc = qeth_l3_setup_netdev(card, carrier_ok);
if (rc)
goto out_remove;
diff --git a/drivers/scsi/Kconfig b/drivers/scsi/Kconfig
index f07444d..640cd1b 100644
--- a/drivers/scsi/Kconfig
+++ b/drivers/scsi/Kconfig
@@ -578,6 +578,7 @@
config SCSI_MYRS
tristate "Mylex DAC960/DAC1100 PCI RAID Controller (SCSI Interface)"
depends on PCI
+ depends on !CPU_BIG_ENDIAN || COMPILE_TEST
select RAID_ATTRS
help
This driver adds support for the Mylex DAC960, AcceleRAID, and
diff --git a/drivers/scsi/NCR5380.c b/drivers/scsi/NCR5380.c
index 8429c85..01c23d2 100644
--- a/drivers/scsi/NCR5380.c
+++ b/drivers/scsi/NCR5380.c
@@ -1198,7 +1198,7 @@ static bool NCR5380_select(struct Scsi_Host *instance, struct scsi_cmnd *cmd)
out:
if (!hostdata->selecting)
- return NULL;
+ return false;
hostdata->selecting = NULL;
return ret;
}
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
index f0e457e..8df822a 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
@@ -904,11 +904,9 @@ static void start_delivery_v1_hw(struct hisi_sas_dq *dq)
{
struct hisi_hba *hisi_hba = dq->hisi_hba;
struct hisi_sas_slot *s, *s1, *s2 = NULL;
- struct list_head *dq_list;
int dlvry_queue = dq->id;
int wp;
- dq_list = &dq->list;
list_for_each_entry_safe(s, s1, &dq->list, delivery) {
if (!s->ready)
break;
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
index cc36b64..77a85ea 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v2_hw.c
@@ -1670,11 +1670,9 @@ static void start_delivery_v2_hw(struct hisi_sas_dq *dq)
{
struct hisi_hba *hisi_hba = dq->hisi_hba;
struct hisi_sas_slot *s, *s1, *s2 = NULL;
- struct list_head *dq_list;
int dlvry_queue = dq->id;
int wp;
- dq_list = &dq->list;
list_for_each_entry_safe(s, s1, &dq->list, delivery) {
if (!s->ready)
break;
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
index bd4ce38..a369450 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v3_hw.c
@@ -886,11 +886,9 @@ static void start_delivery_v3_hw(struct hisi_sas_dq *dq)
{
struct hisi_hba *hisi_hba = dq->hisi_hba;
struct hisi_sas_slot *s, *s1, *s2 = NULL;
- struct list_head *dq_list;
int dlvry_queue = dq->id;
int wp;
- dq_list = &dq->list;
list_for_each_entry_safe(s, s1, &dq->list, delivery) {
if (!s->ready)
break;
diff --git a/drivers/scsi/lpfc/lpfc_debugfs.c b/drivers/scsi/lpfc/lpfc_debugfs.c
index 0c8005b..34d311a 100644
--- a/drivers/scsi/lpfc/lpfc_debugfs.c
+++ b/drivers/scsi/lpfc/lpfc_debugfs.c
@@ -698,6 +698,8 @@ lpfc_debugfs_nodelist_data(struct lpfc_vport *vport, char *buf, int size)
rport = lpfc_ndlp_get_nrport(ndlp);
if (rport)
nrport = rport->remoteport;
+ else
+ nrport = NULL;
spin_unlock(&phba->hbalock);
if (!nrport)
continue;
diff --git a/drivers/scsi/myrb.c b/drivers/scsi/myrb.c
index aeb282f..0642f2d 100644
--- a/drivers/scsi/myrb.c
+++ b/drivers/scsi/myrb.c
@@ -1049,7 +1049,8 @@ static int myrb_get_hba_config(struct myrb_hba *cb)
enquiry2->fw.firmware_type = '0';
enquiry2->fw.turn_id = 0;
}
- sprintf(cb->fw_version, "%d.%02d-%c-%02d",
+ snprintf(cb->fw_version, sizeof(cb->fw_version),
+ "%d.%02d-%c-%02d",
enquiry2->fw.major_version,
enquiry2->fw.minor_version,
enquiry2->fw.firmware_type,
diff --git a/drivers/scsi/myrs.c b/drivers/scsi/myrs.c
index 0264a2e..b8d54ef 100644
--- a/drivers/scsi/myrs.c
+++ b/drivers/scsi/myrs.c
@@ -163,9 +163,12 @@ static unsigned char myrs_get_ctlr_info(struct myrs_hba *cs)
dma_addr_t ctlr_info_addr;
union myrs_sgl *sgl;
unsigned char status;
- struct myrs_ctlr_info old;
+ unsigned short ldev_present, ldev_critical, ldev_offline;
- memcpy(&old, cs->ctlr_info, sizeof(struct myrs_ctlr_info));
+ ldev_present = cs->ctlr_info->ldev_present;
+ ldev_critical = cs->ctlr_info->ldev_critical;
+ ldev_offline = cs->ctlr_info->ldev_offline;
+
ctlr_info_addr = dma_map_single(&cs->pdev->dev, cs->ctlr_info,
sizeof(struct myrs_ctlr_info),
DMA_FROM_DEVICE);
@@ -198,9 +201,9 @@ static unsigned char myrs_get_ctlr_info(struct myrs_hba *cs)
cs->ctlr_info->rbld_active +
cs->ctlr_info->exp_active != 0)
cs->needs_update = true;
- if (cs->ctlr_info->ldev_present != old.ldev_present ||
- cs->ctlr_info->ldev_critical != old.ldev_critical ||
- cs->ctlr_info->ldev_offline != old.ldev_offline)
+ if (cs->ctlr_info->ldev_present != ldev_present ||
+ cs->ctlr_info->ldev_critical != ldev_critical ||
+ cs->ctlr_info->ldev_offline != ldev_offline)
shost_printk(KERN_INFO, cs->host,
"Logical drive count changes (%d/%d/%d)\n",
cs->ctlr_info->ldev_critical,
diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index 6fe20c2..eb59c79 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -4763,6 +4763,7 @@ qla2x00_alloc_fcport(scsi_qla_host_t *vha, gfp_t flags)
fcport->loop_id = FC_NO_LOOP_ID;
qla2x00_set_fcport_state(fcport, FCS_UNCONFIGURED);
fcport->supported_classes = FC_COS_UNSPECIFIED;
+ fcport->fp_speed = PORT_SPEED_UNKNOWN;
fcport->ct_desc.ct_sns = dma_alloc_coherent(&vha->hw->pdev->dev,
sizeof(struct ct_sns_pkt), &fcport->ct_desc.ct_sns_dma,
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index 518f151..20c85ee 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -67,7 +67,7 @@ module_param(ql2xplogiabsentdevice, int, S_IRUGO|S_IWUSR);
MODULE_PARM_DESC(ql2xplogiabsentdevice,
"Option to enable PLOGI to devices that are not present after "
"a Fabric scan. This is needed for several broken switches. "
- "Default is 0 - no PLOGI. 1 - perfom PLOGI.");
+ "Default is 0 - no PLOGI. 1 - perform PLOGI.");
int ql2xloginretrycount = 0;
module_param(ql2xloginretrycount, int, S_IRUGO);
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index c7fccbb..fa6e0c3 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -697,6 +697,12 @@ static bool scsi_end_request(struct request *req, blk_status_t error,
*/
scsi_mq_uninit_cmd(cmd);
+ /*
+ * queue is still alive, so grab the ref for preventing it
+ * from being cleaned up during running queue.
+ */
+ percpu_ref_get(&q->q_usage_counter);
+
__blk_mq_end_request(req, error);
if (scsi_target(sdev)->single_lun ||
@@ -704,6 +710,8 @@ static bool scsi_end_request(struct request *req, blk_status_t error,
kblockd_schedule_work(&sdev->requeue_work);
else
blk_mq_run_hw_queues(q, true);
+
+ percpu_ref_put(&q->q_usage_counter);
} else {
unsigned long flags;
diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 23d7cca..27db55b 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -8099,13 +8099,6 @@ int ufshcd_alloc_host(struct device *dev, struct ufs_hba **hba_handle)
err = -ENOMEM;
goto out_error;
}
-
- /*
- * Do not use blk-mq at this time because blk-mq does not support
- * runtime pm.
- */
- host->use_blk_mq = false;
-
hba = shost_priv(host);
hba->host = host;
hba->dev = dev;
diff --git a/drivers/staging/vboxvideo/vbox_drv.h b/drivers/staging/vboxvideo/vbox_drv.h
index 73395a75..fa933d4 100644
--- a/drivers/staging/vboxvideo/vbox_drv.h
+++ b/drivers/staging/vboxvideo/vbox_drv.h
@@ -99,8 +99,6 @@ struct vbox_private {
int fb_mtrr;
struct {
- struct drm_global_reference mem_global_ref;
- struct ttm_bo_global_ref bo_global_ref;
struct ttm_bo_device bdev;
} ttm;
diff --git a/drivers/staging/vboxvideo/vbox_ttm.c b/drivers/staging/vboxvideo/vbox_ttm.c
index 5ecfa76..b36ec01 100644
--- a/drivers/staging/vboxvideo/vbox_ttm.c
+++ b/drivers/staging/vboxvideo/vbox_ttm.c
@@ -35,61 +35,6 @@ static inline struct vbox_private *vbox_bdev(struct ttm_bo_device *bd)
return container_of(bd, struct vbox_private, ttm.bdev);
}
-static int vbox_ttm_mem_global_init(struct drm_global_reference *ref)
-{
- return ttm_mem_global_init(ref->object);
-}
-
-static void vbox_ttm_mem_global_release(struct drm_global_reference *ref)
-{
- ttm_mem_global_release(ref->object);
-}
-
-/**
- * Adds the vbox memory manager object/structures to the global memory manager.
- */
-static int vbox_ttm_global_init(struct vbox_private *vbox)
-{
- struct drm_global_reference *global_ref;
- int ret;
-
- global_ref = &vbox->ttm.mem_global_ref;
- global_ref->global_type = DRM_GLOBAL_TTM_MEM;
- global_ref->size = sizeof(struct ttm_mem_global);
- global_ref->init = &vbox_ttm_mem_global_init;
- global_ref->release = &vbox_ttm_mem_global_release;
- ret = drm_global_item_ref(global_ref);
- if (ret) {
- DRM_ERROR("Failed setting up TTM memory subsystem.\n");
- return ret;
- }
-
- vbox->ttm.bo_global_ref.mem_glob = vbox->ttm.mem_global_ref.object;
- global_ref = &vbox->ttm.bo_global_ref.ref;
- global_ref->global_type = DRM_GLOBAL_TTM_BO;
- global_ref->size = sizeof(struct ttm_bo_global);
- global_ref->init = &ttm_bo_global_init;
- global_ref->release = &ttm_bo_global_release;
-
- ret = drm_global_item_ref(global_ref);
- if (ret) {
- DRM_ERROR("Failed setting up TTM BO subsystem.\n");
- drm_global_item_unref(&vbox->ttm.mem_global_ref);
- return ret;
- }
-
- return 0;
-}
-
-/**
- * Removes the vbox memory manager object from the global memory manager.
- */
-static void vbox_ttm_global_release(struct vbox_private *vbox)
-{
- drm_global_item_unref(&vbox->ttm.bo_global_ref.ref);
- drm_global_item_unref(&vbox->ttm.mem_global_ref);
-}
-
static void vbox_bo_ttm_destroy(struct ttm_buffer_object *tbo)
{
struct vbox_bo *bo;
@@ -227,18 +172,13 @@ int vbox_mm_init(struct vbox_private *vbox)
struct drm_device *dev = &vbox->ddev;
struct ttm_bo_device *bdev = &vbox->ttm.bdev;
- ret = vbox_ttm_global_init(vbox);
- if (ret)
- return ret;
-
ret = ttm_bo_device_init(&vbox->ttm.bdev,
- vbox->ttm.bo_global_ref.ref.object,
&vbox_bo_driver,
dev->anon_inode->i_mapping,
DRM_FILE_PAGE_OFFSET, true);
if (ret) {
DRM_ERROR("Error initialising bo driver; %d\n", ret);
- goto err_ttm_global_release;
+ return ret;
}
ret = ttm_bo_init_mm(bdev, TTM_PL_VRAM,
@@ -260,8 +200,6 @@ int vbox_mm_init(struct vbox_private *vbox)
err_device_release:
ttm_bo_device_release(&vbox->ttm.bdev);
-err_ttm_global_release:
- vbox_ttm_global_release(vbox);
return ret;
}
@@ -275,7 +213,6 @@ void vbox_mm_fini(struct vbox_private *vbox)
arch_phys_wc_del(vbox->fb_mtrr);
#endif
ttm_bo_device_release(&vbox->ttm.bdev);
- vbox_ttm_global_release(vbox);
}
void vbox_ttm_placement(struct vbox_bo *bo, int domain)
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index e31e4fc..2cfd61d 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -1778,7 +1778,7 @@ EXPORT_SYMBOL(target_submit_tmr);
void transport_generic_request_failure(struct se_cmd *cmd,
sense_reason_t sense_reason)
{
- int ret = 0;
+ int ret = 0, post_ret;
pr_debug("-----[ Storage Engine Exception; sense_reason %d\n",
sense_reason);
@@ -1790,7 +1790,7 @@ void transport_generic_request_failure(struct se_cmd *cmd,
transport_complete_task_attr(cmd);
if (cmd->transport_complete_callback)
- cmd->transport_complete_callback(cmd, false, NULL);
+ cmd->transport_complete_callback(cmd, false, &post_ret);
if (transport_check_aborted_status(cmd, 1))
return;
diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c
index ff6ba6d..cc56cb3 100644
--- a/drivers/tty/serial/sh-sci.c
+++ b/drivers/tty/serial/sh-sci.c
@@ -1614,10 +1614,10 @@ static void sci_request_dma(struct uart_port *port)
hrtimer_init(&s->rx_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
s->rx_timer.function = rx_timer_fn;
+ s->chan_rx_saved = s->chan_rx = chan;
+
if (port->type == PORT_SCIFA || port->type == PORT_SCIFB)
sci_submit_rx(s);
-
- s->chan_rx_saved = s->chan_rx = chan;
}
}
@@ -3102,6 +3102,7 @@ static struct uart_driver sci_uart_driver = {
static int sci_remove(struct platform_device *dev)
{
struct sci_port *port = platform_get_drvdata(dev);
+ unsigned int type = port->port.type; /* uart_remove_... clears it */
sci_ports_in_use &= ~BIT(port->port.line);
uart_remove_one_port(&sci_uart_driver, &port->port);
@@ -3112,8 +3113,7 @@ static int sci_remove(struct platform_device *dev)
sysfs_remove_file(&dev->dev.kobj,
&dev_attr_rx_fifo_trigger.attr);
}
- if (port->port.type == PORT_SCIFA || port->port.type == PORT_SCIFB ||
- port->port.type == PORT_HSCIF) {
+ if (type == PORT_SCIFA || type == PORT_SCIFB || type == PORT_HSCIF) {
sysfs_remove_file(&dev->dev.kobj,
&dev_attr_rx_fifo_timeout.attr);
}
diff --git a/drivers/tty/tty_baudrate.c b/drivers/tty/tty_baudrate.c
index 7576cea..f438eaa 100644
--- a/drivers/tty/tty_baudrate.c
+++ b/drivers/tty/tty_baudrate.c
@@ -77,7 +77,7 @@ speed_t tty_termios_baud_rate(struct ktermios *termios)
else
cbaud += 15;
}
- return baud_table[cbaud];
+ return cbaud >= n_baud_table ? 0 : baud_table[cbaud];
}
EXPORT_SYMBOL(tty_termios_baud_rate);
@@ -113,7 +113,7 @@ speed_t tty_termios_input_baud_rate(struct ktermios *termios)
else
cbaud += 15;
}
- return baud_table[cbaud];
+ return cbaud >= n_baud_table ? 0 : baud_table[cbaud];
#else /* IBSHIFT */
return tty_termios_baud_rate(termios);
#endif /* IBSHIFT */
diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c
index 55370e6..41ec8e5 100644
--- a/drivers/tty/vt/vt.c
+++ b/drivers/tty/vt/vt.c
@@ -1548,7 +1548,7 @@ static void csi_K(struct vc_data *vc, int vpar)
scr_memsetw(start + offset, vc->vc_video_erase_char, 2 * count);
vc->vc_need_wrap = 0;
if (con_should_update(vc))
- do_update_region(vc, (unsigned long) start, count);
+ do_update_region(vc, (unsigned long)(start + offset), count);
}
static void csi_X(struct vc_data *vc, int vpar) /* erase the following vpar positions */
diff --git a/drivers/usb/typec/ucsi/Kconfig b/drivers/usb/typec/ucsi/Kconfig
index e36d6c7..7811888 100644
--- a/drivers/usb/typec/ucsi/Kconfig
+++ b/drivers/usb/typec/ucsi/Kconfig
@@ -23,6 +23,16 @@
if TYPEC_UCSI
+config UCSI_CCG
+ tristate "UCSI Interface Driver for Cypress CCGx"
+ depends on I2C
+ help
+ This driver enables UCSI support on platforms that expose a
+ Cypress CCGx Type-C controller over I2C interface.
+
+ To compile the driver as a module, choose M here: the module will be
+ called ucsi_ccg.
+
config UCSI_ACPI
tristate "UCSI ACPI Interface Driver"
depends on ACPI
diff --git a/drivers/usb/typec/ucsi/Makefile b/drivers/usb/typec/ucsi/Makefile
index 7afbea5..2f4900b 100644
--- a/drivers/usb/typec/ucsi/Makefile
+++ b/drivers/usb/typec/ucsi/Makefile
@@ -8,3 +8,5 @@
typec_ucsi-$(CONFIG_TRACING) += trace.o
obj-$(CONFIG_UCSI_ACPI) += ucsi_acpi.o
+
+obj-$(CONFIG_UCSI_CCG) += ucsi_ccg.o
diff --git a/drivers/usb/typec/ucsi/ucsi_ccg.c b/drivers/usb/typec/ucsi/ucsi_ccg.c
new file mode 100644
index 0000000..de8a43b
--- /dev/null
+++ b/drivers/usb/typec/ucsi/ucsi_ccg.c
@@ -0,0 +1,307 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * UCSI driver for Cypress CCGx Type-C controller
+ *
+ * Copyright (C) 2017-2018 NVIDIA Corporation. All rights reserved.
+ * Author: Ajay Gupta <ajayg@nvidia.com>
+ *
+ * Some code borrowed from drivers/usb/typec/ucsi/ucsi_acpi.c
+ */
+#include <linux/acpi.h>
+#include <linux/delay.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+#include <linux/platform_device.h>
+
+#include <asm/unaligned.h>
+#include "ucsi.h"
+
+struct ucsi_ccg {
+ struct device *dev;
+ struct ucsi *ucsi;
+ struct ucsi_ppm ppm;
+ struct i2c_client *client;
+};
+
+#define CCGX_RAB_INTR_REG 0x06
+#define CCGX_RAB_UCSI_CONTROL 0x39
+#define CCGX_RAB_UCSI_CONTROL_START BIT(0)
+#define CCGX_RAB_UCSI_CONTROL_STOP BIT(1)
+#define CCGX_RAB_UCSI_DATA_BLOCK(offset) (0xf000 | ((offset) & 0xff))
+
+static int ccg_read(struct ucsi_ccg *uc, u16 rab, u8 *data, u32 len)
+{
+ struct i2c_client *client = uc->client;
+ const struct i2c_adapter_quirks *quirks = client->adapter->quirks;
+ unsigned char buf[2];
+ struct i2c_msg msgs[] = {
+ {
+ .addr = client->addr,
+ .flags = 0x0,
+ .len = sizeof(buf),
+ .buf = buf,
+ },
+ {
+ .addr = client->addr,
+ .flags = I2C_M_RD,
+ .buf = data,
+ },
+ };
+ u32 rlen, rem_len = len, max_read_len = len;
+ int status;
+
+ /* check any max_read_len limitation on i2c adapter */
+ if (quirks && quirks->max_read_len)
+ max_read_len = quirks->max_read_len;
+
+ while (rem_len > 0) {
+ msgs[1].buf = &data[len - rem_len];
+ rlen = min_t(u16, rem_len, max_read_len);
+ msgs[1].len = rlen;
+ put_unaligned_le16(rab, buf);
+ status = i2c_transfer(client->adapter, msgs, ARRAY_SIZE(msgs));
+ if (status < 0) {
+ dev_err(uc->dev, "i2c_transfer failed %d\n", status);
+ return status;
+ }
+ rab += rlen;
+ rem_len -= rlen;
+ }
+
+ return 0;
+}
+
+static int ccg_write(struct ucsi_ccg *uc, u16 rab, u8 *data, u32 len)
+{
+ struct i2c_client *client = uc->client;
+ unsigned char *buf;
+ struct i2c_msg msgs[] = {
+ {
+ .addr = client->addr,
+ .flags = 0x0,
+ }
+ };
+ int status;
+
+ buf = kzalloc(len + sizeof(rab), GFP_KERNEL);
+ if (!buf)
+ return -ENOMEM;
+
+ put_unaligned_le16(rab, buf);
+ memcpy(buf + sizeof(rab), data, len);
+
+ msgs[0].len = len + sizeof(rab);
+ msgs[0].buf = buf;
+
+ status = i2c_transfer(client->adapter, msgs, ARRAY_SIZE(msgs));
+ if (status < 0) {
+ dev_err(uc->dev, "i2c_transfer failed %d\n", status);
+ kfree(buf);
+ return status;
+ }
+
+ kfree(buf);
+ return 0;
+}
+
+static int ucsi_ccg_init(struct ucsi_ccg *uc)
+{
+ unsigned int count = 10;
+ u8 data;
+ int status;
+
+ data = CCGX_RAB_UCSI_CONTROL_STOP;
+ status = ccg_write(uc, CCGX_RAB_UCSI_CONTROL, &data, sizeof(data));
+ if (status < 0)
+ return status;
+
+ data = CCGX_RAB_UCSI_CONTROL_START;
+ status = ccg_write(uc, CCGX_RAB_UCSI_CONTROL, &data, sizeof(data));
+ if (status < 0)
+ return status;
+
+ /*
+ * Flush CCGx RESPONSE queue by acking interrupts. Above ucsi control
+ * register write will push response which must be cleared.
+ */
+ do {
+ status = ccg_read(uc, CCGX_RAB_INTR_REG, &data, sizeof(data));
+ if (status < 0)
+ return status;
+
+ if (!data)
+ return 0;
+
+ status = ccg_write(uc, CCGX_RAB_INTR_REG, &data, sizeof(data));
+ if (status < 0)
+ return status;
+
+ usleep_range(10000, 11000);
+ } while (--count);
+
+ return -ETIMEDOUT;
+}
+
+static int ucsi_ccg_send_data(struct ucsi_ccg *uc)
+{
+ u8 *ppm = (u8 *)uc->ppm.data;
+ int status;
+ u16 rab;
+
+ rab = CCGX_RAB_UCSI_DATA_BLOCK(offsetof(struct ucsi_data, message_out));
+ status = ccg_write(uc, rab, ppm +
+ offsetof(struct ucsi_data, message_out),
+ sizeof(uc->ppm.data->message_out));
+ if (status < 0)
+ return status;
+
+ rab = CCGX_RAB_UCSI_DATA_BLOCK(offsetof(struct ucsi_data, ctrl));
+ return ccg_write(uc, rab, ppm + offsetof(struct ucsi_data, ctrl),
+ sizeof(uc->ppm.data->ctrl));
+}
+
+static int ucsi_ccg_recv_data(struct ucsi_ccg *uc)
+{
+ u8 *ppm = (u8 *)uc->ppm.data;
+ int status;
+ u16 rab;
+
+ rab = CCGX_RAB_UCSI_DATA_BLOCK(offsetof(struct ucsi_data, cci));
+ status = ccg_read(uc, rab, ppm + offsetof(struct ucsi_data, cci),
+ sizeof(uc->ppm.data->cci));
+ if (status < 0)
+ return status;
+
+ rab = CCGX_RAB_UCSI_DATA_BLOCK(offsetof(struct ucsi_data, message_in));
+ return ccg_read(uc, rab, ppm + offsetof(struct ucsi_data, message_in),
+ sizeof(uc->ppm.data->message_in));
+}
+
+static int ucsi_ccg_ack_interrupt(struct ucsi_ccg *uc)
+{
+ int status;
+ unsigned char data;
+
+ status = ccg_read(uc, CCGX_RAB_INTR_REG, &data, sizeof(data));
+ if (status < 0)
+ return status;
+
+ return ccg_write(uc, CCGX_RAB_INTR_REG, &data, sizeof(data));
+}
+
+static int ucsi_ccg_sync(struct ucsi_ppm *ppm)
+{
+ struct ucsi_ccg *uc = container_of(ppm, struct ucsi_ccg, ppm);
+ int status;
+
+ status = ucsi_ccg_recv_data(uc);
+ if (status < 0)
+ return status;
+
+ /* ack interrupt to allow next command to run */
+ return ucsi_ccg_ack_interrupt(uc);
+}
+
+static int ucsi_ccg_cmd(struct ucsi_ppm *ppm, struct ucsi_control *ctrl)
+{
+ struct ucsi_ccg *uc = container_of(ppm, struct ucsi_ccg, ppm);
+
+ ppm->data->ctrl.raw_cmd = ctrl->raw_cmd;
+ return ucsi_ccg_send_data(uc);
+}
+
+static irqreturn_t ccg_irq_handler(int irq, void *data)
+{
+ struct ucsi_ccg *uc = data;
+
+ ucsi_notify(uc->ucsi);
+
+ return IRQ_HANDLED;
+}
+
+static int ucsi_ccg_probe(struct i2c_client *client,
+ const struct i2c_device_id *id)
+{
+ struct device *dev = &client->dev;
+ struct ucsi_ccg *uc;
+ int status;
+ u16 rab;
+
+ uc = devm_kzalloc(dev, sizeof(*uc), GFP_KERNEL);
+ if (!uc)
+ return -ENOMEM;
+
+ uc->ppm.data = devm_kzalloc(dev, sizeof(struct ucsi_data), GFP_KERNEL);
+ if (!uc->ppm.data)
+ return -ENOMEM;
+
+ uc->ppm.cmd = ucsi_ccg_cmd;
+ uc->ppm.sync = ucsi_ccg_sync;
+ uc->dev = dev;
+ uc->client = client;
+
+ /* reset ccg device and initialize ucsi */
+ status = ucsi_ccg_init(uc);
+ if (status < 0) {
+ dev_err(uc->dev, "ucsi_ccg_init failed - %d\n", status);
+ return status;
+ }
+
+ status = devm_request_threaded_irq(dev, client->irq, NULL,
+ ccg_irq_handler,
+ IRQF_ONESHOT | IRQF_TRIGGER_HIGH,
+ dev_name(dev), uc);
+ if (status < 0) {
+ dev_err(uc->dev, "request_threaded_irq failed - %d\n", status);
+ return status;
+ }
+
+ uc->ucsi = ucsi_register_ppm(dev, &uc->ppm);
+ if (IS_ERR(uc->ucsi)) {
+ dev_err(uc->dev, "ucsi_register_ppm failed\n");
+ return PTR_ERR(uc->ucsi);
+ }
+
+ rab = CCGX_RAB_UCSI_DATA_BLOCK(offsetof(struct ucsi_data, version));
+ status = ccg_read(uc, rab, (u8 *)(uc->ppm.data) +
+ offsetof(struct ucsi_data, version),
+ sizeof(uc->ppm.data->version));
+ if (status < 0) {
+ ucsi_unregister_ppm(uc->ucsi);
+ return status;
+ }
+
+ i2c_set_clientdata(client, uc);
+ return 0;
+}
+
+static int ucsi_ccg_remove(struct i2c_client *client)
+{
+ struct ucsi_ccg *uc = i2c_get_clientdata(client);
+
+ ucsi_unregister_ppm(uc->ucsi);
+
+ return 0;
+}
+
+static const struct i2c_device_id ucsi_ccg_device_id[] = {
+ {"ccgx-ucsi", 0},
+ {}
+};
+MODULE_DEVICE_TABLE(i2c, ucsi_ccg_device_id);
+
+static struct i2c_driver ucsi_ccg_driver = {
+ .driver = {
+ .name = "ucsi_ccg",
+ },
+ .probe = ucsi_ccg_probe,
+ .remove = ucsi_ccg_remove,
+ .id_table = ucsi_ccg_device_id,
+};
+
+module_i2c_driver(ucsi_ccg_driver);
+
+MODULE_AUTHOR("Ajay Gupta <ajayg@nvidia.com>");
+MODULE_DESCRIPTION("UCSI driver for Cypress CCGx Type-C controller");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c
index f15f89d..7ea6fb6 100644
--- a/drivers/xen/grant-table.c
+++ b/drivers/xen/grant-table.c
@@ -914,7 +914,7 @@ int gnttab_dma_free_pages(struct gnttab_dma_alloc_args *args)
ret = xenmem_reservation_increase(args->nr_pages, args->frames);
if (ret != args->nr_pages) {
- pr_debug("Failed to decrease reservation for DMA buffer\n");
+ pr_debug("Failed to increase reservation for DMA buffer\n");
ret = -EFAULT;
} else {
ret = 0;
diff --git a/drivers/xen/privcmd-buf.c b/drivers/xen/privcmd-buf.c
index df1ed37..de01a6d 100644
--- a/drivers/xen/privcmd-buf.c
+++ b/drivers/xen/privcmd-buf.c
@@ -21,15 +21,9 @@
MODULE_LICENSE("GPL");
-static unsigned int limit = 64;
-module_param(limit, uint, 0644);
-MODULE_PARM_DESC(limit, "Maximum number of pages that may be allocated by "
- "the privcmd-buf device per open file");
-
struct privcmd_buf_private {
struct mutex lock;
struct list_head list;
- unsigned int allocated;
};
struct privcmd_buf_vma_private {
@@ -60,13 +54,10 @@ static void privcmd_buf_vmapriv_free(struct privcmd_buf_vma_private *vma_priv)
{
unsigned int i;
- vma_priv->file_priv->allocated -= vma_priv->n_pages;
-
list_del(&vma_priv->list);
for (i = 0; i < vma_priv->n_pages; i++)
- if (vma_priv->pages[i])
- __free_page(vma_priv->pages[i]);
+ __free_page(vma_priv->pages[i]);
kfree(vma_priv);
}
@@ -146,8 +137,7 @@ static int privcmd_buf_mmap(struct file *file, struct vm_area_struct *vma)
unsigned int i;
int ret = 0;
- if (!(vma->vm_flags & VM_SHARED) || count > limit ||
- file_priv->allocated + count > limit)
+ if (!(vma->vm_flags & VM_SHARED))
return -EINVAL;
vma_priv = kzalloc(sizeof(*vma_priv) + count * sizeof(void *),
@@ -155,19 +145,15 @@ static int privcmd_buf_mmap(struct file *file, struct vm_area_struct *vma)
if (!vma_priv)
return -ENOMEM;
- vma_priv->n_pages = count;
- count = 0;
- for (i = 0; i < vma_priv->n_pages; i++) {
+ for (i = 0; i < count; i++) {
vma_priv->pages[i] = alloc_page(GFP_KERNEL | __GFP_ZERO);
if (!vma_priv->pages[i])
break;
- count++;
+ vma_priv->n_pages++;
}
mutex_lock(&file_priv->lock);
- file_priv->allocated += count;
-
vma_priv->file_priv = file_priv;
vma_priv->users = 1;
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 8095352..68f322f 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -3163,6 +3163,9 @@ void btrfs_destroy_inode(struct inode *inode);
int btrfs_drop_inode(struct inode *inode);
int __init btrfs_init_cachep(void);
void __cold btrfs_destroy_cachep(void);
+struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
+ struct btrfs_root *root, int *new,
+ struct btrfs_path *path);
struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
struct btrfs_root *root, int *was_new);
struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index b0ab41d..3f0b6d1 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1664,9 +1664,8 @@ static int cleaner_kthread(void *arg)
struct btrfs_root *root = arg;
struct btrfs_fs_info *fs_info = root->fs_info;
int again;
- struct btrfs_trans_handle *trans;
- do {
+ while (1) {
again = 0;
/* Make the cleaner go to sleep early. */
@@ -1715,42 +1714,16 @@ static int cleaner_kthread(void *arg)
*/
btrfs_delete_unused_bgs(fs_info);
sleep:
+ if (kthread_should_park())
+ kthread_parkme();
+ if (kthread_should_stop())
+ return 0;
if (!again) {
set_current_state(TASK_INTERRUPTIBLE);
- if (!kthread_should_stop())
- schedule();
+ schedule();
__set_current_state(TASK_RUNNING);
}
- } while (!kthread_should_stop());
-
- /*
- * Transaction kthread is stopped before us and wakes us up.
- * However we might have started a new transaction and COWed some
- * tree blocks when deleting unused block groups for example. So
- * make sure we commit the transaction we started to have a clean
- * shutdown when evicting the btree inode - if it has dirty pages
- * when we do the final iput() on it, eviction will trigger a
- * writeback for it which will fail with null pointer dereferences
- * since work queues and other resources were already released and
- * destroyed by the time the iput/eviction/writeback is made.
- */
- trans = btrfs_attach_transaction(root);
- if (IS_ERR(trans)) {
- if (PTR_ERR(trans) != -ENOENT)
- btrfs_err(fs_info,
- "cleaner transaction attach returned %ld",
- PTR_ERR(trans));
- } else {
- int ret;
-
- ret = btrfs_commit_transaction(trans);
- if (ret)
- btrfs_err(fs_info,
- "cleaner open transaction commit returned %d",
- ret);
}
-
- return 0;
}
static int transaction_kthread(void *arg)
@@ -3931,6 +3904,13 @@ void close_ctree(struct btrfs_fs_info *fs_info)
int ret;
set_bit(BTRFS_FS_CLOSING_START, &fs_info->flags);
+ /*
+ * We don't want the cleaner to start new transactions, add more delayed
+ * iputs, etc. while we're closing. We can't use kthread_stop() yet
+ * because that frees the task_struct, and the transaction kthread might
+ * still try to wake up the cleaner.
+ */
+ kthread_park(fs_info->cleaner_kthread);
/* wait for the qgroup rescan worker to stop */
btrfs_qgroup_wait_for_completion(fs_info, false);
@@ -3958,9 +3938,8 @@ void close_ctree(struct btrfs_fs_info *fs_info)
if (!sb_rdonly(fs_info->sb)) {
/*
- * If the cleaner thread is stopped and there are
- * block groups queued for removal, the deletion will be
- * skipped when we quit the cleaner thread.
+ * The cleaner kthread is stopped, so do one final pass over
+ * unused block groups.
*/
btrfs_delete_unused_bgs(fs_info);
@@ -4359,13 +4338,23 @@ static int btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info,
unpin = pinned_extents;
again:
while (1) {
+ /*
+ * The btrfs_finish_extent_commit() may get the same range as
+ * ours between find_first_extent_bit and clear_extent_dirty.
+ * Hence, hold the unused_bg_unpin_mutex to avoid double unpin
+ * the same extent range.
+ */
+ mutex_lock(&fs_info->unused_bg_unpin_mutex);
ret = find_first_extent_bit(unpin, 0, &start, &end,
EXTENT_DIRTY, NULL);
- if (ret)
+ if (ret) {
+ mutex_unlock(&fs_info->unused_bg_unpin_mutex);
break;
+ }
clear_extent_dirty(unpin, start, end);
btrfs_error_unpin_extent_range(fs_info, start, end);
+ mutex_unlock(&fs_info->unused_bg_unpin_mutex);
cond_resched();
}
diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c
index 4ba0aed..74aa552 100644
--- a/fs/btrfs/free-space-cache.c
+++ b/fs/btrfs/free-space-cache.c
@@ -75,7 +75,8 @@ static struct inode *__lookup_free_space_inode(struct btrfs_root *root,
* sure NOFS is set to keep us from deadlocking.
*/
nofs_flag = memalloc_nofs_save();
- inode = btrfs_iget(fs_info->sb, &location, root, NULL);
+ inode = btrfs_iget_path(fs_info->sb, &location, root, NULL, path);
+ btrfs_release_path(path);
memalloc_nofs_restore(nofs_flag);
if (IS_ERR(inode))
return inode;
@@ -838,6 +839,25 @@ int load_free_space_cache(struct btrfs_fs_info *fs_info,
path->search_commit_root = 1;
path->skip_locking = 1;
+ /*
+ * We must pass a path with search_commit_root set to btrfs_iget in
+ * order to avoid a deadlock when allocating extents for the tree root.
+ *
+ * When we are COWing an extent buffer from the tree root, when looking
+ * for a free extent, at extent-tree.c:find_free_extent(), we can find
+ * block group without its free space cache loaded. When we find one
+ * we must load its space cache which requires reading its free space
+ * cache's inode item from the root tree. If this inode item is located
+ * in the same leaf that we started COWing before, then we end up in
+ * deadlock on the extent buffer (trying to read lock it when we
+ * previously write locked it).
+ *
+ * It's safe to read the inode item using the commit root because
+ * block groups, once loaded, stay in memory forever (until they are
+ * removed) as well as their space caches once loaded. New block groups
+ * once created get their ->cached field set to BTRFS_CACHE_FINISHED so
+ * we will never try to read their inode item while the fs is mounted.
+ */
inode = lookup_free_space_inode(fs_info, block_group, path);
if (IS_ERR(inode)) {
btrfs_free_path(path);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index d3df5b5..9ea4c6f 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -1531,12 +1531,11 @@ static noinline int run_delalloc_nocow(struct inode *inode,
}
btrfs_release_path(path);
- if (cur_offset <= end && cow_start == (u64)-1) {
+ if (cur_offset <= end && cow_start == (u64)-1)
cow_start = cur_offset;
- cur_offset = end;
- }
if (cow_start != (u64)-1) {
+ cur_offset = end;
ret = cow_file_range(inode, locked_page, cow_start, end, end,
page_started, nr_written, 1, NULL);
if (ret)
@@ -3570,10 +3569,11 @@ static noinline int acls_after_inode_item(struct extent_buffer *leaf,
/*
* read an inode from the btree into the in-memory inode
*/
-static int btrfs_read_locked_inode(struct inode *inode)
+static int btrfs_read_locked_inode(struct inode *inode,
+ struct btrfs_path *in_path)
{
struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
- struct btrfs_path *path;
+ struct btrfs_path *path = in_path;
struct extent_buffer *leaf;
struct btrfs_inode_item *inode_item;
struct btrfs_root *root = BTRFS_I(inode)->root;
@@ -3589,15 +3589,18 @@ static int btrfs_read_locked_inode(struct inode *inode)
if (!ret)
filled = true;
- path = btrfs_alloc_path();
- if (!path)
- return -ENOMEM;
+ if (!path) {
+ path = btrfs_alloc_path();
+ if (!path)
+ return -ENOMEM;
+ }
memcpy(&location, &BTRFS_I(inode)->location, sizeof(location));
ret = btrfs_lookup_inode(NULL, root, path, &location, 0);
if (ret) {
- btrfs_free_path(path);
+ if (path != in_path)
+ btrfs_free_path(path);
return ret;
}
@@ -3722,7 +3725,8 @@ static int btrfs_read_locked_inode(struct inode *inode)
btrfs_ino(BTRFS_I(inode)),
root->root_key.objectid, ret);
}
- btrfs_free_path(path);
+ if (path != in_path)
+ btrfs_free_path(path);
if (!maybe_acls)
cache_no_acl(inode);
@@ -5644,8 +5648,9 @@ static struct inode *btrfs_iget_locked(struct super_block *s,
/* Get an inode object given its location and corresponding root.
* Returns in *is_new if the inode was read from disk
*/
-struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
- struct btrfs_root *root, int *new)
+struct inode *btrfs_iget_path(struct super_block *s, struct btrfs_key *location,
+ struct btrfs_root *root, int *new,
+ struct btrfs_path *path)
{
struct inode *inode;
@@ -5656,7 +5661,7 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
if (inode->i_state & I_NEW) {
int ret;
- ret = btrfs_read_locked_inode(inode);
+ ret = btrfs_read_locked_inode(inode, path);
if (!ret) {
inode_tree_add(inode);
unlock_new_inode(inode);
@@ -5678,6 +5683,12 @@ struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
return inode;
}
+struct inode *btrfs_iget(struct super_block *s, struct btrfs_key *location,
+ struct btrfs_root *root, int *new)
+{
+ return btrfs_iget_path(s, location, root, new, NULL);
+}
+
static struct inode *new_simple_dir(struct super_block *s,
struct btrfs_key *key,
struct btrfs_root *root)
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 3ca6943..802a628 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -3488,6 +3488,8 @@ static int btrfs_extent_same_range(struct inode *src, u64 loff, u64 olen,
const u64 sz = BTRFS_I(src)->root->fs_info->sectorsize;
len = round_down(i_size_read(src), sz) - loff;
+ if (len == 0)
+ return 0;
olen = len;
}
}
@@ -4257,9 +4259,17 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src,
goto out_unlock;
if (len == 0)
olen = len = src->i_size - off;
- /* if we extend to eof, continue to block boundary */
- if (off + len == src->i_size)
+ /*
+ * If we extend to eof, continue to block boundary if and only if the
+ * destination end offset matches the destination file's size, otherwise
+ * we would be corrupting data by placing the eof block into the middle
+ * of a file.
+ */
+ if (off + len == src->i_size) {
+ if (!IS_ALIGNED(len, bs) && destoff + len < inode->i_size)
+ goto out_unlock;
len = ALIGN(src->i_size, bs) - off;
+ }
if (len == 0) {
ret = 0;
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index b362b45..cbc9d0d 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -1916,7 +1916,7 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data)
}
/* Used to sort the devices by max_avail(descending sort) */
-static int btrfs_cmp_device_free_bytes(const void *dev_info1,
+static inline int btrfs_cmp_device_free_bytes(const void *dev_info1,
const void *dev_info2)
{
if (((struct btrfs_device_info *)dev_info1)->max_avail >
@@ -1945,8 +1945,8 @@ static inline void btrfs_descending_sort_devices(
* The helper to calc the free space on the devices that can be used to store
* file data.
*/
-static int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
- u64 *free_bytes)
+static inline int btrfs_calc_avail_data_space(struct btrfs_fs_info *fs_info,
+ u64 *free_bytes)
{
struct btrfs_device_info *devices_info;
struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index cab0b1f..efcf89a8 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -440,7 +440,7 @@ static int check_block_group_item(struct btrfs_fs_info *fs_info,
type != (BTRFS_BLOCK_GROUP_METADATA |
BTRFS_BLOCK_GROUP_DATA)) {
block_group_err(fs_info, leaf, slot,
-"invalid type, have 0x%llx (%lu bits set) expect either 0x%llx, 0x%llx, 0x%llu or 0x%llx",
+"invalid type, have 0x%llx (%lu bits set) expect either 0x%llx, 0x%llx, 0x%llx or 0x%llx",
type, hweight64(type),
BTRFS_BLOCK_GROUP_DATA, BTRFS_BLOCK_GROUP_METADATA,
BTRFS_BLOCK_GROUP_SYSTEM,
diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c
index e07f337..a5ce99a 100644
--- a/fs/btrfs/tree-log.c
+++ b/fs/btrfs/tree-log.c
@@ -4396,6 +4396,23 @@ static int btrfs_log_changed_extents(struct btrfs_trans_handle *trans,
logged_end = end;
list_for_each_entry_safe(em, n, &tree->modified_extents, list) {
+ /*
+ * Skip extents outside our logging range. It's important to do
+ * it for correctness because if we don't ignore them, we may
+ * log them before their ordered extent completes, and therefore
+ * we could log them without logging their respective checksums
+ * (the checksum items are added to the csum tree at the very
+ * end of btrfs_finish_ordered_io()). Also leave such extents
+ * outside of our range in the list, since we may have another
+ * ranged fsync in the near future that needs them. If an extent
+ * outside our range corresponds to a hole, log it to avoid
+ * leaving gaps between extents (fsck will complain when we are
+ * not using the NO_HOLES feature).
+ */
+ if ((em->start > end || em->start + em->len <= start) &&
+ em->block_start != EXTENT_MAP_HOLE)
+ continue;
+
list_del_init(&em->list);
/*
* Just an arbitrary number, this can be really CPU intensive
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 27cad84..189df66 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -1931,10 +1931,17 @@ static ssize_t ceph_copy_file_range(struct file *src_file, loff_t src_off,
if (!prealloc_cf)
return -ENOMEM;
- /* Start by sync'ing the source file */
+ /* Start by sync'ing the source and destination files */
ret = file_write_and_wait_range(src_file, src_off, (src_off + len));
- if (ret < 0)
+ if (ret < 0) {
+ dout("failed to write src file (%zd)\n", ret);
goto out;
+ }
+ ret = file_write_and_wait_range(dst_file, dst_off, (dst_off + len));
+ if (ret < 0) {
+ dout("failed to write dst file (%zd)\n", ret);
+ goto out;
+ }
/*
* We need FILE_WR caps for dst_ci and FILE_RD for src_ci as other
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index 67a9aeb..bd13a32 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -80,12 +80,8 @@ static int parse_reply_info_in(void **p, void *end,
info->symlink = *p;
*p += info->symlink_len;
- if (features & CEPH_FEATURE_DIRLAYOUTHASH)
- ceph_decode_copy_safe(p, end, &info->dir_layout,
- sizeof(info->dir_layout), bad);
- else
- memset(&info->dir_layout, 0, sizeof(info->dir_layout));
-
+ ceph_decode_copy_safe(p, end, &info->dir_layout,
+ sizeof(info->dir_layout), bad);
ceph_decode_32_safe(p, end, info->xattr_len, bad);
ceph_decode_need(p, end, info->xattr_len, bad);
info->xattr_data = *p;
@@ -3182,10 +3178,8 @@ static void send_mds_reconnect(struct ceph_mds_client *mdsc,
recon_state.pagelist = pagelist;
if (session->s_con.peer_features & CEPH_FEATURE_MDSENC)
recon_state.msg_version = 3;
- else if (session->s_con.peer_features & CEPH_FEATURE_FLOCK)
- recon_state.msg_version = 2;
else
- recon_state.msg_version = 1;
+ recon_state.msg_version = 2;
err = iterate_session_caps(session, encode_caps_cb, &recon_state);
if (err < 0)
goto fail;
diff --git a/fs/ceph/quota.c b/fs/ceph/quota.c
index 32d4f13..03f4d24 100644
--- a/fs/ceph/quota.c
+++ b/fs/ceph/quota.c
@@ -237,7 +237,8 @@ static bool check_quota_exceeded(struct inode *inode, enum quota_check_op op,
ceph_put_snap_realm(mdsc, realm);
realm = next;
}
- ceph_put_snap_realm(mdsc, realm);
+ if (realm)
+ ceph_put_snap_realm(mdsc, realm);
up_read(&mdsc->snap_rwsem);
return exceeded;
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 05f01fb..22a9d81 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -5835,9 +5835,10 @@ int ext4_mark_iloc_dirty(handle_t *handle,
{
int err = 0;
- if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
+ if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb)))) {
+ put_bh(iloc->bh);
return -EIO;
-
+ }
if (IS_I_VERSION(inode))
inode_inc_iversion(inode);
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 17adcb1..437f71f 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -126,6 +126,7 @@ static struct buffer_head *__ext4_read_dirblock(struct inode *inode,
if (!is_dx_block && type == INDEX) {
ext4_error_inode(inode, func, line, block,
"directory leaf block found instead of index block");
+ brelse(bh);
return ERR_PTR(-EFSCORRUPTED);
}
if (!ext4_has_metadata_csum(inode->i_sb) ||
@@ -2811,7 +2812,9 @@ int ext4_orphan_add(handle_t *handle, struct inode *inode)
list_del_init(&EXT4_I(inode)->i_orphan);
mutex_unlock(&sbi->s_orphan_lock);
}
- }
+ } else
+ brelse(iloc.bh);
+
jbd_debug(4, "superblock will point to %lu\n", inode->i_ino);
jbd_debug(4, "orphan inode %lu will point to %d\n",
inode->i_ino, NEXT_ORPHAN(inode));
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index ebbc663..a5efee3 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -459,16 +459,18 @@ static int set_flexbg_block_bitmap(struct super_block *sb, handle_t *handle,
BUFFER_TRACE(bh, "get_write_access");
err = ext4_journal_get_write_access(handle, bh);
- if (err)
+ if (err) {
+ brelse(bh);
return err;
+ }
ext4_debug("mark block bitmap %#04llx (+%llu/%u)\n",
first_cluster, first_cluster - start, count2);
ext4_set_bits(bh->b_data, first_cluster - start, count2);
err = ext4_handle_dirty_metadata(handle, NULL, bh);
+ brelse(bh);
if (unlikely(err))
return err;
- brelse(bh);
}
return 0;
@@ -605,7 +607,6 @@ static int setup_new_flex_group_blocks(struct super_block *sb,
bh = bclean(handle, sb, block);
if (IS_ERR(bh)) {
err = PTR_ERR(bh);
- bh = NULL;
goto out;
}
overhead = ext4_group_overhead_blocks(sb, group);
@@ -618,9 +619,9 @@ static int setup_new_flex_group_blocks(struct super_block *sb,
ext4_mark_bitmap_end(EXT4_B2C(sbi, group_data[i].blocks_count),
sb->s_blocksize * 8, bh->b_data);
err = ext4_handle_dirty_metadata(handle, NULL, bh);
+ brelse(bh);
if (err)
goto out;
- brelse(bh);
handle_ib:
if (bg_flags[i] & EXT4_BG_INODE_UNINIT)
@@ -635,18 +636,16 @@ static int setup_new_flex_group_blocks(struct super_block *sb,
bh = bclean(handle, sb, block);
if (IS_ERR(bh)) {
err = PTR_ERR(bh);
- bh = NULL;
goto out;
}
ext4_mark_bitmap_end(EXT4_INODES_PER_GROUP(sb),
sb->s_blocksize * 8, bh->b_data);
err = ext4_handle_dirty_metadata(handle, NULL, bh);
+ brelse(bh);
if (err)
goto out;
- brelse(bh);
}
- bh = NULL;
/* Mark group tables in block bitmap */
for (j = 0; j < GROUP_TABLE_COUNT; j++) {
@@ -685,7 +684,6 @@ static int setup_new_flex_group_blocks(struct super_block *sb,
}
out:
- brelse(bh);
err2 = ext4_journal_stop(handle);
if (err2 && !err)
err = err2;
@@ -873,6 +871,7 @@ static int add_new_gdb(handle_t *handle, struct inode *inode,
err = ext4_handle_dirty_metadata(handle, NULL, gdb_bh);
if (unlikely(err)) {
ext4_std_error(sb, err);
+ iloc.bh = NULL;
goto exit_inode;
}
brelse(dind);
@@ -924,6 +923,7 @@ static int add_new_gdb_meta_bg(struct super_block *sb,
sizeof(struct buffer_head *),
GFP_NOFS);
if (!n_group_desc) {
+ brelse(gdb_bh);
err = -ENOMEM;
ext4_warning(sb, "not enough memory for %lu groups",
gdb_num + 1);
@@ -939,8 +939,6 @@ static int add_new_gdb_meta_bg(struct super_block *sb,
kvfree(o_group_desc);
BUFFER_TRACE(gdb_bh, "get_write_access");
err = ext4_journal_get_write_access(handle, gdb_bh);
- if (unlikely(err))
- brelse(gdb_bh);
return err;
}
@@ -1124,8 +1122,10 @@ static void update_backups(struct super_block *sb, sector_t blk_off, char *data,
backup_block, backup_block -
ext4_group_first_block_no(sb, group));
BUFFER_TRACE(bh, "get_write_access");
- if ((err = ext4_journal_get_write_access(handle, bh)))
+ if ((err = ext4_journal_get_write_access(handle, bh))) {
+ brelse(bh);
break;
+ }
lock_buffer(bh);
memcpy(bh->b_data, data, size);
if (rest)
@@ -2023,7 +2023,7 @@ int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count)
err = ext4_alloc_flex_bg_array(sb, n_group + 1);
if (err)
- return err;
+ goto out;
err = ext4_mb_alloc_groupinfo(sb, n_group + 1);
if (err)
@@ -2059,6 +2059,10 @@ int ext4_resize_fs(struct super_block *sb, ext4_fsblk_t n_blocks_count)
n_blocks_count_retry = 0;
free_flex_gd(flex_gd);
flex_gd = NULL;
+ if (resize_inode) {
+ iput(resize_inode);
+ resize_inode = NULL;
+ }
goto retry;
}
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index a221f1c..53ff6c2 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -4075,6 +4075,14 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
sbi->s_groups_count = blocks_count;
sbi->s_blockfile_groups = min_t(ext4_group_t, sbi->s_groups_count,
(EXT4_MAX_BLOCK_FILE_PHYS / EXT4_BLOCKS_PER_GROUP(sb)));
+ if (((u64)sbi->s_groups_count * sbi->s_inodes_per_group) !=
+ le32_to_cpu(es->s_inodes_count)) {
+ ext4_msg(sb, KERN_ERR, "inodes count not valid: %u vs %llu",
+ le32_to_cpu(es->s_inodes_count),
+ ((u64)sbi->s_groups_count * sbi->s_inodes_per_group));
+ ret = -EINVAL;
+ goto failed_mount;
+ }
db_count = (sbi->s_groups_count + EXT4_DESC_PER_BLOCK(sb) - 1) /
EXT4_DESC_PER_BLOCK(sb);
if (ext4_has_feature_meta_bg(sb)) {
@@ -4094,14 +4102,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
ret = -ENOMEM;
goto failed_mount;
}
- if (((u64)sbi->s_groups_count * sbi->s_inodes_per_group) !=
- le32_to_cpu(es->s_inodes_count)) {
- ext4_msg(sb, KERN_ERR, "inodes count not valid: %u vs %llu",
- le32_to_cpu(es->s_inodes_count),
- ((u64)sbi->s_groups_count * sbi->s_inodes_per_group));
- ret = -EINVAL;
- goto failed_mount;
- }
bgl_lock_init(sbi->s_blockgroup_lock);
@@ -4510,6 +4510,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
percpu_counter_destroy(&sbi->s_freeinodes_counter);
percpu_counter_destroy(&sbi->s_dirs_counter);
percpu_counter_destroy(&sbi->s_dirtyclusters_counter);
+ percpu_free_rwsem(&sbi->s_journal_flag_rwsem);
failed_mount5:
ext4_ext_release(sb);
ext4_release_system_zone(sb);
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index f36fc5d5..7643d52 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -1031,10 +1031,8 @@ static int ext4_xattr_inode_update_ref(handle_t *handle, struct inode *ea_inode,
inode_lock(ea_inode);
ret = ext4_reserve_inode_write(handle, ea_inode, &iloc);
- if (ret) {
- iloc.bh = NULL;
+ if (ret)
goto out;
- }
ref_count = ext4_xattr_inode_get_ref(ea_inode);
ref_count += ref_change;
@@ -1080,12 +1078,10 @@ static int ext4_xattr_inode_update_ref(handle_t *handle, struct inode *ea_inode,
}
ret = ext4_mark_iloc_dirty(handle, ea_inode, &iloc);
- iloc.bh = NULL;
if (ret)
ext4_warning_inode(ea_inode,
"ext4_mark_iloc_dirty() failed ret=%d", ret);
out:
- brelse(iloc.bh);
inode_unlock(ea_inode);
return ret;
}
@@ -1388,6 +1384,12 @@ static int ext4_xattr_inode_write(handle_t *handle, struct inode *ea_inode,
bh = ext4_getblk(handle, ea_inode, block, 0);
if (IS_ERR(bh))
return PTR_ERR(bh);
+ if (!bh) {
+ WARN_ON_ONCE(1);
+ EXT4_ERROR_INODE(ea_inode,
+ "ext4_getblk() return bh = NULL");
+ return -EFSCORRUPTED;
+ }
ret = ext4_journal_get_write_access(handle, bh);
if (ret)
goto out;
@@ -2276,8 +2278,10 @@ static struct buffer_head *ext4_xattr_get_block(struct inode *inode)
if (!bh)
return ERR_PTR(-EIO);
error = ext4_xattr_check_block(inode, bh);
- if (error)
+ if (error) {
+ brelse(bh);
return ERR_PTR(error);
+ }
return bh;
}
@@ -2397,6 +2401,8 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index,
error = ext4_xattr_block_set(handle, inode, &i, &bs);
} else if (error == -ENOSPC) {
if (EXT4_I(inode)->i_file_acl && !bs.s.base) {
+ brelse(bs.bh);
+ bs.bh = NULL;
error = ext4_xattr_block_find(inode, &i, &bs);
if (error)
goto cleanup;
@@ -2617,6 +2623,8 @@ static int ext4_xattr_move_to_block(handle_t *handle, struct inode *inode,
kfree(buffer);
if (is)
brelse(is->iloc.bh);
+ if (bs)
+ brelse(bs->bh);
kfree(is);
kfree(bs);
@@ -2696,7 +2704,6 @@ int ext4_expand_extra_isize_ea(struct inode *inode, int new_extra_isize,
struct ext4_inode *raw_inode, handle_t *handle)
{
struct ext4_xattr_ibody_header *header;
- struct buffer_head *bh;
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
static unsigned int mnt_count;
size_t min_offs;
@@ -2737,13 +2744,17 @@ int ext4_expand_extra_isize_ea(struct inode *inode, int new_extra_isize,
* EA block can hold new_extra_isize bytes.
*/
if (EXT4_I(inode)->i_file_acl) {
+ struct buffer_head *bh;
+
bh = sb_bread(inode->i_sb, EXT4_I(inode)->i_file_acl);
error = -EIO;
if (!bh)
goto cleanup;
error = ext4_xattr_check_block(inode, bh);
- if (error)
+ if (error) {
+ brelse(bh);
goto cleanup;
+ }
base = BHDR(bh);
end = bh->b_data + bh->b_size;
min_offs = end - base;
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index ae813e6..a5e516a 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -165,9 +165,13 @@ static bool fuse_block_alloc(struct fuse_conn *fc, bool for_background)
static void fuse_drop_waiting(struct fuse_conn *fc)
{
- if (fc->connected) {
- atomic_dec(&fc->num_waiting);
- } else if (atomic_dec_and_test(&fc->num_waiting)) {
+ /*
+ * lockess check of fc->connected is okay, because atomic_dec_and_test()
+ * provides a memory barrier mached with the one in fuse_wait_aborted()
+ * to ensure no wake-up is missed.
+ */
+ if (atomic_dec_and_test(&fc->num_waiting) &&
+ !READ_ONCE(fc->connected)) {
/* wake up aborters */
wake_up_all(&fc->blocked_waitq);
}
@@ -1768,8 +1772,10 @@ static int fuse_retrieve(struct fuse_conn *fc, struct inode *inode,
req->in.args[1].size = total_len;
err = fuse_request_send_notify_reply(fc, req, outarg->notify_unique);
- if (err)
+ if (err) {
fuse_retrieve_end(fc, req);
+ fuse_put_request(fc, req);
+ }
return err;
}
@@ -2219,6 +2225,8 @@ EXPORT_SYMBOL_GPL(fuse_abort_conn);
void fuse_wait_aborted(struct fuse_conn *fc)
{
+ /* matches implicit memory barrier in fuse_drop_waiting() */
+ smp_mb();
wait_event(fc->blocked_waitq, atomic_read(&fc->num_waiting) == 0);
}
diff --git a/fs/fuse/file.c b/fs/fuse/file.c
index cc2121b..b52f9ba 100644
--- a/fs/fuse/file.c
+++ b/fs/fuse/file.c
@@ -2924,10 +2924,12 @@ fuse_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
}
if (io->async) {
+ bool blocking = io->blocking;
+
fuse_aio_complete(io, ret < 0 ? ret : 0, -1);
/* we have a non-extending, async request, so return */
- if (!io->blocking)
+ if (!blocking)
return -EIOCBQUEUED;
wait_for_completion(&wait);
diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c
index a683d9b..9a4a15d 100644
--- a/fs/gfs2/bmap.c
+++ b/fs/gfs2/bmap.c
@@ -826,7 +826,7 @@ static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length,
ret = gfs2_meta_inode_buffer(ip, &dibh);
if (ret)
goto unlock;
- iomap->private = dibh;
+ mp->mp_bh[0] = dibh;
if (gfs2_is_stuffed(ip)) {
if (flags & IOMAP_WRITE) {
@@ -863,9 +863,6 @@ static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length,
len = lblock_stop - lblock + 1;
iomap->length = len << inode->i_blkbits;
- get_bh(dibh);
- mp->mp_bh[0] = dibh;
-
height = ip->i_height;
while ((lblock + 1) * sdp->sd_sb.sb_bsize > sdp->sd_heightsize[height])
height++;
@@ -898,8 +895,6 @@ static int gfs2_iomap_get(struct inode *inode, loff_t pos, loff_t length,
iomap->bdev = inode->i_sb->s_bdev;
unlock:
up_read(&ip->i_rw_mutex);
- if (ret && dibh)
- brelse(dibh);
return ret;
do_alloc:
@@ -980,9 +975,9 @@ static void gfs2_iomap_journaled_page_done(struct inode *inode, loff_t pos,
static int gfs2_iomap_begin_write(struct inode *inode, loff_t pos,
loff_t length, unsigned flags,
- struct iomap *iomap)
+ struct iomap *iomap,
+ struct metapath *mp)
{
- struct metapath mp = { .mp_aheight = 1, };
struct gfs2_inode *ip = GFS2_I(inode);
struct gfs2_sbd *sdp = GFS2_SB(inode);
unsigned int data_blocks = 0, ind_blocks = 0, rblocks;
@@ -996,9 +991,9 @@ static int gfs2_iomap_begin_write(struct inode *inode, loff_t pos,
unstuff = gfs2_is_stuffed(ip) &&
pos + length > gfs2_max_stuffed_size(ip);
- ret = gfs2_iomap_get(inode, pos, length, flags, iomap, &mp);
+ ret = gfs2_iomap_get(inode, pos, length, flags, iomap, mp);
if (ret)
- goto out_release;
+ goto out_unlock;
alloc_required = unstuff || iomap->type == IOMAP_HOLE;
@@ -1013,7 +1008,7 @@ static int gfs2_iomap_begin_write(struct inode *inode, loff_t pos,
ret = gfs2_quota_lock_check(ip, &ap);
if (ret)
- goto out_release;
+ goto out_unlock;
ret = gfs2_inplace_reserve(ip, &ap);
if (ret)
@@ -1038,17 +1033,15 @@ static int gfs2_iomap_begin_write(struct inode *inode, loff_t pos,
ret = gfs2_unstuff_dinode(ip, NULL);
if (ret)
goto out_trans_end;
- release_metapath(&mp);
- brelse(iomap->private);
- iomap->private = NULL;
+ release_metapath(mp);
ret = gfs2_iomap_get(inode, iomap->offset, iomap->length,
- flags, iomap, &mp);
+ flags, iomap, mp);
if (ret)
goto out_trans_end;
}
if (iomap->type == IOMAP_HOLE) {
- ret = gfs2_iomap_alloc(inode, iomap, flags, &mp);
+ ret = gfs2_iomap_alloc(inode, iomap, flags, mp);
if (ret) {
gfs2_trans_end(sdp);
gfs2_inplace_release(ip);
@@ -1056,7 +1049,6 @@ static int gfs2_iomap_begin_write(struct inode *inode, loff_t pos,
goto out_qunlock;
}
}
- release_metapath(&mp);
if (!gfs2_is_stuffed(ip) && gfs2_is_jdata(ip))
iomap->page_done = gfs2_iomap_journaled_page_done;
return 0;
@@ -1069,10 +1061,7 @@ static int gfs2_iomap_begin_write(struct inode *inode, loff_t pos,
out_qunlock:
if (alloc_required)
gfs2_quota_unlock(ip);
-out_release:
- if (iomap->private)
- brelse(iomap->private);
- release_metapath(&mp);
+out_unlock:
gfs2_write_unlock(inode);
return ret;
}
@@ -1088,10 +1077,10 @@ static int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length,
trace_gfs2_iomap_start(ip, pos, length, flags);
if ((flags & IOMAP_WRITE) && !(flags & IOMAP_DIRECT)) {
- ret = gfs2_iomap_begin_write(inode, pos, length, flags, iomap);
+ ret = gfs2_iomap_begin_write(inode, pos, length, flags, iomap, &mp);
} else {
ret = gfs2_iomap_get(inode, pos, length, flags, iomap, &mp);
- release_metapath(&mp);
+
/*
* Silently fall back to buffered I/O for stuffed files or if
* we've hot a hole (see gfs2_file_direct_write).
@@ -1100,6 +1089,11 @@ static int gfs2_iomap_begin(struct inode *inode, loff_t pos, loff_t length,
iomap->type != IOMAP_MAPPED)
ret = -ENOTBLK;
}
+ if (!ret) {
+ get_bh(mp.mp_bh[0]);
+ iomap->private = mp.mp_bh[0];
+ }
+ release_metapath(&mp);
trace_gfs2_iomap_end(ip, iomap, ret);
return ret;
}
@@ -1908,10 +1902,16 @@ static int punch_hole(struct gfs2_inode *ip, u64 offset, u64 length)
if (ret < 0)
goto out;
- /* issue read-ahead on metadata */
- if (mp.mp_aheight > 1) {
- for (; ret > 1; ret--) {
- metapointer_range(&mp, mp.mp_aheight - ret,
+ /* On the first pass, issue read-ahead on metadata. */
+ if (mp.mp_aheight > 1 && strip_h == ip->i_height - 1) {
+ unsigned int height = mp.mp_aheight - 1;
+
+ /* No read-ahead for data blocks. */
+ if (mp.mp_aheight - 1 == strip_h)
+ height--;
+
+ for (; height >= mp.mp_aheight - ret; height--) {
+ metapointer_range(&mp, height,
start_list, start_aligned,
end_list, end_aligned,
&start, &end);
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index ffe3032..b08a530 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -733,6 +733,7 @@ void gfs2_clear_rgrpd(struct gfs2_sbd *sdp)
if (gl) {
glock_clear_object(gl, rgd);
+ gfs2_rgrp_brelse(rgd);
gfs2_glock_put(gl);
}
@@ -1174,7 +1175,7 @@ static u32 count_unlinked(struct gfs2_rgrpd *rgd)
* @rgd: the struct gfs2_rgrpd describing the RG to read in
*
* Read in all of a Resource Group's header and bitmap blocks.
- * Caller must eventually call gfs2_rgrp_relse() to free the bitmaps.
+ * Caller must eventually call gfs2_rgrp_brelse() to free the bitmaps.
*
* Returns: errno
*/
diff --git a/fs/inode.c b/fs/inode.c
index 9e198f0..35d2108 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -730,8 +730,11 @@ static enum lru_status inode_lru_isolate(struct list_head *item,
return LRU_REMOVED;
}
- /* recently referenced inodes get one more pass */
- if (inode->i_state & I_REFERENCED) {
+ /*
+ * Recently referenced inodes and inodes with many attached pages
+ * get one more pass.
+ */
+ if (inode->i_state & I_REFERENCED || inode->i_data.nrpages > 1) {
inode->i_state &= ~I_REFERENCED;
spin_unlock(&inode->i_lock);
return LRU_ROTATE;
diff --git a/fs/namespace.c b/fs/namespace.c
index 98d27da..a7f9126 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -695,9 +695,6 @@ static struct mountpoint *lookup_mountpoint(struct dentry *dentry)
hlist_for_each_entry(mp, chain, m_hash) {
if (mp->m_dentry == dentry) {
- /* might be worth a WARN_ON() */
- if (d_unlinked(dentry))
- return ERR_PTR(-ENOENT);
mp->m_count++;
return mp;
}
@@ -711,6 +708,9 @@ static struct mountpoint *get_mountpoint(struct dentry *dentry)
int ret;
if (d_mountpoint(dentry)) {
+ /* might be worth a WARN_ON() */
+ if (d_unlinked(dentry))
+ return ERR_PTR(-ENOENT);
mountpoint:
read_seqlock_excl(&mount_lock);
mp = lookup_mountpoint(dentry);
@@ -1540,8 +1540,13 @@ static int do_umount(struct mount *mnt, int flags)
namespace_lock();
lock_mount_hash();
- event++;
+ /* Recheck MNT_LOCKED with the locks held */
+ retval = -EINVAL;
+ if (mnt->mnt.mnt_flags & MNT_LOCKED)
+ goto out;
+
+ event++;
if (flags & MNT_DETACH) {
if (!list_empty(&mnt->mnt_list))
umount_tree(mnt, UMOUNT_PROPAGATE);
@@ -1555,6 +1560,7 @@ static int do_umount(struct mount *mnt, int flags)
retval = 0;
}
}
+out:
unlock_mount_hash();
namespace_unlock();
return retval;
@@ -1645,7 +1651,7 @@ int ksys_umount(char __user *name, int flags)
goto dput_and_out;
if (!check_mnt(mnt))
goto dput_and_out;
- if (mnt->mnt.mnt_flags & MNT_LOCKED)
+ if (mnt->mnt.mnt_flags & MNT_LOCKED) /* Check optimistically */
goto dput_and_out;
retval = -EPERM;
if (flags & MNT_FORCE && !capable(CAP_SYS_ADMIN))
@@ -1728,8 +1734,14 @@ struct mount *copy_tree(struct mount *mnt, struct dentry *dentry,
for (s = r; s; s = next_mnt(s, r)) {
if (!(flag & CL_COPY_UNBINDABLE) &&
IS_MNT_UNBINDABLE(s)) {
- s = skip_mnt_tree(s);
- continue;
+ if (s->mnt.mnt_flags & MNT_LOCKED) {
+ /* Both unbindable and locked. */
+ q = ERR_PTR(-EPERM);
+ goto out;
+ } else {
+ s = skip_mnt_tree(s);
+ continue;
+ }
}
if (!(flag & CL_COPY_MNT_NS_FILE) &&
is_mnt_ns_file(s->mnt.mnt_root)) {
@@ -1782,7 +1794,7 @@ void drop_collected_mounts(struct vfsmount *mnt)
{
namespace_lock();
lock_mount_hash();
- umount_tree(real_mount(mnt), UMOUNT_SYNC);
+ umount_tree(real_mount(mnt), 0);
unlock_mount_hash();
namespace_unlock();
}
diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c
index fa515d5..7b861bb 100644
--- a/fs/nfs/callback_proc.c
+++ b/fs/nfs/callback_proc.c
@@ -66,7 +66,7 @@ __be32 nfs4_callback_getattr(void *argp, void *resp,
out_iput:
rcu_read_unlock();
trace_nfs4_cb_getattr(cps->clp, &args->fh, inode, -ntohl(res->status));
- iput(inode);
+ nfs_iput_and_deactive(inode);
out:
dprintk("%s: exit with status = %d\n", __func__, ntohl(res->status));
return res->status;
@@ -108,7 +108,7 @@ __be32 nfs4_callback_recall(void *argp, void *resp,
}
trace_nfs4_cb_recall(cps->clp, &args->fh, inode,
&args->stateid, -ntohl(res));
- iput(inode);
+ nfs_iput_and_deactive(inode);
out:
dprintk("%s: exit with status = %d\n", __func__, ntohl(res));
return res;
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 07b8395..6ec2f78 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -850,16 +850,23 @@ nfs_delegation_find_inode_server(struct nfs_server *server,
const struct nfs_fh *fhandle)
{
struct nfs_delegation *delegation;
- struct inode *res = NULL;
+ struct inode *freeme, *res = NULL;
list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
spin_lock(&delegation->lock);
if (delegation->inode != NULL &&
nfs_compare_fh(fhandle, &NFS_I(delegation->inode)->fh) == 0) {
- res = igrab(delegation->inode);
+ freeme = igrab(delegation->inode);
+ if (freeme && nfs_sb_active(freeme->i_sb))
+ res = freeme;
spin_unlock(&delegation->lock);
if (res != NULL)
return res;
+ if (freeme) {
+ rcu_read_unlock();
+ iput(freeme);
+ rcu_read_lock();
+ }
return ERR_PTR(-EAGAIN);
}
spin_unlock(&delegation->lock);
diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
index 62ae0fd..ffea578 100644
--- a/fs/nfs/nfs4state.c
+++ b/fs/nfs/nfs4state.c
@@ -2601,11 +2601,12 @@ static void nfs4_state_manager(struct nfs_client *clp)
nfs4_clear_state_manager_bit(clp);
/* Did we race with an attempt to give us more work? */
if (clp->cl_state == 0)
- break;
+ return;
if (test_and_set_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state) != 0)
- break;
- } while (refcount_read(&clp->cl_count) > 1);
- return;
+ return;
+ } while (refcount_read(&clp->cl_count) > 1 && !signalled());
+ goto out_drain;
+
out_error:
if (strlen(section))
section_sep = ": ";
@@ -2613,6 +2614,7 @@ static void nfs4_state_manager(struct nfs_client *clp)
" with error %d\n", section_sep, section,
clp->cl_hostname, -status);
ssleep(1);
+out_drain:
nfs4_end_drain_session(clp);
nfs4_clear_state_manager_bit(clp);
}
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index edff074..d505990 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -1038,6 +1038,9 @@ nfsd4_verify_copy(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate,
{
__be32 status;
+ if (!cstate->save_fh.fh_dentry)
+ return nfserr_nofilehandle;
+
status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->save_fh,
src_stateid, RD_STATE, src, NULL);
if (status) {
diff --git a/fs/notify/fanotify/fanotify.c b/fs/notify/fanotify/fanotify.c
index 5769cf3..e08a664 100644
--- a/fs/notify/fanotify/fanotify.c
+++ b/fs/notify/fanotify/fanotify.c
@@ -115,12 +115,12 @@ static bool fanotify_should_send_event(struct fsnotify_iter_info *iter_info,
continue;
mark = iter_info->marks[type];
/*
- * if the event is for a child and this inode doesn't care about
- * events on the child, don't send it!
+ * If the event is for a child and this mark doesn't care about
+ * events on a child, don't send it!
*/
- if (type == FSNOTIFY_OBJ_TYPE_INODE &&
- (event_mask & FS_EVENT_ON_CHILD) &&
- !(mark->mask & FS_EVENT_ON_CHILD))
+ if (event_mask & FS_EVENT_ON_CHILD &&
+ (type != FSNOTIFY_OBJ_TYPE_INODE ||
+ !(mark->mask & FS_EVENT_ON_CHILD)))
continue;
marks_mask |= mark->mask;
diff --git a/fs/notify/fsnotify.c b/fs/notify/fsnotify.c
index 2172ba5..d2c3490 100644
--- a/fs/notify/fsnotify.c
+++ b/fs/notify/fsnotify.c
@@ -167,9 +167,9 @@ int __fsnotify_parent(const struct path *path, struct dentry *dentry, __u32 mask
parent = dget_parent(dentry);
p_inode = parent->d_inode;
- if (unlikely(!fsnotify_inode_watches_children(p_inode)))
+ if (unlikely(!fsnotify_inode_watches_children(p_inode))) {
__fsnotify_update_child_dentry_flags(p_inode);
- else if (p_inode->i_fsnotify_mask & mask) {
+ } else if (p_inode->i_fsnotify_mask & mask & ALL_FSNOTIFY_EVENTS) {
struct name_snapshot name;
/* we are notifying a parent so come up with the new mask which
@@ -339,6 +339,9 @@ int fsnotify(struct inode *to_tell, __u32 mask, const void *data, int data_is,
sb = mnt->mnt.mnt_sb;
mnt_or_sb_mask = mnt->mnt_fsnotify_mask | sb->s_fsnotify_mask;
}
+ /* An event "on child" is not intended for a mount/sb mark */
+ if (mask & FS_EVENT_ON_CHILD)
+ mnt_or_sb_mask = 0;
/*
* Optimization: srcu_read_lock() has a memory barrier which can
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index da578ad..eb1ce30 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -2411,8 +2411,16 @@ static int ocfs2_dio_end_io(struct kiocb *iocb,
/* this io's submitter should not have unlocked this before we could */
BUG_ON(!ocfs2_iocb_is_rw_locked(iocb));
- if (bytes > 0 && private)
- ret = ocfs2_dio_end_io_write(inode, private, offset, bytes);
+ if (bytes <= 0)
+ mlog_ratelimited(ML_ERROR, "Direct IO failed, bytes = %lld",
+ (long long)bytes);
+ if (private) {
+ if (bytes > 0)
+ ret = ocfs2_dio_end_io_write(inode, private, offset,
+ bytes);
+ else
+ ocfs2_dio_free_write_ctx(inode, private);
+ }
ocfs2_iocb_clear_rw_locked(iocb);
diff --git a/fs/ocfs2/cluster/masklog.h b/fs/ocfs2/cluster/masklog.h
index 308ea0e..a396096 100644
--- a/fs/ocfs2/cluster/masklog.h
+++ b/fs/ocfs2/cluster/masklog.h
@@ -178,6 +178,15 @@ do { \
##__VA_ARGS__); \
} while (0)
+#define mlog_ratelimited(mask, fmt, ...) \
+do { \
+ static DEFINE_RATELIMIT_STATE(_rs, \
+ DEFAULT_RATELIMIT_INTERVAL, \
+ DEFAULT_RATELIMIT_BURST); \
+ if (__ratelimit(&_rs)) \
+ mlog(mask, fmt, ##__VA_ARGS__); \
+} while (0)
+
#define mlog_errno(st) ({ \
int _st = (st); \
if (_st != -ERESTARTSYS && _st != -EINTR && \
diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c
index 6fc5425..2652d00 100644
--- a/fs/xfs/libxfs/xfs_attr_leaf.c
+++ b/fs/xfs/libxfs/xfs_attr_leaf.c
@@ -243,7 +243,7 @@ xfs_attr3_leaf_verify(
struct xfs_mount *mp = bp->b_target->bt_mount;
struct xfs_attr_leafblock *leaf = bp->b_addr;
struct xfs_attr_leaf_entry *entries;
- uint16_t end;
+ uint32_t end; /* must be 32bit - see below */
int i;
xfs_attr3_leaf_hdr_from_disk(mp->m_attr_geo, &ichdr, leaf);
@@ -293,6 +293,11 @@ xfs_attr3_leaf_verify(
/*
* Quickly check the freemap information. Attribute data has to be
* aligned to 4-byte boundaries, and likewise for the free space.
+ *
+ * Note that for 64k block size filesystems, the freemap entries cannot
+ * overflow as they are only be16 fields. However, when checking end
+ * pointer of the freemap, we have to be careful to detect overflows and
+ * so use uint32_t for those checks.
*/
for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; i++) {
if (ichdr.freemap[i].base > mp->m_attr_geo->blksize)
@@ -303,7 +308,9 @@ xfs_attr3_leaf_verify(
return __this_address;
if (ichdr.freemap[i].size & 0x3)
return __this_address;
- end = ichdr.freemap[i].base + ichdr.freemap[i].size;
+
+ /* be care of 16 bit overflows here */
+ end = (uint32_t)ichdr.freemap[i].base + ichdr.freemap[i].size;
if (end < ichdr.freemap[i].base)
return __this_address;
if (end > mp->m_attr_geo->blksize)
diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c
index 6e2c08f..6ecdbb3 100644
--- a/fs/xfs/xfs_ioctl.c
+++ b/fs/xfs/xfs_ioctl.c
@@ -1608,7 +1608,7 @@ xfs_ioc_getbmap(
error = 0;
out_free_buf:
kmem_free(buf);
- return 0;
+ return error;
}
struct getfsmap_info {
diff --git a/fs/xfs/xfs_message.c b/fs/xfs/xfs_message.c
index 576c375..6b736ea 100644
--- a/fs/xfs/xfs_message.c
+++ b/fs/xfs/xfs_message.c
@@ -107,5 +107,5 @@ assfail(char *expr, char *file, int line)
void
xfs_hex_dump(void *p, int length)
{
- print_hex_dump(KERN_ALERT, "", DUMP_PREFIX_ADDRESS, 16, 1, p, length, 1);
+ print_hex_dump(KERN_ALERT, "", DUMP_PREFIX_OFFSET, 16, 1, p, length, 1);
}
diff --git a/include/asm-generic/4level-fixup.h b/include/asm-generic/4level-fixup.h
index 89f3b03..e3667c9 100644
--- a/include/asm-generic/4level-fixup.h
+++ b/include/asm-generic/4level-fixup.h
@@ -3,7 +3,7 @@
#define _4LEVEL_FIXUP_H
#define __ARCH_HAS_4LEVEL_HACK
-#define __PAGETABLE_PUD_FOLDED
+#define __PAGETABLE_PUD_FOLDED 1
#define PUD_SHIFT PGDIR_SHIFT
#define PUD_SIZE PGDIR_SIZE
diff --git a/include/asm-generic/5level-fixup.h b/include/asm-generic/5level-fixup.h
index 9c2e070..73474bb 100644
--- a/include/asm-generic/5level-fixup.h
+++ b/include/asm-generic/5level-fixup.h
@@ -3,7 +3,7 @@
#define _5LEVEL_FIXUP_H
#define __ARCH_HAS_5LEVEL_HACK
-#define __PAGETABLE_P4D_FOLDED
+#define __PAGETABLE_P4D_FOLDED 1
#define P4D_SHIFT PGDIR_SHIFT
#define P4D_SIZE PGDIR_SIZE
diff --git a/include/asm-generic/pgtable-nop4d-hack.h b/include/asm-generic/pgtable-nop4d-hack.h
index 0c34215..1d6dd38 100644
--- a/include/asm-generic/pgtable-nop4d-hack.h
+++ b/include/asm-generic/pgtable-nop4d-hack.h
@@ -5,7 +5,7 @@
#ifndef __ASSEMBLY__
#include <asm-generic/5level-fixup.h>
-#define __PAGETABLE_PUD_FOLDED
+#define __PAGETABLE_PUD_FOLDED 1
/*
* Having the pud type consist of a pgd gets the size right, and allows
diff --git a/include/asm-generic/pgtable-nop4d.h b/include/asm-generic/pgtable-nop4d.h
index 1a29b2a..04cb913 100644
--- a/include/asm-generic/pgtable-nop4d.h
+++ b/include/asm-generic/pgtable-nop4d.h
@@ -4,7 +4,7 @@
#ifndef __ASSEMBLY__
-#define __PAGETABLE_P4D_FOLDED
+#define __PAGETABLE_P4D_FOLDED 1
typedef struct { pgd_t pgd; } p4d_t;
diff --git a/include/asm-generic/pgtable-nopmd.h b/include/asm-generic/pgtable-nopmd.h
index f35f6e8..b85b827 100644
--- a/include/asm-generic/pgtable-nopmd.h
+++ b/include/asm-generic/pgtable-nopmd.h
@@ -8,7 +8,7 @@
struct mm_struct;
-#define __PAGETABLE_PMD_FOLDED
+#define __PAGETABLE_PMD_FOLDED 1
/*
* Having the pmd type consist of a pud gets the size right, and allows
diff --git a/include/asm-generic/pgtable-nopud.h b/include/asm-generic/pgtable-nopud.h
index e950b9c..9bef475 100644
--- a/include/asm-generic/pgtable-nopud.h
+++ b/include/asm-generic/pgtable-nopud.h
@@ -9,7 +9,7 @@
#else
#include <asm-generic/pgtable-nop4d.h>
-#define __PAGETABLE_PUD_FOLDED
+#define __PAGETABLE_PUD_FOLDED 1
/*
* Having the pud type consist of a p4d gets the size right, and allows
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 5657a20..359fb93 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -1127,4 +1127,20 @@ static inline bool arch_has_pfn_modify_check(void)
#endif
#endif
+/*
+ * On some architectures it depends on the mm if the p4d/pud or pmd
+ * layer of the page table hierarchy is folded or not.
+ */
+#ifndef mm_p4d_folded
+#define mm_p4d_folded(mm) __is_defined(__PAGETABLE_P4D_FOLDED)
+#endif
+
+#ifndef mm_pud_folded
+#define mm_pud_folded(mm) __is_defined(__PAGETABLE_PUD_FOLDED)
+#endif
+
+#ifndef mm_pmd_folded
+#define mm_pmd_folded(mm) __is_defined(__PAGETABLE_PMD_FOLDED)
+#endif
+
#endif /* _ASM_GENERIC_PGTABLE_H */
diff --git a/include/drm/drmP.h b/include/drm/drmP.h
index 514beb2..bdb0d55 100644
--- a/include/drm/drmP.h
+++ b/include/drm/drmP.h
@@ -68,7 +68,6 @@
#include <drm/drm_agpsupport.h>
#include <drm/drm_crtc.h>
#include <drm/drm_fourcc.h>
-#include <drm/drm_global.h>
#include <drm/drm_hashtab.h>
#include <drm/drm_mm.h>
#include <drm/drm_os_linux.h>
diff --git a/include/drm/drm_global.h b/include/drm/drm_global.h
deleted file mode 100644
index 3a83060..0000000
--- a/include/drm/drm_global.h
+++ /dev/null
@@ -1,53 +0,0 @@
-/**************************************************************************
- *
- * Copyright 2008-2009 VMware, Inc., Palo Alto, CA., USA
- * All Rights Reserved.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the
- * "Software"), to deal in the Software without restriction, including
- * without limitation the rights to use, copy, modify, merge, publish,
- * distribute, sub license, and/or sell copies of the Software, and to
- * permit persons to whom the Software is furnished to do so, subject to
- * the following conditions:
- *
- * The above copyright notice and this permission notice (including the
- * next paragraph) shall be included in all copies or substantial portions
- * of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. IN NO EVENT SHALL
- * THE COPYRIGHT HOLDERS, AUTHORS AND/OR ITS SUPPLIERS BE LIABLE FOR ANY CLAIM,
- * DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
- * OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE
- * USE OR OTHER DEALINGS IN THE SOFTWARE.
- *
- **************************************************************************/
-/*
- * Authors: Thomas Hellstrom <thellstrom-at-vmware-dot-com>
- */
-
-#ifndef _DRM_GLOBAL_H_
-#define _DRM_GLOBAL_H_
-enum drm_global_types {
- DRM_GLOBAL_TTM_MEM = 0,
- DRM_GLOBAL_TTM_BO,
- DRM_GLOBAL_TTM_OBJECT,
- DRM_GLOBAL_NUM
-};
-
-struct drm_global_reference {
- enum drm_global_types global_type;
- size_t size;
- void *object;
- int (*init) (struct drm_global_reference *);
- void (*release) (struct drm_global_reference *);
-};
-
-void drm_global_init(void);
-void drm_global_release(void);
-int drm_global_item_ref(struct drm_global_reference *ref);
-void drm_global_item_unref(struct drm_global_reference *ref);
-
-#endif
diff --git a/include/drm/gpu_scheduler.h b/include/drm/gpu_scheduler.h
index d87b268..926379d 100644
--- a/include/drm/gpu_scheduler.h
+++ b/include/drm/gpu_scheduler.h
@@ -264,6 +264,7 @@ struct drm_sched_backend_ops {
* @hang_limit: once the hangs by a job crosses this limit then it is marked
* guilty and it will be considered for scheduling further.
* @num_jobs: the number of jobs in queue in the scheduler
+ * @ready: marks if the underlying HW is ready to work
*
* One scheduler is implemented for each hardware ring.
*/
@@ -283,22 +284,26 @@ struct drm_gpu_scheduler {
spinlock_t job_list_lock;
int hang_limit;
atomic_t num_jobs;
+ bool ready;
};
int drm_sched_init(struct drm_gpu_scheduler *sched,
const struct drm_sched_backend_ops *ops,
uint32_t hw_submission, unsigned hang_limit, long timeout,
const char *name);
+
void drm_sched_fini(struct drm_gpu_scheduler *sched);
int drm_sched_job_init(struct drm_sched_job *job,
struct drm_sched_entity *entity,
void *owner);
+void drm_sched_job_cleanup(struct drm_sched_job *job);
void drm_sched_wakeup(struct drm_gpu_scheduler *sched);
void drm_sched_hw_job_reset(struct drm_gpu_scheduler *sched,
struct drm_sched_job *job);
void drm_sched_job_recovery(struct drm_gpu_scheduler *sched);
bool drm_sched_dependency_optimized(struct dma_fence* fence,
struct drm_sched_entity *entity);
+void drm_sched_fault(struct drm_gpu_scheduler *sched);
void drm_sched_job_kickout(struct drm_sched_job *s_job);
void drm_sched_rq_add_entity(struct drm_sched_rq *rq,
diff --git a/include/drm/ttm/ttm_bo_driver.h b/include/drm/ttm/ttm_bo_driver.h
index e4fee8e..1021106 100644
--- a/include/drm/ttm/ttm_bo_driver.h
+++ b/include/drm/ttm/ttm_bo_driver.h
@@ -31,7 +31,6 @@
#define _TTM_BO_DRIVER_H_
#include <drm/drm_mm.h>
-#include <drm/drm_global.h>
#include <drm/drm_vma_manager.h>
#include <linux/workqueue.h>
#include <linux/fs.h>
@@ -385,15 +384,6 @@ struct ttm_bo_driver {
};
/**
- * struct ttm_bo_global_ref - Argument to initialize a struct ttm_bo_global.
- */
-
-struct ttm_bo_global_ref {
- struct drm_global_reference ref;
- struct ttm_mem_global *mem_glob;
-};
-
-/**
* struct ttm_bo_global - Buffer object driver global data.
*
* @mem_glob: Pointer to a struct ttm_mem_global object for accounting.
@@ -407,7 +397,7 @@ struct ttm_bo_global_ref {
* @swap_lru: Lru list of buffer objects used for swapping.
*/
-struct ttm_bo_global {
+extern struct ttm_bo_global {
/**
* Constant after init.
@@ -416,12 +406,12 @@ struct ttm_bo_global {
struct kobject kobj;
struct ttm_mem_global *mem_glob;
struct page *dummy_read_page;
- struct mutex device_list_mutex;
spinlock_t lru_lock;
/**
- * Protected by device_list_mutex.
+ * Protected by ttm_global_mutex.
*/
+ unsigned int use_count;
struct list_head device_list;
/**
@@ -433,7 +423,7 @@ struct ttm_bo_global {
* Internal protection.
*/
atomic_t bo_count;
-};
+} ttm_bo_glob;
#define TTM_NUM_MEM_TYPES 8
@@ -578,9 +568,6 @@ void ttm_bo_mem_put(struct ttm_buffer_object *bo, struct ttm_mem_reg *mem);
void ttm_bo_mem_put_locked(struct ttm_buffer_object *bo,
struct ttm_mem_reg *mem);
-void ttm_bo_global_release(struct drm_global_reference *ref);
-int ttm_bo_global_init(struct drm_global_reference *ref);
-
int ttm_bo_device_release(struct ttm_bo_device *bdev);
/**
@@ -598,7 +585,7 @@ int ttm_bo_device_release(struct ttm_bo_device *bdev);
* Returns:
* !0: Failure.
*/
-int ttm_bo_device_init(struct ttm_bo_device *bdev, struct ttm_bo_global *glob,
+int ttm_bo_device_init(struct ttm_bo_device *bdev,
struct ttm_bo_driver *driver,
struct address_space *mapping,
uint64_t file_page_offset, bool need_dma32);
diff --git a/include/drm/ttm/ttm_memory.h b/include/drm/ttm/ttm_memory.h
index 737b5fe..3ff48a0 100644
--- a/include/drm/ttm/ttm_memory.h
+++ b/include/drm/ttm/ttm_memory.h
@@ -63,7 +63,7 @@
#define TTM_MEM_MAX_ZONES 2
struct ttm_mem_zone;
-struct ttm_mem_global {
+extern struct ttm_mem_global {
struct kobject kobj;
struct ttm_bo_global *bo_glob;
struct workqueue_struct *swap_queue;
@@ -78,7 +78,7 @@ struct ttm_mem_global {
#else
struct ttm_mem_zone *zone_dma32;
#endif
-};
+} ttm_mem_glob;
extern int ttm_mem_global_init(struct ttm_mem_global *glob);
extern void ttm_mem_global_release(struct ttm_mem_global *glob);
diff --git a/include/linux/ceph/ceph_features.h b/include/linux/ceph/ceph_features.h
index 6b92b33..65a38c4 100644
--- a/include/linux/ceph/ceph_features.h
+++ b/include/linux/ceph/ceph_features.h
@@ -213,12 +213,6 @@ DEFINE_CEPH_FEATURE_DEPRECATED(63, 1, RESERVED_BROKEN, LUMINOUS) // client-facin
CEPH_FEATURE_NEW_OSDOPREPLY_ENCODING | \
CEPH_FEATURE_CEPHX_V2)
-#define CEPH_FEATURES_REQUIRED_DEFAULT \
- (CEPH_FEATURE_NOSRCADDR | \
- CEPH_FEATURE_SUBSCRIBE2 | \
- CEPH_FEATURE_RECONNECT_SEQ | \
- CEPH_FEATURE_PGID64 | \
- CEPH_FEATURE_PGPOOL3 | \
- CEPH_FEATURE_OSDENC)
+#define CEPH_FEATURES_REQUIRED_DEFAULT 0
#endif
diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h
index c0f5db3..2010493 100644
--- a/include/linux/compiler-gcc.h
+++ b/include/linux/compiler-gcc.h
@@ -143,18 +143,6 @@
#define KASAN_ABI_VERSION 3
#endif
-/*
- * Because __no_sanitize_address conflicts with inlining:
- * https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67368
- * we do one or the other.
- */
-#ifdef CONFIG_KASAN
-#define __no_sanitize_address_or_inline \
- __no_sanitize_address __maybe_unused notrace
-#else
-#define __no_sanitize_address_or_inline inline
-#endif
-
#if GCC_VERSION >= 50100
#define COMPILER_HAS_GENERIC_BUILTIN_OVERFLOW 1
#endif
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 18c80cf..06396c1 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -189,7 +189,7 @@ void __read_once_size(const volatile void *p, void *res, int size)
* https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67368
* '__maybe_unused' allows us to avoid defined-but-not-used warnings.
*/
-# define __no_kasan_or_inline __no_sanitize_address __maybe_unused
+# define __no_kasan_or_inline __no_sanitize_address notrace __maybe_unused
#else
# define __no_kasan_or_inline __always_inline
#endif
diff --git a/include/linux/compiler_attributes.h b/include/linux/compiler_attributes.h
index 6b28c1b..f8c400b 100644
--- a/include/linux/compiler_attributes.h
+++ b/include/linux/compiler_attributes.h
@@ -4,22 +4,26 @@
/*
* The attributes in this file are unconditionally defined and they directly
- * map to compiler attribute(s) -- except those that are optional.
+ * map to compiler attribute(s), unless one of the compilers does not support
+ * the attribute. In that case, __has_attribute is used to check for support
+ * and the reason is stated in its comment ("Optional: ...").
*
* Any other "attributes" (i.e. those that depend on a configuration option,
* on a compiler, on an architecture, on plugins, on other attributes...)
* should be defined elsewhere (e.g. compiler_types.h or compiler-*.h).
+ * The intention is to keep this file as simple as possible, as well as
+ * compiler- and version-agnostic (e.g. avoiding GCC_VERSION checks).
*
* This file is meant to be sorted (by actual attribute name,
* not by #define identifier). Use the __attribute__((__name__)) syntax
* (i.e. with underscores) to avoid future collisions with other macros.
- * If an attribute is optional, state the reason in the comment.
+ * Provide links to the documentation of each supported compiler, if it exists.
*/
/*
- * To check for optional attributes, we use __has_attribute, which is supported
- * on gcc >= 5, clang >= 2.9 and icc >= 17. In the meantime, to support
- * 4.6 <= gcc < 5, we implement __has_attribute by hand.
+ * __has_attribute is supported on gcc >= 5, clang >= 2.9 and icc >= 17.
+ * In the meantime, to support 4.6 <= gcc < 5, we implement __has_attribute
+ * by hand.
*
* sparse does not support __has_attribute (yet) and defines __GNUC_MINOR__
* depending on the compiler used to build it; however, these attributes have
diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h
index 3439d7d..4a3f9c0 100644
--- a/include/linux/compiler_types.h
+++ b/include/linux/compiler_types.h
@@ -130,6 +130,10 @@ struct ftrace_likely_data {
# define randomized_struct_fields_end
#endif
+#ifndef asm_volatile_goto
+#define asm_volatile_goto(x...) asm goto(x)
+#endif
+
/* Are two types/vars the same type (ignoring qualifiers)? */
#define __same_type(a, b) __builtin_types_compatible_p(typeof(a), typeof(b))
diff --git a/include/linux/efi.h b/include/linux/efi.h
index 845174e..100ce4a 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -1167,6 +1167,8 @@ static inline bool efi_enabled(int feature)
extern void efi_reboot(enum reboot_mode reboot_mode, const char *__unused);
extern bool efi_is_table_address(unsigned long phys_addr);
+
+extern int efi_apply_persistent_mem_reservations(void);
#else
static inline bool efi_enabled(int feature)
{
@@ -1185,6 +1187,11 @@ static inline bool efi_is_table_address(unsigned long phys_addr)
{
return false;
}
+
+static inline int efi_apply_persistent_mem_reservations(void)
+{
+ return 0;
+}
#endif
extern int efi_status_to_err(efi_status_t status);
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 2827b87..387c70d 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -722,8 +722,8 @@ struct hid_usage_id {
* input will not be passed to raw_event unless hid_device_io_start is
* called.
*
- * raw_event and event should return 0 on no action performed, 1 when no
- * further processing should be done and negative on error
+ * raw_event and event should return negative on error, any other value will
+ * pass the event on to .event() typically return 0 for success.
*
* input_mapping shall return a negative value to completely ignore this usage
* (e.g. doubled or invalid usage), zero to continue with parsing of this
diff --git a/include/linux/i8253.h b/include/linux/i8253.h
index e6bb36a..8336b2f 100644
--- a/include/linux/i8253.h
+++ b/include/linux/i8253.h
@@ -21,6 +21,7 @@
#define PIT_LATCH ((PIT_TICK_RATE + HZ/2) / HZ)
extern raw_spinlock_t i8253_lock;
+extern bool i8253_clear_counter_on_shutdown;
extern struct clock_event_device i8253_clockevent;
extern void clockevent_i8253_init(bool oneshot);
diff --git a/include/linux/mm.h b/include/linux/mm.h
index fcf9cc9..5411de9 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -1744,11 +1744,15 @@ int __pud_alloc(struct mm_struct *mm, p4d_t *p4d, unsigned long address);
static inline void mm_inc_nr_puds(struct mm_struct *mm)
{
+ if (mm_pud_folded(mm))
+ return;
atomic_long_add(PTRS_PER_PUD * sizeof(pud_t), &mm->pgtables_bytes);
}
static inline void mm_dec_nr_puds(struct mm_struct *mm)
{
+ if (mm_pud_folded(mm))
+ return;
atomic_long_sub(PTRS_PER_PUD * sizeof(pud_t), &mm->pgtables_bytes);
}
#endif
@@ -1768,11 +1772,15 @@ int __pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long address);
static inline void mm_inc_nr_pmds(struct mm_struct *mm)
{
+ if (mm_pmd_folded(mm))
+ return;
atomic_long_add(PTRS_PER_PMD * sizeof(pmd_t), &mm->pgtables_bytes);
}
static inline void mm_dec_nr_pmds(struct mm_struct *mm)
{
+ if (mm_pmd_folded(mm))
+ return;
atomic_long_sub(PTRS_PER_PMD * sizeof(pmd_t), &mm->pgtables_bytes);
}
#endif
diff --git a/include/linux/mtd/nand.h b/include/linux/mtd/nand.h
index abe975c..7f53ece 100644
--- a/include/linux/mtd/nand.h
+++ b/include/linux/mtd/nand.h
@@ -324,9 +324,8 @@ static inline unsigned int nanddev_ntargets(const struct nand_device *nand)
*/
static inline unsigned int nanddev_neraseblocks(const struct nand_device *nand)
{
- return (u64)nand->memorg.luns_per_target *
- nand->memorg.eraseblocks_per_lun *
- nand->memorg.pages_per_eraseblock;
+ return nand->memorg.ntargets * nand->memorg.luns_per_target *
+ nand->memorg.eraseblocks_per_lun;
}
/**
@@ -569,7 +568,7 @@ static inline void nanddev_pos_next_eraseblock(struct nand_device *nand,
}
/**
- * nanddev_pos_next_eraseblock() - Move a position to the next page
+ * nanddev_pos_next_page() - Move a position to the next page
* @nand: NAND device
* @pos: the position to update
*
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index dc1d9ed..857f8ab 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3190,6 +3190,26 @@ static inline void netdev_tx_sent_queue(struct netdev_queue *dev_queue,
#endif
}
+/* Variant of netdev_tx_sent_queue() for drivers that are aware
+ * that they should not test BQL status themselves.
+ * We do want to change __QUEUE_STATE_STACK_XOFF only for the last
+ * skb of a batch.
+ * Returns true if the doorbell must be used to kick the NIC.
+ */
+static inline bool __netdev_tx_sent_queue(struct netdev_queue *dev_queue,
+ unsigned int bytes,
+ bool xmit_more)
+{
+ if (xmit_more) {
+#ifdef CONFIG_BQL
+ dql_queued(&dev_queue->dql, bytes);
+#endif
+ return netif_tx_queue_stopped(dev_queue);
+ }
+ netdev_tx_sent_queue(dev_queue, bytes);
+ return true;
+}
+
/**
* netdev_sent_queue - report the number of bytes queued to hardware
* @dev: network device
diff --git a/include/linux/netfilter/ipset/ip_set.h b/include/linux/netfilter/ipset/ip_set.h
index 34fc80f..1d100ef 100644
--- a/include/linux/netfilter/ipset/ip_set.h
+++ b/include/linux/netfilter/ipset/ip_set.h
@@ -314,7 +314,7 @@ enum {
extern ip_set_id_t ip_set_get_byname(struct net *net,
const char *name, struct ip_set **set);
extern void ip_set_put_byindex(struct net *net, ip_set_id_t index);
-extern const char *ip_set_name_byindex(struct net *net, ip_set_id_t index);
+extern void ip_set_name_byindex(struct net *net, ip_set_id_t index, char *name);
extern ip_set_id_t ip_set_nfnl_get_byindex(struct net *net, ip_set_id_t index);
extern void ip_set_nfnl_put(struct net *net, ip_set_id_t index);
diff --git a/include/linux/netfilter/ipset/ip_set_comment.h b/include/linux/netfilter/ipset/ip_set_comment.h
index 8e2bab1..70877f8d 100644
--- a/include/linux/netfilter/ipset/ip_set_comment.h
+++ b/include/linux/netfilter/ipset/ip_set_comment.h
@@ -43,11 +43,11 @@ ip_set_init_comment(struct ip_set *set, struct ip_set_comment *comment,
rcu_assign_pointer(comment->c, c);
}
-/* Used only when dumping a set, protected by rcu_read_lock_bh() */
+/* Used only when dumping a set, protected by rcu_read_lock() */
static inline int
ip_set_put_comment(struct sk_buff *skb, const struct ip_set_comment *comment)
{
- struct ip_set_comment_rcu *c = rcu_dereference_bh(comment->c);
+ struct ip_set_comment_rcu *c = rcu_dereference(comment->c);
if (!c)
return 0;
diff --git a/include/linux/nmi.h b/include/linux/nmi.h
index 08f9247..9003e29 100644
--- a/include/linux/nmi.h
+++ b/include/linux/nmi.h
@@ -119,6 +119,8 @@ static inline int hardlockup_detector_perf_init(void) { return 0; }
void watchdog_nmi_stop(void);
void watchdog_nmi_start(void);
int watchdog_nmi_probe(void);
+int watchdog_nmi_enable(unsigned int cpu);
+void watchdog_nmi_disable(unsigned int cpu);
/**
* touch_nmi_watchdog - restart NMI watchdog timeout.
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 14b789a..1656c59 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -317,6 +317,8 @@ bool ipv6_chk_acast_addr(struct net *net, struct net_device *dev,
const struct in6_addr *addr);
bool ipv6_chk_acast_addr_src(struct net *net, struct net_device *dev,
const struct in6_addr *addr);
+int ipv6_anycast_init(void);
+void ipv6_anycast_cleanup(void);
/* Device notifier */
int register_inet6addr_notifier(struct notifier_block *nb);
diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h
index d7578cf..c9c78c1 100644
--- a/include/net/if_inet6.h
+++ b/include/net/if_inet6.h
@@ -146,10 +146,12 @@ struct ifacaddr6 {
struct in6_addr aca_addr;
struct fib6_info *aca_rt;
struct ifacaddr6 *aca_next;
+ struct hlist_node aca_addr_lst;
int aca_users;
refcount_t aca_refcnt;
unsigned long aca_cstamp;
unsigned long aca_tstamp;
+ struct rcu_head rcu;
};
#define IFA_HOST IPV6_ADDR_LOOPBACK
diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h
index eed04af..ae7b86f 100644
--- a/include/net/netfilter/nf_conntrack_l4proto.h
+++ b/include/net/netfilter/nf_conntrack_l4proto.h
@@ -153,4 +153,43 @@ void nf_ct_l4proto_log_invalid(const struct sk_buff *skb,
const char *fmt, ...) { }
#endif /* CONFIG_SYSCTL */
+static inline struct nf_generic_net *nf_generic_pernet(struct net *net)
+{
+ return &net->ct.nf_ct_proto.generic;
+}
+
+static inline struct nf_tcp_net *nf_tcp_pernet(struct net *net)
+{
+ return &net->ct.nf_ct_proto.tcp;
+}
+
+static inline struct nf_udp_net *nf_udp_pernet(struct net *net)
+{
+ return &net->ct.nf_ct_proto.udp;
+}
+
+static inline struct nf_icmp_net *nf_icmp_pernet(struct net *net)
+{
+ return &net->ct.nf_ct_proto.icmp;
+}
+
+static inline struct nf_icmp_net *nf_icmpv6_pernet(struct net *net)
+{
+ return &net->ct.nf_ct_proto.icmpv6;
+}
+
+#ifdef CONFIG_NF_CT_PROTO_DCCP
+static inline struct nf_dccp_net *nf_dccp_pernet(struct net *net)
+{
+ return &net->ct.nf_ct_proto.dccp;
+}
+#endif
+
+#ifdef CONFIG_NF_CT_PROTO_SCTP
+static inline struct nf_sctp_net *nf_sctp_pernet(struct net *net)
+{
+ return &net->ct.nf_ct_proto.sctp;
+}
+#endif
+
#endif /*_NF_CONNTRACK_PROTOCOL_H*/
diff --git a/include/trace/events/kyber.h b/include/trace/events/kyber.h
index a9834c3..c0e7d24 100644
--- a/include/trace/events/kyber.h
+++ b/include/trace/events/kyber.h
@@ -31,8 +31,8 @@ TRACE_EVENT(kyber_latency,
TP_fast_assign(
__entry->dev = disk_devt(dev_to_disk(kobj_to_dev(q->kobj.parent)));
- strlcpy(__entry->domain, domain, DOMAIN_LEN);
- strlcpy(__entry->type, type, DOMAIN_LEN);
+ strlcpy(__entry->domain, domain, sizeof(__entry->domain));
+ strlcpy(__entry->type, type, sizeof(__entry->type));
__entry->percentile = percentile;
__entry->numerator = numerator;
__entry->denominator = denominator;
@@ -60,7 +60,7 @@ TRACE_EVENT(kyber_adjust,
TP_fast_assign(
__entry->dev = disk_devt(dev_to_disk(kobj_to_dev(q->kobj.parent)));
- strlcpy(__entry->domain, domain, DOMAIN_LEN);
+ strlcpy(__entry->domain, domain, sizeof(__entry->domain));
__entry->depth = depth;
),
@@ -82,7 +82,7 @@ TRACE_EVENT(kyber_throttled,
TP_fast_assign(
__entry->dev = disk_devt(dev_to_disk(kobj_to_dev(q->kobj.parent)));
- strlcpy(__entry->domain, domain, DOMAIN_LEN);
+ strlcpy(__entry->domain, domain, sizeof(__entry->domain));
),
TP_printk("%d,%d %s", MAJOR(__entry->dev), MINOR(__entry->dev),
diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h
index 370e9a5..be84e43 100644
--- a/include/uapi/drm/amdgpu_drm.h
+++ b/include/uapi/drm/amdgpu_drm.h
@@ -326,6 +326,12 @@ struct drm_amdgpu_gem_userptr {
/* GFX9 and later: */
#define AMDGPU_TILING_SWIZZLE_MODE_SHIFT 0
#define AMDGPU_TILING_SWIZZLE_MODE_MASK 0x1f
+#define AMDGPU_TILING_DCC_OFFSET_256B_SHIFT 5
+#define AMDGPU_TILING_DCC_OFFSET_256B_MASK 0xFFFFFF
+#define AMDGPU_TILING_DCC_PITCH_MAX_SHIFT 29
+#define AMDGPU_TILING_DCC_PITCH_MAX_MASK 0x3FFF
+#define AMDGPU_TILING_DCC_INDEPENDENT_64B_SHIFT 43
+#define AMDGPU_TILING_DCC_INDEPENDENT_64B_MASK 0x1
/* Set/Get helpers for tiling flags. */
#define AMDGPU_TILING_SET(field, value) \
diff --git a/include/uapi/linux/kfd_ioctl.h b/include/uapi/linux/kfd_ioctl.h
index f5ff8a7..b01eb50 100644
--- a/include/uapi/linux/kfd_ioctl.h
+++ b/include/uapi/linux/kfd_ioctl.h
@@ -83,11 +83,11 @@ struct kfd_ioctl_set_cu_mask_args {
};
struct kfd_ioctl_get_queue_wave_state_args {
- uint64_t ctl_stack_address; /* to KFD */
- uint32_t ctl_stack_used_size; /* from KFD */
- uint32_t save_area_used_size; /* from KFD */
- uint32_t queue_id; /* to KFD */
- uint32_t pad;
+ __u64 ctl_stack_address; /* to KFD */
+ __u32 ctl_stack_used_size; /* from KFD */
+ __u32 save_area_used_size; /* from KFD */
+ __u32 queue_id; /* to KFD */
+ __u32 pad;
};
/* For kfd_ioctl_set_memory_policy_args.default_policy and alternate_policy */
@@ -255,10 +255,10 @@ struct kfd_hsa_memory_exception_data {
/* hw exception data */
struct kfd_hsa_hw_exception_data {
- uint32_t reset_type;
- uint32_t reset_cause;
- uint32_t memory_lost;
- uint32_t gpu_id;
+ __u32 reset_type;
+ __u32 reset_cause;
+ __u32 memory_lost;
+ __u32 gpu_id;
};
/* Event data */
diff --git a/include/uapi/linux/netfilter/nf_tables.h b/include/uapi/linux/netfilter/nf_tables.h
index 579974b..7de4f1b 100644
--- a/include/uapi/linux/netfilter/nf_tables.h
+++ b/include/uapi/linux/netfilter/nf_tables.h
@@ -1635,8 +1635,8 @@ enum nft_ng_attributes {
NFTA_NG_MODULUS,
NFTA_NG_TYPE,
NFTA_NG_OFFSET,
- NFTA_NG_SET_NAME,
- NFTA_NG_SET_ID,
+ NFTA_NG_SET_NAME, /* deprecated */
+ NFTA_NG_SET_ID, /* deprecated */
__NFTA_NG_MAX
};
#define NFTA_NG_MAX (__NFTA_NG_MAX - 1)
diff --git a/include/uapi/linux/netfilter_bridge.h b/include/uapi/linux/netfilter_bridge.h
index 156ccd0..1610fdb 100644
--- a/include/uapi/linux/netfilter_bridge.h
+++ b/include/uapi/linux/netfilter_bridge.h
@@ -11,6 +11,10 @@
#include <linux/if_vlan.h>
#include <linux/if_pppox.h>
+#ifndef __KERNEL__
+#include <limits.h> /* for INT_MIN, INT_MAX */
+#endif
+
/* Bridge Hooks */
/* After promisc drops, checksum checks. */
#define NF_BR_PRE_ROUTING 0
diff --git a/include/uapi/linux/sctp.h b/include/uapi/linux/sctp.h
index 34dd3d4..c81feb3 100644
--- a/include/uapi/linux/sctp.h
+++ b/include/uapi/linux/sctp.h
@@ -568,6 +568,8 @@ struct sctp_assoc_reset_event {
#define SCTP_ASSOC_CHANGE_DENIED 0x0004
#define SCTP_ASSOC_CHANGE_FAILED 0x0008
+#define SCTP_STREAM_CHANGE_DENIED SCTP_ASSOC_CHANGE_DENIED
+#define SCTP_STREAM_CHANGE_FAILED SCTP_ASSOC_CHANGE_FAILED
struct sctp_stream_change_event {
__u16 strchange_type;
__u16 strchange_flags;
@@ -1151,6 +1153,7 @@ struct sctp_add_streams {
/* SCTP Stream schedulers */
enum sctp_sched_type {
SCTP_SS_FCFS,
+ SCTP_SS_DEFAULT = SCTP_SS_FCFS,
SCTP_SS_PRIO,
SCTP_SS_RR,
SCTP_SS_MAX = SCTP_SS_RR
diff --git a/include/xen/xen-ops.h b/include/xen/xen-ops.h
index 18803ff..4969817 100644
--- a/include/xen/xen-ops.h
+++ b/include/xen/xen-ops.h
@@ -42,16 +42,12 @@ int xen_setup_shutdown_event(void);
extern unsigned long *xen_contiguous_bitmap;
-#ifdef CONFIG_XEN_PV
+#if defined(CONFIG_XEN_PV) || defined(CONFIG_ARM) || defined(CONFIG_ARM64)
int xen_create_contiguous_region(phys_addr_t pstart, unsigned int order,
unsigned int address_bits,
dma_addr_t *dma_handle);
void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order);
-
-int xen_remap_pfn(struct vm_area_struct *vma, unsigned long addr,
- xen_pfn_t *pfn, int nr, int *err_ptr, pgprot_t prot,
- unsigned int domid, bool no_translate, struct page **pages);
#else
static inline int xen_create_contiguous_region(phys_addr_t pstart,
unsigned int order,
@@ -63,7 +59,13 @@ static inline int xen_create_contiguous_region(phys_addr_t pstart,
static inline void xen_destroy_contiguous_region(phys_addr_t pstart,
unsigned int order) { }
+#endif
+#if defined(CONFIG_XEN_PV)
+int xen_remap_pfn(struct vm_area_struct *vma, unsigned long addr,
+ xen_pfn_t *pfn, int nr, int *err_ptr, pgprot_t prot,
+ unsigned int domid, bool no_translate, struct page **pages);
+#else
static inline int xen_remap_pfn(struct vm_area_struct *vma, unsigned long addr,
xen_pfn_t *pfn, int nr, int *err_ptr,
pgprot_t prot, unsigned int domid,
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 6377225..1a796e0 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -553,7 +553,6 @@ bool is_bpf_text_address(unsigned long addr)
int bpf_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
char *sym)
{
- unsigned long symbol_start, symbol_end;
struct bpf_prog_aux *aux;
unsigned int it = 0;
int ret = -ERANGE;
@@ -566,10 +565,9 @@ int bpf_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
if (it++ != symnum)
continue;
- bpf_get_prog_addr_region(aux->prog, &symbol_start, &symbol_end);
bpf_get_prog_name(aux->prog, sym);
- *value = symbol_start;
+ *value = (unsigned long)aux->prog->bpf_func;
*type = BPF_SYM_ELF_TYPE;
ret = 0;
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index ccb9327..cf5040f 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2078,6 +2078,7 @@ static int bpf_prog_get_info_by_fd(struct bpf_prog *prog,
info.jited_prog_len = 0;
info.xlated_prog_len = 0;
info.nr_jited_ksyms = 0;
+ info.nr_jited_func_lens = 0;
goto done;
}
@@ -2158,11 +2159,11 @@ static int bpf_prog_get_info_by_fd(struct bpf_prog *prog,
}
ulen = info.nr_jited_ksyms;
- info.nr_jited_ksyms = prog->aux->func_cnt;
+ info.nr_jited_ksyms = prog->aux->func_cnt ? : 1;
if (info.nr_jited_ksyms && ulen) {
if (bpf_dump_raw_ok()) {
+ unsigned long ksym_addr;
u64 __user *user_ksyms;
- ulong ksym_addr;
u32 i;
/* copy the address of the kernel symbol
@@ -2170,10 +2171,17 @@ static int bpf_prog_get_info_by_fd(struct bpf_prog *prog,
*/
ulen = min_t(u32, info.nr_jited_ksyms, ulen);
user_ksyms = u64_to_user_ptr(info.jited_ksyms);
- for (i = 0; i < ulen; i++) {
- ksym_addr = (ulong) prog->aux->func[i]->bpf_func;
- ksym_addr &= PAGE_MASK;
- if (put_user((u64) ksym_addr, &user_ksyms[i]))
+ if (prog->aux->func_cnt) {
+ for (i = 0; i < ulen; i++) {
+ ksym_addr = (unsigned long)
+ prog->aux->func[i]->bpf_func;
+ if (put_user((u64) ksym_addr,
+ &user_ksyms[i]))
+ return -EFAULT;
+ }
+ } else {
+ ksym_addr = (unsigned long) prog->bpf_func;
+ if (put_user((u64) ksym_addr, &user_ksyms[0]))
return -EFAULT;
}
} else {
@@ -2182,7 +2190,7 @@ static int bpf_prog_get_info_by_fd(struct bpf_prog *prog,
}
ulen = info.nr_jited_func_lens;
- info.nr_jited_func_lens = prog->aux->func_cnt;
+ info.nr_jited_func_lens = prog->aux->func_cnt ? : 1;
if (info.nr_jited_func_lens && ulen) {
if (bpf_dump_raw_ok()) {
u32 __user *user_lens;
@@ -2191,9 +2199,16 @@ static int bpf_prog_get_info_by_fd(struct bpf_prog *prog,
/* copy the JITed image lengths for each function */
ulen = min_t(u32, info.nr_jited_func_lens, ulen);
user_lens = u64_to_user_ptr(info.jited_func_lens);
- for (i = 0; i < ulen; i++) {
- func_len = prog->aux->func[i]->jited_len;
- if (put_user(func_len, &user_lens[i]))
+ if (prog->aux->func_cnt) {
+ for (i = 0; i < ulen; i++) {
+ func_len =
+ prog->aux->func[i]->jited_len;
+ if (put_user(func_len, &user_lens[i]))
+ return -EFAULT;
+ }
+ } else {
+ func_len = prog->jited_len;
+ if (put_user(func_len, &user_lens[0]))
return -EFAULT;
}
} else {
diff --git a/kernel/debug/kdb/kdb_bt.c b/kernel/debug/kdb/kdb_bt.c
index 6ad4a9f..7921ae4 100644
--- a/kernel/debug/kdb/kdb_bt.c
+++ b/kernel/debug/kdb/kdb_bt.c
@@ -179,14 +179,14 @@ kdb_bt(int argc, const char **argv)
kdb_printf("no process for cpu %ld\n", cpu);
return 0;
}
- sprintf(buf, "btt 0x%p\n", KDB_TSK(cpu));
+ sprintf(buf, "btt 0x%px\n", KDB_TSK(cpu));
kdb_parse(buf);
return 0;
}
kdb_printf("btc: cpu status: ");
kdb_parse("cpu\n");
for_each_online_cpu(cpu) {
- sprintf(buf, "btt 0x%p\n", KDB_TSK(cpu));
+ sprintf(buf, "btt 0x%px\n", KDB_TSK(cpu));
kdb_parse(buf);
touch_nmi_watchdog();
}
diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c
index ed5d349..6a4b414 100644
--- a/kernel/debug/kdb/kdb_io.c
+++ b/kernel/debug/kdb/kdb_io.c
@@ -216,7 +216,7 @@ static char *kdb_read(char *buffer, size_t bufsize)
int count;
int i;
int diag, dtab_count;
- int key;
+ int key, buf_size, ret;
diag = kdbgetintenv("DTABCOUNT", &dtab_count);
@@ -336,9 +336,8 @@ static char *kdb_read(char *buffer, size_t bufsize)
else
p_tmp = tmpbuffer;
len = strlen(p_tmp);
- count = kallsyms_symbol_complete(p_tmp,
- sizeof(tmpbuffer) -
- (p_tmp - tmpbuffer));
+ buf_size = sizeof(tmpbuffer) - (p_tmp - tmpbuffer);
+ count = kallsyms_symbol_complete(p_tmp, buf_size);
if (tab == 2 && count > 0) {
kdb_printf("\n%d symbols are found.", count);
if (count > dtab_count) {
@@ -350,9 +349,13 @@ static char *kdb_read(char *buffer, size_t bufsize)
}
kdb_printf("\n");
for (i = 0; i < count; i++) {
- if (WARN_ON(!kallsyms_symbol_next(p_tmp, i)))
+ ret = kallsyms_symbol_next(p_tmp, i, buf_size);
+ if (WARN_ON(!ret))
break;
- kdb_printf("%s ", p_tmp);
+ if (ret != -E2BIG)
+ kdb_printf("%s ", p_tmp);
+ else
+ kdb_printf("%s... ", p_tmp);
*(p_tmp + len) = '\0';
}
if (i >= dtab_count)
diff --git a/kernel/debug/kdb/kdb_keyboard.c b/kernel/debug/kdb/kdb_keyboard.c
index 118527a..750497b 100644
--- a/kernel/debug/kdb/kdb_keyboard.c
+++ b/kernel/debug/kdb/kdb_keyboard.c
@@ -173,11 +173,11 @@ int kdb_get_kbd_char(void)
case KT_LATIN:
if (isprint(keychar))
break; /* printable characters */
- /* drop through */
+ /* fall through */
case KT_SPEC:
if (keychar == K_ENTER)
break;
- /* drop through */
+ /* fall through */
default:
return -1; /* ignore unprintables */
}
diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c
index bb4fe4e..d72b32c66 100644
--- a/kernel/debug/kdb/kdb_main.c
+++ b/kernel/debug/kdb/kdb_main.c
@@ -1192,7 +1192,7 @@ static int kdb_local(kdb_reason_t reason, int error, struct pt_regs *regs,
if (reason == KDB_REASON_DEBUG) {
/* special case below */
} else {
- kdb_printf("\nEntering kdb (current=0x%p, pid %d) ",
+ kdb_printf("\nEntering kdb (current=0x%px, pid %d) ",
kdb_current, kdb_current ? kdb_current->pid : 0);
#if defined(CONFIG_SMP)
kdb_printf("on processor %d ", raw_smp_processor_id());
@@ -1208,7 +1208,7 @@ static int kdb_local(kdb_reason_t reason, int error, struct pt_regs *regs,
*/
switch (db_result) {
case KDB_DB_BPT:
- kdb_printf("\nEntering kdb (0x%p, pid %d) ",
+ kdb_printf("\nEntering kdb (0x%px, pid %d) ",
kdb_current, kdb_current->pid);
#if defined(CONFIG_SMP)
kdb_printf("on processor %d ", raw_smp_processor_id());
@@ -1493,6 +1493,7 @@ static void kdb_md_line(const char *fmtstr, unsigned long addr,
char cbuf[32];
char *c = cbuf;
int i;
+ int j;
unsigned long word;
memset(cbuf, '\0', sizeof(cbuf));
@@ -1538,25 +1539,9 @@ static void kdb_md_line(const char *fmtstr, unsigned long addr,
wc.word = word;
#define printable_char(c) \
({unsigned char __c = c; isascii(__c) && isprint(__c) ? __c : '.'; })
- switch (bytesperword) {
- case 8:
+ for (j = 0; j < bytesperword; j++)
*c++ = printable_char(*cp++);
- *c++ = printable_char(*cp++);
- *c++ = printable_char(*cp++);
- *c++ = printable_char(*cp++);
- addr += 4;
- case 4:
- *c++ = printable_char(*cp++);
- *c++ = printable_char(*cp++);
- addr += 2;
- case 2:
- *c++ = printable_char(*cp++);
- addr++;
- case 1:
- *c++ = printable_char(*cp++);
- addr++;
- break;
- }
+ addr += bytesperword;
#undef printable_char
}
}
@@ -2048,7 +2033,7 @@ static int kdb_lsmod(int argc, const char **argv)
if (mod->state == MODULE_STATE_UNFORMED)
continue;
- kdb_printf("%-20s%8u 0x%p ", mod->name,
+ kdb_printf("%-20s%8u 0x%px ", mod->name,
mod->core_layout.size, (void *)mod);
#ifdef CONFIG_MODULE_UNLOAD
kdb_printf("%4d ", module_refcount(mod));
@@ -2059,7 +2044,7 @@ static int kdb_lsmod(int argc, const char **argv)
kdb_printf(" (Loading)");
else
kdb_printf(" (Live)");
- kdb_printf(" 0x%p", mod->core_layout.base);
+ kdb_printf(" 0x%px", mod->core_layout.base);
#ifdef CONFIG_MODULE_UNLOAD
{
@@ -2341,7 +2326,7 @@ void kdb_ps1(const struct task_struct *p)
return;
cpu = kdb_process_cpu(p);
- kdb_printf("0x%p %8d %8d %d %4d %c 0x%p %c%s\n",
+ kdb_printf("0x%px %8d %8d %d %4d %c 0x%px %c%s\n",
(void *)p, p->pid, p->parent->pid,
kdb_task_has_cpu(p), kdb_process_cpu(p),
kdb_task_state_char(p),
@@ -2354,7 +2339,7 @@ void kdb_ps1(const struct task_struct *p)
} else {
if (KDB_TSK(cpu) != p)
kdb_printf(" Error: does not match running "
- "process table (0x%p)\n", KDB_TSK(cpu));
+ "process table (0x%px)\n", KDB_TSK(cpu));
}
}
}
@@ -2687,7 +2672,7 @@ int kdb_register_flags(char *cmd,
for_each_kdbcmd(kp, i) {
if (kp->cmd_name && (strcmp(kp->cmd_name, cmd) == 0)) {
kdb_printf("Duplicate kdb command registered: "
- "%s, func %p help %s\n", cmd, func, help);
+ "%s, func %px help %s\n", cmd, func, help);
return 1;
}
}
diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h
index 1e5a502..2118d82 100644
--- a/kernel/debug/kdb/kdb_private.h
+++ b/kernel/debug/kdb/kdb_private.h
@@ -83,7 +83,7 @@ typedef struct __ksymtab {
unsigned long sym_start;
unsigned long sym_end;
} kdb_symtab_t;
-extern int kallsyms_symbol_next(char *prefix_name, int flag);
+extern int kallsyms_symbol_next(char *prefix_name, int flag, int buf_size);
extern int kallsyms_symbol_complete(char *prefix_name, int max_len);
/* Exported Symbols for kernel loadable modules to use. */
diff --git a/kernel/debug/kdb/kdb_support.c b/kernel/debug/kdb/kdb_support.c
index 990b3cc..50bf9b1 100644
--- a/kernel/debug/kdb/kdb_support.c
+++ b/kernel/debug/kdb/kdb_support.c
@@ -40,7 +40,7 @@
int kdbgetsymval(const char *symname, kdb_symtab_t *symtab)
{
if (KDB_DEBUG(AR))
- kdb_printf("kdbgetsymval: symname=%s, symtab=%p\n", symname,
+ kdb_printf("kdbgetsymval: symname=%s, symtab=%px\n", symname,
symtab);
memset(symtab, 0, sizeof(*symtab));
symtab->sym_start = kallsyms_lookup_name(symname);
@@ -88,7 +88,7 @@ int kdbnearsym(unsigned long addr, kdb_symtab_t *symtab)
char *knt1 = NULL;
if (KDB_DEBUG(AR))
- kdb_printf("kdbnearsym: addr=0x%lx, symtab=%p\n", addr, symtab);
+ kdb_printf("kdbnearsym: addr=0x%lx, symtab=%px\n", addr, symtab);
memset(symtab, 0, sizeof(*symtab));
if (addr < 4096)
@@ -149,7 +149,7 @@ int kdbnearsym(unsigned long addr, kdb_symtab_t *symtab)
symtab->mod_name = "kernel";
if (KDB_DEBUG(AR))
kdb_printf("kdbnearsym: returns %d symtab->sym_start=0x%lx, "
- "symtab->mod_name=%p, symtab->sym_name=%p (%s)\n", ret,
+ "symtab->mod_name=%px, symtab->sym_name=%px (%s)\n", ret,
symtab->sym_start, symtab->mod_name, symtab->sym_name,
symtab->sym_name);
@@ -221,11 +221,13 @@ int kallsyms_symbol_complete(char *prefix_name, int max_len)
* Parameters:
* prefix_name prefix of a symbol name to lookup
* flag 0 means search from the head, 1 means continue search.
+ * buf_size maximum length that can be written to prefix_name
+ * buffer
* Returns:
* 1 if a symbol matches the given prefix.
* 0 if no string found
*/
-int kallsyms_symbol_next(char *prefix_name, int flag)
+int kallsyms_symbol_next(char *prefix_name, int flag, int buf_size)
{
int prefix_len = strlen(prefix_name);
static loff_t pos;
@@ -235,10 +237,8 @@ int kallsyms_symbol_next(char *prefix_name, int flag)
pos = 0;
while ((name = kdb_walk_kallsyms(&pos))) {
- if (strncmp(name, prefix_name, prefix_len) == 0) {
- strncpy(prefix_name, name, strlen(name)+1);
- return 1;
- }
+ if (!strncmp(name, prefix_name, prefix_len))
+ return strscpy(prefix_name, name, buf_size);
}
return 0;
}
@@ -432,7 +432,7 @@ int kdb_getphysword(unsigned long *word, unsigned long addr, size_t size)
*word = w8;
break;
}
- /* drop through */
+ /* fall through */
default:
diag = KDB_BADWIDTH;
kdb_printf("kdb_getphysword: bad width %ld\n", (long) size);
@@ -481,7 +481,7 @@ int kdb_getword(unsigned long *word, unsigned long addr, size_t size)
*word = w8;
break;
}
- /* drop through */
+ /* fall through */
default:
diag = KDB_BADWIDTH;
kdb_printf("kdb_getword: bad width %ld\n", (long) size);
@@ -525,7 +525,7 @@ int kdb_putword(unsigned long addr, unsigned long word, size_t size)
diag = kdb_putarea(addr, w8);
break;
}
- /* drop through */
+ /* fall through */
default:
diag = KDB_BADWIDTH;
kdb_printf("kdb_putword: bad width %ld\n", (long) size);
@@ -887,13 +887,13 @@ void debug_kusage(void)
__func__, dah_first);
if (dah_first) {
h_used = (struct debug_alloc_header *)debug_alloc_pool;
- kdb_printf("%s: h_used %p size %d\n", __func__, h_used,
+ kdb_printf("%s: h_used %px size %d\n", __func__, h_used,
h_used->size);
}
do {
h_used = (struct debug_alloc_header *)
((char *)h_free + dah_overhead + h_free->size);
- kdb_printf("%s: h_used %p size %d caller %p\n",
+ kdb_printf("%s: h_used %px size %d caller %px\n",
__func__, h_used, h_used->size, h_used->caller);
h_free = (struct debug_alloc_header *)
(debug_alloc_pool + h_free->next);
@@ -902,7 +902,7 @@ void debug_kusage(void)
((char *)h_free + dah_overhead + h_free->size);
if ((char *)h_used - debug_alloc_pool !=
sizeof(debug_alloc_pool_aligned))
- kdb_printf("%s: h_used %p size %d caller %p\n",
+ kdb_printf("%s: h_used %px size %d caller %px\n",
__func__, h_used, h_used->size, h_used->caller);
out:
spin_unlock(&dap_lock);
diff --git a/kernel/resource.c b/kernel/resource.c
index b3a3a1f..b0fbf68 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -319,16 +319,23 @@ int release_resource(struct resource *old)
EXPORT_SYMBOL(release_resource);
/**
- * Finds the lowest iomem resource that covers part of [start..end]. The
- * caller must specify start, end, flags, and desc (which may be
+ * Finds the lowest iomem resource that covers part of [@start..@end]. The
+ * caller must specify @start, @end, @flags, and @desc (which may be
* IORES_DESC_NONE).
*
- * If a resource is found, returns 0 and *res is overwritten with the part
- * of the resource that's within [start..end]; if none is found, returns
- * -1.
+ * If a resource is found, returns 0 and @*res is overwritten with the part
+ * of the resource that's within [@start..@end]; if none is found, returns
+ * -1 or -EINVAL for other invalid parameters.
*
* This function walks the whole tree and not just first level children
* unless @first_lvl is true.
+ *
+ * @start: start address of the resource searched for
+ * @end: end address of same resource
+ * @flags: flags which the resource must have
+ * @desc: descriptor the resource must have
+ * @first_lvl: walk only the first level children, if set
+ * @res: return ptr, if resource found
*/
static int find_next_iomem_res(resource_size_t start, resource_size_t end,
unsigned long flags, unsigned long desc,
@@ -399,6 +406,8 @@ static int __walk_iomem_res_desc(resource_size_t start, resource_size_t end,
* @flags: I/O resource flags
* @start: start addr
* @end: end addr
+ * @arg: function argument for the callback @func
+ * @func: callback function that is called for each qualifying resource area
*
* NOTE: For a new descriptor search, define a new IORES_DESC in
* <linux/ioport.h> and set it in 'desc' of a target resource entry.
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index f12225f..091e089 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -5851,11 +5851,14 @@ void __init sched_init_smp(void)
/*
* There's no userspace yet to cause hotplug operations; hence all the
* CPU masks are stable and all blatant races in the below code cannot
- * happen.
+ * happen. The hotplug lock is nevertheless taken to satisfy lockdep,
+ * but there won't be any contention on it.
*/
+ cpus_read_lock();
mutex_lock(&sched_domains_mutex);
sched_init_domains(cpu_active_mask);
mutex_unlock(&sched_domains_mutex);
+ cpus_read_unlock();
/* Move init over to a non-isolated CPU */
if (set_cpus_allowed_ptr(current, housekeeping_cpumask(HK_FLAG_DOMAIN)) < 0)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index ee271bb..ac855b2 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2400,8 +2400,8 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags)
local = 1;
/*
- * Retry task to preferred node migration periodically, in case it
- * case it previously failed, or the scheduler moved us.
+ * Retry to migrate task to preferred node periodically, in case it
+ * previously failed, or the scheduler moved us.
*/
if (time_after(jiffies, p->numa_migrate_retry)) {
task_numa_placement(p);
@@ -5674,11 +5674,11 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p,
return target;
}
-static unsigned long cpu_util_wake(int cpu, struct task_struct *p);
+static unsigned long cpu_util_without(int cpu, struct task_struct *p);
-static unsigned long capacity_spare_wake(int cpu, struct task_struct *p)
+static unsigned long capacity_spare_without(int cpu, struct task_struct *p)
{
- return max_t(long, capacity_of(cpu) - cpu_util_wake(cpu, p), 0);
+ return max_t(long, capacity_of(cpu) - cpu_util_without(cpu, p), 0);
}
/*
@@ -5738,7 +5738,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p,
avg_load += cfs_rq_load_avg(&cpu_rq(i)->cfs);
- spare_cap = capacity_spare_wake(i, p);
+ spare_cap = capacity_spare_without(i, p);
if (spare_cap > max_spare_cap)
max_spare_cap = spare_cap;
@@ -5889,8 +5889,8 @@ static inline int find_idlest_cpu(struct sched_domain *sd, struct task_struct *p
return prev_cpu;
/*
- * We need task's util for capacity_spare_wake, sync it up to prev_cpu's
- * last_update_time.
+ * We need task's util for capacity_spare_without, sync it up to
+ * prev_cpu's last_update_time.
*/
if (!(sd_flag & SD_BALANCE_FORK))
sync_entity_load_avg(&p->se);
@@ -6216,10 +6216,19 @@ static inline unsigned long cpu_util(int cpu)
}
/*
- * cpu_util_wake: Compute CPU utilization with any contributions from
- * the waking task p removed.
+ * cpu_util_without: compute cpu utilization without any contributions from *p
+ * @cpu: the CPU which utilization is requested
+ * @p: the task which utilization should be discounted
+ *
+ * The utilization of a CPU is defined by the utilization of tasks currently
+ * enqueued on that CPU as well as tasks which are currently sleeping after an
+ * execution on that CPU.
+ *
+ * This method returns the utilization of the specified CPU by discounting the
+ * utilization of the specified task, whenever the task is currently
+ * contributing to the CPU utilization.
*/
-static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
+static unsigned long cpu_util_without(int cpu, struct task_struct *p)
{
struct cfs_rq *cfs_rq;
unsigned int util;
@@ -6231,7 +6240,7 @@ static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
cfs_rq = &cpu_rq(cpu)->cfs;
util = READ_ONCE(cfs_rq->avg.util_avg);
- /* Discount task's blocked util from CPU's util */
+ /* Discount task's util from CPU's util */
util -= min_t(unsigned int, util, task_util(p));
/*
@@ -6240,14 +6249,14 @@ static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
* a) if *p is the only task sleeping on this CPU, then:
* cpu_util (== task_util) > util_est (== 0)
* and thus we return:
- * cpu_util_wake = (cpu_util - task_util) = 0
+ * cpu_util_without = (cpu_util - task_util) = 0
*
* b) if other tasks are SLEEPING on this CPU, which is now exiting
* IDLE, then:
* cpu_util >= task_util
* cpu_util > util_est (== 0)
* and thus we discount *p's blocked utilization to return:
- * cpu_util_wake = (cpu_util - task_util) >= 0
+ * cpu_util_without = (cpu_util - task_util) >= 0
*
* c) if other tasks are RUNNABLE on that CPU and
* util_est > cpu_util
@@ -6260,8 +6269,33 @@ static unsigned long cpu_util_wake(int cpu, struct task_struct *p)
* covered by the following code when estimated utilization is
* enabled.
*/
- if (sched_feat(UTIL_EST))
- util = max(util, READ_ONCE(cfs_rq->avg.util_est.enqueued));
+ if (sched_feat(UTIL_EST)) {
+ unsigned int estimated =
+ READ_ONCE(cfs_rq->avg.util_est.enqueued);
+
+ /*
+ * Despite the following checks we still have a small window
+ * for a possible race, when an execl's select_task_rq_fair()
+ * races with LB's detach_task():
+ *
+ * detach_task()
+ * p->on_rq = TASK_ON_RQ_MIGRATING;
+ * ---------------------------------- A
+ * deactivate_task() \
+ * dequeue_task() + RaceTime
+ * util_est_dequeue() /
+ * ---------------------------------- B
+ *
+ * The additional check on "current == p" it's required to
+ * properly fix the execl regression and it helps in further
+ * reducing the chances for the above race.
+ */
+ if (unlikely(task_on_rq_queued(p) || current == p)) {
+ estimated -= min_t(unsigned int, estimated,
+ (_task_util_est(p) | UTIL_AVG_UNCHANGED));
+ }
+ util = max(util, estimated);
+ }
/*
* Utilization (estimated) can exceed the CPU capacity, thus let's
diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c
index 7cdecfc..3d7355d 100644
--- a/kernel/sched/psi.c
+++ b/kernel/sched/psi.c
@@ -633,38 +633,39 @@ void psi_cgroup_free(struct cgroup *cgroup)
*/
void cgroup_move_task(struct task_struct *task, struct css_set *to)
{
- bool move_psi = !psi_disabled;
unsigned int task_flags = 0;
struct rq_flags rf;
struct rq *rq;
- if (move_psi) {
- rq = task_rq_lock(task, &rf);
-
- if (task_on_rq_queued(task))
- task_flags = TSK_RUNNING;
- else if (task->in_iowait)
- task_flags = TSK_IOWAIT;
-
- if (task->flags & PF_MEMSTALL)
- task_flags |= TSK_MEMSTALL;
-
- if (task_flags)
- psi_task_change(task, task_flags, 0);
+ if (psi_disabled) {
+ /*
+ * Lame to do this here, but the scheduler cannot be locked
+ * from the outside, so we move cgroups from inside sched/.
+ */
+ rcu_assign_pointer(task->cgroups, to);
+ return;
}
- /*
- * Lame to do this here, but the scheduler cannot be locked
- * from the outside, so we move cgroups from inside sched/.
- */
+ rq = task_rq_lock(task, &rf);
+
+ if (task_on_rq_queued(task))
+ task_flags = TSK_RUNNING;
+ else if (task->in_iowait)
+ task_flags = TSK_IOWAIT;
+
+ if (task->flags & PF_MEMSTALL)
+ task_flags |= TSK_MEMSTALL;
+
+ if (task_flags)
+ psi_task_change(task, task_flags, 0);
+
+ /* See comment above */
rcu_assign_pointer(task->cgroups, to);
- if (move_psi) {
- if (task_flags)
- psi_task_change(task, 0, task_flags);
+ if (task_flags)
+ psi_task_change(task, 0, task_flags);
- task_rq_unlock(rq, task, &rf);
- }
+ task_rq_unlock(rq, task, &rf);
}
#endif /* CONFIG_CGROUPS */
diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c
index ce32cf7..8f0644a 100644
--- a/kernel/time/posix-cpu-timers.c
+++ b/kernel/time/posix-cpu-timers.c
@@ -917,9 +917,6 @@ static void check_process_timers(struct task_struct *tsk,
struct task_cputime cputime;
unsigned long soft;
- if (dl_task(tsk))
- check_dl_overrun(tsk);
-
/*
* If cputimer is not running, then there are no active
* process wide timers (POSIX 1.b, itimers, RLIMIT_CPU).
diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c
index 3ef15a6..bd30e93 100644
--- a/kernel/trace/trace_probe.c
+++ b/kernel/trace/trace_probe.c
@@ -535,7 +535,7 @@ int traceprobe_update_arg(struct probe_arg *arg)
if (code[1].op != FETCH_OP_IMM)
return -EINVAL;
- tmp = strpbrk("+-", code->data);
+ tmp = strpbrk(code->data, "+-");
if (tmp)
c = *tmp;
ret = traceprobe_split_symbol_offset(code->data,
diff --git a/kernel/user_namespace.c b/kernel/user_namespace.c
index e5222b5..923414a 100644
--- a/kernel/user_namespace.c
+++ b/kernel/user_namespace.c
@@ -974,10 +974,6 @@ static ssize_t map_write(struct file *file, const char __user *buf,
if (!new_idmap_permitted(file, ns, cap_setid, &new_map))
goto out;
- ret = sort_idmaps(&new_map);
- if (ret < 0)
- goto out;
-
ret = -EPERM;
/* Map the lower ids from the parent user namespace to the
* kernel global id space.
@@ -1004,6 +1000,14 @@ static ssize_t map_write(struct file *file, const char __user *buf,
e->lower_first = lower_first;
}
+ /*
+ * If we want to use binary search for lookup, this clones the extent
+ * array and sorts both copies.
+ */
+ ret = sort_idmaps(&new_map);
+ if (ret < 0)
+ goto out;
+
/* Install the map */
if (new_map.nr_extents <= UID_GID_MAP_MAX_BASE_EXTENTS) {
memcpy(map->extent, new_map.extent,
diff --git a/lib/raid6/test/Makefile b/lib/raid6/test/Makefile
index 5d73f5c..7977764 100644
--- a/lib/raid6/test/Makefile
+++ b/lib/raid6/test/Makefile
@@ -27,7 +27,7 @@
CFLAGS += -I../../../arch/arm/include -mfpu=neon
HAS_NEON = yes
endif
-ifeq ($(ARCH),arm64)
+ifeq ($(ARCH),aarch64)
CFLAGS += -I../../../arch/arm64/include
HAS_NEON = yes
endif
@@ -41,7 +41,7 @@
gcc -c -x assembler - >&/dev/null && \
rm ./-.o && echo -DCONFIG_AS_AVX512=1)
else ifeq ($(HAS_NEON),yes)
- OBJS += neon.o neon1.o neon2.o neon4.o neon8.o
+ OBJS += neon.o neon1.o neon2.o neon4.o neon8.o recov_neon.o recov_neon_inner.o
CFLAGS += -DCONFIG_KERNEL_MODE_NEON=1
else
HAS_ALTIVEC := $(shell printf '\#include <altivec.h>\nvector int a;\n' |\
diff --git a/lib/ubsan.c b/lib/ubsan.c
index 59fee96..e4162f5 100644
--- a/lib/ubsan.c
+++ b/lib/ubsan.c
@@ -427,8 +427,7 @@ void __ubsan_handle_shift_out_of_bounds(struct shift_out_of_bounds_data *data,
EXPORT_SYMBOL(__ubsan_handle_shift_out_of_bounds);
-void __noreturn
-__ubsan_handle_builtin_unreachable(struct unreachable_data *data)
+void __ubsan_handle_builtin_unreachable(struct unreachable_data *data)
{
unsigned long flags;
diff --git a/mm/gup.c b/mm/gup.c
index f76e77a..aa43620 100644
--- a/mm/gup.c
+++ b/mm/gup.c
@@ -385,11 +385,17 @@ static struct page *follow_p4d_mask(struct vm_area_struct *vma,
* @vma: vm_area_struct mapping @address
* @address: virtual address to look up
* @flags: flags modifying lookup behaviour
- * @page_mask: on output, *page_mask is set according to the size of the page
+ * @ctx: contains dev_pagemap for %ZONE_DEVICE memory pinning and a
+ * pointer to output page_mask
*
* @flags can have FOLL_ flags set, defined in <linux/mm.h>
*
- * Returns the mapped (struct page *), %NULL if no mapping exists, or
+ * When getting pages from ZONE_DEVICE memory, the @ctx->pgmap caches
+ * the device's dev_pagemap metadata to avoid repeating expensive lookups.
+ *
+ * On output, the @ctx->page_mask is set according to the size of the page.
+ *
+ * Return: the mapped (struct page *), %NULL if no mapping exists, or
* an error pointer if there is a mapping to something not represented
* by a page descriptor (see also vm_normal_page()).
*/
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index c007fb5..7f2a28a 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -3233,7 +3233,7 @@ static int is_hugetlb_entry_hwpoisoned(pte_t pte)
int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
struct vm_area_struct *vma)
{
- pte_t *src_pte, *dst_pte, entry;
+ pte_t *src_pte, *dst_pte, entry, dst_entry;
struct page *ptepage;
unsigned long addr;
int cow;
@@ -3261,15 +3261,30 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
break;
}
- /* If the pagetables are shared don't copy or take references */
- if (dst_pte == src_pte)
+ /*
+ * If the pagetables are shared don't copy or take references.
+ * dst_pte == src_pte is the common case of src/dest sharing.
+ *
+ * However, src could have 'unshared' and dst shares with
+ * another vma. If dst_pte !none, this implies sharing.
+ * Check here before taking page table lock, and once again
+ * after taking the lock below.
+ */
+ dst_entry = huge_ptep_get(dst_pte);
+ if ((dst_pte == src_pte) || !huge_pte_none(dst_entry))
continue;
dst_ptl = huge_pte_lock(h, dst, dst_pte);
src_ptl = huge_pte_lockptr(h, src, src_pte);
spin_lock_nested(src_ptl, SINGLE_DEPTH_NESTING);
entry = huge_ptep_get(src_pte);
- if (huge_pte_none(entry)) { /* skip none entry */
+ dst_entry = huge_ptep_get(dst_pte);
+ if (huge_pte_none(entry) || !huge_pte_none(dst_entry)) {
+ /*
+ * Skip if src entry none. Also, skip in the
+ * unlikely case dst entry !none as this implies
+ * sharing with another vma.
+ */
;
} else if (unlikely(is_hugetlb_entry_migration(entry) ||
is_hugetlb_entry_hwpoisoned(entry))) {
diff --git a/mm/memblock.c b/mm/memblock.c
index 7df468c..9a2d5ae 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -1179,7 +1179,7 @@ void __init_memblock __next_mem_range_rev(u64 *idx, int nid,
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP
/*
- * Common iterator interface used to define for_each_mem_range().
+ * Common iterator interface used to define for_each_mem_pfn_range().
*/
void __init_memblock __next_mem_pfn_range(int *idx, int nid,
unsigned long *out_start_pfn,
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index a919ba5..6847177 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -4061,17 +4061,6 @@ __alloc_pages_slowpath(gfp_t gfp_mask, unsigned int order,
int reserve_flags;
/*
- * In the slowpath, we sanity check order to avoid ever trying to
- * reclaim >= MAX_ORDER areas which will never succeed. Callers may
- * be using allocators in order of preference for an area that is
- * too large.
- */
- if (order >= MAX_ORDER) {
- WARN_ON_ONCE(!(gfp_mask & __GFP_NOWARN));
- return NULL;
- }
-
- /*
* We also sanity check to catch abuse of atomic reserves being used by
* callers that are not in atomic context.
*/
@@ -4364,6 +4353,15 @@ __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, int preferred_nid,
gfp_t alloc_mask; /* The gfp_t that was actually used for allocation */
struct alloc_context ac = { };
+ /*
+ * There are several places where we assume that the order value is sane
+ * so bail out early if the request is out of bound.
+ */
+ if (unlikely(order >= MAX_ORDER)) {
+ WARN_ON_ONCE(!(gfp_mask & __GFP_NOWARN));
+ return NULL;
+ }
+
gfp_mask &= gfp_allowed_mask;
alloc_mask = gfp_mask;
if (!prepare_alloc_pages(gfp_mask, order, preferred_nid, nodemask, &ac, &alloc_mask, &alloc_flags))
@@ -7789,6 +7787,14 @@ bool has_unmovable_pages(struct zone *zone, struct page *page, int count,
goto unmovable;
/*
+ * If the zone is movable and we have ruled out all reserved
+ * pages then it should be reasonably safe to assume the rest
+ * is movable.
+ */
+ if (zone_idx(zone) == ZONE_MOVABLE)
+ continue;
+
+ /*
* Hugepages are not in LRU lists, but they're movable.
* We need not scan over tail pages bacause we don't
* handle each tail page individually in migration.
diff --git a/mm/shmem.c b/mm/shmem.c
index ea26d7a..d44991e 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -2563,9 +2563,7 @@ static loff_t shmem_file_llseek(struct file *file, loff_t offset, int whence)
inode_lock(inode);
/* We're holding i_mutex so we can access i_size directly */
- if (offset < 0)
- offset = -EINVAL;
- else if (offset >= inode->i_size)
+ if (offset < 0 || offset >= inode->i_size)
offset = -ENXIO;
else {
start = offset >> PAGE_SHIFT;
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 644f746..8688ae6 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -2813,7 +2813,7 @@ static struct swap_info_struct *alloc_swap_info(void)
unsigned int type;
int i;
- p = kzalloc(sizeof(*p), GFP_KERNEL);
+ p = kvzalloc(sizeof(*p), GFP_KERNEL);
if (!p)
return ERR_PTR(-ENOMEM);
@@ -2824,7 +2824,7 @@ static struct swap_info_struct *alloc_swap_info(void)
}
if (type >= MAX_SWAPFILES) {
spin_unlock(&swap_lock);
- kfree(p);
+ kvfree(p);
return ERR_PTR(-EPERM);
}
if (type >= nr_swapfiles) {
@@ -2838,7 +2838,7 @@ static struct swap_info_struct *alloc_swap_info(void)
smp_wmb();
nr_swapfiles++;
} else {
- kfree(p);
+ kvfree(p);
p = swap_info[type];
/*
* Do not memset this entry: a racing procfs swap_next()
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 6038ce5..9c62459 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -1827,12 +1827,13 @@ static bool need_update(int cpu)
/*
* The fast way of checking if there are any vmstat diffs.
- * This works because the diffs are byte sized items.
*/
- if (memchr_inv(p->vm_stat_diff, 0, NR_VM_ZONE_STAT_ITEMS))
+ if (memchr_inv(p->vm_stat_diff, 0, NR_VM_ZONE_STAT_ITEMS *
+ sizeof(p->vm_stat_diff[0])))
return true;
#ifdef CONFIG_NUMA
- if (memchr_inv(p->vm_numa_stat_diff, 0, NR_VM_NUMA_STAT_ITEMS))
+ if (memchr_inv(p->vm_numa_stat_diff, 0, NR_VM_NUMA_STAT_ITEMS *
+ sizeof(p->vm_numa_stat_diff[0])))
return true;
#endif
}
diff --git a/mm/z3fold.c b/mm/z3fold.c
index 4b366d1..aee9b0b 100644
--- a/mm/z3fold.c
+++ b/mm/z3fold.c
@@ -99,6 +99,7 @@ struct z3fold_header {
#define NCHUNKS ((PAGE_SIZE - ZHDR_SIZE_ALIGNED) >> CHUNK_SHIFT)
#define BUDDY_MASK (0x3)
+#define BUDDY_SHIFT 2
/**
* struct z3fold_pool - stores metadata for each z3fold pool
@@ -145,7 +146,7 @@ enum z3fold_page_flags {
MIDDLE_CHUNK_MAPPED,
NEEDS_COMPACTING,
PAGE_STALE,
- UNDER_RECLAIM
+ PAGE_CLAIMED, /* by either reclaim or free */
};
/*****************
@@ -174,7 +175,7 @@ static struct z3fold_header *init_z3fold_page(struct page *page,
clear_bit(MIDDLE_CHUNK_MAPPED, &page->private);
clear_bit(NEEDS_COMPACTING, &page->private);
clear_bit(PAGE_STALE, &page->private);
- clear_bit(UNDER_RECLAIM, &page->private);
+ clear_bit(PAGE_CLAIMED, &page->private);
spin_lock_init(&zhdr->page_lock);
kref_init(&zhdr->refcount);
@@ -223,8 +224,11 @@ static unsigned long encode_handle(struct z3fold_header *zhdr, enum buddy bud)
unsigned long handle;
handle = (unsigned long)zhdr;
- if (bud != HEADLESS)
- handle += (bud + zhdr->first_num) & BUDDY_MASK;
+ if (bud != HEADLESS) {
+ handle |= (bud + zhdr->first_num) & BUDDY_MASK;
+ if (bud == LAST)
+ handle |= (zhdr->last_chunks << BUDDY_SHIFT);
+ }
return handle;
}
@@ -234,6 +238,12 @@ static struct z3fold_header *handle_to_z3fold_header(unsigned long handle)
return (struct z3fold_header *)(handle & PAGE_MASK);
}
+/* only for LAST bud, returns zero otherwise */
+static unsigned short handle_to_chunks(unsigned long handle)
+{
+ return (handle & ~PAGE_MASK) >> BUDDY_SHIFT;
+}
+
/*
* (handle & BUDDY_MASK) < zhdr->first_num is possible in encode_handle
* but that doesn't matter. because the masking will result in the
@@ -720,37 +730,39 @@ static void z3fold_free(struct z3fold_pool *pool, unsigned long handle)
page = virt_to_page(zhdr);
if (test_bit(PAGE_HEADLESS, &page->private)) {
- /* HEADLESS page stored */
- bud = HEADLESS;
- } else {
- z3fold_page_lock(zhdr);
- bud = handle_to_buddy(handle);
-
- switch (bud) {
- case FIRST:
- zhdr->first_chunks = 0;
- break;
- case MIDDLE:
- zhdr->middle_chunks = 0;
- zhdr->start_middle = 0;
- break;
- case LAST:
- zhdr->last_chunks = 0;
- break;
- default:
- pr_err("%s: unknown bud %d\n", __func__, bud);
- WARN_ON(1);
- z3fold_page_unlock(zhdr);
- return;
+ /* if a headless page is under reclaim, just leave.
+ * NB: we use test_and_set_bit for a reason: if the bit
+ * has not been set before, we release this page
+ * immediately so we don't care about its value any more.
+ */
+ if (!test_and_set_bit(PAGE_CLAIMED, &page->private)) {
+ spin_lock(&pool->lock);
+ list_del(&page->lru);
+ spin_unlock(&pool->lock);
+ free_z3fold_page(page);
+ atomic64_dec(&pool->pages_nr);
}
+ return;
}
- if (bud == HEADLESS) {
- spin_lock(&pool->lock);
- list_del(&page->lru);
- spin_unlock(&pool->lock);
- free_z3fold_page(page);
- atomic64_dec(&pool->pages_nr);
+ /* Non-headless case */
+ z3fold_page_lock(zhdr);
+ bud = handle_to_buddy(handle);
+
+ switch (bud) {
+ case FIRST:
+ zhdr->first_chunks = 0;
+ break;
+ case MIDDLE:
+ zhdr->middle_chunks = 0;
+ break;
+ case LAST:
+ zhdr->last_chunks = 0;
+ break;
+ default:
+ pr_err("%s: unknown bud %d\n", __func__, bud);
+ WARN_ON(1);
+ z3fold_page_unlock(zhdr);
return;
}
@@ -758,7 +770,7 @@ static void z3fold_free(struct z3fold_pool *pool, unsigned long handle)
atomic64_dec(&pool->pages_nr);
return;
}
- if (test_bit(UNDER_RECLAIM, &page->private)) {
+ if (test_bit(PAGE_CLAIMED, &page->private)) {
z3fold_page_unlock(zhdr);
return;
}
@@ -836,20 +848,30 @@ static int z3fold_reclaim_page(struct z3fold_pool *pool, unsigned int retries)
}
list_for_each_prev(pos, &pool->lru) {
page = list_entry(pos, struct page, lru);
- if (test_bit(PAGE_HEADLESS, &page->private))
- /* candidate found */
- break;
+
+ /* this bit could have been set by free, in which case
+ * we pass over to the next page in the pool.
+ */
+ if (test_and_set_bit(PAGE_CLAIMED, &page->private))
+ continue;
zhdr = page_address(page);
- if (!z3fold_page_trylock(zhdr))
+ if (test_bit(PAGE_HEADLESS, &page->private))
+ break;
+
+ if (!z3fold_page_trylock(zhdr)) {
+ zhdr = NULL;
continue; /* can't evict at this point */
+ }
kref_get(&zhdr->refcount);
list_del_init(&zhdr->buddy);
zhdr->cpu = -1;
- set_bit(UNDER_RECLAIM, &page->private);
break;
}
+ if (!zhdr)
+ break;
+
list_del_init(&page->lru);
spin_unlock(&pool->lock);
@@ -898,6 +920,7 @@ static int z3fold_reclaim_page(struct z3fold_pool *pool, unsigned int retries)
if (test_bit(PAGE_HEADLESS, &page->private)) {
if (ret == 0) {
free_z3fold_page(page);
+ atomic64_dec(&pool->pages_nr);
return 0;
}
spin_lock(&pool->lock);
@@ -905,7 +928,7 @@ static int z3fold_reclaim_page(struct z3fold_pool *pool, unsigned int retries)
spin_unlock(&pool->lock);
} else {
z3fold_page_lock(zhdr);
- clear_bit(UNDER_RECLAIM, &page->private);
+ clear_bit(PAGE_CLAIMED, &page->private);
if (kref_put(&zhdr->refcount,
release_z3fold_page_locked)) {
atomic64_dec(&pool->pages_nr);
@@ -964,7 +987,7 @@ static void *z3fold_map(struct z3fold_pool *pool, unsigned long handle)
set_bit(MIDDLE_CHUNK_MAPPED, &page->private);
break;
case LAST:
- addr += PAGE_SIZE - (zhdr->last_chunks << CHUNK_SHIFT);
+ addr += PAGE_SIZE - (handle_to_chunks(handle) << CHUNK_SHIFT);
break;
default:
pr_err("unknown buddy id %d\n", buddy);
diff --git a/net/core/dev.c b/net/core/dev.c
index 77d43ae..0ffcbdd 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -3272,7 +3272,7 @@ struct sk_buff *dev_hard_start_xmit(struct sk_buff *first, struct net_device *de
}
skb = next;
- if (netif_xmit_stopped(txq) && skb) {
+ if (netif_tx_queue_stopped(txq) && skb) {
rc = NETDEV_TX_BUSY;
break;
}
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 676f3ad..588f475 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -1166,8 +1166,8 @@ bool __skb_flow_dissect(const struct sk_buff *skb,
break;
}
- if (dissector_uses_key(flow_dissector,
- FLOW_DISSECTOR_KEY_PORTS)) {
+ if (dissector_uses_key(flow_dissector, FLOW_DISSECTOR_KEY_PORTS) &&
+ !(key_control->flags & FLOW_DIS_IS_FRAGMENT)) {
key_ports = skb_flow_dissector_target(flow_dissector,
FLOW_DISSECTOR_KEY_PORTS,
target_container);
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 5da9552..2b9fdbc 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -717,7 +717,8 @@ int netpoll_setup(struct netpoll *np)
read_lock_bh(&idev->lock);
list_for_each_entry(ifp, &idev->addr_list, if_list) {
- if (ipv6_addr_type(&ifp->addr) & IPV6_ADDR_LINKLOCAL)
+ if (!!(ipv6_addr_type(&ifp->addr) & IPV6_ADDR_LINKLOCAL) !=
+ !!(ipv6_addr_type(&np->remote_ip.in6) & IPV6_ADDR_LINKLOCAL))
continue;
np->local_ip.in6 = ifp->addr;
err = 0;
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index e01274b..33d9227 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -3367,7 +3367,7 @@ static int rtnl_dump_all(struct sk_buff *skb, struct netlink_callback *cb)
cb->seq = 0;
}
ret = dumpit(skb, cb);
- if (ret < 0)
+ if (ret)
break;
}
cb->family = idx;
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 946de0e..b4ee5c8 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -4944,6 +4944,8 @@ static unsigned int skb_gso_mac_seglen(const struct sk_buff *skb)
*
* This is a helper to do that correctly considering GSO_BY_FRAGS.
*
+ * @skb: GSO skb
+ *
* @seg_len: The segmented length (from skb_gso_*_seglen). In the
* GSO_BY_FRAGS case this will be [header sizes + GSO_BY_FRAGS].
*
diff --git a/net/core/sock.c b/net/core/sock.c
index 6fcc4bc..080a880 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -3279,6 +3279,7 @@ int sock_load_diag_module(int family, int protocol)
#ifdef CONFIG_INET
if (family == AF_INET &&
+ protocol != IPPROTO_RAW &&
!rcu_access_pointer(inet_protos[protocol]))
return -ENOENT;
#endif
diff --git a/net/ipv4/inet_fragment.c b/net/ipv4/inet_fragment.c
index bcb11f3..760a9e5 100644
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -178,21 +178,22 @@ static struct inet_frag_queue *inet_frag_alloc(struct netns_frags *nf,
}
static struct inet_frag_queue *inet_frag_create(struct netns_frags *nf,
- void *arg)
+ void *arg,
+ struct inet_frag_queue **prev)
{
struct inet_frags *f = nf->f;
struct inet_frag_queue *q;
- int err;
q = inet_frag_alloc(nf, f, arg);
- if (!q)
+ if (!q) {
+ *prev = ERR_PTR(-ENOMEM);
return NULL;
-
+ }
mod_timer(&q->timer, jiffies + nf->timeout);
- err = rhashtable_insert_fast(&nf->rhashtable, &q->node,
- f->rhash_params);
- if (err < 0) {
+ *prev = rhashtable_lookup_get_insert_key(&nf->rhashtable, &q->key,
+ &q->node, f->rhash_params);
+ if (*prev) {
q->flags |= INET_FRAG_COMPLETE;
inet_frag_kill(q);
inet_frag_destroy(q);
@@ -204,22 +205,22 @@ static struct inet_frag_queue *inet_frag_create(struct netns_frags *nf,
/* TODO : call from rcu_read_lock() and no longer use refcount_inc_not_zero() */
struct inet_frag_queue *inet_frag_find(struct netns_frags *nf, void *key)
{
- struct inet_frag_queue *fq;
+ struct inet_frag_queue *fq = NULL, *prev;
if (!nf->high_thresh || frag_mem_limit(nf) > nf->high_thresh)
return NULL;
rcu_read_lock();
- fq = rhashtable_lookup(&nf->rhashtable, key, nf->f->rhash_params);
- if (fq) {
+ prev = rhashtable_lookup(&nf->rhashtable, key, nf->f->rhash_params);
+ if (!prev)
+ fq = inet_frag_create(nf, key, &prev);
+ if (prev && !IS_ERR(prev)) {
+ fq = prev;
if (!refcount_inc_not_zero(&fq->refcnt))
fq = NULL;
- rcu_read_unlock();
- return fq;
}
rcu_read_unlock();
-
- return inet_frag_create(nf, key);
+ return fq;
}
EXPORT_SYMBOL(inet_frag_find);
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index 9b0158f..d6ee343 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -722,10 +722,14 @@ struct sk_buff *ip_check_defrag(struct net *net, struct sk_buff *skb, u32 user)
if (ip_is_fragment(&iph)) {
skb = skb_share_check(skb, GFP_ATOMIC);
if (skb) {
- if (!pskb_may_pull(skb, netoff + iph.ihl * 4))
- return skb;
- if (pskb_trim_rcsum(skb, netoff + len))
- return skb;
+ if (!pskb_may_pull(skb, netoff + iph.ihl * 4)) {
+ kfree_skb(skb);
+ return NULL;
+ }
+ if (pskb_trim_rcsum(skb, netoff + len)) {
+ kfree_skb(skb);
+ return NULL;
+ }
memset(IPCB(skb), 0, sizeof(struct inet_skb_parm));
if (ip_defrag(net, skb, user))
return NULL;
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 26c36cc..fffcc13 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -1246,7 +1246,7 @@ int ip_setsockopt(struct sock *sk, int level,
return -ENOPROTOOPT;
err = do_ip_setsockopt(sk, level, optname, optval, optlen);
-#ifdef CONFIG_BPFILTER
+#if IS_ENABLED(CONFIG_BPFILTER_UMH)
if (optname >= BPFILTER_IPT_SO_SET_REPLACE &&
optname < BPFILTER_IPT_SET_MAX)
err = bpfilter_ip_set_sockopt(sk, optname, optval, optlen);
@@ -1559,7 +1559,7 @@ int ip_getsockopt(struct sock *sk, int level,
int err;
err = do_ip_getsockopt(sk, level, optname, optval, optlen, 0);
-#ifdef CONFIG_BPFILTER
+#if IS_ENABLED(CONFIG_BPFILTER_UMH)
if (optname >= BPFILTER_IPT_SO_GET_INFO &&
optname < BPFILTER_IPT_GET_MAX)
err = bpfilter_ip_get_sockopt(sk, optname, optval, optlen);
@@ -1596,7 +1596,7 @@ int compat_ip_getsockopt(struct sock *sk, int level, int optname,
err = do_ip_getsockopt(sk, level, optname, optval, optlen,
MSG_CMSG_COMPAT);
-#ifdef CONFIG_BPFILTER
+#if IS_ENABLED(CONFIG_BPFILTER_UMH)
if (optname >= BPFILTER_IPT_SO_GET_INFO &&
optname < BPFILTER_IPT_GET_MAX)
err = bpfilter_ip_get_sockopt(sk, optname, optval, optlen);
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 3f4d610..f0cd291 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -1001,6 +1001,9 @@ static int __init inet6_init(void)
err = ip6_flowlabel_init();
if (err)
goto ip6_flowlabel_fail;
+ err = ipv6_anycast_init();
+ if (err)
+ goto ipv6_anycast_fail;
err = addrconf_init();
if (err)
goto addrconf_fail;
@@ -1091,6 +1094,8 @@ static int __init inet6_init(void)
ipv6_exthdrs_fail:
addrconf_cleanup();
addrconf_fail:
+ ipv6_anycast_cleanup();
+ipv6_anycast_fail:
ip6_flowlabel_cleanup();
ip6_flowlabel_fail:
ndisc_late_cleanup();
diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c
index 4e0ff70..9499905 100644
--- a/net/ipv6/anycast.c
+++ b/net/ipv6/anycast.c
@@ -44,8 +44,22 @@
#include <net/checksum.h>
+#define IN6_ADDR_HSIZE_SHIFT 8
+#define IN6_ADDR_HSIZE BIT(IN6_ADDR_HSIZE_SHIFT)
+/* anycast address hash table
+ */
+static struct hlist_head inet6_acaddr_lst[IN6_ADDR_HSIZE];
+static DEFINE_SPINLOCK(acaddr_hash_lock);
+
static int ipv6_dev_ac_dec(struct net_device *dev, const struct in6_addr *addr);
+static u32 inet6_acaddr_hash(struct net *net, const struct in6_addr *addr)
+{
+ u32 val = ipv6_addr_hash(addr) ^ net_hash_mix(net);
+
+ return hash_32(val, IN6_ADDR_HSIZE_SHIFT);
+}
+
/*
* socket join an anycast group
*/
@@ -204,16 +218,39 @@ void ipv6_sock_ac_close(struct sock *sk)
rtnl_unlock();
}
+static void ipv6_add_acaddr_hash(struct net *net, struct ifacaddr6 *aca)
+{
+ unsigned int hash = inet6_acaddr_hash(net, &aca->aca_addr);
+
+ spin_lock(&acaddr_hash_lock);
+ hlist_add_head_rcu(&aca->aca_addr_lst, &inet6_acaddr_lst[hash]);
+ spin_unlock(&acaddr_hash_lock);
+}
+
+static void ipv6_del_acaddr_hash(struct ifacaddr6 *aca)
+{
+ spin_lock(&acaddr_hash_lock);
+ hlist_del_init_rcu(&aca->aca_addr_lst);
+ spin_unlock(&acaddr_hash_lock);
+}
+
static void aca_get(struct ifacaddr6 *aca)
{
refcount_inc(&aca->aca_refcnt);
}
+static void aca_free_rcu(struct rcu_head *h)
+{
+ struct ifacaddr6 *aca = container_of(h, struct ifacaddr6, rcu);
+
+ fib6_info_release(aca->aca_rt);
+ kfree(aca);
+}
+
static void aca_put(struct ifacaddr6 *ac)
{
if (refcount_dec_and_test(&ac->aca_refcnt)) {
- fib6_info_release(ac->aca_rt);
- kfree(ac);
+ call_rcu(&ac->rcu, aca_free_rcu);
}
}
@@ -229,6 +266,7 @@ static struct ifacaddr6 *aca_alloc(struct fib6_info *f6i,
aca->aca_addr = *addr;
fib6_info_hold(f6i);
aca->aca_rt = f6i;
+ INIT_HLIST_NODE(&aca->aca_addr_lst);
aca->aca_users = 1;
/* aca_tstamp should be updated upon changes */
aca->aca_cstamp = aca->aca_tstamp = jiffies;
@@ -285,6 +323,8 @@ int __ipv6_dev_ac_inc(struct inet6_dev *idev, const struct in6_addr *addr)
aca_get(aca);
write_unlock_bh(&idev->lock);
+ ipv6_add_acaddr_hash(net, aca);
+
ip6_ins_rt(net, f6i);
addrconf_join_solict(idev->dev, &aca->aca_addr);
@@ -325,6 +365,7 @@ int __ipv6_dev_ac_dec(struct inet6_dev *idev, const struct in6_addr *addr)
else
idev->ac_list = aca->aca_next;
write_unlock_bh(&idev->lock);
+ ipv6_del_acaddr_hash(aca);
addrconf_leave_solict(idev, &aca->aca_addr);
ip6_del_rt(dev_net(idev->dev), aca->aca_rt);
@@ -352,6 +393,8 @@ void ipv6_ac_destroy_dev(struct inet6_dev *idev)
idev->ac_list = aca->aca_next;
write_unlock_bh(&idev->lock);
+ ipv6_del_acaddr_hash(aca);
+
addrconf_leave_solict(idev, &aca->aca_addr);
ip6_del_rt(dev_net(idev->dev), aca->aca_rt);
@@ -390,17 +433,25 @@ static bool ipv6_chk_acast_dev(struct net_device *dev, const struct in6_addr *ad
bool ipv6_chk_acast_addr(struct net *net, struct net_device *dev,
const struct in6_addr *addr)
{
+ unsigned int hash = inet6_acaddr_hash(net, addr);
+ struct net_device *nh_dev;
+ struct ifacaddr6 *aca;
bool found = false;
rcu_read_lock();
if (dev)
found = ipv6_chk_acast_dev(dev, addr);
else
- for_each_netdev_rcu(net, dev)
- if (ipv6_chk_acast_dev(dev, addr)) {
+ hlist_for_each_entry_rcu(aca, &inet6_acaddr_lst[hash],
+ aca_addr_lst) {
+ nh_dev = fib6_info_nh_dev(aca->aca_rt);
+ if (!nh_dev || !net_eq(dev_net(nh_dev), net))
+ continue;
+ if (ipv6_addr_equal(&aca->aca_addr, addr)) {
found = true;
break;
}
+ }
rcu_read_unlock();
return found;
}
@@ -540,3 +591,24 @@ void ac6_proc_exit(struct net *net)
remove_proc_entry("anycast6", net->proc_net);
}
#endif
+
+/* Init / cleanup code
+ */
+int __init ipv6_anycast_init(void)
+{
+ int i;
+
+ for (i = 0; i < IN6_ADDR_HSIZE; i++)
+ INIT_HLIST_HEAD(&inet6_acaddr_lst[i]);
+ return 0;
+}
+
+void ipv6_anycast_cleanup(void)
+{
+ int i;
+
+ spin_lock(&acaddr_hash_lock);
+ for (i = 0; i < IN6_ADDR_HSIZE; i++)
+ WARN_ON(!hlist_empty(&inet6_acaddr_lst[i]));
+ spin_unlock(&acaddr_hash_lock);
+}
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index 1b8bc00..ae37861 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -591,7 +591,7 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
/* fib entries are never clones */
if (arg.filter.flags & RTM_F_CLONED)
- return skb->len;
+ goto out;
w = (void *)cb->args[2];
if (!w) {
@@ -621,7 +621,7 @@ static int inet6_dump_fib(struct sk_buff *skb, struct netlink_callback *cb)
tb = fib6_get_table(net, arg.filter.table_id);
if (!tb) {
if (arg.filter.dump_all_families)
- return skb->len;
+ goto out;
NL_SET_ERR_MSG_MOD(cb->extack, "FIB table does not exist");
return -ENOENT;
diff --git a/net/ipv6/netfilter/nf_conntrack_reasm.c b/net/ipv6/netfilter/nf_conntrack_reasm.c
index b8ac369..d219979 100644
--- a/net/ipv6/netfilter/nf_conntrack_reasm.c
+++ b/net/ipv6/netfilter/nf_conntrack_reasm.c
@@ -587,11 +587,16 @@ int nf_ct_frag6_gather(struct net *net, struct sk_buff *skb, u32 user)
*/
ret = -EINPROGRESS;
if (fq->q.flags == (INET_FRAG_FIRST_IN | INET_FRAG_LAST_IN) &&
- fq->q.meat == fq->q.len &&
- nf_ct_frag6_reasm(fq, skb, dev))
- ret = 0;
- else
+ fq->q.meat == fq->q.len) {
+ unsigned long orefdst = skb->_skb_refdst;
+
+ skb->_skb_refdst = 0UL;
+ if (nf_ct_frag6_reasm(fq, skb, dev))
+ ret = 0;
+ skb->_skb_refdst = orefdst;
+ } else {
skb_dst_drop(skb);
+ }
out_unlock:
spin_unlock_bh(&fq->q.lock);
diff --git a/net/netfilter/ipset/ip_set_core.c b/net/netfilter/ipset/ip_set_core.c
index bc4bd247..1577f2f 100644
--- a/net/netfilter/ipset/ip_set_core.c
+++ b/net/netfilter/ipset/ip_set_core.c
@@ -55,11 +55,15 @@ MODULE_AUTHOR("Jozsef Kadlecsik <kadlec@blackhole.kfki.hu>");
MODULE_DESCRIPTION("core IP set support");
MODULE_ALIAS_NFNL_SUBSYS(NFNL_SUBSYS_IPSET);
-/* When the nfnl mutex is held: */
+/* When the nfnl mutex or ip_set_ref_lock is held: */
#define ip_set_dereference(p) \
- rcu_dereference_protected(p, lockdep_nfnl_is_held(NFNL_SUBSYS_IPSET))
+ rcu_dereference_protected(p, \
+ lockdep_nfnl_is_held(NFNL_SUBSYS_IPSET) || \
+ lockdep_is_held(&ip_set_ref_lock))
#define ip_set(inst, id) \
ip_set_dereference((inst)->ip_set_list)[id]
+#define ip_set_ref_netlink(inst,id) \
+ rcu_dereference_raw((inst)->ip_set_list)[id]
/* The set types are implemented in modules and registered set types
* can be found in ip_set_type_list. Adding/deleting types is
@@ -693,21 +697,20 @@ ip_set_put_byindex(struct net *net, ip_set_id_t index)
EXPORT_SYMBOL_GPL(ip_set_put_byindex);
/* Get the name of a set behind a set index.
- * We assume the set is referenced, so it does exist and
- * can't be destroyed. The set cannot be renamed due to
- * the referencing either.
- *
+ * Set itself is protected by RCU, but its name isn't: to protect against
+ * renaming, grab ip_set_ref_lock as reader (see ip_set_rename()) and copy the
+ * name.
*/
-const char *
-ip_set_name_byindex(struct net *net, ip_set_id_t index)
+void
+ip_set_name_byindex(struct net *net, ip_set_id_t index, char *name)
{
- const struct ip_set *set = ip_set_rcu_get(net, index);
+ struct ip_set *set = ip_set_rcu_get(net, index);
BUG_ON(!set);
- BUG_ON(set->ref == 0);
- /* Referenced, so it's safe */
- return set->name;
+ read_lock_bh(&ip_set_ref_lock);
+ strncpy(name, set->name, IPSET_MAXNAMELEN);
+ read_unlock_bh(&ip_set_ref_lock);
}
EXPORT_SYMBOL_GPL(ip_set_name_byindex);
@@ -961,7 +964,7 @@ static int ip_set_create(struct net *net, struct sock *ctnl,
/* Wraparound */
goto cleanup;
- list = kcalloc(i, sizeof(struct ip_set *), GFP_KERNEL);
+ list = kvcalloc(i, sizeof(struct ip_set *), GFP_KERNEL);
if (!list)
goto cleanup;
/* nfnl mutex is held, both lists are valid */
@@ -973,7 +976,7 @@ static int ip_set_create(struct net *net, struct sock *ctnl,
/* Use new list */
index = inst->ip_set_max;
inst->ip_set_max = i;
- kfree(tmp);
+ kvfree(tmp);
ret = 0;
} else if (ret) {
goto cleanup;
@@ -1153,7 +1156,7 @@ static int ip_set_rename(struct net *net, struct sock *ctnl,
if (!set)
return -ENOENT;
- read_lock_bh(&ip_set_ref_lock);
+ write_lock_bh(&ip_set_ref_lock);
if (set->ref != 0) {
ret = -IPSET_ERR_REFERENCED;
goto out;
@@ -1170,7 +1173,7 @@ static int ip_set_rename(struct net *net, struct sock *ctnl,
strncpy(set->name, name2, IPSET_MAXNAMELEN);
out:
- read_unlock_bh(&ip_set_ref_lock);
+ write_unlock_bh(&ip_set_ref_lock);
return ret;
}
@@ -1252,7 +1255,7 @@ ip_set_dump_done(struct netlink_callback *cb)
struct ip_set_net *inst =
(struct ip_set_net *)cb->args[IPSET_CB_NET];
ip_set_id_t index = (ip_set_id_t)cb->args[IPSET_CB_INDEX];
- struct ip_set *set = ip_set(inst, index);
+ struct ip_set *set = ip_set_ref_netlink(inst, index);
if (set->variant->uref)
set->variant->uref(set, cb, false);
@@ -1441,7 +1444,7 @@ ip_set_dump_start(struct sk_buff *skb, struct netlink_callback *cb)
release_refcount:
/* If there was an error or set is done, release set */
if (ret || !cb->args[IPSET_CB_ARG0]) {
- set = ip_set(inst, index);
+ set = ip_set_ref_netlink(inst, index);
if (set->variant->uref)
set->variant->uref(set, cb, false);
pr_debug("release set %s\n", set->name);
@@ -2059,7 +2062,7 @@ ip_set_net_init(struct net *net)
if (inst->ip_set_max >= IPSET_INVALID_ID)
inst->ip_set_max = IPSET_INVALID_ID - 1;
- list = kcalloc(inst->ip_set_max, sizeof(struct ip_set *), GFP_KERNEL);
+ list = kvcalloc(inst->ip_set_max, sizeof(struct ip_set *), GFP_KERNEL);
if (!list)
return -ENOMEM;
inst->is_deleted = false;
@@ -2087,7 +2090,7 @@ ip_set_net_exit(struct net *net)
}
}
nfnl_unlock(NFNL_SUBSYS_IPSET);
- kfree(rcu_dereference_protected(inst->ip_set_list, 1));
+ kvfree(rcu_dereference_protected(inst->ip_set_list, 1));
}
static struct pernet_operations ip_set_net_ops = {
diff --git a/net/netfilter/ipset/ip_set_hash_netportnet.c b/net/netfilter/ipset/ip_set_hash_netportnet.c
index d391485..613e18e 100644
--- a/net/netfilter/ipset/ip_set_hash_netportnet.c
+++ b/net/netfilter/ipset/ip_set_hash_netportnet.c
@@ -213,13 +213,13 @@ hash_netportnet4_uadt(struct ip_set *set, struct nlattr *tb[],
if (tb[IPSET_ATTR_CIDR]) {
e.cidr[0] = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (!e.cidr[0] || e.cidr[0] > HOST_MASK)
+ if (e.cidr[0] > HOST_MASK)
return -IPSET_ERR_INVALID_CIDR;
}
if (tb[IPSET_ATTR_CIDR2]) {
e.cidr[1] = nla_get_u8(tb[IPSET_ATTR_CIDR2]);
- if (!e.cidr[1] || e.cidr[1] > HOST_MASK)
+ if (e.cidr[1] > HOST_MASK)
return -IPSET_ERR_INVALID_CIDR;
}
@@ -493,13 +493,13 @@ hash_netportnet6_uadt(struct ip_set *set, struct nlattr *tb[],
if (tb[IPSET_ATTR_CIDR]) {
e.cidr[0] = nla_get_u8(tb[IPSET_ATTR_CIDR]);
- if (!e.cidr[0] || e.cidr[0] > HOST_MASK)
+ if (e.cidr[0] > HOST_MASK)
return -IPSET_ERR_INVALID_CIDR;
}
if (tb[IPSET_ATTR_CIDR2]) {
e.cidr[1] = nla_get_u8(tb[IPSET_ATTR_CIDR2]);
- if (!e.cidr[1] || e.cidr[1] > HOST_MASK)
+ if (e.cidr[1] > HOST_MASK)
return -IPSET_ERR_INVALID_CIDR;
}
diff --git a/net/netfilter/ipset/ip_set_list_set.c b/net/netfilter/ipset/ip_set_list_set.c
index 072a658..4eef55d 100644
--- a/net/netfilter/ipset/ip_set_list_set.c
+++ b/net/netfilter/ipset/ip_set_list_set.c
@@ -148,9 +148,7 @@ __list_set_del_rcu(struct rcu_head * rcu)
{
struct set_elem *e = container_of(rcu, struct set_elem, rcu);
struct ip_set *set = e->set;
- struct list_set *map = set->data;
- ip_set_put_byindex(map->net, e->id);
ip_set_ext_destroy(set, e);
kfree(e);
}
@@ -158,15 +156,21 @@ __list_set_del_rcu(struct rcu_head * rcu)
static inline void
list_set_del(struct ip_set *set, struct set_elem *e)
{
+ struct list_set *map = set->data;
+
set->elements--;
list_del_rcu(&e->list);
+ ip_set_put_byindex(map->net, e->id);
call_rcu(&e->rcu, __list_set_del_rcu);
}
static inline void
-list_set_replace(struct set_elem *e, struct set_elem *old)
+list_set_replace(struct ip_set *set, struct set_elem *e, struct set_elem *old)
{
+ struct list_set *map = set->data;
+
list_replace_rcu(&old->list, &e->list);
+ ip_set_put_byindex(map->net, old->id);
call_rcu(&old->rcu, __list_set_del_rcu);
}
@@ -298,7 +302,7 @@ list_set_uadd(struct ip_set *set, void *value, const struct ip_set_ext *ext,
INIT_LIST_HEAD(&e->list);
list_set_init_extensions(set, ext, e);
if (n)
- list_set_replace(e, n);
+ list_set_replace(set, e, n);
else if (next)
list_add_tail_rcu(&e->list, &next->list);
else if (prev)
@@ -486,6 +490,7 @@ list_set_list(const struct ip_set *set,
const struct list_set *map = set->data;
struct nlattr *atd, *nested;
u32 i = 0, first = cb->args[IPSET_CB_ARG0];
+ char name[IPSET_MAXNAMELEN];
struct set_elem *e;
int ret = 0;
@@ -504,8 +509,8 @@ list_set_list(const struct ip_set *set,
nested = ipset_nest_start(skb, IPSET_ATTR_DATA);
if (!nested)
goto nla_put_failure;
- if (nla_put_string(skb, IPSET_ATTR_NAME,
- ip_set_name_byindex(map->net, e->id)))
+ ip_set_name_byindex(map->net, e->id, name);
+ if (nla_put_string(skb, IPSET_ATTR_NAME, name))
goto nla_put_failure;
if (ip_set_put_extensions(skb, set, e, true))
goto nla_put_failure;
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index ca1168d..e92e749 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1073,19 +1073,22 @@ static unsigned int early_drop_list(struct net *net,
return drops;
}
-static noinline int early_drop(struct net *net, unsigned int _hash)
+static noinline int early_drop(struct net *net, unsigned int hash)
{
- unsigned int i;
+ unsigned int i, bucket;
for (i = 0; i < NF_CT_EVICTION_RANGE; i++) {
struct hlist_nulls_head *ct_hash;
- unsigned int hash, hsize, drops;
+ unsigned int hsize, drops;
rcu_read_lock();
nf_conntrack_get_ht(&ct_hash, &hsize);
- hash = reciprocal_scale(_hash++, hsize);
+ if (!i)
+ bucket = reciprocal_scale(hash, hsize);
+ else
+ bucket = (bucket + 1) % hsize;
- drops = early_drop_list(net, &ct_hash[hash]);
+ drops = early_drop_list(net, &ct_hash[bucket]);
rcu_read_unlock();
if (drops) {
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
index 171e9e1..023c144 100644
--- a/net/netfilter/nf_conntrack_proto_dccp.c
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -384,11 +384,6 @@ dccp_state_table[CT_DCCP_ROLE_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] =
},
};
-static inline struct nf_dccp_net *dccp_pernet(struct net *net)
-{
- return &net->ct.nf_ct_proto.dccp;
-}
-
static noinline bool
dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
const struct dccp_hdr *dh)
@@ -401,7 +396,7 @@ dccp_new(struct nf_conn *ct, const struct sk_buff *skb,
state = dccp_state_table[CT_DCCP_ROLE_CLIENT][dh->dccph_type][CT_DCCP_NONE];
switch (state) {
default:
- dn = dccp_pernet(net);
+ dn = nf_dccp_pernet(net);
if (dn->dccp_loose == 0) {
msg = "not picking up existing connection ";
goto out_invalid;
@@ -568,7 +563,7 @@ static int dccp_packet(struct nf_conn *ct, struct sk_buff *skb,
timeouts = nf_ct_timeout_lookup(ct);
if (!timeouts)
- timeouts = dccp_pernet(nf_ct_net(ct))->dccp_timeout;
+ timeouts = nf_dccp_pernet(nf_ct_net(ct))->dccp_timeout;
nf_ct_refresh_acct(ct, ctinfo, skb, timeouts[new_state]);
return NF_ACCEPT;
@@ -681,7 +676,7 @@ static int nlattr_to_dccp(struct nlattr *cda[], struct nf_conn *ct)
static int dccp_timeout_nlattr_to_obj(struct nlattr *tb[],
struct net *net, void *data)
{
- struct nf_dccp_net *dn = dccp_pernet(net);
+ struct nf_dccp_net *dn = nf_dccp_pernet(net);
unsigned int *timeouts = data;
int i;
@@ -814,7 +809,7 @@ static int dccp_kmemdup_sysctl_table(struct net *net, struct nf_proto_net *pn,
static int dccp_init_net(struct net *net)
{
- struct nf_dccp_net *dn = dccp_pernet(net);
+ struct nf_dccp_net *dn = nf_dccp_pernet(net);
struct nf_proto_net *pn = &dn->pn;
if (!pn->users) {
diff --git a/net/netfilter/nf_conntrack_proto_generic.c b/net/netfilter/nf_conntrack_proto_generic.c
index e10e867..5da19d5 100644
--- a/net/netfilter/nf_conntrack_proto_generic.c
+++ b/net/netfilter/nf_conntrack_proto_generic.c
@@ -27,11 +27,6 @@ static bool nf_generic_should_process(u8 proto)
}
}
-static inline struct nf_generic_net *generic_pernet(struct net *net)
-{
- return &net->ct.nf_ct_proto.generic;
-}
-
static bool generic_pkt_to_tuple(const struct sk_buff *skb,
unsigned int dataoff,
struct net *net, struct nf_conntrack_tuple *tuple)
@@ -58,7 +53,7 @@ static int generic_packet(struct nf_conn *ct,
}
if (!timeout)
- timeout = &generic_pernet(nf_ct_net(ct))->timeout;
+ timeout = &nf_generic_pernet(nf_ct_net(ct))->timeout;
nf_ct_refresh_acct(ct, ctinfo, skb, *timeout);
return NF_ACCEPT;
@@ -72,7 +67,7 @@ static int generic_packet(struct nf_conn *ct,
static int generic_timeout_nlattr_to_obj(struct nlattr *tb[],
struct net *net, void *data)
{
- struct nf_generic_net *gn = generic_pernet(net);
+ struct nf_generic_net *gn = nf_generic_pernet(net);
unsigned int *timeout = data;
if (!timeout)
@@ -138,7 +133,7 @@ static int generic_kmemdup_sysctl_table(struct nf_proto_net *pn,
static int generic_init_net(struct net *net)
{
- struct nf_generic_net *gn = generic_pernet(net);
+ struct nf_generic_net *gn = nf_generic_pernet(net);
struct nf_proto_net *pn = &gn->pn;
gn->timeout = nf_ct_generic_timeout;
diff --git a/net/netfilter/nf_conntrack_proto_icmp.c b/net/netfilter/nf_conntrack_proto_icmp.c
index 3598520..de64d8a 100644
--- a/net/netfilter/nf_conntrack_proto_icmp.c
+++ b/net/netfilter/nf_conntrack_proto_icmp.c
@@ -25,11 +25,6 @@
static const unsigned int nf_ct_icmp_timeout = 30*HZ;
-static inline struct nf_icmp_net *icmp_pernet(struct net *net)
-{
- return &net->ct.nf_ct_proto.icmp;
-}
-
static bool icmp_pkt_to_tuple(const struct sk_buff *skb, unsigned int dataoff,
struct net *net, struct nf_conntrack_tuple *tuple)
{
@@ -103,7 +98,7 @@ static int icmp_packet(struct nf_conn *ct,
}
if (!timeout)
- timeout = &icmp_pernet(nf_ct_net(ct))->timeout;
+ timeout = &nf_icmp_pernet(nf_ct_net(ct))->timeout;
nf_ct_refresh_acct(ct, ctinfo, skb, *timeout);
return NF_ACCEPT;
@@ -275,7 +270,7 @@ static int icmp_timeout_nlattr_to_obj(struct nlattr *tb[],
struct net *net, void *data)
{
unsigned int *timeout = data;
- struct nf_icmp_net *in = icmp_pernet(net);
+ struct nf_icmp_net *in = nf_icmp_pernet(net);
if (tb[CTA_TIMEOUT_ICMP_TIMEOUT]) {
if (!timeout)
@@ -337,7 +332,7 @@ static int icmp_kmemdup_sysctl_table(struct nf_proto_net *pn,
static int icmp_init_net(struct net *net)
{
- struct nf_icmp_net *in = icmp_pernet(net);
+ struct nf_icmp_net *in = nf_icmp_pernet(net);
struct nf_proto_net *pn = &in->pn;
in->timeout = nf_ct_icmp_timeout;
diff --git a/net/netfilter/nf_conntrack_proto_icmpv6.c b/net/netfilter/nf_conntrack_proto_icmpv6.c
index 378618f..a15eefb 100644
--- a/net/netfilter/nf_conntrack_proto_icmpv6.c
+++ b/net/netfilter/nf_conntrack_proto_icmpv6.c
@@ -30,11 +30,6 @@
static const unsigned int nf_ct_icmpv6_timeout = 30*HZ;
-static inline struct nf_icmp_net *icmpv6_pernet(struct net *net)
-{
- return &net->ct.nf_ct_proto.icmpv6;
-}
-
static bool icmpv6_pkt_to_tuple(const struct sk_buff *skb,
unsigned int dataoff,
struct net *net,
@@ -87,7 +82,7 @@ static bool icmpv6_invert_tuple(struct nf_conntrack_tuple *tuple,
static unsigned int *icmpv6_get_timeouts(struct net *net)
{
- return &icmpv6_pernet(net)->timeout;
+ return &nf_icmpv6_pernet(net)->timeout;
}
/* Returns verdict for packet, or -1 for invalid. */
@@ -286,7 +281,7 @@ static int icmpv6_timeout_nlattr_to_obj(struct nlattr *tb[],
struct net *net, void *data)
{
unsigned int *timeout = data;
- struct nf_icmp_net *in = icmpv6_pernet(net);
+ struct nf_icmp_net *in = nf_icmpv6_pernet(net);
if (!timeout)
timeout = icmpv6_get_timeouts(net);
@@ -348,7 +343,7 @@ static int icmpv6_kmemdup_sysctl_table(struct nf_proto_net *pn,
static int icmpv6_init_net(struct net *net)
{
- struct nf_icmp_net *in = icmpv6_pernet(net);
+ struct nf_icmp_net *in = nf_icmpv6_pernet(net);
struct nf_proto_net *pn = &in->pn;
in->timeout = nf_ct_icmpv6_timeout;
diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c
index 3d719d3..d53e3e7 100644
--- a/net/netfilter/nf_conntrack_proto_sctp.c
+++ b/net/netfilter/nf_conntrack_proto_sctp.c
@@ -146,11 +146,6 @@ static const u8 sctp_conntracks[2][11][SCTP_CONNTRACK_MAX] = {
}
};
-static inline struct nf_sctp_net *sctp_pernet(struct net *net)
-{
- return &net->ct.nf_ct_proto.sctp;
-}
-
#ifdef CONFIG_NF_CONNTRACK_PROCFS
/* Print out the private part of the conntrack. */
static void sctp_print_conntrack(struct seq_file *s, struct nf_conn *ct)
@@ -480,7 +475,7 @@ static int sctp_packet(struct nf_conn *ct,
timeouts = nf_ct_timeout_lookup(ct);
if (!timeouts)
- timeouts = sctp_pernet(nf_ct_net(ct))->timeouts;
+ timeouts = nf_sctp_pernet(nf_ct_net(ct))->timeouts;
nf_ct_refresh_acct(ct, ctinfo, skb, timeouts[new_state]);
@@ -599,7 +594,7 @@ static int sctp_timeout_nlattr_to_obj(struct nlattr *tb[],
struct net *net, void *data)
{
unsigned int *timeouts = data;
- struct nf_sctp_net *sn = sctp_pernet(net);
+ struct nf_sctp_net *sn = nf_sctp_pernet(net);
int i;
/* set default SCTP timeouts. */
@@ -736,7 +731,7 @@ static int sctp_kmemdup_sysctl_table(struct nf_proto_net *pn,
static int sctp_init_net(struct net *net)
{
- struct nf_sctp_net *sn = sctp_pernet(net);
+ struct nf_sctp_net *sn = nf_sctp_pernet(net);
struct nf_proto_net *pn = &sn->pn;
if (!pn->users) {
diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c
index 1bcf998..4dcbd51 100644
--- a/net/netfilter/nf_conntrack_proto_tcp.c
+++ b/net/netfilter/nf_conntrack_proto_tcp.c
@@ -272,11 +272,6 @@ static const u8 tcp_conntracks[2][6][TCP_CONNTRACK_MAX] = {
}
};
-static inline struct nf_tcp_net *tcp_pernet(struct net *net)
-{
- return &net->ct.nf_ct_proto.tcp;
-}
-
#ifdef CONFIG_NF_CONNTRACK_PROCFS
/* Print out the private part of the conntrack. */
static void tcp_print_conntrack(struct seq_file *s, struct nf_conn *ct)
@@ -475,7 +470,7 @@ static bool tcp_in_window(const struct nf_conn *ct,
const struct tcphdr *tcph)
{
struct net *net = nf_ct_net(ct);
- struct nf_tcp_net *tn = tcp_pernet(net);
+ struct nf_tcp_net *tn = nf_tcp_pernet(net);
struct ip_ct_tcp_state *sender = &state->seen[dir];
struct ip_ct_tcp_state *receiver = &state->seen[!dir];
const struct nf_conntrack_tuple *tuple = &ct->tuplehash[dir].tuple;
@@ -767,7 +762,7 @@ static noinline bool tcp_new(struct nf_conn *ct, const struct sk_buff *skb,
{
enum tcp_conntrack new_state;
struct net *net = nf_ct_net(ct);
- const struct nf_tcp_net *tn = tcp_pernet(net);
+ const struct nf_tcp_net *tn = nf_tcp_pernet(net);
const struct ip_ct_tcp_state *sender = &ct->proto.tcp.seen[0];
const struct ip_ct_tcp_state *receiver = &ct->proto.tcp.seen[1];
@@ -841,7 +836,7 @@ static int tcp_packet(struct nf_conn *ct,
const struct nf_hook_state *state)
{
struct net *net = nf_ct_net(ct);
- struct nf_tcp_net *tn = tcp_pernet(net);
+ struct nf_tcp_net *tn = nf_tcp_pernet(net);
struct nf_conntrack_tuple *tuple;
enum tcp_conntrack new_state, old_state;
unsigned int index, *timeouts;
@@ -1283,7 +1278,7 @@ static unsigned int tcp_nlattr_tuple_size(void)
static int tcp_timeout_nlattr_to_obj(struct nlattr *tb[],
struct net *net, void *data)
{
- struct nf_tcp_net *tn = tcp_pernet(net);
+ struct nf_tcp_net *tn = nf_tcp_pernet(net);
unsigned int *timeouts = data;
int i;
@@ -1508,7 +1503,7 @@ static int tcp_kmemdup_sysctl_table(struct nf_proto_net *pn,
static int tcp_init_net(struct net *net)
{
- struct nf_tcp_net *tn = tcp_pernet(net);
+ struct nf_tcp_net *tn = nf_tcp_pernet(net);
struct nf_proto_net *pn = &tn->pn;
if (!pn->users) {
diff --git a/net/netfilter/nf_conntrack_proto_udp.c b/net/netfilter/nf_conntrack_proto_udp.c
index a7aa703..c879d8d 100644
--- a/net/netfilter/nf_conntrack_proto_udp.c
+++ b/net/netfilter/nf_conntrack_proto_udp.c
@@ -32,14 +32,9 @@ static const unsigned int udp_timeouts[UDP_CT_MAX] = {
[UDP_CT_REPLIED] = 180*HZ,
};
-static inline struct nf_udp_net *udp_pernet(struct net *net)
-{
- return &net->ct.nf_ct_proto.udp;
-}
-
static unsigned int *udp_get_timeouts(struct net *net)
{
- return udp_pernet(net)->timeouts;
+ return nf_udp_pernet(net)->timeouts;
}
static void udp_error_log(const struct sk_buff *skb,
@@ -212,7 +207,7 @@ static int udp_timeout_nlattr_to_obj(struct nlattr *tb[],
struct net *net, void *data)
{
unsigned int *timeouts = data;
- struct nf_udp_net *un = udp_pernet(net);
+ struct nf_udp_net *un = nf_udp_pernet(net);
if (!timeouts)
timeouts = un->timeouts;
@@ -292,7 +287,7 @@ static int udp_kmemdup_sysctl_table(struct nf_proto_net *pn,
static int udp_init_net(struct net *net)
{
- struct nf_udp_net *un = udp_pernet(net);
+ struct nf_udp_net *un = nf_udp_pernet(net);
struct nf_proto_net *pn = &un->pn;
if (!pn->users) {
diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c
index e7a50af..a518eb1 100644
--- a/net/netfilter/nfnetlink_cttimeout.c
+++ b/net/netfilter/nfnetlink_cttimeout.c
@@ -382,7 +382,8 @@ static int cttimeout_default_set(struct net *net, struct sock *ctnl,
static int
cttimeout_default_fill_info(struct net *net, struct sk_buff *skb, u32 portid,
u32 seq, u32 type, int event, u16 l3num,
- const struct nf_conntrack_l4proto *l4proto)
+ const struct nf_conntrack_l4proto *l4proto,
+ const unsigned int *timeouts)
{
struct nlmsghdr *nlh;
struct nfgenmsg *nfmsg;
@@ -408,7 +409,7 @@ cttimeout_default_fill_info(struct net *net, struct sk_buff *skb, u32 portid,
if (!nest_parms)
goto nla_put_failure;
- ret = l4proto->ctnl_timeout.obj_to_nlattr(skb, NULL);
+ ret = l4proto->ctnl_timeout.obj_to_nlattr(skb, timeouts);
if (ret < 0)
goto nla_put_failure;
@@ -430,6 +431,7 @@ static int cttimeout_default_get(struct net *net, struct sock *ctnl,
struct netlink_ext_ack *extack)
{
const struct nf_conntrack_l4proto *l4proto;
+ unsigned int *timeouts = NULL;
struct sk_buff *skb2;
int ret, err;
__u16 l3num;
@@ -442,12 +444,44 @@ static int cttimeout_default_get(struct net *net, struct sock *ctnl,
l4num = nla_get_u8(cda[CTA_TIMEOUT_L4PROTO]);
l4proto = nf_ct_l4proto_find_get(l4num);
- /* This protocol is not supported, skip. */
- if (l4proto->l4proto != l4num) {
- err = -EOPNOTSUPP;
+ err = -EOPNOTSUPP;
+ if (l4proto->l4proto != l4num)
goto err;
+
+ switch (l4proto->l4proto) {
+ case IPPROTO_ICMP:
+ timeouts = &nf_icmp_pernet(net)->timeout;
+ break;
+ case IPPROTO_TCP:
+ timeouts = nf_tcp_pernet(net)->timeouts;
+ break;
+ case IPPROTO_UDP:
+ timeouts = nf_udp_pernet(net)->timeouts;
+ break;
+ case IPPROTO_DCCP:
+#ifdef CONFIG_NF_CT_PROTO_DCCP
+ timeouts = nf_dccp_pernet(net)->dccp_timeout;
+#endif
+ break;
+ case IPPROTO_ICMPV6:
+ timeouts = &nf_icmpv6_pernet(net)->timeout;
+ break;
+ case IPPROTO_SCTP:
+#ifdef CONFIG_NF_CT_PROTO_SCTP
+ timeouts = nf_sctp_pernet(net)->timeouts;
+#endif
+ break;
+ case 255:
+ timeouts = &nf_generic_pernet(net)->timeout;
+ break;
+ default:
+ WARN_ON_ONCE(1);
+ break;
}
+ if (!timeouts)
+ goto err;
+
skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (skb2 == NULL) {
err = -ENOMEM;
@@ -458,8 +492,7 @@ static int cttimeout_default_get(struct net *net, struct sock *ctnl,
nlh->nlmsg_seq,
NFNL_MSG_TYPE(nlh->nlmsg_type),
IPCTNL_MSG_TIMEOUT_DEFAULT_SET,
- l3num,
- l4proto);
+ l3num, l4proto, timeouts);
if (ret <= 0) {
kfree_skb(skb2);
err = -ENOMEM;
diff --git a/net/netfilter/nft_compat.c b/net/netfilter/nft_compat.c
index 768292e..9d0ede4 100644
--- a/net/netfilter/nft_compat.c
+++ b/net/netfilter/nft_compat.c
@@ -54,9 +54,11 @@ static bool nft_xt_put(struct nft_xt *xt)
return false;
}
-static int nft_compat_chain_validate_dependency(const char *tablename,
- const struct nft_chain *chain)
+static int nft_compat_chain_validate_dependency(const struct nft_ctx *ctx,
+ const char *tablename)
{
+ enum nft_chain_types type = NFT_CHAIN_T_DEFAULT;
+ const struct nft_chain *chain = ctx->chain;
const struct nft_base_chain *basechain;
if (!tablename ||
@@ -64,9 +66,12 @@ static int nft_compat_chain_validate_dependency(const char *tablename,
return 0;
basechain = nft_base_chain(chain);
- if (strcmp(tablename, "nat") == 0 &&
- basechain->type->type != NFT_CHAIN_T_NAT)
- return -EINVAL;
+ if (strcmp(tablename, "nat") == 0) {
+ if (ctx->family != NFPROTO_BRIDGE)
+ type = NFT_CHAIN_T_NAT;
+ if (basechain->type->type != type)
+ return -EINVAL;
+ }
return 0;
}
@@ -342,8 +347,7 @@ static int nft_target_validate(const struct nft_ctx *ctx,
if (target->hooks && !(hook_mask & target->hooks))
return -EINVAL;
- ret = nft_compat_chain_validate_dependency(target->table,
- ctx->chain);
+ ret = nft_compat_chain_validate_dependency(ctx, target->table);
if (ret < 0)
return ret;
}
@@ -590,8 +594,7 @@ static int nft_match_validate(const struct nft_ctx *ctx,
if (match->hooks && !(hook_mask & match->hooks))
return -EINVAL;
- ret = nft_compat_chain_validate_dependency(match->table,
- ctx->chain);
+ ret = nft_compat_chain_validate_dependency(ctx, match->table);
if (ret < 0)
return ret;
}
diff --git a/net/netfilter/nft_numgen.c b/net/netfilter/nft_numgen.c
index 649d170..3cc1b3d 100644
--- a/net/netfilter/nft_numgen.c
+++ b/net/netfilter/nft_numgen.c
@@ -24,7 +24,6 @@ struct nft_ng_inc {
u32 modulus;
atomic_t counter;
u32 offset;
- struct nft_set *map;
};
static u32 nft_ng_inc_gen(struct nft_ng_inc *priv)
@@ -48,34 +47,11 @@ static void nft_ng_inc_eval(const struct nft_expr *expr,
regs->data[priv->dreg] = nft_ng_inc_gen(priv);
}
-static void nft_ng_inc_map_eval(const struct nft_expr *expr,
- struct nft_regs *regs,
- const struct nft_pktinfo *pkt)
-{
- struct nft_ng_inc *priv = nft_expr_priv(expr);
- const struct nft_set *map = priv->map;
- const struct nft_set_ext *ext;
- u32 result;
- bool found;
-
- result = nft_ng_inc_gen(priv);
- found = map->ops->lookup(nft_net(pkt), map, &result, &ext);
-
- if (!found)
- return;
-
- nft_data_copy(®s->data[priv->dreg],
- nft_set_ext_data(ext), map->dlen);
-}
-
static const struct nla_policy nft_ng_policy[NFTA_NG_MAX + 1] = {
[NFTA_NG_DREG] = { .type = NLA_U32 },
[NFTA_NG_MODULUS] = { .type = NLA_U32 },
[NFTA_NG_TYPE] = { .type = NLA_U32 },
[NFTA_NG_OFFSET] = { .type = NLA_U32 },
- [NFTA_NG_SET_NAME] = { .type = NLA_STRING,
- .len = NFT_SET_MAXNAMELEN - 1 },
- [NFTA_NG_SET_ID] = { .type = NLA_U32 },
};
static int nft_ng_inc_init(const struct nft_ctx *ctx,
@@ -101,22 +77,6 @@ static int nft_ng_inc_init(const struct nft_ctx *ctx,
NFT_DATA_VALUE, sizeof(u32));
}
-static int nft_ng_inc_map_init(const struct nft_ctx *ctx,
- const struct nft_expr *expr,
- const struct nlattr * const tb[])
-{
- struct nft_ng_inc *priv = nft_expr_priv(expr);
- u8 genmask = nft_genmask_next(ctx->net);
-
- nft_ng_inc_init(ctx, expr, tb);
-
- priv->map = nft_set_lookup_global(ctx->net, ctx->table,
- tb[NFTA_NG_SET_NAME],
- tb[NFTA_NG_SET_ID], genmask);
-
- return PTR_ERR_OR_ZERO(priv->map);
-}
-
static int nft_ng_dump(struct sk_buff *skb, enum nft_registers dreg,
u32 modulus, enum nft_ng_types type, u32 offset)
{
@@ -143,27 +103,10 @@ static int nft_ng_inc_dump(struct sk_buff *skb, const struct nft_expr *expr)
priv->offset);
}
-static int nft_ng_inc_map_dump(struct sk_buff *skb,
- const struct nft_expr *expr)
-{
- const struct nft_ng_inc *priv = nft_expr_priv(expr);
-
- if (nft_ng_dump(skb, priv->dreg, priv->modulus,
- NFT_NG_INCREMENTAL, priv->offset) ||
- nla_put_string(skb, NFTA_NG_SET_NAME, priv->map->name))
- goto nla_put_failure;
-
- return 0;
-
-nla_put_failure:
- return -1;
-}
-
struct nft_ng_random {
enum nft_registers dreg:8;
u32 modulus;
u32 offset;
- struct nft_set *map;
};
static u32 nft_ng_random_gen(struct nft_ng_random *priv)
@@ -183,25 +126,6 @@ static void nft_ng_random_eval(const struct nft_expr *expr,
regs->data[priv->dreg] = nft_ng_random_gen(priv);
}
-static void nft_ng_random_map_eval(const struct nft_expr *expr,
- struct nft_regs *regs,
- const struct nft_pktinfo *pkt)
-{
- struct nft_ng_random *priv = nft_expr_priv(expr);
- const struct nft_set *map = priv->map;
- const struct nft_set_ext *ext;
- u32 result;
- bool found;
-
- result = nft_ng_random_gen(priv);
- found = map->ops->lookup(nft_net(pkt), map, &result, &ext);
- if (!found)
- return;
-
- nft_data_copy(®s->data[priv->dreg],
- nft_set_ext_data(ext), map->dlen);
-}
-
static int nft_ng_random_init(const struct nft_ctx *ctx,
const struct nft_expr *expr,
const struct nlattr * const tb[])
@@ -226,21 +150,6 @@ static int nft_ng_random_init(const struct nft_ctx *ctx,
NFT_DATA_VALUE, sizeof(u32));
}
-static int nft_ng_random_map_init(const struct nft_ctx *ctx,
- const struct nft_expr *expr,
- const struct nlattr * const tb[])
-{
- struct nft_ng_random *priv = nft_expr_priv(expr);
- u8 genmask = nft_genmask_next(ctx->net);
-
- nft_ng_random_init(ctx, expr, tb);
- priv->map = nft_set_lookup_global(ctx->net, ctx->table,
- tb[NFTA_NG_SET_NAME],
- tb[NFTA_NG_SET_ID], genmask);
-
- return PTR_ERR_OR_ZERO(priv->map);
-}
-
static int nft_ng_random_dump(struct sk_buff *skb, const struct nft_expr *expr)
{
const struct nft_ng_random *priv = nft_expr_priv(expr);
@@ -249,22 +158,6 @@ static int nft_ng_random_dump(struct sk_buff *skb, const struct nft_expr *expr)
priv->offset);
}
-static int nft_ng_random_map_dump(struct sk_buff *skb,
- const struct nft_expr *expr)
-{
- const struct nft_ng_random *priv = nft_expr_priv(expr);
-
- if (nft_ng_dump(skb, priv->dreg, priv->modulus,
- NFT_NG_RANDOM, priv->offset) ||
- nla_put_string(skb, NFTA_NG_SET_NAME, priv->map->name))
- goto nla_put_failure;
-
- return 0;
-
-nla_put_failure:
- return -1;
-}
-
static struct nft_expr_type nft_ng_type;
static const struct nft_expr_ops nft_ng_inc_ops = {
.type = &nft_ng_type,
@@ -274,14 +167,6 @@ static const struct nft_expr_ops nft_ng_inc_ops = {
.dump = nft_ng_inc_dump,
};
-static const struct nft_expr_ops nft_ng_inc_map_ops = {
- .type = &nft_ng_type,
- .size = NFT_EXPR_SIZE(sizeof(struct nft_ng_inc)),
- .eval = nft_ng_inc_map_eval,
- .init = nft_ng_inc_map_init,
- .dump = nft_ng_inc_map_dump,
-};
-
static const struct nft_expr_ops nft_ng_random_ops = {
.type = &nft_ng_type,
.size = NFT_EXPR_SIZE(sizeof(struct nft_ng_random)),
@@ -290,14 +175,6 @@ static const struct nft_expr_ops nft_ng_random_ops = {
.dump = nft_ng_random_dump,
};
-static const struct nft_expr_ops nft_ng_random_map_ops = {
- .type = &nft_ng_type,
- .size = NFT_EXPR_SIZE(sizeof(struct nft_ng_random)),
- .eval = nft_ng_random_map_eval,
- .init = nft_ng_random_map_init,
- .dump = nft_ng_random_map_dump,
-};
-
static const struct nft_expr_ops *
nft_ng_select_ops(const struct nft_ctx *ctx, const struct nlattr * const tb[])
{
@@ -312,12 +189,8 @@ nft_ng_select_ops(const struct nft_ctx *ctx, const struct nlattr * const tb[])
switch (type) {
case NFT_NG_INCREMENTAL:
- if (tb[NFTA_NG_SET_NAME])
- return &nft_ng_inc_map_ops;
return &nft_ng_inc_ops;
case NFT_NG_RANDOM:
- if (tb[NFTA_NG_SET_NAME])
- return &nft_ng_random_map_ops;
return &nft_ng_random_ops;
}
diff --git a/net/netfilter/nft_osf.c b/net/netfilter/nft_osf.c
index ca5e5d8..b13618c 100644
--- a/net/netfilter/nft_osf.c
+++ b/net/netfilter/nft_osf.c
@@ -50,7 +50,7 @@ static int nft_osf_init(const struct nft_ctx *ctx,
int err;
u8 ttl;
- if (nla_get_u8(tb[NFTA_OSF_TTL])) {
+ if (tb[NFTA_OSF_TTL]) {
ttl = nla_get_u8(tb[NFTA_OSF_TTL]);
if (ttl > 2)
return -EINVAL;
diff --git a/net/netfilter/xt_IDLETIMER.c b/net/netfilter/xt_IDLETIMER.c
index c6acfc2..eb4cbd2 100644
--- a/net/netfilter/xt_IDLETIMER.c
+++ b/net/netfilter/xt_IDLETIMER.c
@@ -114,6 +114,22 @@ static void idletimer_tg_expired(struct timer_list *t)
schedule_work(&timer->work);
}
+static int idletimer_check_sysfs_name(const char *name, unsigned int size)
+{
+ int ret;
+
+ ret = xt_check_proc_name(name, size);
+ if (ret < 0)
+ return ret;
+
+ if (!strcmp(name, "power") ||
+ !strcmp(name, "subsystem") ||
+ !strcmp(name, "uevent"))
+ return -EINVAL;
+
+ return 0;
+}
+
static int idletimer_tg_create(struct idletimer_tg_info *info)
{
int ret;
@@ -124,6 +140,10 @@ static int idletimer_tg_create(struct idletimer_tg_info *info)
goto out;
}
+ ret = idletimer_check_sysfs_name(info->label, sizeof(info->label));
+ if (ret < 0)
+ goto out_free_timer;
+
sysfs_attr_init(&info->timer->attr.attr);
info->timer->attr.attr.name = kstrdup(info->label, GFP_KERNEL);
if (!info->timer->attr.attr.name) {
diff --git a/net/openvswitch/conntrack.c b/net/openvswitch/conntrack.c
index 6bec37a..a4660c4 100644
--- a/net/openvswitch/conntrack.c
+++ b/net/openvswitch/conntrack.c
@@ -1203,7 +1203,8 @@ static int ovs_ct_commit(struct net *net, struct sw_flow_key *key,
&info->labels.mask);
if (err)
return err;
- } else if (labels_nonzero(&info->labels.mask)) {
+ } else if (IS_ENABLED(CONFIG_NF_CONNTRACK_LABELS) &&
+ labels_nonzero(&info->labels.mask)) {
err = ovs_ct_set_labels(ct, key, &info->labels.value,
&info->labels.mask);
if (err)
diff --git a/net/rxrpc/ar-internal.h b/net/rxrpc/ar-internal.h
index 382196e..bc628ac 100644
--- a/net/rxrpc/ar-internal.h
+++ b/net/rxrpc/ar-internal.h
@@ -611,6 +611,7 @@ struct rxrpc_call {
* not hard-ACK'd packet follows this.
*/
rxrpc_seq_t tx_top; /* Highest Tx slot allocated. */
+ u16 tx_backoff; /* Delay to insert due to Tx failure */
/* TCP-style slow-start congestion control [RFC5681]. Since the SMSS
* is fixed, we keep these numbers in terms of segments (ie. DATA
diff --git a/net/rxrpc/call_event.c b/net/rxrpc/call_event.c
index 8e7434e..468efc3 100644
--- a/net/rxrpc/call_event.c
+++ b/net/rxrpc/call_event.c
@@ -123,6 +123,7 @@ static void __rxrpc_propose_ACK(struct rxrpc_call *call, u8 ack_reason,
else
ack_at = expiry;
+ ack_at += READ_ONCE(call->tx_backoff);
ack_at += now;
if (time_before(ack_at, call->ack_at)) {
WRITE_ONCE(call->ack_at, ack_at);
@@ -311,6 +312,7 @@ void rxrpc_process_call(struct work_struct *work)
container_of(work, struct rxrpc_call, processor);
rxrpc_serial_t *send_ack;
unsigned long now, next, t;
+ unsigned int iterations = 0;
rxrpc_see_call(call);
@@ -319,6 +321,11 @@ void rxrpc_process_call(struct work_struct *work)
call->debug_id, rxrpc_call_states[call->state], call->events);
recheck_state:
+ /* Limit the number of times we do this before returning to the manager */
+ iterations++;
+ if (iterations > 5)
+ goto requeue;
+
if (test_and_clear_bit(RXRPC_CALL_EV_ABORT, &call->events)) {
rxrpc_send_abort_packet(call);
goto recheck_state;
@@ -447,13 +454,16 @@ void rxrpc_process_call(struct work_struct *work)
rxrpc_reduce_call_timer(call, next, now, rxrpc_timer_restart);
/* other events may have been raised since we started checking */
- if (call->events && call->state < RXRPC_CALL_COMPLETE) {
- __rxrpc_queue_call(call);
- goto out;
- }
+ if (call->events && call->state < RXRPC_CALL_COMPLETE)
+ goto requeue;
out_put:
rxrpc_put_call(call, rxrpc_call_put);
out:
_leave("");
+ return;
+
+requeue:
+ __rxrpc_queue_call(call);
+ goto out;
}
diff --git a/net/rxrpc/output.c b/net/rxrpc/output.c
index 1894188..736aa92 100644
--- a/net/rxrpc/output.c
+++ b/net/rxrpc/output.c
@@ -35,6 +35,21 @@ struct rxrpc_abort_buffer {
static const char rxrpc_keepalive_string[] = "";
/*
+ * Increase Tx backoff on transmission failure and clear it on success.
+ */
+static void rxrpc_tx_backoff(struct rxrpc_call *call, int ret)
+{
+ if (ret < 0) {
+ u16 tx_backoff = READ_ONCE(call->tx_backoff);
+
+ if (tx_backoff < HZ)
+ WRITE_ONCE(call->tx_backoff, tx_backoff + 1);
+ } else {
+ WRITE_ONCE(call->tx_backoff, 0);
+ }
+}
+
+/*
* Arrange for a keepalive ping a certain time after we last transmitted. This
* lets the far side know we're still interested in this call and helps keep
* the route through any intervening firewall open.
@@ -210,6 +225,7 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping,
else
trace_rxrpc_tx_packet(call->debug_id, &pkt->whdr,
rxrpc_tx_point_call_ack);
+ rxrpc_tx_backoff(call, ret);
if (call->state < RXRPC_CALL_COMPLETE) {
if (ret < 0) {
@@ -218,7 +234,7 @@ int rxrpc_send_ack_packet(struct rxrpc_call *call, bool ping,
rxrpc_propose_ACK(call, pkt->ack.reason,
ntohs(pkt->ack.maxSkew),
ntohl(pkt->ack.serial),
- true, true,
+ false, true,
rxrpc_propose_ack_retry_tx);
} else {
spin_lock_bh(&call->lock);
@@ -300,7 +316,7 @@ int rxrpc_send_abort_packet(struct rxrpc_call *call)
else
trace_rxrpc_tx_packet(call->debug_id, &pkt.whdr,
rxrpc_tx_point_call_abort);
-
+ rxrpc_tx_backoff(call, ret);
rxrpc_put_connection(conn);
return ret;
@@ -413,6 +429,7 @@ int rxrpc_send_data_packet(struct rxrpc_call *call, struct sk_buff *skb,
else
trace_rxrpc_tx_packet(call->debug_id, &whdr,
rxrpc_tx_point_call_data_nofrag);
+ rxrpc_tx_backoff(call, ret);
if (ret == -EMSGSIZE)
goto send_fragmentable;
@@ -445,9 +462,18 @@ int rxrpc_send_data_packet(struct rxrpc_call *call, struct sk_buff *skb,
rxrpc_reduce_call_timer(call, expect_rx_by, nowj,
rxrpc_timer_set_for_normal);
}
- }
- rxrpc_set_keepalive(call);
+ rxrpc_set_keepalive(call);
+ } else {
+ /* Cancel the call if the initial transmission fails,
+ * particularly if that's due to network routing issues that
+ * aren't going away anytime soon. The layer above can arrange
+ * the retransmission.
+ */
+ if (!test_and_set_bit(RXRPC_CALL_BEGAN_RX_TIMER, &call->flags))
+ rxrpc_set_call_completion(call, RXRPC_CALL_LOCAL_ERROR,
+ RX_USER_ABORT, ret);
+ }
_leave(" = %d [%u]", ret, call->peer->maxdata);
return ret;
@@ -506,6 +532,7 @@ int rxrpc_send_data_packet(struct rxrpc_call *call, struct sk_buff *skb,
else
trace_rxrpc_tx_packet(call->debug_id, &whdr,
rxrpc_tx_point_call_data_frag);
+ rxrpc_tx_backoff(call, ret);
up_write(&conn->params.local->defrag_sem);
goto done;
diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c
index 1dae5f2..c8cf4d1 100644
--- a/net/sched/act_mirred.c
+++ b/net/sched/act_mirred.c
@@ -258,7 +258,8 @@ static int tcf_mirred_act(struct sk_buff *skb, const struct tc_action *a,
if (is_redirect) {
skb2->tc_redirected = 1;
skb2->tc_from_ingress = skb2->tc_at_ingress;
-
+ if (skb2->tc_from_ingress)
+ skb2->tstamp = 0;
/* let's the caller reinsert the packet, if possible */
if (use_reinsert) {
res->ingress = want_ingress;
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 9aada2d..c6c3278 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -709,11 +709,23 @@ static int fl_set_enc_opt(struct nlattr **tb, struct fl_flow_key *key,
struct netlink_ext_ack *extack)
{
const struct nlattr *nla_enc_key, *nla_opt_key, *nla_opt_msk = NULL;
- int option_len, key_depth, msk_depth = 0;
+ int err, option_len, key_depth, msk_depth = 0;
+
+ err = nla_validate_nested(tb[TCA_FLOWER_KEY_ENC_OPTS],
+ TCA_FLOWER_KEY_ENC_OPTS_MAX,
+ enc_opts_policy, extack);
+ if (err)
+ return err;
nla_enc_key = nla_data(tb[TCA_FLOWER_KEY_ENC_OPTS]);
if (tb[TCA_FLOWER_KEY_ENC_OPTS_MASK]) {
+ err = nla_validate_nested(tb[TCA_FLOWER_KEY_ENC_OPTS_MASK],
+ TCA_FLOWER_KEY_ENC_OPTS_MAX,
+ enc_opts_policy, extack);
+ if (err)
+ return err;
+
nla_opt_msk = nla_data(tb[TCA_FLOWER_KEY_ENC_OPTS_MASK]);
msk_depth = nla_len(tb[TCA_FLOWER_KEY_ENC_OPTS_MASK]);
}
diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c
index 57b3ad9..2c38e3d 100644
--- a/net/sched/sch_netem.c
+++ b/net/sched/sch_netem.c
@@ -648,15 +648,6 @@ static struct sk_buff *netem_dequeue(struct Qdisc *sch)
*/
skb->dev = qdisc_dev(sch);
-#ifdef CONFIG_NET_CLS_ACT
- /*
- * If it's at ingress let's pretend the delay is
- * from the network (tstamp will be updated).
- */
- if (skb->tc_redirected && skb->tc_from_ingress)
- skb->tstamp = 0;
-#endif
-
if (q->slot.slot_next) {
q->slot.packets_left--;
q->slot.bytes_left -= qdisc_pkt_len(skb);
diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
index 9cb854b..c37e1c2 100644
--- a/net/sctp/outqueue.c
+++ b/net/sctp/outqueue.c
@@ -212,7 +212,7 @@ void sctp_outq_init(struct sctp_association *asoc, struct sctp_outq *q)
INIT_LIST_HEAD(&q->retransmit);
INIT_LIST_HEAD(&q->sacked);
INIT_LIST_HEAD(&q->abandoned);
- sctp_sched_set_sched(asoc, SCTP_SS_FCFS);
+ sctp_sched_set_sched(asoc, SCTP_SS_DEFAULT);
}
/* Free the outqueue structure and any related pending chunks.
diff --git a/net/sunrpc/auth_generic.c b/net/sunrpc/auth_generic.c
index d8831b9..ab4a3be 100644
--- a/net/sunrpc/auth_generic.c
+++ b/net/sunrpc/auth_generic.c
@@ -281,13 +281,7 @@ static bool generic_key_to_expire(struct rpc_cred *cred)
{
struct auth_cred *acred = &container_of(cred, struct generic_cred,
gc_base)->acred;
- bool ret;
-
- get_rpccred(cred);
- ret = test_bit(RPC_CRED_KEY_EXPIRE_SOON, &acred->ac_flags);
- put_rpccred(cred);
-
- return ret;
+ return test_bit(RPC_CRED_KEY_EXPIRE_SOON, &acred->ac_flags);
}
static const struct rpc_credops generic_credops = {
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index 30f970c..5d3f252 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -1239,36 +1239,59 @@ gss_create(const struct rpc_auth_create_args *args, struct rpc_clnt *clnt)
return &gss_auth->rpc_auth;
}
+static struct gss_cred *
+gss_dup_cred(struct gss_auth *gss_auth, struct gss_cred *gss_cred)
+{
+ struct gss_cred *new;
+
+ /* Make a copy of the cred so that we can reference count it */
+ new = kzalloc(sizeof(*gss_cred), GFP_NOIO);
+ if (new) {
+ struct auth_cred acred = {
+ .uid = gss_cred->gc_base.cr_uid,
+ };
+ struct gss_cl_ctx *ctx =
+ rcu_dereference_protected(gss_cred->gc_ctx, 1);
+
+ rpcauth_init_cred(&new->gc_base, &acred,
+ &gss_auth->rpc_auth,
+ &gss_nullops);
+ new->gc_base.cr_flags = 1UL << RPCAUTH_CRED_UPTODATE;
+ new->gc_service = gss_cred->gc_service;
+ new->gc_principal = gss_cred->gc_principal;
+ kref_get(&gss_auth->kref);
+ rcu_assign_pointer(new->gc_ctx, ctx);
+ gss_get_ctx(ctx);
+ }
+ return new;
+}
+
/*
- * gss_destroying_context will cause the RPCSEC_GSS to send a NULL RPC call
+ * gss_send_destroy_context will cause the RPCSEC_GSS to send a NULL RPC call
* to the server with the GSS control procedure field set to
* RPC_GSS_PROC_DESTROY. This should normally cause the server to release
* all RPCSEC_GSS state associated with that context.
*/
-static int
-gss_destroying_context(struct rpc_cred *cred)
+static void
+gss_send_destroy_context(struct rpc_cred *cred)
{
struct gss_cred *gss_cred = container_of(cred, struct gss_cred, gc_base);
struct gss_auth *gss_auth = container_of(cred->cr_auth, struct gss_auth, rpc_auth);
struct gss_cl_ctx *ctx = rcu_dereference_protected(gss_cred->gc_ctx, 1);
+ struct gss_cred *new;
struct rpc_task *task;
- if (test_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags) == 0)
- return 0;
+ new = gss_dup_cred(gss_auth, gss_cred);
+ if (new) {
+ ctx->gc_proc = RPC_GSS_PROC_DESTROY;
- ctx->gc_proc = RPC_GSS_PROC_DESTROY;
- cred->cr_ops = &gss_nullops;
+ task = rpc_call_null(gss_auth->client, &new->gc_base,
+ RPC_TASK_ASYNC|RPC_TASK_SOFT);
+ if (!IS_ERR(task))
+ rpc_put_task(task);
- /* Take a reference to ensure the cred will be destroyed either
- * by the RPC call or by the put_rpccred() below */
- get_rpccred(cred);
-
- task = rpc_call_null(gss_auth->client, cred, RPC_TASK_ASYNC|RPC_TASK_SOFT);
- if (!IS_ERR(task))
- rpc_put_task(task);
-
- put_rpccred(cred);
- return 1;
+ put_rpccred(&new->gc_base);
+ }
}
/* gss_destroy_cred (and gss_free_ctx) are used to clean up after failure
@@ -1330,8 +1353,8 @@ static void
gss_destroy_cred(struct rpc_cred *cred)
{
- if (gss_destroying_context(cred))
- return;
+ if (test_and_clear_bit(RPCAUTH_CRED_UPTODATE, &cred->cr_flags) != 0)
+ gss_send_destroy_context(cred);
gss_destroy_nullcred(cred);
}
diff --git a/net/sunrpc/xdr.c b/net/sunrpc/xdr.c
index 2bbb8d3..f302c6e 100644
--- a/net/sunrpc/xdr.c
+++ b/net/sunrpc/xdr.c
@@ -546,7 +546,7 @@ EXPORT_SYMBOL_GPL(xdr_commit_encode);
static __be32 *xdr_get_next_encode_buffer(struct xdr_stream *xdr,
size_t nbytes)
{
- static __be32 *p;
+ __be32 *p;
int space_left;
int frag1bytes, frag2bytes;
@@ -673,11 +673,10 @@ void xdr_truncate_encode(struct xdr_stream *xdr, size_t len)
WARN_ON_ONCE(xdr->iov);
return;
}
- if (fraglen) {
+ if (fraglen)
xdr->end = head->iov_base + head->iov_len;
- xdr->page_ptr--;
- }
/* (otherwise assume xdr->end is already set) */
+ xdr->page_ptr--;
head->iov_len = len;
buf->len = len;
xdr->p = head->iov_base + head->iov_len;
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 201c3b5..836727e 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1594,14 +1594,17 @@ static int tipc_link_proto_rcv(struct tipc_link *l, struct sk_buff *skb,
if (in_range(peers_prio, l->priority + 1, TIPC_MAX_LINK_PRI))
l->priority = peers_prio;
- /* ACTIVATE_MSG serves as PEER_RESET if link is already down */
- if (msg_peer_stopping(hdr))
+ /* If peer is going down we want full re-establish cycle */
+ if (msg_peer_stopping(hdr)) {
rc = tipc_link_fsm_evt(l, LINK_FAILURE_EVT);
- else if ((mtyp == RESET_MSG) || !link_is_up(l))
+ break;
+ }
+ /* ACTIVATE_MSG serves as PEER_RESET if link is already down */
+ if (mtyp == RESET_MSG || !link_is_up(l))
rc = tipc_link_fsm_evt(l, LINK_PEER_RESET_EVT);
/* ACTIVATE_MSG takes up link if it was already locally reset */
- if ((mtyp == ACTIVATE_MSG) && (l->state == LINK_ESTABLISHING))
+ if (mtyp == ACTIVATE_MSG && l->state == LINK_ESTABLISHING)
rc = TIPC_LINK_UP_EVT;
l->peer_session = msg_session(hdr);
diff --git a/scripts/faddr2line b/scripts/faddr2line
index a0149db..6c6439f 100755
--- a/scripts/faddr2line
+++ b/scripts/faddr2line
@@ -71,7 +71,7 @@
# Try to figure out the source directory prefix so we can remove it from the
# addr2line output. HACK ALERT: This assumes that start_kernel() is in
-# kernel/init.c! This only works for vmlinux. Otherwise it falls back to
+# init/main.c! This only works for vmlinux. Otherwise it falls back to
# printing the absolute path.
find_dir_prefix() {
local objfile=$1
diff --git a/scripts/kconfig/merge_config.sh b/scripts/kconfig/merge_config.sh
index da66e77..0ef9064 100755
--- a/scripts/kconfig/merge_config.sh
+++ b/scripts/kconfig/merge_config.sh
@@ -102,7 +102,8 @@
fi
MERGE_LIST=$*
-SED_CONFIG_EXP="s/^\(# \)\{0,1\}\(${CONFIG_PREFIX}[a-zA-Z0-9_]*\)[= ].*/\2/p"
+SED_CONFIG_EXP1="s/^\(${CONFIG_PREFIX}[a-zA-Z0-9_]*\)=.*/\1/p"
+SED_CONFIG_EXP2="s/^# \(${CONFIG_PREFIX}[a-zA-Z0-9_]*\) is not set$/\1/p"
TMP_FILE=$(mktemp ./.tmp.config.XXXXXXXXXX)
@@ -116,7 +117,7 @@
echo "The merge file '$MERGE_FILE' does not exist. Exit." >&2
exit 1
fi
- CFG_LIST=$(sed -n "$SED_CONFIG_EXP" $MERGE_FILE)
+ CFG_LIST=$(sed -n -e "$SED_CONFIG_EXP1" -e "$SED_CONFIG_EXP2" $MERGE_FILE)
for CFG in $CFG_LIST ; do
grep -q -w $CFG $TMP_FILE || continue
@@ -159,7 +160,7 @@
# Check all specified config values took (might have missed-dependency issues)
-for CFG in $(sed -n "$SED_CONFIG_EXP" $TMP_FILE); do
+for CFG in $(sed -n -e "$SED_CONFIG_EXP1" -e "$SED_CONFIG_EXP2" $TMP_FILE); do
REQUESTED_VAL=$(grep -w -e "$CFG" $TMP_FILE)
ACTUAL_VAL=$(grep -w -e "$CFG" "$KCONFIG_CONFIG")
diff --git a/scripts/package/builddeb b/scripts/package/builddeb
index 90c9a8a..f43a274 100755
--- a/scripts/package/builddeb
+++ b/scripts/package/builddeb
@@ -81,11 +81,11 @@
cp System.map "$tmpdir/boot/System.map-$version"
cp $KCONFIG_CONFIG "$tmpdir/boot/config-$version"
fi
-cp "$($MAKE -s image_name)" "$tmpdir/$installed_image_path"
+cp "$($MAKE -s -f $srctree/Makefile image_name)" "$tmpdir/$installed_image_path"
-if grep -q "^CONFIG_OF=y" $KCONFIG_CONFIG ; then
+if grep -q "^CONFIG_OF_EARLY_FLATTREE=y" $KCONFIG_CONFIG ; then
# Only some architectures with OF support have this target
- if grep -q dtbs_install "${srctree}/arch/$SRCARCH/Makefile"; then
+ if [ -d "${srctree}/arch/$SRCARCH/boot/dts" ]; then
$MAKE KBUILD_SRC= INSTALL_DTBS_PATH="$tmpdir/usr/lib/$packagename" dtbs_install
fi
fi
diff --git a/scripts/package/mkdebian b/scripts/package/mkdebian
index 663a7f3..edcad61 100755
--- a/scripts/package/mkdebian
+++ b/scripts/package/mkdebian
@@ -88,6 +88,7 @@
version=$KERNELRELEASE
if [ -n "$KDEB_PKGVERSION" ]; then
packageversion=$KDEB_PKGVERSION
+ revision=${packageversion##*-}
else
revision=$(cat .version 2>/dev/null||echo 1)
packageversion=$version-$revision
@@ -205,10 +206,12 @@
#!$(command -v $MAKE) -f
build:
- \$(MAKE) KERNELRELEASE=${version} ARCH=${ARCH} KBUILD_SRC=
+ \$(MAKE) KERNELRELEASE=${version} ARCH=${ARCH} \
+ KBUILD_BUILD_VERSION=${revision} KBUILD_SRC=
binary-arch:
- \$(MAKE) KERNELRELEASE=${version} ARCH=${ARCH} KBUILD_SRC= intdeb-pkg
+ \$(MAKE) KERNELRELEASE=${version} ARCH=${ARCH} \
+ KBUILD_BUILD_VERSION=${revision} KBUILD_SRC= intdeb-pkg
clean:
rm -rf debian/*tmp debian/files
diff --git a/scripts/package/mkspec b/scripts/package/mkspec
index e05646d..009147d 100755
--- a/scripts/package/mkspec
+++ b/scripts/package/mkspec
@@ -12,6 +12,7 @@
# how we were called determines which rpms we build and how we build them
if [ "$1" = prebuilt ]; then
S=DEL
+ MAKE="$MAKE -f $srctree/Makefile"
else
S=
fi
@@ -78,19 +79,19 @@
$S %setup -q
$S
$S %build
-$S make %{?_smp_mflags} KBUILD_BUILD_VERSION=%{release}
+$S $MAKE %{?_smp_mflags} KBUILD_BUILD_VERSION=%{release}
$S
%install
mkdir -p %{buildroot}/boot
%ifarch ia64
mkdir -p %{buildroot}/boot/efi
- cp \$(make image_name) %{buildroot}/boot/efi/vmlinuz-$KERNELRELEASE
+ cp \$($MAKE image_name) %{buildroot}/boot/efi/vmlinuz-$KERNELRELEASE
ln -s efi/vmlinuz-$KERNELRELEASE %{buildroot}/boot/
%else
- cp \$(make image_name) %{buildroot}/boot/vmlinuz-$KERNELRELEASE
+ cp \$($MAKE image_name) %{buildroot}/boot/vmlinuz-$KERNELRELEASE
%endif
-$M make %{?_smp_mflags} INSTALL_MOD_PATH=%{buildroot} KBUILD_SRC= modules_install
- make %{?_smp_mflags} INSTALL_HDR_PATH=%{buildroot}/usr KBUILD_SRC= headers_install
+$M $MAKE %{?_smp_mflags} INSTALL_MOD_PATH=%{buildroot} modules_install
+ $MAKE %{?_smp_mflags} INSTALL_HDR_PATH=%{buildroot}/usr headers_install
cp System.map %{buildroot}/boot/System.map-$KERNELRELEASE
cp .config %{buildroot}/boot/config-$KERNELRELEASE
bzip2 -9 --keep vmlinux
diff --git a/scripts/setlocalversion b/scripts/setlocalversion
index 79f7dd5..71f3941 100755
--- a/scripts/setlocalversion
+++ b/scripts/setlocalversion
@@ -74,7 +74,7 @@
fi
# Check for uncommitted changes
- if git status -uno --porcelain | grep -qv '^.. scripts/package'; then
+ if git diff-index --name-only HEAD | grep -qv "^scripts/package"; then
printf '%s' -dirty
fi
diff --git a/scripts/spdxcheck.py b/scripts/spdxcheck.py
index 839e190..5056fb3 100755
--- a/scripts/spdxcheck.py
+++ b/scripts/spdxcheck.py
@@ -168,7 +168,6 @@
self.curline = 0
try:
for line in fd:
- line = line.decode(locale.getpreferredencoding(False), errors='ignore')
self.curline += 1
if self.curline > maxlines:
break
diff --git a/security/integrity/digsig_asymmetric.c b/security/integrity/digsig_asymmetric.c
index 6dc0751..d775e03 100644
--- a/security/integrity/digsig_asymmetric.c
+++ b/security/integrity/digsig_asymmetric.c
@@ -106,6 +106,7 @@ int asymmetric_verify(struct key *keyring, const char *sig,
pks.pkey_algo = "rsa";
pks.hash_algo = hash_algo_name[hdr->hash_algo];
+ pks.encoding = "pkcs1";
pks.digest = (u8 *)data;
pks.digest_size = datalen;
pks.s = hdr->sig;
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 7ce6832..a67459e 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -5318,6 +5318,9 @@ static int selinux_sctp_bind_connect(struct sock *sk, int optname,
addr_buf = address;
while (walk_size < addrlen) {
+ if (walk_size + sizeof(sa_family_t) > addrlen)
+ return -EINVAL;
+
addr = addr_buf;
switch (addr->sa_family) {
case AF_UNSPEC:
diff --git a/security/selinux/ss/mls.c b/security/selinux/ss/mls.c
index 2fe459df..b7efa22 100644
--- a/security/selinux/ss/mls.c
+++ b/security/selinux/ss/mls.c
@@ -245,9 +245,13 @@ int mls_context_to_sid(struct policydb *pol,
char *rangep[2];
if (!pol->mls_enabled) {
- if ((def_sid != SECSID_NULL && oldc) || (*scontext) == '\0')
- return 0;
- return -EINVAL;
+ /*
+ * With no MLS, only return -EINVAL if there is a MLS field
+ * and it did not come from an xattr.
+ */
+ if (oldc && def_sid == SECSID_NULL)
+ return -EINVAL;
+ return 0;
}
/*
diff --git a/sound/pci/hda/thinkpad_helper.c b/sound/pci/hda/thinkpad_helper.c
index 97f49b7..568575b 100644
--- a/sound/pci/hda/thinkpad_helper.c
+++ b/sound/pci/hda/thinkpad_helper.c
@@ -58,8 +58,8 @@ static void hda_fixup_thinkpad_acpi(struct hda_codec *codec,
removefunc = false;
}
if (led_set_func(TPACPI_LED_MICMUTE, false) >= 0 &&
- snd_hda_gen_add_micmute_led(codec,
- update_tpacpi_micmute) > 0)
+ !snd_hda_gen_add_micmute_led(codec,
+ update_tpacpi_micmute))
removefunc = false;
}
diff --git a/tools/arch/arm64/include/asm/barrier.h b/tools/arch/arm64/include/asm/barrier.h
index 12835ea..378c051 100644
--- a/tools/arch/arm64/include/asm/barrier.h
+++ b/tools/arch/arm64/include/asm/barrier.h
@@ -14,74 +14,75 @@
#define wmb() asm volatile("dmb ishst" ::: "memory")
#define rmb() asm volatile("dmb ishld" ::: "memory")
-#define smp_store_release(p, v) \
-do { \
- union { typeof(*p) __val; char __c[1]; } __u = \
- { .__val = (__force typeof(*p)) (v) }; \
- \
- switch (sizeof(*p)) { \
- case 1: \
- asm volatile ("stlrb %w1, %0" \
- : "=Q" (*p) \
- : "r" (*(__u8 *)__u.__c) \
- : "memory"); \
- break; \
- case 2: \
- asm volatile ("stlrh %w1, %0" \
- : "=Q" (*p) \
- : "r" (*(__u16 *)__u.__c) \
- : "memory"); \
- break; \
- case 4: \
- asm volatile ("stlr %w1, %0" \
- : "=Q" (*p) \
- : "r" (*(__u32 *)__u.__c) \
- : "memory"); \
- break; \
- case 8: \
- asm volatile ("stlr %1, %0" \
- : "=Q" (*p) \
- : "r" (*(__u64 *)__u.__c) \
- : "memory"); \
- break; \
- default: \
- /* Only to shut up gcc ... */ \
- mb(); \
- break; \
- } \
+#define smp_store_release(p, v) \
+do { \
+ union { typeof(*p) __val; char __c[1]; } __u = \
+ { .__val = (v) }; \
+ \
+ switch (sizeof(*p)) { \
+ case 1: \
+ asm volatile ("stlrb %w1, %0" \
+ : "=Q" (*p) \
+ : "r" (*(__u8_alias_t *)__u.__c) \
+ : "memory"); \
+ break; \
+ case 2: \
+ asm volatile ("stlrh %w1, %0" \
+ : "=Q" (*p) \
+ : "r" (*(__u16_alias_t *)__u.__c) \
+ : "memory"); \
+ break; \
+ case 4: \
+ asm volatile ("stlr %w1, %0" \
+ : "=Q" (*p) \
+ : "r" (*(__u32_alias_t *)__u.__c) \
+ : "memory"); \
+ break; \
+ case 8: \
+ asm volatile ("stlr %1, %0" \
+ : "=Q" (*p) \
+ : "r" (*(__u64_alias_t *)__u.__c) \
+ : "memory"); \
+ break; \
+ default: \
+ /* Only to shut up gcc ... */ \
+ mb(); \
+ break; \
+ } \
} while (0)
-#define smp_load_acquire(p) \
-({ \
- union { typeof(*p) __val; char __c[1]; } __u; \
- \
- switch (sizeof(*p)) { \
- case 1: \
- asm volatile ("ldarb %w0, %1" \
- : "=r" (*(__u8 *)__u.__c) \
- : "Q" (*p) : "memory"); \
- break; \
- case 2: \
- asm volatile ("ldarh %w0, %1" \
- : "=r" (*(__u16 *)__u.__c) \
- : "Q" (*p) : "memory"); \
- break; \
- case 4: \
- asm volatile ("ldar %w0, %1" \
- : "=r" (*(__u32 *)__u.__c) \
- : "Q" (*p) : "memory"); \
- break; \
- case 8: \
- asm volatile ("ldar %0, %1" \
- : "=r" (*(__u64 *)__u.__c) \
- : "Q" (*p) : "memory"); \
- break; \
- default: \
- /* Only to shut up gcc ... */ \
- mb(); \
- break; \
- } \
- __u.__val; \
+#define smp_load_acquire(p) \
+({ \
+ union { typeof(*p) __val; char __c[1]; } __u = \
+ { .__c = { 0 } }; \
+ \
+ switch (sizeof(*p)) { \
+ case 1: \
+ asm volatile ("ldarb %w0, %1" \
+ : "=r" (*(__u8_alias_t *)__u.__c) \
+ : "Q" (*p) : "memory"); \
+ break; \
+ case 2: \
+ asm volatile ("ldarh %w0, %1" \
+ : "=r" (*(__u16_alias_t *)__u.__c) \
+ : "Q" (*p) : "memory"); \
+ break; \
+ case 4: \
+ asm volatile ("ldar %w0, %1" \
+ : "=r" (*(__u32_alias_t *)__u.__c) \
+ : "Q" (*p) : "memory"); \
+ break; \
+ case 8: \
+ asm volatile ("ldar %0, %1" \
+ : "=r" (*(__u64_alias_t *)__u.__c) \
+ : "Q" (*p) : "memory"); \
+ break; \
+ default: \
+ /* Only to shut up gcc ... */ \
+ mb(); \
+ break; \
+ } \
+ __u.__val; \
})
#endif /* _TOOLS_LINUX_ASM_AARCH64_BARRIER_H */
diff --git a/tools/perf/Documentation/perf-list.txt b/tools/perf/Documentation/perf-list.txt
index 236b9b97..667c14e 100644
--- a/tools/perf/Documentation/perf-list.txt
+++ b/tools/perf/Documentation/perf-list.txt
@@ -55,7 +55,6 @@
S - read sample value (PERF_SAMPLE_READ)
D - pin the event to the PMU
W - group is weak and will fallback to non-group if not schedulable,
- only supported in 'perf stat' for now.
The 'p' modifier can be used for specifying how precise the instruction
address should be. The 'p' modifier can be specified multiple times:
diff --git a/tools/perf/Makefile.perf b/tools/perf/Makefile.perf
index 3ccb4f0..d956554 100644
--- a/tools/perf/Makefile.perf
+++ b/tools/perf/Makefile.perf
@@ -387,7 +387,7 @@
linux_uapi_dir := $(srctree)/tools/include/uapi/linux
asm_generic_uapi_dir := $(srctree)/tools/include/uapi/asm-generic
-arch_asm_uapi_dir := $(srctree)/tools/arch/$(ARCH)/include/uapi/asm/
+arch_asm_uapi_dir := $(srctree)/tools/arch/$(SRCARCH)/include/uapi/asm/
beauty_outdir := $(OUTPUT)trace/beauty/generated
beauty_ioctl_outdir := $(beauty_outdir)/ioctl
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 10cf889..488779b 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -391,7 +391,12 @@ static int record__open(struct record *rec)
ui__warning("%s\n", msg);
goto try_again;
}
-
+ if ((errno == EINVAL || errno == EBADF) &&
+ pos->leader != pos &&
+ pos->weak_group) {
+ pos = perf_evlist__reset_weak_group(evlist, pos);
+ goto try_again;
+ }
rc = -errno;
perf_evsel__open_strerror(pos, &opts->target,
errno, msg, sizeof(msg));
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index d1028d7..a635abf 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -383,32 +383,6 @@ static bool perf_evsel__should_store_id(struct perf_evsel *counter)
return STAT_RECORD || counter->attr.read_format & PERF_FORMAT_ID;
}
-static struct perf_evsel *perf_evsel__reset_weak_group(struct perf_evsel *evsel)
-{
- struct perf_evsel *c2, *leader;
- bool is_open = true;
-
- leader = evsel->leader;
- pr_debug("Weak group for %s/%d failed\n",
- leader->name, leader->nr_members);
-
- /*
- * for_each_group_member doesn't work here because it doesn't
- * include the first entry.
- */
- evlist__for_each_entry(evsel_list, c2) {
- if (c2 == evsel)
- is_open = false;
- if (c2->leader == leader) {
- if (is_open)
- perf_evsel__close(c2);
- c2->leader = c2;
- c2->nr_members = 0;
- }
- }
- return leader;
-}
-
static bool is_target_alive(struct target *_target,
struct thread_map *threads)
{
@@ -477,7 +451,7 @@ static int __run_perf_stat(int argc, const char **argv, int run_idx)
if ((errno == EINVAL || errno == EBADF) &&
counter->leader != counter &&
counter->weak_group) {
- counter = perf_evsel__reset_weak_group(counter);
+ counter = perf_evlist__reset_weak_group(evsel_list, counter);
goto try_again;
}
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index b2838de..aa0c73e 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -1429,6 +1429,9 @@ int cmd_top(int argc, const char **argv)
}
}
+ if (opts->branch_stack && callchain_param.enabled)
+ symbol_conf.show_branchflag_count = true;
+
sort__mode = SORT_MODE__TOP;
/* display thread wants entries to be collapsed in a different tree */
perf_hpp_list.need_collapse = 1;
diff --git a/tools/perf/builtin-trace.c b/tools/perf/builtin-trace.c
index dc8a6c4..8356194 100644
--- a/tools/perf/builtin-trace.c
+++ b/tools/perf/builtin-trace.c
@@ -108,6 +108,7 @@ struct trace {
} stats;
unsigned int max_stack;
unsigned int min_stack;
+ bool raw_augmented_syscalls;
bool not_ev_qualifier;
bool live;
bool full_time;
@@ -1724,13 +1725,28 @@ static int trace__fprintf_sample(struct trace *trace, struct perf_evsel *evsel,
return printed;
}
-static void *syscall__augmented_args(struct syscall *sc, struct perf_sample *sample, int *augmented_args_size)
+static void *syscall__augmented_args(struct syscall *sc, struct perf_sample *sample, int *augmented_args_size, bool raw_augmented)
{
void *augmented_args = NULL;
+ /*
+ * For now with BPF raw_augmented we hook into raw_syscalls:sys_enter
+ * and there we get all 6 syscall args plus the tracepoint common
+ * fields (sizeof(long)) and the syscall_nr (another long). So we check
+ * if that is the case and if so don't look after the sc->args_size,
+ * but always after the full raw_syscalls:sys_enter payload, which is
+ * fixed.
+ *
+ * We'll revisit this later to pass s->args_size to the BPF augmenter
+ * (now tools/perf/examples/bpf/augmented_raw_syscalls.c, so that it
+ * copies only what we need for each syscall, like what happens when we
+ * use syscalls:sys_enter_NAME, so that we reduce the kernel/userspace
+ * traffic to just what is needed for each syscall.
+ */
+ int args_size = raw_augmented ? (8 * (int)sizeof(long)) : sc->args_size;
- *augmented_args_size = sample->raw_size - sc->args_size;
+ *augmented_args_size = sample->raw_size - args_size;
if (*augmented_args_size > 0)
- augmented_args = sample->raw_data + sc->args_size;
+ augmented_args = sample->raw_data + args_size;
return augmented_args;
}
@@ -1780,7 +1796,7 @@ static int trace__sys_enter(struct trace *trace, struct perf_evsel *evsel,
* here and avoid using augmented syscalls when the evsel is the raw_syscalls one.
*/
if (evsel != trace->syscalls.events.sys_enter)
- augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size);
+ augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size, trace->raw_augmented_syscalls);
ttrace->entry_time = sample->time;
msg = ttrace->entry_str;
printed += scnprintf(msg + printed, trace__entry_str_size - printed, "%s(", sc->name);
@@ -1833,7 +1849,7 @@ static int trace__fprintf_sys_enter(struct trace *trace, struct perf_evsel *evse
goto out_put;
args = perf_evsel__sc_tp_ptr(evsel, args, sample);
- augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size);
+ augmented_args = syscall__augmented_args(sc, sample, &augmented_args_size, trace->raw_augmented_syscalls);
syscall__scnprintf_args(sc, msg, sizeof(msg), args, augmented_args, augmented_args_size, trace, thread);
fprintf(trace->output, "%s", msg);
err = 0;
@@ -3501,7 +3517,15 @@ int cmd_trace(int argc, const char **argv)
evsel->handler = trace__sys_enter;
evlist__for_each_entry(trace.evlist, evsel) {
+ bool raw_syscalls_sys_exit = strcmp(perf_evsel__name(evsel), "raw_syscalls:sys_exit") == 0;
+
+ if (raw_syscalls_sys_exit) {
+ trace.raw_augmented_syscalls = true;
+ goto init_augmented_syscall_tp;
+ }
+
if (strstarts(perf_evsel__name(evsel), "syscalls:sys_exit_")) {
+init_augmented_syscall_tp:
perf_evsel__init_augmented_syscall_tp(evsel);
perf_evsel__init_augmented_syscall_tp_ret(evsel);
evsel->handler = trace__sys_exit;
diff --git a/tools/perf/examples/bpf/augmented_raw_syscalls.c b/tools/perf/examples/bpf/augmented_raw_syscalls.c
new file mode 100644
index 0000000..90a1933
--- /dev/null
+++ b/tools/perf/examples/bpf/augmented_raw_syscalls.c
@@ -0,0 +1,131 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Augment the raw_syscalls tracepoints with the contents of the pointer arguments.
+ *
+ * Test it with:
+ *
+ * perf trace -e tools/perf/examples/bpf/augmented_raw_syscalls.c cat /etc/passwd > /dev/null
+ *
+ * This exactly matches what is marshalled into the raw_syscall:sys_enter
+ * payload expected by the 'perf trace' beautifiers.
+ *
+ * For now it just uses the existing tracepoint augmentation code in 'perf
+ * trace', in the next csets we'll hook up these with the sys_enter/sys_exit
+ * code that will combine entry/exit in a strace like way.
+ */
+
+#include <stdio.h>
+#include <linux/socket.h>
+
+/* bpf-output associated map */
+struct bpf_map SEC("maps") __augmented_syscalls__ = {
+ .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY,
+ .key_size = sizeof(int),
+ .value_size = sizeof(u32),
+ .max_entries = __NR_CPUS__,
+};
+
+struct syscall_enter_args {
+ unsigned long long common_tp_fields;
+ long syscall_nr;
+ unsigned long args[6];
+};
+
+struct syscall_exit_args {
+ unsigned long long common_tp_fields;
+ long syscall_nr;
+ long ret;
+};
+
+struct augmented_filename {
+ unsigned int size;
+ int reserved;
+ char value[256];
+};
+
+#define SYS_OPEN 2
+#define SYS_OPENAT 257
+
+SEC("raw_syscalls:sys_enter")
+int sys_enter(struct syscall_enter_args *args)
+{
+ struct {
+ struct syscall_enter_args args;
+ struct augmented_filename filename;
+ } augmented_args;
+ unsigned int len = sizeof(augmented_args);
+ const void *filename_arg = NULL;
+
+ probe_read(&augmented_args.args, sizeof(augmented_args.args), args);
+ /*
+ * Yonghong and Edward Cree sayz:
+ *
+ * https://www.spinics.net/lists/netdev/msg531645.html
+ *
+ * >> R0=inv(id=0) R1=inv2 R6=ctx(id=0,off=0,imm=0) R7=inv64 R10=fp0,call_-1
+ * >> 10: (bf) r1 = r6
+ * >> 11: (07) r1 += 16
+ * >> 12: (05) goto pc+2
+ * >> 15: (79) r3 = *(u64 *)(r1 +0)
+ * >> dereference of modified ctx ptr R1 off=16 disallowed
+ * > Aha, we at least got a different error message this time.
+ * > And indeed llvm has done that optimisation, rather than the more obvious
+ * > 11: r3 = *(u64 *)(r1 +16)
+ * > because it wants to have lots of reads share a single insn. You may be able
+ * > to defeat that optimisation by adding compiler barriers, idk. Maybe someone
+ * > with llvm knowledge can figure out how to stop it (ideally, llvm would know
+ * > when it's generating for bpf backend and not do that). -O0? ¯\_(ツ)_/¯
+ *
+ * The optimization mostly likes below:
+ *
+ * br1:
+ * ...
+ * r1 += 16
+ * goto merge
+ * br2:
+ * ...
+ * r1 += 20
+ * goto merge
+ * merge:
+ * *(u64 *)(r1 + 0)
+ *
+ * The compiler tries to merge common loads. There is no easy way to
+ * stop this compiler optimization without turning off a lot of other
+ * optimizations. The easiest way is to add barriers:
+ *
+ * __asm__ __volatile__("": : :"memory")
+ *
+ * after the ctx memory access to prevent their down stream merging.
+ */
+ switch (augmented_args.args.syscall_nr) {
+ case SYS_OPEN: filename_arg = (const void *)args->args[0];
+ __asm__ __volatile__("": : :"memory");
+ break;
+ case SYS_OPENAT: filename_arg = (const void *)args->args[1];
+ break;
+ }
+
+ if (filename_arg != NULL) {
+ augmented_args.filename.reserved = 0;
+ augmented_args.filename.size = probe_read_str(&augmented_args.filename.value,
+ sizeof(augmented_args.filename.value),
+ filename_arg);
+ if (augmented_args.filename.size < sizeof(augmented_args.filename.value)) {
+ len -= sizeof(augmented_args.filename.value) - augmented_args.filename.size;
+ len &= sizeof(augmented_args.filename.value) - 1;
+ }
+ } else {
+ len = sizeof(augmented_args.args);
+ }
+
+ perf_event_output(args, &__augmented_syscalls__, BPF_F_CURRENT_CPU, &augmented_args, len);
+ return 0;
+}
+
+SEC("raw_syscalls:sys_exit")
+int sys_exit(struct syscall_exit_args *args)
+{
+ return 1; /* 0 as soon as we start copying data returned by the kernel, e.g. 'read' */
+}
+
+license(GPL);
diff --git a/tools/perf/jvmti/jvmti_agent.c b/tools/perf/jvmti/jvmti_agent.c
index ac1bcdc..f7eb63c 100644
--- a/tools/perf/jvmti/jvmti_agent.c
+++ b/tools/perf/jvmti/jvmti_agent.c
@@ -125,7 +125,7 @@ perf_get_timestamp(void)
}
static int
-debug_cache_init(void)
+create_jit_cache_dir(void)
{
char str[32];
char *base, *p;
@@ -144,8 +144,13 @@ debug_cache_init(void)
strftime(str, sizeof(str), JIT_LANG"-jit-%Y%m%d", &tm);
- snprintf(jit_path, PATH_MAX - 1, "%s/.debug/", base);
-
+ ret = snprintf(jit_path, PATH_MAX, "%s/.debug/", base);
+ if (ret >= PATH_MAX) {
+ warnx("jvmti: cannot generate jit cache dir because %s/.debug/"
+ " is too long, please check the cwd, JITDUMPDIR, and"
+ " HOME variables", base);
+ return -1;
+ }
ret = mkdir(jit_path, 0755);
if (ret == -1) {
if (errno != EEXIST) {
@@ -154,20 +159,32 @@ debug_cache_init(void)
}
}
- snprintf(jit_path, PATH_MAX - 1, "%s/.debug/jit", base);
+ ret = snprintf(jit_path, PATH_MAX, "%s/.debug/jit", base);
+ if (ret >= PATH_MAX) {
+ warnx("jvmti: cannot generate jit cache dir because"
+ " %s/.debug/jit is too long, please check the cwd,"
+ " JITDUMPDIR, and HOME variables", base);
+ return -1;
+ }
ret = mkdir(jit_path, 0755);
if (ret == -1) {
if (errno != EEXIST) {
- warn("cannot create jit cache dir %s", jit_path);
+ warn("jvmti: cannot create jit cache dir %s", jit_path);
return -1;
}
}
- snprintf(jit_path, PATH_MAX - 1, "%s/.debug/jit/%s.XXXXXXXX", base, str);
-
+ ret = snprintf(jit_path, PATH_MAX, "%s/.debug/jit/%s.XXXXXXXX", base, str);
+ if (ret >= PATH_MAX) {
+ warnx("jvmti: cannot generate jit cache dir because"
+ " %s/.debug/jit/%s.XXXXXXXX is too long, please check"
+ " the cwd, JITDUMPDIR, and HOME variables",
+ base, str);
+ return -1;
+ }
p = mkdtemp(jit_path);
if (p != jit_path) {
- warn("cannot create jit cache dir %s", jit_path);
+ warn("jvmti: cannot create jit cache dir %s", jit_path);
return -1;
}
@@ -228,7 +245,7 @@ void *jvmti_open(void)
{
char dump_path[PATH_MAX];
struct jitheader header;
- int fd;
+ int fd, ret;
FILE *fp;
init_arch_timestamp();
@@ -245,12 +262,22 @@ void *jvmti_open(void)
memset(&header, 0, sizeof(header));
- debug_cache_init();
+ /*
+ * jitdump file dir
+ */
+ if (create_jit_cache_dir() < 0)
+ return NULL;
/*
* jitdump file name
*/
- scnprintf(dump_path, PATH_MAX, "%s/jit-%i.dump", jit_path, getpid());
+ ret = snprintf(dump_path, PATH_MAX, "%s/jit-%i.dump", jit_path, getpid());
+ if (ret >= PATH_MAX) {
+ warnx("jvmti: cannot generate jitdump file full path because"
+ " %s/jit-%i.dump is too long, please check the cwd,"
+ " JITDUMPDIR, and HOME variables", jit_path, getpid());
+ return NULL;
+ }
fd = open(dump_path, O_CREAT|O_TRUNC|O_RDWR, 0666);
if (fd == -1)
diff --git a/tools/perf/scripts/python/exported-sql-viewer.py b/tools/perf/scripts/python/exported-sql-viewer.py
index 24cb0bd..f278ce5 100755
--- a/tools/perf/scripts/python/exported-sql-viewer.py
+++ b/tools/perf/scripts/python/exported-sql-viewer.py
@@ -119,6 +119,14 @@
return "[kernel]"
return name
+def findnth(s, sub, n, offs=0):
+ pos = s.find(sub)
+ if pos < 0:
+ return pos
+ if n <= 1:
+ return offs + pos
+ return findnth(s[pos + 1:], sub, n - 1, offs + pos + 1)
+
# Percent to one decimal place
def PercentToOneDP(n, d):
@@ -1464,6 +1472,317 @@
else:
self.find_bar.NotFound()
+# Dialog data item converted and validated using a SQL table
+
+class SQLTableDialogDataItem():
+
+ def __init__(self, glb, label, placeholder_text, table_name, match_column, column_name1, column_name2, parent):
+ self.glb = glb
+ self.label = label
+ self.placeholder_text = placeholder_text
+ self.table_name = table_name
+ self.match_column = match_column
+ self.column_name1 = column_name1
+ self.column_name2 = column_name2
+ self.parent = parent
+
+ self.value = ""
+
+ self.widget = QLineEdit()
+ self.widget.editingFinished.connect(self.Validate)
+ self.widget.textChanged.connect(self.Invalidate)
+ self.red = False
+ self.error = ""
+ self.validated = True
+
+ self.last_id = 0
+ self.first_time = 0
+ self.last_time = 2 ** 64
+ if self.table_name == "<timeranges>":
+ query = QSqlQuery(self.glb.db)
+ QueryExec(query, "SELECT id, time FROM samples ORDER BY id DESC LIMIT 1")
+ if query.next():
+ self.last_id = int(query.value(0))
+ self.last_time = int(query.value(1))
+ QueryExec(query, "SELECT time FROM samples WHERE time != 0 ORDER BY id LIMIT 1")
+ if query.next():
+ self.first_time = int(query.value(0))
+ if placeholder_text:
+ placeholder_text += ", between " + str(self.first_time) + " and " + str(self.last_time)
+
+ if placeholder_text:
+ self.widget.setPlaceholderText(placeholder_text)
+
+ def ValueToIds(self, value):
+ ids = []
+ query = QSqlQuery(self.glb.db)
+ stmt = "SELECT id FROM " + self.table_name + " WHERE " + self.match_column + " = '" + value + "'"
+ ret = query.exec_(stmt)
+ if ret:
+ while query.next():
+ ids.append(str(query.value(0)))
+ return ids
+
+ def IdBetween(self, query, lower_id, higher_id, order):
+ QueryExec(query, "SELECT id FROM samples WHERE id > " + str(lower_id) + " AND id < " + str(higher_id) + " ORDER BY id " + order + " LIMIT 1")
+ if query.next():
+ return True, int(query.value(0))
+ else:
+ return False, 0
+
+ def BinarySearchTime(self, lower_id, higher_id, target_time, get_floor):
+ query = QSqlQuery(self.glb.db)
+ while True:
+ next_id = int((lower_id + higher_id) / 2)
+ QueryExec(query, "SELECT time FROM samples WHERE id = " + str(next_id))
+ if not query.next():
+ ok, dbid = self.IdBetween(query, lower_id, next_id, "DESC")
+ if not ok:
+ ok, dbid = self.IdBetween(query, next_id, higher_id, "")
+ if not ok:
+ return str(higher_id)
+ next_id = dbid
+ QueryExec(query, "SELECT time FROM samples WHERE id = " + str(next_id))
+ next_time = int(query.value(0))
+ if get_floor:
+ if target_time > next_time:
+ lower_id = next_id
+ else:
+ higher_id = next_id
+ if higher_id <= lower_id + 1:
+ return str(higher_id)
+ else:
+ if target_time >= next_time:
+ lower_id = next_id
+ else:
+ higher_id = next_id
+ if higher_id <= lower_id + 1:
+ return str(lower_id)
+
+ def ConvertRelativeTime(self, val):
+ print "val ", val
+ mult = 1
+ suffix = val[-2:]
+ if suffix == "ms":
+ mult = 1000000
+ elif suffix == "us":
+ mult = 1000
+ elif suffix == "ns":
+ mult = 1
+ else:
+ return val
+ val = val[:-2].strip()
+ if not self.IsNumber(val):
+ return val
+ val = int(val) * mult
+ if val >= 0:
+ val += self.first_time
+ else:
+ val += self.last_time
+ return str(val)
+
+ def ConvertTimeRange(self, vrange):
+ print "vrange ", vrange
+ if vrange[0] == "":
+ vrange[0] = str(self.first_time)
+ if vrange[1] == "":
+ vrange[1] = str(self.last_time)
+ vrange[0] = self.ConvertRelativeTime(vrange[0])
+ vrange[1] = self.ConvertRelativeTime(vrange[1])
+ print "vrange2 ", vrange
+ if not self.IsNumber(vrange[0]) or not self.IsNumber(vrange[1]):
+ return False
+ print "ok1"
+ beg_range = max(int(vrange[0]), self.first_time)
+ end_range = min(int(vrange[1]), self.last_time)
+ if beg_range > self.last_time or end_range < self.first_time:
+ return False
+ print "ok2"
+ vrange[0] = self.BinarySearchTime(0, self.last_id, beg_range, True)
+ vrange[1] = self.BinarySearchTime(1, self.last_id + 1, end_range, False)
+ print "vrange3 ", vrange
+ return True
+
+ def AddTimeRange(self, value, ranges):
+ print "value ", value
+ n = value.count("-")
+ if n == 1:
+ pass
+ elif n == 2:
+ if value.split("-")[1].strip() == "":
+ n = 1
+ elif n == 3:
+ n = 2
+ else:
+ return False
+ pos = findnth(value, "-", n)
+ vrange = [value[:pos].strip() ,value[pos+1:].strip()]
+ if self.ConvertTimeRange(vrange):
+ ranges.append(vrange)
+ return True
+ return False
+
+ def InvalidValue(self, value):
+ self.value = ""
+ palette = QPalette()
+ palette.setColor(QPalette.Text,Qt.red)
+ self.widget.setPalette(palette)
+ self.red = True
+ self.error = self.label + " invalid value '" + value + "'"
+ self.parent.ShowMessage(self.error)
+
+ def IsNumber(self, value):
+ try:
+ x = int(value)
+ except:
+ x = 0
+ return str(x) == value
+
+ def Invalidate(self):
+ self.validated = False
+
+ def Validate(self):
+ input_string = self.widget.text()
+ self.validated = True
+ if self.red:
+ palette = QPalette()
+ self.widget.setPalette(palette)
+ self.red = False
+ if not len(input_string.strip()):
+ self.error = ""
+ self.value = ""
+ return
+ if self.table_name == "<timeranges>":
+ ranges = []
+ for value in [x.strip() for x in input_string.split(",")]:
+ if not self.AddTimeRange(value, ranges):
+ return self.InvalidValue(value)
+ ranges = [("(" + self.column_name1 + " >= " + r[0] + " AND " + self.column_name1 + " <= " + r[1] + ")") for r in ranges]
+ self.value = " OR ".join(ranges)
+ elif self.table_name == "<ranges>":
+ singles = []
+ ranges = []
+ for value in [x.strip() for x in input_string.split(",")]:
+ if "-" in value:
+ vrange = value.split("-")
+ if len(vrange) != 2 or not self.IsNumber(vrange[0]) or not self.IsNumber(vrange[1]):
+ return self.InvalidValue(value)
+ ranges.append(vrange)
+ else:
+ if not self.IsNumber(value):
+ return self.InvalidValue(value)
+ singles.append(value)
+ ranges = [("(" + self.column_name1 + " >= " + r[0] + " AND " + self.column_name1 + " <= " + r[1] + ")") for r in ranges]
+ if len(singles):
+ ranges.append(self.column_name1 + " IN (" + ",".join(singles) + ")")
+ self.value = " OR ".join(ranges)
+ elif self.table_name:
+ all_ids = []
+ for value in [x.strip() for x in input_string.split(",")]:
+ ids = self.ValueToIds(value)
+ if len(ids):
+ all_ids.extend(ids)
+ else:
+ return self.InvalidValue(value)
+ self.value = self.column_name1 + " IN (" + ",".join(all_ids) + ")"
+ if self.column_name2:
+ self.value = "( " + self.value + " OR " + self.column_name2 + " IN (" + ",".join(all_ids) + ") )"
+ else:
+ self.value = input_string.strip()
+ self.error = ""
+ self.parent.ClearMessage()
+
+ def IsValid(self):
+ if not self.validated:
+ self.Validate()
+ if len(self.error):
+ self.parent.ShowMessage(self.error)
+ return False
+ return True
+
+# Selected branch report creation dialog
+
+class SelectedBranchDialog(QDialog):
+
+ def __init__(self, glb, parent=None):
+ super(SelectedBranchDialog, self).__init__(parent)
+
+ self.glb = glb
+
+ self.name = ""
+ self.where_clause = ""
+
+ self.setWindowTitle("Selected Branches")
+ self.setMinimumWidth(600)
+
+ items = (
+ ("Report name:", "Enter a name to appear in the window title bar", "", "", "", ""),
+ ("Time ranges:", "Enter time ranges", "<timeranges>", "", "samples.id", ""),
+ ("CPUs:", "Enter CPUs or ranges e.g. 0,5-6", "<ranges>", "", "cpu", ""),
+ ("Commands:", "Only branches with these commands will be included", "comms", "comm", "comm_id", ""),
+ ("PIDs:", "Only branches with these process IDs will be included", "threads", "pid", "thread_id", ""),
+ ("TIDs:", "Only branches with these thread IDs will be included", "threads", "tid", "thread_id", ""),
+ ("DSOs:", "Only branches with these DSOs will be included", "dsos", "short_name", "samples.dso_id", "to_dso_id"),
+ ("Symbols:", "Only branches with these symbols will be included", "symbols", "name", "symbol_id", "to_symbol_id"),
+ ("Raw SQL clause: ", "Enter a raw SQL WHERE clause", "", "", "", ""),
+ )
+ self.data_items = [SQLTableDialogDataItem(glb, *x, parent=self) for x in items]
+
+ self.grid = QGridLayout()
+
+ for row in xrange(len(self.data_items)):
+ self.grid.addWidget(QLabel(self.data_items[row].label), row, 0)
+ self.grid.addWidget(self.data_items[row].widget, row, 1)
+
+ self.status = QLabel()
+
+ self.ok_button = QPushButton("Ok", self)
+ self.ok_button.setDefault(True)
+ self.ok_button.released.connect(self.Ok)
+ self.ok_button.setSizePolicy(QSizePolicy.Fixed, QSizePolicy.Fixed)
+
+ self.cancel_button = QPushButton("Cancel", self)
+ self.cancel_button.released.connect(self.reject)
+ self.cancel_button.setSizePolicy(QSizePolicy.Fixed, QSizePolicy.Fixed)
+
+ self.hbox = QHBoxLayout()
+ #self.hbox.addStretch()
+ self.hbox.addWidget(self.status)
+ self.hbox.addWidget(self.ok_button)
+ self.hbox.addWidget(self.cancel_button)
+
+ self.vbox = QVBoxLayout()
+ self.vbox.addLayout(self.grid)
+ self.vbox.addLayout(self.hbox)
+
+ self.setLayout(self.vbox);
+
+ def Ok(self):
+ self.name = self.data_items[0].value
+ if not self.name:
+ self.ShowMessage("Report name is required")
+ return
+ for d in self.data_items:
+ if not d.IsValid():
+ return
+ for d in self.data_items[1:]:
+ if len(d.value):
+ if len(self.where_clause):
+ self.where_clause += " AND "
+ self.where_clause += d.value
+ if len(self.where_clause):
+ self.where_clause = " AND ( " + self.where_clause + " ) "
+ else:
+ self.ShowMessage("No selection")
+ return
+ self.accept()
+
+ def ShowMessage(self, msg):
+ self.status.setText("<font color=#FF0000>" + msg)
+
+ def ClearMessage(self):
+ self.status.setText("")
+
# Event list
def GetEventList(db):
@@ -1656,7 +1975,7 @@
def FindDone(self, row):
self.find_bar.Idle()
if row >= 0:
- self.view.setCurrentIndex(self.model.index(row, 0, QModelIndex()))
+ self.view.setCurrentIndex(self.model.mapFromSource(self.data_model.index(row, 0, QModelIndex())))
else:
self.find_bar.NotFound()
@@ -1765,6 +2084,149 @@
def setActiveSubWindow(self, nr):
self.mdi_area.setActiveSubWindow(self.mdi_area.subWindowList()[nr - 1])
+# Help text
+
+glb_help_text = """
+<h1>Contents</h1>
+<style>
+p.c1 {
+ text-indent: 40px;
+}
+p.c2 {
+ text-indent: 80px;
+}
+}
+</style>
+<p class=c1><a href=#reports>1. Reports</a></p>
+<p class=c2><a href=#callgraph>1.1 Context-Sensitive Call Graph</a></p>
+<p class=c2><a href=#allbranches>1.2 All branches</a></p>
+<p class=c2><a href=#selectedbranches>1.3 Selected branches</a></p>
+<p class=c1><a href=#tables>2. Tables</a></p>
+<h1 id=reports>1. Reports</h1>
+<h2 id=callgraph>1.1 Context-Sensitive Call Graph</h2>
+The result is a GUI window with a tree representing a context-sensitive
+call-graph. Expanding a couple of levels of the tree and adjusting column
+widths to suit will display something like:
+<pre>
+ Call Graph: pt_example
+Call Path Object Count Time(ns) Time(%) Branch Count Branch Count(%)
+v- ls
+ v- 2638:2638
+ v- _start ld-2.19.so 1 10074071 100.0 211135 100.0
+ |- unknown unknown 1 13198 0.1 1 0.0
+ >- _dl_start ld-2.19.so 1 1400980 13.9 19637 9.3
+ >- _d_linit_internal ld-2.19.so 1 448152 4.4 11094 5.3
+ v-__libc_start_main@plt ls 1 8211741 81.5 180397 85.4
+ >- _dl_fixup ld-2.19.so 1 7607 0.1 108 0.1
+ >- __cxa_atexit libc-2.19.so 1 11737 0.1 10 0.0
+ >- __libc_csu_init ls 1 10354 0.1 10 0.0
+ |- _setjmp libc-2.19.so 1 0 0.0 4 0.0
+ v- main ls 1 8182043 99.6 180254 99.9
+</pre>
+<h3>Points to note:</h3>
+<ul>
+<li>The top level is a command name (comm)</li>
+<li>The next level is a thread (pid:tid)</li>
+<li>Subsequent levels are functions</li>
+<li>'Count' is the number of calls</li>
+<li>'Time' is the elapsed time until the function returns</li>
+<li>Percentages are relative to the level above</li>
+<li>'Branch Count' is the total number of branches for that function and all functions that it calls
+</ul>
+<h3>Find</h3>
+Ctrl-F displays a Find bar which finds function names by either an exact match or a pattern match.
+The pattern matching symbols are ? for any character and * for zero or more characters.
+<h2 id=allbranches>1.2 All branches</h2>
+The All branches report displays all branches in chronological order.
+Not all data is fetched immediately. More records can be fetched using the Fetch bar provided.
+<h3>Disassembly</h3>
+Open a branch to display disassembly. This only works if:
+<ol>
+<li>The disassembler is available. Currently, only Intel XED is supported - see <a href=#xed>Intel XED Setup</a></li>
+<li>The object code is available. Currently, only the perf build ID cache is searched for object code.
+The default directory ~/.debug can be overridden by setting environment variable PERF_BUILDID_DIR.
+One exception is kcore where the DSO long name is used (refer dsos_view on the Tables menu),
+or alternatively, set environment variable PERF_KCORE to the kcore file name.</li>
+</ol>
+<h4 id=xed>Intel XED Setup</h4>
+To use Intel XED, libxed.so must be present. To build and install libxed.so:
+<pre>
+git clone https://github.com/intelxed/mbuild.git mbuild
+git clone https://github.com/intelxed/xed
+cd xed
+./mfile.py --share
+sudo ./mfile.py --prefix=/usr/local install
+sudo ldconfig
+</pre>
+<h3>Find</h3>
+Ctrl-F displays a Find bar which finds substrings by either an exact match or a regular expression match.
+Refer to Python documentation for the regular expression syntax.
+All columns are searched, but only currently fetched rows are searched.
+<h2 id=selectedbranches>1.3 Selected branches</h2>
+This is the same as the <a href=#allbranches>All branches</a> report but with the data reduced
+by various selection criteria. A dialog box displays available criteria which are AND'ed together.
+<h3>1.3.1 Time ranges</h3>
+The time ranges hint text shows the total time range. Relative time ranges can also be entered in
+ms, us or ns. Also, negative values are relative to the end of trace. Examples:
+<pre>
+ 81073085947329-81073085958238 From 81073085947329 to 81073085958238
+ 100us-200us From 100us to 200us
+ 10ms- From 10ms to the end
+ -100ns The first 100ns
+ -10ms- The last 10ms
+</pre>
+N.B. Due to the granularity of timestamps, there could be no branches in any given time range.
+<h1 id=tables>2. Tables</h1>
+The Tables menu shows all tables and views in the database. Most tables have an associated view
+which displays the information in a more friendly way. Not all data for large tables is fetched
+immediately. More records can be fetched using the Fetch bar provided. Columns can be sorted,
+but that can be slow for large tables.
+<p>There are also tables of database meta-information.
+For SQLite3 databases, the sqlite_master table is included.
+For PostgreSQL databases, information_schema.tables/views/columns are included.
+<h3>Find</h3>
+Ctrl-F displays a Find bar which finds substrings by either an exact match or a regular expression match.
+Refer to Python documentation for the regular expression syntax.
+All columns are searched, but only currently fetched rows are searched.
+<p>N.B. Results are found in id order, so if the table is re-ordered, find-next and find-previous
+will go to the next/previous result in id order, instead of display order.
+"""
+
+# Help window
+
+class HelpWindow(QMdiSubWindow):
+
+ def __init__(self, glb, parent=None):
+ super(HelpWindow, self).__init__(parent)
+
+ self.text = QTextBrowser()
+ self.text.setHtml(glb_help_text)
+ self.text.setReadOnly(True)
+ self.text.setOpenExternalLinks(True)
+
+ self.setWidget(self.text)
+
+ AddSubWindow(glb.mainwindow.mdi_area, self, "Exported SQL Viewer Help")
+
+# Main window that only displays the help text
+
+class HelpOnlyWindow(QMainWindow):
+
+ def __init__(self, parent=None):
+ super(HelpOnlyWindow, self).__init__(parent)
+
+ self.setMinimumSize(200, 100)
+ self.resize(800, 600)
+ self.setWindowTitle("Exported SQL Viewer Help")
+ self.setWindowIcon(self.style().standardIcon(QStyle.SP_MessageBoxInformation))
+
+ self.text = QTextBrowser()
+ self.text.setHtml(glb_help_text)
+ self.text.setReadOnly(True)
+ self.text.setOpenExternalLinks(True)
+
+ self.setCentralWidget(self.text)
+
# Font resize
def ResizeFont(widget, diff):
@@ -1851,6 +2313,9 @@
self.window_menu = WindowMenu(self.mdi_area, menu)
+ help_menu = menu.addMenu("&Help")
+ help_menu.addAction(CreateAction("&Exported SQL Viewer Help", "Helpful information", self.Help, self, QKeySequence.HelpContents))
+
def Find(self):
win = self.mdi_area.activeSubWindow()
if win:
@@ -1888,6 +2353,8 @@
if event == "branches":
label = "All branches" if branches_events == 1 else "All branches " + "(id=" + dbid + ")"
reports_menu.addAction(CreateAction(label, "Create a new window displaying branch events", lambda x=dbid: self.NewBranchView(x), self))
+ label = "Selected branches" if branches_events == 1 else "Selected branches " + "(id=" + dbid + ")"
+ reports_menu.addAction(CreateAction(label, "Create a new window displaying branch events", lambda x=dbid: self.NewSelectedBranchView(x), self))
def TableMenu(self, tables, menu):
table_menu = menu.addMenu("&Tables")
@@ -1900,9 +2367,18 @@
def NewBranchView(self, event_id):
BranchWindow(self.glb, event_id, "", "", self)
+ def NewSelectedBranchView(self, event_id):
+ dialog = SelectedBranchDialog(self.glb, self)
+ ret = dialog.exec_()
+ if ret:
+ BranchWindow(self.glb, event_id, dialog.name, dialog.where_clause, self)
+
def NewTableView(self, table_name):
TableWindow(self.glb, table_name, self)
+ def Help(self):
+ HelpWindow(self.glb, self)
+
# XED Disassembler
class xed_state_t(Structure):
@@ -1929,7 +2405,12 @@
class LibXED():
def __init__(self):
- self.libxed = CDLL("libxed.so")
+ try:
+ self.libxed = CDLL("libxed.so")
+ except:
+ self.libxed = None
+ if not self.libxed:
+ self.libxed = CDLL("/usr/local/lib/libxed.so")
self.xed_tables_init = self.libxed.xed_tables_init
self.xed_tables_init.restype = None
@@ -2097,10 +2578,16 @@
def Main():
if (len(sys.argv) < 2):
- print >> sys.stderr, "Usage is: exported-sql-viewer.py <database name>"
+ print >> sys.stderr, "Usage is: exported-sql-viewer.py {<database name> | --help-only}"
raise Exception("Too few arguments")
dbname = sys.argv[1]
+ if dbname == "--help-only":
+ app = QApplication(sys.argv)
+ mainwindow = HelpOnlyWindow()
+ mainwindow.show()
+ err = app.exec_()
+ sys.exit(err)
is_sqlite3 = False
try:
diff --git a/tools/perf/tests/attr/test-record-group-sampling b/tools/perf/tests/attr/test-record-group-sampling
index 8a33ca4..f0729c4 100644
--- a/tools/perf/tests/attr/test-record-group-sampling
+++ b/tools/perf/tests/attr/test-record-group-sampling
@@ -37,4 +37,3 @@
sample_period=0
freq=0
write_backward=0
-sample_id_all=0
diff --git a/tools/perf/util/evlist.c b/tools/perf/util/evlist.c
index e88e6f9..668d2a9 100644
--- a/tools/perf/util/evlist.c
+++ b/tools/perf/util/evlist.c
@@ -1810,3 +1810,30 @@ void perf_evlist__force_leader(struct perf_evlist *evlist)
leader->forced_leader = true;
}
}
+
+struct perf_evsel *perf_evlist__reset_weak_group(struct perf_evlist *evsel_list,
+ struct perf_evsel *evsel)
+{
+ struct perf_evsel *c2, *leader;
+ bool is_open = true;
+
+ leader = evsel->leader;
+ pr_debug("Weak group for %s/%d failed\n",
+ leader->name, leader->nr_members);
+
+ /*
+ * for_each_group_member doesn't work here because it doesn't
+ * include the first entry.
+ */
+ evlist__for_each_entry(evsel_list, c2) {
+ if (c2 == evsel)
+ is_open = false;
+ if (c2->leader == leader) {
+ if (is_open)
+ perf_evsel__close(c2);
+ c2->leader = c2;
+ c2->nr_members = 0;
+ }
+ }
+ return leader;
+}
diff --git a/tools/perf/util/evlist.h b/tools/perf/util/evlist.h
index dc66436..9919eed 100644
--- a/tools/perf/util/evlist.h
+++ b/tools/perf/util/evlist.h
@@ -312,4 +312,7 @@ bool perf_evlist__exclude_kernel(struct perf_evlist *evlist);
void perf_evlist__force_leader(struct perf_evlist *evlist);
+struct perf_evsel *perf_evlist__reset_weak_group(struct perf_evlist *evlist,
+ struct perf_evsel *evsel);
+
#endif /* __PERF_EVLIST_H */
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 6d18705..d37bb15 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -956,7 +956,6 @@ void perf_evsel__config(struct perf_evsel *evsel, struct record_opts *opts,
attr->sample_freq = 0;
attr->sample_period = 0;
attr->write_backward = 0;
- attr->sample_id_all = 0;
}
if (opts->no_samples)
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
index 58f6a9c..4503f3c 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
@@ -1474,6 +1474,8 @@ static void intel_pt_calc_mtc_timestamp(struct intel_pt_decoder *decoder)
decoder->have_calc_cyc_to_tsc = false;
intel_pt_calc_cyc_to_tsc(decoder, true);
}
+
+ intel_pt_log_to("Setting timestamp", decoder->timestamp);
}
static void intel_pt_calc_cbr(struct intel_pt_decoder *decoder)
@@ -1514,6 +1516,8 @@ static void intel_pt_calc_cyc_timestamp(struct intel_pt_decoder *decoder)
decoder->timestamp = timestamp;
decoder->timestamp_insn_cnt = 0;
+
+ intel_pt_log_to("Setting timestamp", decoder->timestamp);
}
/* Walk PSB+ packets when already in sync. */
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-log.c b/tools/perf/util/intel-pt-decoder/intel-pt-log.c
index e02bc7b..5e64da27 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-log.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-log.c
@@ -31,6 +31,11 @@ static FILE *f;
static char log_name[MAX_LOG_NAME];
bool intel_pt_enable_logging;
+void *intel_pt_log_fp(void)
+{
+ return f;
+}
+
void intel_pt_log_enable(void)
{
intel_pt_enable_logging = true;
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-log.h b/tools/perf/util/intel-pt-decoder/intel-pt-log.h
index 45b64f9..cc08493 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-log.h
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-log.h
@@ -22,6 +22,7 @@
struct intel_pt_pkt;
+void *intel_pt_log_fp(void);
void intel_pt_log_enable(void);
void intel_pt_log_disable(void);
void intel_pt_log_set_name(const char *name);
diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c
index 86cc9a6..149ff36 100644
--- a/tools/perf/util/intel-pt.c
+++ b/tools/perf/util/intel-pt.c
@@ -206,6 +206,16 @@ static void intel_pt_dump_event(struct intel_pt *pt, unsigned char *buf,
intel_pt_dump(pt, buf, len);
}
+static void intel_pt_log_event(union perf_event *event)
+{
+ FILE *f = intel_pt_log_fp();
+
+ if (!intel_pt_enable_logging || !f)
+ return;
+
+ perf_event__fprintf(event, f);
+}
+
static int intel_pt_do_fix_overlap(struct intel_pt *pt, struct auxtrace_buffer *a,
struct auxtrace_buffer *b)
{
@@ -2010,9 +2020,9 @@ static int intel_pt_process_event(struct perf_session *session,
event->header.type == PERF_RECORD_SWITCH_CPU_WIDE)
err = intel_pt_context_switch(pt, event, sample);
- intel_pt_log("event %s (%u): cpu %d time %"PRIu64" tsc %#"PRIx64"\n",
- perf_event__name(event->header.type), event->header.type,
- sample->cpu, sample->time, timestamp);
+ intel_pt_log("event %u: cpu %d time %"PRIu64" tsc %#"PRIx64" ",
+ event->header.type, sample->cpu, sample->time, timestamp);
+ intel_pt_log_event(event);
return err;
}
diff --git a/tools/perf/util/pmu.c b/tools/perf/util/pmu.c
index 7799788..7e49baa 100644
--- a/tools/perf/util/pmu.c
+++ b/tools/perf/util/pmu.c
@@ -773,7 +773,7 @@ static void pmu_add_cpu_aliases(struct list_head *head, struct perf_pmu *pmu)
if (!is_arm_pmu_core(name)) {
pname = pe->pmu ? pe->pmu : "cpu";
- if (strncmp(pname, name, strlen(pname)))
+ if (strcmp(pname, name))
continue;
}
diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c
index 9527d47..01ec04b 100644
--- a/tools/testing/nvdimm/test/nfit.c
+++ b/tools/testing/nvdimm/test/nfit.c
@@ -140,8 +140,8 @@ static u32 handle[] = {
[6] = NFIT_DIMM_HANDLE(1, 0, 0, 0, 1),
};
-static unsigned long dimm_fail_cmd_flags[NUM_DCR];
-static int dimm_fail_cmd_code[NUM_DCR];
+static unsigned long dimm_fail_cmd_flags[ARRAY_SIZE(handle)];
+static int dimm_fail_cmd_code[ARRAY_SIZE(handle)];
static const struct nd_intel_smart smart_def = {
.flags = ND_INTEL_SMART_HEALTH_VALID
@@ -205,7 +205,7 @@ struct nfit_test {
unsigned long deadline;
spinlock_t lock;
} ars_state;
- struct device *dimm_dev[NUM_DCR];
+ struct device *dimm_dev[ARRAY_SIZE(handle)];
struct nd_intel_smart *smart;
struct nd_intel_smart_threshold *smart_threshold;
struct badrange badrange;
@@ -2680,7 +2680,7 @@ static int nfit_test_probe(struct platform_device *pdev)
u32 nfit_handle = __to_nfit_memdev(nfit_mem)->device_handle;
int i;
- for (i = 0; i < NUM_DCR; i++)
+ for (i = 0; i < ARRAY_SIZE(handle); i++)
if (nfit_handle == handle[i])
dev_set_drvdata(nfit_test->dimm_dev[i],
nfit_mem);
diff --git a/tools/testing/selftests/powerpc/mm/wild_bctr.c b/tools/testing/selftests/powerpc/mm/wild_bctr.c
index 1b0e9e9..f2fa101 100644
--- a/tools/testing/selftests/powerpc/mm/wild_bctr.c
+++ b/tools/testing/selftests/powerpc/mm/wild_bctr.c
@@ -47,8 +47,9 @@ static int ok(void)
return 0;
}
-#define REG_POISON 0x5a5aUL
-#define POISONED_REG(n) ((REG_POISON << 48) | ((n) << 32) | (REG_POISON << 16) | (n))
+#define REG_POISON 0x5a5a
+#define POISONED_REG(n) ((((unsigned long)REG_POISON) << 48) | ((n) << 32) | \
+ (((unsigned long)REG_POISON) << 16) | (n))
static inline void poison_regs(void)
{
@@ -105,6 +106,20 @@ static void dump_regs(void)
}
}
+#ifdef _CALL_AIXDESC
+struct opd {
+ unsigned long ip;
+ unsigned long toc;
+ unsigned long env;
+};
+static struct opd bad_opd = {
+ .ip = BAD_NIP,
+};
+#define BAD_FUNC (&bad_opd)
+#else
+#define BAD_FUNC BAD_NIP
+#endif
+
int test_wild_bctr(void)
{
int (*func_ptr)(void);
@@ -133,7 +148,7 @@ int test_wild_bctr(void)
poison_regs();
- func_ptr = (int (*)(void))BAD_NIP;
+ func_ptr = (int (*)(void))BAD_FUNC;
func_ptr();
FAIL_IF(1); /* we didn't segv? */