Merge branches 'iommu/fixes', 'arm/omap', 'arm/smmu', 'arm/shmobile', 'x86/amd', 'arm/exynos', 'arm/renesas', 'ppc/pamu' and 'arm/msm' into next
diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl
index 677a025..ba60d93 100644
--- a/Documentation/DocBook/drm.tmpl
+++ b/Documentation/DocBook/drm.tmpl
@@ -79,7 +79,7 @@
   <partintro>
     <para>
       This first part of the DRM Developer's Guide documents core DRM code,
-      helper libraries for writting drivers and generic userspace interfaces
+      helper libraries for writing drivers and generic userspace interfaces
       exposed by DRM drivers.
     </para>
   </partintro>
@@ -459,7 +459,7 @@
       providing a solution to every graphics memory-related problems, GEM
       identified common code between drivers and created a support library to
       share it. GEM has simpler initialization and execution requirements than
-      TTM, but has no video RAM management capabitilies and is thus limited to
+      TTM, but has no video RAM management capabilities and is thus limited to
       UMA devices.
     </para>
     <sect2>
@@ -889,7 +889,7 @@
 	    vice versa. Drivers must use the kernel dma-buf buffer sharing framework
 	    to manage the PRIME file descriptors. Similar to the mode setting
 	    API PRIME is agnostic to the underlying buffer object manager, as
-	    long as handles are 32bit unsinged integers.
+	    long as handles are 32bit unsigned integers.
 	  </para>
 	  <para>
 	    While non-GEM drivers must implement the operations themselves, GEM
@@ -2356,7 +2356,7 @@
       first create properties and then create and associate individual instances
       of those properties to objects. A property can be instantiated multiple
       times and associated with different objects. Values are stored in property
-      instances, and all other property information are stored in the propery
+      instances, and all other property information are stored in the property
       and shared between all instances of the property.
     </para>
     <para>
@@ -2697,10 +2697,10 @@
   <sect1>
     <title>Legacy Support Code</title>
     <para>
-      The section very brievely covers some of the old legacy support code which
+      The section very briefly covers some of the old legacy support code which
       is only used by old DRM drivers which have done a so-called shadow-attach
       to the underlying device instead of registering as a real driver. This
-      also includes some of the old generic buffer mangement and command
+      also includes some of the old generic buffer management and command
       submission code. Do not use any of this in new and modern drivers.
     </para>
 
diff --git a/Documentation/DocBook/media/Makefile b/Documentation/DocBook/media/Makefile
index f9fd615..1d27f0a 100644
--- a/Documentation/DocBook/media/Makefile
+++ b/Documentation/DocBook/media/Makefile
@@ -195,7 +195,7 @@
 #
 
 install_media_images = \
-	$(Q)cp $(OBJIMGFILES) $(MEDIA_SRC_DIR)/v4l/*.svg $(MEDIA_OBJ_DIR)/media_api
+	$(Q)-cp $(OBJIMGFILES) $(MEDIA_SRC_DIR)/v4l/*.svg $(MEDIA_OBJ_DIR)/media_api
 
 $(MEDIA_OBJ_DIR)/%: $(MEDIA_SRC_DIR)/%.b64
 	$(Q)base64 -d $< >$@
diff --git a/Documentation/devicetree/bindings/iommu/samsung,sysmmu.txt b/Documentation/devicetree/bindings/iommu/samsung,sysmmu.txt
new file mode 100644
index 0000000..6fa4c73
--- /dev/null
+++ b/Documentation/devicetree/bindings/iommu/samsung,sysmmu.txt
@@ -0,0 +1,70 @@
+Samsung Exynos IOMMU H/W, System MMU (System Memory Management Unit)
+
+Samsung's Exynos architecture contains System MMUs that enables scattered
+physical memory chunks visible as a contiguous region to DMA-capable peripheral
+devices like MFC, FIMC, FIMD, GScaler, FIMC-IS and so forth.
+
+System MMU is an IOMMU and supports identical translation table format to
+ARMv7 translation tables with minimum set of page properties including access
+permissions, shareability and security protection. In addition, System MMU has
+another capabilities like L2 TLB or block-fetch buffers to minimize translation
+latency.
+
+System MMUs are in many to one relation with peripheral devices, i.e. single
+peripheral device might have multiple System MMUs (usually one for each bus
+master), but one System MMU can handle transactions from only one peripheral
+device. The relation between a System MMU and the peripheral device needs to be
+defined in device node of the peripheral device.
+
+MFC in all Exynos SoCs and FIMD, M2M Scalers and G2D in Exynos5420 has 2 System
+MMUs.
+* MFC has one System MMU on its left and right bus.
+* FIMD in Exynos5420 has one System MMU for window 0 and 4, the other system MMU
+  for window 1, 2 and 3.
+* M2M Scalers and G2D in Exynos5420 has one System MMU on the read channel and
+  the other System MMU on the write channel.
+The drivers must consider how to handle those System MMUs. One of the idea is
+to implement child devices or sub-devices which are the client devices of the
+System MMU.
+
+Note:
+The current DT binding for the Exynos System MMU is incomplete.
+The following properties can be removed or changed, if found incompatible with
+the "Generic IOMMU Binding" support for attaching devices to the IOMMU.
+
+Required properties:
+- compatible: Should be "samsung,exynos-sysmmu"
+- reg: A tuple of base address and size of System MMU registers.
+- interrupt-parent: The phandle of the interrupt controller of System MMU
+- interrupts: An interrupt specifier for interrupt signal of System MMU,
+	      according to the format defined by a particular interrupt
+	      controller.
+- clock-names: Should be "sysmmu" if the System MMU is needed to gate its clock.
+	       Optional "master" if the clock to the System MMU is gated by
+	       another gate clock other than "sysmmu".
+	       Exynos4 SoCs, there needs no "master" clock.
+	       Exynos5 SoCs, some System MMUs must have "master" clocks.
+- clocks: Required if the System MMU is needed to gate its clock.
+- samsung,power-domain: Required if the System MMU is needed to gate its power.
+	  Please refer to the following document:
+	  Documentation/devicetree/bindings/arm/exynos/power_domain.txt
+
+Examples:
+	gsc_0: gsc@13e00000 {
+		compatible = "samsung,exynos5-gsc";
+		reg = <0x13e00000 0x1000>;
+		interrupts = <0 85 0>;
+		samsung,power-domain = <&pd_gsc>;
+		clocks = <&clock CLK_GSCL0>;
+		clock-names = "gscl";
+	};
+
+	sysmmu_gsc0: sysmmu@13E80000 {
+		compatible = "samsung,exynos-sysmmu";
+		reg = <0x13E80000 0x1000>;
+		interrupt-parent = <&combiner>;
+		interrupts = <2 0>;
+		clock-names = "sysmmu", "master";
+		clocks = <&clock CLK_SMMU_GSCL0>, <&clock CLK_GSCL0>;
+		samsung,power-domain = <&pd_gsc>;
+	};
diff --git a/Documentation/devicetree/bindings/net/mdio-gpio.txt b/Documentation/devicetree/bindings/net/mdio-gpio.txt
index c79bab0..8dbcf82 100644
--- a/Documentation/devicetree/bindings/net/mdio-gpio.txt
+++ b/Documentation/devicetree/bindings/net/mdio-gpio.txt
@@ -14,7 +14,7 @@
 Example:
 
 aliases {
-	mdio-gpio0 = <&mdio0>;
+	mdio-gpio0 = &mdio0;
 };
 
 mdio0: mdio {
diff --git a/Documentation/email-clients.txt b/Documentation/email-clients.txt
index e9f5dac..4e30eba 100644
--- a/Documentation/email-clients.txt
+++ b/Documentation/email-clients.txt
@@ -201,20 +201,15 @@
 
 - Edit your Thunderbird config settings so that it won't use format=flowed.
   Go to "edit->preferences->advanced->config editor" to bring up the
-  thunderbird's registry editor, and set "mailnews.send_plaintext_flowed" to
-  "false".
+  thunderbird's registry editor.
 
-- Disable HTML Format: Set "mail.identity.id1.compose_html" to "false".
+- Set "mailnews.send_plaintext_flowed" to "false"
 
-- Enable "preformat" mode: Set "editor.quotesPreformatted" to "true".
+- Set "mailnews.wraplength" from "72" to "0"
 
-- Enable UTF8: Set "prefs.converted-to-utf8" to "true".
+- "View" > "Message Body As" > "Plain Text"
 
-- Install the "toggle wordwrap" extension.  Download the file from:
-    https://addons.mozilla.org/thunderbird/addon/2351/
-  Then go to "tools->add ons", select "install" at the bottom of the screen,
-  and browse to where you saved the .xul file.  This adds an "Enable
-  Wordwrap" entry under the Options menu of the message composer.
+- "View" > "Character Encoding" > "Unicode (UTF-8)"
 
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 TkRat (GUI)
diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index 8b9cd8e..264bcde 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -1245,8 +1245,9 @@
 
 The "intr" line gives counts of interrupts  serviced since boot time, for each
 of the  possible system interrupts.   The first  column  is the  total of  all
-interrupts serviced; each  subsequent column is the  total for that particular
-interrupt.
+interrupts serviced  including  unnumbered  architecture specific  interrupts;
+each  subsequent column is the  total for that particular numbered interrupt.
+Unnumbered interrupts are not shown, only summed into the total.
 
 The "ctxt" line gives the total number of context switches across all CPUs.
 
diff --git a/Documentation/hwmon/sysfs-interface b/Documentation/hwmon/sysfs-interface
index 79f8257..2cc95ad 100644
--- a/Documentation/hwmon/sysfs-interface
+++ b/Documentation/hwmon/sysfs-interface
@@ -327,6 +327,13 @@
 		from the max value.
 		RW
 
+temp[1-*]_min_hyst
+		Temperature hysteresis value for min limit.
+		Unit: millidegree Celsius
+		Must be reported as an absolute temperature, NOT a delta
+		from the min value.
+		RW
+
 temp[1-*]_input Temperature input value.
 		Unit: millidegree Celsius
 		RO
@@ -362,6 +369,13 @@
 		Unit: millidegree Celsius
 		RW
 
+temp[1-*]_lcrit_hyst
+		Temperature hysteresis value for critical min limit.
+		Unit: millidegree Celsius
+		Must be reported as an absolute temperature, NOT a delta
+		from the critical min value.
+		RW
+
 temp[1-*]_offset
 		Temperature offset which is added to the temperature reading
 		by the chip.
diff --git a/Documentation/java.txt b/Documentation/java.txt
index e6a7232..4180205 100644
--- a/Documentation/java.txt
+++ b/Documentation/java.txt
@@ -188,6 +188,9 @@
 #define CP_METHODREF 10
 #define CP_INTERFACEMETHODREF 11
 #define CP_NAMEANDTYPE 12
+#define CP_METHODHANDLE 15
+#define CP_METHODTYPE 16
+#define CP_INVOKEDYNAMIC 18
 
 /* Define some commonly used error messages */
 
@@ -242,14 +245,19 @@
 		break;
 	case CP_CLASS:
 	case CP_STRING:
+	case CP_METHODTYPE:
 		seekerr = fseek(classfile, 2, SEEK_CUR);
 		break;
+	case CP_METHODHANDLE:
+		seekerr = fseek(classfile, 3, SEEK_CUR);
+		break;
 	case CP_INTEGER:
 	case CP_FLOAT:
 	case CP_FIELDREF:
 	case CP_METHODREF:
 	case CP_INTERFACEMETHODREF:
 	case CP_NAMEANDTYPE:
+	case CP_INVOKEDYNAMIC:
 		seekerr = fseek(classfile, 4, SEEK_CUR);
 		break;
 	case CP_LONG:
diff --git a/Documentation/networking/filter.txt b/Documentation/networking/filter.txt
index 81f940f..e3ba753 100644
--- a/Documentation/networking/filter.txt
+++ b/Documentation/networking/filter.txt
@@ -277,7 +277,7 @@
   mark                                  skb->mark
   queue                                 skb->queue_mapping
   hatype                                skb->dev->type
-  rxhash                                skb->rxhash
+  rxhash                                skb->hash
   cpu                                   raw_smp_processor_id()
   vlan_tci                              vlan_tx_tag_get(skb)
   vlan_pr                               vlan_tx_tag_present(skb)
diff --git a/Documentation/networking/packet_mmap.txt b/Documentation/networking/packet_mmap.txt
index 6fea79e..38112d5 100644
--- a/Documentation/networking/packet_mmap.txt
+++ b/Documentation/networking/packet_mmap.txt
@@ -578,7 +578,7 @@
 
 Currently implemented fanout policies are:
 
-  - PACKET_FANOUT_HASH: schedule to socket by skb's rxhash
+  - PACKET_FANOUT_HASH: schedule to socket by skb's packet hash
   - PACKET_FANOUT_LB: schedule to socket by round-robin
   - PACKET_FANOUT_CPU: schedule to socket by CPU packet arrives on
   - PACKET_FANOUT_RND: schedule to socket by random selection
diff --git a/MAINTAINERS b/MAINTAINERS
index 6846c7c..cc45111 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -537,7 +537,7 @@
 F:	arch/alpha/
 
 ALTERA TRIPLE SPEED ETHERNET DRIVER
-M:	Vince Bridgers <vbridgers2013@gmail.com
+M:	Vince Bridgers <vbridgers2013@gmail.com>
 L:	netdev@vger.kernel.org
 L:	nios2-dev@lists.rocketboards.org (moderated for non-subscribers)
 S:	Maintained
@@ -6514,10 +6514,10 @@
 F:	arch/openrisc/
 
 OPENVSWITCH
-M:	Jesse Gross <jesse@nicira.com>
+M:	Pravin Shelar <pshelar@nicira.com>
 L:	dev@openvswitch.org
 W:	http://openvswitch.org
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/jesse/openvswitch.git
+T:	git git://git.kernel.org/pub/scm/linux/kernel/git/pshelar/openvswitch.git
 S:	Maintained
 F:	net/openvswitch/
 
diff --git a/Makefile b/Makefile
index 9d99378..cf3412d7 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 3
 PATCHLEVEL = 15
 SUBLEVEL = 0
-EXTRAVERSION = -rc6
+EXTRAVERSION = -rc7
 NAME = Shuffling Zombie Juror
 
 # *DOCUMENTATION*
diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S
index f1432da..c5fa7a6 100644
--- a/arch/parisc/kernel/syscall_table.S
+++ b/arch/parisc/kernel/syscall_table.S
@@ -432,7 +432,7 @@
 	ENTRY_SAME(sched_setattr)
 	ENTRY_SAME(sched_getattr)	/* 335 */
 	ENTRY_COMP(utimes)
-	ENTRY_COMP(renameat2)
+	ENTRY_SAME(renameat2)
 
 	/* Nothing yet */
 
diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c
index 452d3eb..e9f8fa9 100644
--- a/arch/s390/net/bpf_jit_comp.c
+++ b/arch/s390/net/bpf_jit_comp.c
@@ -811,7 +811,7 @@
 		return NULL;
 	memset(header, 0, sz);
 	header->pages = sz / PAGE_SIZE;
-	hole = sz - (bpfsize + sizeof(*header));
+	hole = min(sz - (bpfsize + sizeof(*header)), PAGE_SIZE - sizeof(*header));
 	/* Insert random number of illegal instructions before BPF code
 	 * and make sure the first instruction starts at an even address.
 	 */
diff --git a/arch/sparc/include/asm/pgtable_64.h b/arch/sparc/include/asm/pgtable_64.h
index fde5aba..1a49ffd 100644
--- a/arch/sparc/include/asm/pgtable_64.h
+++ b/arch/sparc/include/asm/pgtable_64.h
@@ -24,7 +24,8 @@
 
 /* The kernel image occupies 0x4000000 to 0x6000000 (4MB --> 96MB).
  * The page copy blockops can use 0x6000000 to 0x8000000.
- * The TSB is mapped in the 0x8000000 to 0xa000000 range.
+ * The 8K TSB is mapped in the 0x8000000 to 0x8400000 range.
+ * The 4M TSB is mapped in the 0x8400000 to 0x8800000 range.
  * The PROM resides in an area spanning 0xf0000000 to 0x100000000.
  * The vmalloc area spans 0x100000000 to 0x200000000.
  * Since modules need to be in the lowest 32-bits of the address space,
@@ -33,7 +34,8 @@
  * 0x400000000.
  */
 #define	TLBTEMP_BASE		_AC(0x0000000006000000,UL)
-#define	TSBMAP_BASE		_AC(0x0000000008000000,UL)
+#define	TSBMAP_8K_BASE		_AC(0x0000000008000000,UL)
+#define	TSBMAP_4M_BASE		_AC(0x0000000008400000,UL)
 #define MODULES_VADDR		_AC(0x0000000010000000,UL)
 #define MODULES_LEN		_AC(0x00000000e0000000,UL)
 #define MODULES_END		_AC(0x00000000f0000000,UL)
diff --git a/arch/sparc/kernel/sysfs.c b/arch/sparc/kernel/sysfs.c
index a364000c..7f41d40 100644
--- a/arch/sparc/kernel/sysfs.c
+++ b/arch/sparc/kernel/sysfs.c
@@ -151,7 +151,7 @@
 			size_t count)
 {
 	unsigned long val, err;
-	int ret = sscanf(buf, "%ld", &val);
+	int ret = sscanf(buf, "%lu", &val);
 
 	if (ret != 1)
 		return -EINVAL;
diff --git a/arch/sparc/lib/NG2memcpy.S b/arch/sparc/lib/NG2memcpy.S
index 2c20ad6..30eee6e 100644
--- a/arch/sparc/lib/NG2memcpy.S
+++ b/arch/sparc/lib/NG2memcpy.S
@@ -236,6 +236,7 @@
 	 */
 	VISEntryHalf
 
+	membar		#Sync
 	alignaddr	%o1, %g0, %g0
 
 	add		%o1, (64 - 1), %o4
diff --git a/arch/sparc/mm/fault_64.c b/arch/sparc/mm/fault_64.c
index a8ff0d1..4ced3fc 100644
--- a/arch/sparc/mm/fault_64.c
+++ b/arch/sparc/mm/fault_64.c
@@ -281,18 +281,6 @@
 	show_regs(regs);
 }
 
-static void noinline __kprobes bogus_32bit_fault_address(struct pt_regs *regs,
-							 unsigned long addr)
-{
-	static int times;
-
-	if (times++ < 10)
-		printk(KERN_ERR "FAULT[%s:%d]: 32-bit process "
-		       "reports 64-bit fault address [%lx]\n",
-		       current->comm, current->pid, addr);
-	show_regs(regs);
-}
-
 asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs)
 {
 	enum ctx_state prev_state = exception_enter();
@@ -322,10 +310,8 @@
 				goto intr_or_no_mm;
 			}
 		}
-		if (unlikely((address >> 32) != 0)) {
-			bogus_32bit_fault_address(regs, address);
+		if (unlikely((address >> 32) != 0))
 			goto intr_or_no_mm;
-		}
 	}
 
 	if (regs->tstate & TSTATE_PRIV) {
diff --git a/arch/sparc/mm/tsb.c b/arch/sparc/mm/tsb.c
index f5d506f..fe19b81 100644
--- a/arch/sparc/mm/tsb.c
+++ b/arch/sparc/mm/tsb.c
@@ -133,7 +133,19 @@
 	mm->context.tsb_block[tsb_idx].tsb_nentries =
 		tsb_bytes / sizeof(struct tsb);
 
-	base = TSBMAP_BASE;
+	switch (tsb_idx) {
+	case MM_TSB_BASE:
+		base = TSBMAP_8K_BASE;
+		break;
+#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
+	case MM_TSB_HUGE:
+		base = TSBMAP_4M_BASE;
+		break;
+#endif
+	default:
+		BUG();
+	}
+
 	tte = pgprot_val(PAGE_KERNEL_LOCKED);
 	tsb_paddr = __pa(mm->context.tsb_block[tsb_idx].tsb);
 	BUG_ON(tsb_paddr & (tsb_bytes - 1UL));
diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c
index aa333d9..adb02aa 100644
--- a/arch/x86/kernel/cpu/perf_event_intel.c
+++ b/arch/x86/kernel/cpu/perf_event_intel.c
@@ -169,7 +169,6 @@
 {
 	FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */
 	FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */
-	FIXED_EVENT_CONSTRAINT(0x013c, 2), /* CPU_CLK_UNHALTED.REF */
 	FIXED_EVENT_CONSTRAINT(0x0300, 2), /* pseudo CPU_CLK_UNHALTED.REF */
 	EVENT_CONSTRAINT_END
 };
diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c
index dc01773..6d5663a 100644
--- a/arch/x86/net/bpf_jit_comp.c
+++ b/arch/x86/net/bpf_jit_comp.c
@@ -171,7 +171,7 @@
 	memset(header, 0xcc, sz); /* fill whole space with int3 instructions */
 
 	header->pages = sz / PAGE_SIZE;
-	hole = sz - (proglen + sizeof(*header));
+	hole = min(sz - (proglen + sizeof(*header)), PAGE_SIZE - sizeof(*header));
 
 	/* insert a random number of int3 instructions before BPF code */
 	*image_ptr = &header->image[prandom_u32() % hole];
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index a886713..d5d30ed 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -1009,6 +1009,7 @@
 		dma_unmap_page(dev, unmap->addr[i], unmap->len,
 			       DMA_BIDIRECTIONAL);
 	}
+	cnt = unmap->map_cnt;
 	mempool_free(unmap, __get_unmap_pool(cnt)->pool);
 }
 
@@ -1074,6 +1075,7 @@
 	memset(unmap, 0, sizeof(*unmap));
 	kref_init(&unmap->kref);
 	unmap->dev = dev;
+	unmap->map_cnt = nr;
 
 	return unmap;
 }
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index 766b68e..394cbc5 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -191,12 +191,10 @@
 
 static void mv_chan_activate(struct mv_xor_chan *chan)
 {
-	u32 activation;
-
 	dev_dbg(mv_chan_to_devp(chan), " activate chan.\n");
-	activation = readl_relaxed(XOR_ACTIVATION(chan));
-	activation |= 0x1;
-	writel_relaxed(activation, XOR_ACTIVATION(chan));
+
+	/* writel ensures all descriptors are flushed before activation */
+	writel(BIT(0), XOR_ACTIVATION(chan));
 }
 
 static char mv_chan_is_busy(struct mv_xor_chan *chan)
diff --git a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
index 7762665..876de9a 100644
--- a/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/engine/disp/nvd0.c
@@ -1009,7 +1009,7 @@
 	}
 
 	if (outp == 8)
-		return false;
+		return conf;
 
 	data = exec_lookup(priv, head, outp, ctrl, dcb, &ver, &hdr, &cnt, &len, &info1);
 	if (data == 0x0000)
diff --git a/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c b/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c
index 43fec17..bbf117b 100644
--- a/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c
+++ b/drivers/gpu/drm/nouveau/core/subdev/therm/nvd0.c
@@ -40,6 +40,7 @@
 		case 0x00: return 2;
 		case 0x19: return 1;
 		case 0x1c: return 0;
+		case 0x1e: return 2;
 		default:
 			break;
 		}
diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
index 6852861..8149e7c 100644
--- a/drivers/gpu/drm/radeon/radeon.h
+++ b/drivers/gpu/drm/radeon/radeon.h
@@ -1642,6 +1642,7 @@
 	unsigned		fb_version;
 	atomic_t		handles[RADEON_MAX_VCE_HANDLES];
 	struct drm_file		*filp[RADEON_MAX_VCE_HANDLES];
+	unsigned		img_size[RADEON_MAX_VCE_HANDLES];
 	struct delayed_work	idle_work;
 };
 
@@ -1655,7 +1656,7 @@
 			       uint32_t handle, struct radeon_fence **fence);
 void radeon_vce_free_handles(struct radeon_device *rdev, struct drm_file *filp);
 void radeon_vce_note_usage(struct radeon_device *rdev);
-int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi);
+int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi, unsigned size);
 int radeon_vce_cs_parse(struct radeon_cs_parser *p);
 bool radeon_vce_semaphore_emit(struct radeon_device *rdev,
 			       struct radeon_ring *ring,
@@ -2640,7 +2641,8 @@
 #define ASIC_IS_DCE8(rdev) ((rdev->family >= CHIP_BONAIRE))
 #define ASIC_IS_DCE81(rdev) ((rdev->family == CHIP_KAVERI))
 #define ASIC_IS_DCE82(rdev) ((rdev->family == CHIP_BONAIRE))
-#define ASIC_IS_DCE83(rdev) ((rdev->family == CHIP_KABINI))
+#define ASIC_IS_DCE83(rdev) ((rdev->family == CHIP_KABINI) || \
+			     (rdev->family == CHIP_MULLINS))
 
 #define ASIC_IS_LOMBOK(rdev) ((rdev->ddev->pdev->device == 0x6849) || \
 			      (rdev->ddev->pdev->device == 0x6850) || \
diff --git a/drivers/gpu/drm/radeon/radeon_bios.c b/drivers/gpu/drm/radeon/radeon_bios.c
index b3633d9..9ab3097 100644
--- a/drivers/gpu/drm/radeon/radeon_bios.c
+++ b/drivers/gpu/drm/radeon/radeon_bios.c
@@ -196,6 +196,20 @@
 		}
 	}
 
+	if (!found) {
+		while ((pdev = pci_get_class(PCI_CLASS_DISPLAY_OTHER << 8, pdev)) != NULL) {
+			dhandle = ACPI_HANDLE(&pdev->dev);
+			if (!dhandle)
+				continue;
+
+			status = acpi_get_handle(dhandle, "ATRM", &atrm_handle);
+			if (!ACPI_FAILURE(status)) {
+				found = true;
+				break;
+			}
+		}
+	}
+
 	if (!found)
 		return false;
 
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index 408b6ac..f00dbbf 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -999,7 +999,7 @@
 
 	/* avoid high jitter with small fractional dividers */
 	if (pll->flags & RADEON_PLL_USE_FRAC_FB_DIV && (fb_div % 10)) {
-		fb_div_min = max(fb_div_min, (9 - (fb_div % 10)) * 20 + 60);
+		fb_div_min = max(fb_div_min, (9 - (fb_div % 10)) * 20 + 50);
 		if (fb_div < fb_div_min) {
 			unsigned tmp = DIV_ROUND_UP(fb_div_min, fb_div);
 			fb_div *= tmp;
diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c
index 0cc47f1..eaaedba 100644
--- a/drivers/gpu/drm/radeon/radeon_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_kms.c
@@ -577,28 +577,29 @@
 			return r;
 		}
 
-		r = radeon_bo_reserve(rdev->ring_tmp_bo.bo, false);
-		if (r) {
-			radeon_vm_fini(rdev, &fpriv->vm);
-			kfree(fpriv);
-			return r;
+		if (rdev->accel_working) {
+			r = radeon_bo_reserve(rdev->ring_tmp_bo.bo, false);
+			if (r) {
+				radeon_vm_fini(rdev, &fpriv->vm);
+				kfree(fpriv);
+				return r;
+			}
+
+			/* map the ib pool buffer read only into
+			 * virtual address space */
+			bo_va = radeon_vm_bo_add(rdev, &fpriv->vm,
+						 rdev->ring_tmp_bo.bo);
+			r = radeon_vm_bo_set_addr(rdev, bo_va, RADEON_VA_IB_OFFSET,
+						  RADEON_VM_PAGE_READABLE |
+						  RADEON_VM_PAGE_SNOOPED);
+
+			radeon_bo_unreserve(rdev->ring_tmp_bo.bo);
+			if (r) {
+				radeon_vm_fini(rdev, &fpriv->vm);
+				kfree(fpriv);
+				return r;
+			}
 		}
-
-		/* map the ib pool buffer read only into
-		 * virtual address space */
-		bo_va = radeon_vm_bo_add(rdev, &fpriv->vm,
-					 rdev->ring_tmp_bo.bo);
-		r = radeon_vm_bo_set_addr(rdev, bo_va, RADEON_VA_IB_OFFSET,
-					  RADEON_VM_PAGE_READABLE |
-					  RADEON_VM_PAGE_SNOOPED);
-
-		radeon_bo_unreserve(rdev->ring_tmp_bo.bo);
-		if (r) {
-			radeon_vm_fini(rdev, &fpriv->vm);
-			kfree(fpriv);
-			return r;
-		}
-
 		file_priv->driver_priv = fpriv;
 	}
 
@@ -626,13 +627,15 @@
 		struct radeon_bo_va *bo_va;
 		int r;
 
-		r = radeon_bo_reserve(rdev->ring_tmp_bo.bo, false);
-		if (!r) {
-			bo_va = radeon_vm_bo_find(&fpriv->vm,
-						  rdev->ring_tmp_bo.bo);
-			if (bo_va)
-				radeon_vm_bo_rmv(rdev, bo_va);
-			radeon_bo_unreserve(rdev->ring_tmp_bo.bo);
+		if (rdev->accel_working) {
+			r = radeon_bo_reserve(rdev->ring_tmp_bo.bo, false);
+			if (!r) {
+				bo_va = radeon_vm_bo_find(&fpriv->vm,
+							  rdev->ring_tmp_bo.bo);
+				if (bo_va)
+					radeon_vm_bo_rmv(rdev, bo_va);
+				radeon_bo_unreserve(rdev->ring_tmp_bo.bo);
+			}
 		}
 
 		radeon_vm_fini(rdev, &fpriv->vm);
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index 19bec0d..4faa4d6 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -458,7 +458,7 @@
 			 * into account. We don't want to disallow buffer moves
 			 * completely.
 			 */
-			if (current_domain != RADEON_GEM_DOMAIN_CPU &&
+			if ((lobj->alt_domain & current_domain) != 0 &&
 			    (domain & current_domain) == 0 && /* will be moved */
 			    bytes_moved > bytes_moved_threshold) {
 				/* don't move it */
@@ -699,22 +699,30 @@
 	rbo = container_of(bo, struct radeon_bo, tbo);
 	radeon_bo_check_tiling(rbo, 0, 0);
 	rdev = rbo->rdev;
-	if (bo->mem.mem_type == TTM_PL_VRAM) {
-		size = bo->mem.num_pages << PAGE_SHIFT;
-		offset = bo->mem.start << PAGE_SHIFT;
-		if ((offset + size) > rdev->mc.visible_vram_size) {
-			/* hurrah the memory is not visible ! */
-			radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_VRAM);
-			rbo->placement.lpfn = rdev->mc.visible_vram_size >> PAGE_SHIFT;
-			r = ttm_bo_validate(bo, &rbo->placement, false, false);
-			if (unlikely(r != 0))
-				return r;
-			offset = bo->mem.start << PAGE_SHIFT;
-			/* this should not happen */
-			if ((offset + size) > rdev->mc.visible_vram_size)
-				return -EINVAL;
-		}
+	if (bo->mem.mem_type != TTM_PL_VRAM)
+		return 0;
+
+	size = bo->mem.num_pages << PAGE_SHIFT;
+	offset = bo->mem.start << PAGE_SHIFT;
+	if ((offset + size) <= rdev->mc.visible_vram_size)
+		return 0;
+
+	/* hurrah the memory is not visible ! */
+	radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_VRAM);
+	rbo->placement.lpfn = rdev->mc.visible_vram_size >> PAGE_SHIFT;
+	r = ttm_bo_validate(bo, &rbo->placement, false, false);
+	if (unlikely(r == -ENOMEM)) {
+		radeon_ttm_placement_from_domain(rbo, RADEON_GEM_DOMAIN_GTT);
+		return ttm_bo_validate(bo, &rbo->placement, false, false);
+	} else if (unlikely(r != 0)) {
+		return r;
 	}
+
+	offset = bo->mem.start << PAGE_SHIFT;
+	/* this should never happen */
+	if ((offset + size) > rdev->mc.visible_vram_size)
+		return -EINVAL;
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index f30b842..53d6e1b 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -361,6 +361,11 @@
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct radeon_device *rdev = ddev->dev_private;
 
+	/* Can't set profile when the card is off */
+	if  ((rdev->flags & RADEON_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
 	mutex_lock(&rdev->pm.mutex);
 	if (rdev->pm.pm_method == PM_METHOD_PROFILE) {
 		if (strncmp("default", buf, strlen("default")) == 0)
@@ -409,6 +414,13 @@
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct radeon_device *rdev = ddev->dev_private;
 
+	/* Can't set method when the card is off */
+	if  ((rdev->flags & RADEON_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON)) {
+		count = -EINVAL;
+		goto fail;
+	}
+
 	/* we don't support the legacy modes with dpm */
 	if (rdev->pm.pm_method == PM_METHOD_DPM) {
 		count = -EINVAL;
@@ -446,6 +458,10 @@
 	struct radeon_device *rdev = ddev->dev_private;
 	enum radeon_pm_state_type pm = rdev->pm.dpm.user_state;
 
+	if  ((rdev->flags & RADEON_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return snprintf(buf, PAGE_SIZE, "off\n");
+
 	return snprintf(buf, PAGE_SIZE, "%s\n",
 			(pm == POWER_STATE_TYPE_BATTERY) ? "battery" :
 			(pm == POWER_STATE_TYPE_BALANCED) ? "balanced" : "performance");
@@ -459,6 +475,11 @@
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct radeon_device *rdev = ddev->dev_private;
 
+	/* Can't set dpm state when the card is off */
+	if  ((rdev->flags & RADEON_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
 	mutex_lock(&rdev->pm.mutex);
 	if (strncmp("battery", buf, strlen("battery")) == 0)
 		rdev->pm.dpm.user_state = POWER_STATE_TYPE_BATTERY;
@@ -485,6 +506,10 @@
 	struct radeon_device *rdev = ddev->dev_private;
 	enum radeon_dpm_forced_level level = rdev->pm.dpm.forced_level;
 
+	if  ((rdev->flags & RADEON_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return snprintf(buf, PAGE_SIZE, "off\n");
+
 	return snprintf(buf, PAGE_SIZE, "%s\n",
 			(level == RADEON_DPM_FORCED_LEVEL_AUTO) ? "auto" :
 			(level == RADEON_DPM_FORCED_LEVEL_LOW) ? "low" : "high");
@@ -500,6 +525,11 @@
 	enum radeon_dpm_forced_level level;
 	int ret = 0;
 
+	/* Can't force performance level when the card is off */
+	if  ((rdev->flags & RADEON_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
 	mutex_lock(&rdev->pm.mutex);
 	if (strncmp("low", buf, strlen("low")) == 0) {
 		level = RADEON_DPM_FORCED_LEVEL_LOW;
@@ -538,8 +568,14 @@
 				      char *buf)
 {
 	struct radeon_device *rdev = dev_get_drvdata(dev);
+	struct drm_device *ddev = rdev->ddev;
 	int temp;
 
+	/* Can't get temperature when the card is off */
+	if  ((rdev->flags & RADEON_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
 	if (rdev->asic->pm.get_temperature)
 		temp = radeon_get_temperature(rdev);
 	else
@@ -1614,8 +1650,12 @@
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct radeon_device *rdev = dev->dev_private;
+	struct drm_device *ddev = rdev->ddev;
 
-	if (rdev->pm.dpm_enabled) {
+	if  ((rdev->flags & RADEON_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON)) {
+		seq_printf(m, "PX asic powered off\n");
+	} else if (rdev->pm.dpm_enabled) {
 		mutex_lock(&rdev->pm.mutex);
 		if (rdev->asic->dpm.debugfs_print_current_performance_level)
 			radeon_dpm_debugfs_print_current_performance_level(rdev, m);
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c
index f73324c..3971d96 100644
--- a/drivers/gpu/drm/radeon/radeon_vce.c
+++ b/drivers/gpu/drm/radeon/radeon_vce.c
@@ -443,13 +443,16 @@
  * @p: parser context
  * @lo: address of lower dword
  * @hi: address of higher dword
+ * @size: size of checker for relocation buffer
  *
  * Patch relocation inside command stream with real buffer address
  */
-int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi)
+int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi,
+			unsigned size)
 {
 	struct radeon_cs_chunk *relocs_chunk;
-	uint64_t offset;
+	struct radeon_cs_reloc *reloc;
+	uint64_t start, end, offset;
 	unsigned idx;
 
 	relocs_chunk = &p->chunks[p->chunk_relocs_idx];
@@ -462,15 +465,60 @@
 		return -EINVAL;
 	}
 
-	offset += p->relocs_ptr[(idx / 4)]->gpu_offset;
+	reloc = p->relocs_ptr[(idx / 4)];
+	start = reloc->gpu_offset;
+	end = start + radeon_bo_size(reloc->robj);
+	start += offset;
 
-        p->ib.ptr[lo] = offset & 0xFFFFFFFF;
-        p->ib.ptr[hi] = offset >> 32;
+	p->ib.ptr[lo] = start & 0xFFFFFFFF;
+	p->ib.ptr[hi] = start >> 32;
+
+	if (end <= start) {
+		DRM_ERROR("invalid reloc offset %llX!\n", offset);
+		return -EINVAL;
+	}
+	if ((end - start) < size) {
+		DRM_ERROR("buffer to small (%d / %d)!\n",
+			(unsigned)(end - start), size);
+		return -EINVAL;
+	}
 
 	return 0;
 }
 
 /**
+ * radeon_vce_validate_handle - validate stream handle
+ *
+ * @p: parser context
+ * @handle: handle to validate
+ *
+ * Validates the handle and return the found session index or -EINVAL
+ * we we don't have another free session index.
+ */
+int radeon_vce_validate_handle(struct radeon_cs_parser *p, uint32_t handle)
+{
+	unsigned i;
+
+	/* validate the handle */
+	for (i = 0; i < RADEON_MAX_VCE_HANDLES; ++i) {
+		if (atomic_read(&p->rdev->vce.handles[i]) == handle)
+			return i;
+	}
+
+	/* handle not found try to alloc a new one */
+	for (i = 0; i < RADEON_MAX_VCE_HANDLES; ++i) {
+		if (!atomic_cmpxchg(&p->rdev->vce.handles[i], 0, handle)) {
+			p->rdev->vce.filp[i] = p->filp;
+			p->rdev->vce.img_size[i] = 0;
+			return i;
+		}
+	}
+
+	DRM_ERROR("No more free VCE handles!\n");
+	return -EINVAL;
+}
+
+/**
  * radeon_vce_cs_parse - parse and validate the command stream
  *
  * @p: parser context
@@ -478,8 +526,10 @@
  */
 int radeon_vce_cs_parse(struct radeon_cs_parser *p)
 {
-	uint32_t handle = 0;
-	bool destroy = false;
+	int session_idx = -1;
+	bool destroyed = false;
+	uint32_t tmp, handle = 0;
+	uint32_t *size = &tmp;
 	int i, r;
 
 	while (p->idx < p->chunks[p->chunk_ib_idx].length_dw) {
@@ -491,13 +541,29 @@
                 	return -EINVAL;
 		}
 
+		if (destroyed) {
+			DRM_ERROR("No other command allowed after destroy!\n");
+			return -EINVAL;
+		}
+
 		switch (cmd) {
 		case 0x00000001: // session
 			handle = radeon_get_ib_value(p, p->idx + 2);
+			session_idx = radeon_vce_validate_handle(p, handle);
+			if (session_idx < 0)
+				return session_idx;
+			size = &p->rdev->vce.img_size[session_idx];
 			break;
 
 		case 0x00000002: // task info
+			break;
+
 		case 0x01000001: // create
+			*size = radeon_get_ib_value(p, p->idx + 8) *
+				radeon_get_ib_value(p, p->idx + 10) *
+				8 * 3 / 2;
+			break;
+
 		case 0x04000001: // config extension
 		case 0x04000002: // pic control
 		case 0x04000005: // rate control
@@ -506,23 +572,39 @@
 			break;
 
 		case 0x03000001: // encode
-			r = radeon_vce_cs_reloc(p, p->idx + 10, p->idx + 9);
+			r = radeon_vce_cs_reloc(p, p->idx + 10, p->idx + 9,
+						*size);
 			if (r)
 				return r;
 
-			r = radeon_vce_cs_reloc(p, p->idx + 12, p->idx + 11);
+			r = radeon_vce_cs_reloc(p, p->idx + 12, p->idx + 11,
+						*size / 3);
 			if (r)
 				return r;
 			break;
 
 		case 0x02000001: // destroy
-			destroy = true;
+			destroyed = true;
 			break;
 
 		case 0x05000001: // context buffer
+			r = radeon_vce_cs_reloc(p, p->idx + 3, p->idx + 2,
+						*size * 2);
+			if (r)
+				return r;
+			break;
+
 		case 0x05000004: // video bitstream buffer
+			tmp = radeon_get_ib_value(p, p->idx + 4);
+			r = radeon_vce_cs_reloc(p, p->idx + 3, p->idx + 2,
+						tmp);
+			if (r)
+				return r;
+			break;
+
 		case 0x05000005: // feedback buffer
-			r = radeon_vce_cs_reloc(p, p->idx + 3, p->idx + 2);
+			r = radeon_vce_cs_reloc(p, p->idx + 3, p->idx + 2,
+						4096);
 			if (r)
 				return r;
 			break;
@@ -532,33 +614,21 @@
 			return -EINVAL;
 		}
 
+		if (session_idx == -1) {
+			DRM_ERROR("no session command at start of IB\n");
+			return -EINVAL;
+		}
+
 		p->idx += len / 4;
 	}
 
-	if (destroy) {
+	if (destroyed) {
 		/* IB contains a destroy msg, free the handle */
 		for (i = 0; i < RADEON_MAX_VCE_HANDLES; ++i)
 			atomic_cmpxchg(&p->rdev->vce.handles[i], handle, 0);
-
-		return 0;
-        }
-
-	/* create or encode, validate the handle */
-	for (i = 0; i < RADEON_MAX_VCE_HANDLES; ++i) {
-		if (atomic_read(&p->rdev->vce.handles[i]) == handle)
-			return 0;
 	}
 
-	/* handle not found try to alloc a new one */
-	for (i = 0; i < RADEON_MAX_VCE_HANDLES; ++i) {
-		if (!atomic_cmpxchg(&p->rdev->vce.handles[i], 0, handle)) {
-			p->rdev->vce.filp[i] = p->filp;
-			return 0;
-		}
-	}
-
-	DRM_ERROR("No more free VCE handles!\n");
-	return -EINVAL;
+	return 0;
 }
 
 /**
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index 2aae6ce..d9ab99f 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -595,7 +595,7 @@
 	ndw = 64;
 
 	/* assume the worst case */
-	ndw += vm->max_pde_used * 12;
+	ndw += vm->max_pde_used * 16;
 
 	/* update too big for an IB */
 	if (ndw > 0xfffff)
diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h
index 683532f..7321283 100644
--- a/drivers/gpu/drm/radeon/sid.h
+++ b/drivers/gpu/drm/radeon/sid.h
@@ -107,8 +107,8 @@
 #define		SPLL_CHG_STATUS				(1 << 1)
 #define	SPLL_CNTL_MODE					0x618
 #define		SPLL_SW_DIR_CONTROL			(1 << 0)
-#	define SPLL_REFCLK_SEL(x)			((x) << 8)
-#	define SPLL_REFCLK_SEL_MASK			0xFF00
+#	define SPLL_REFCLK_SEL(x)			((x) << 26)
+#	define SPLL_REFCLK_SEL_MASK			(3 << 26)
 
 #define	CG_SPLL_SPREAD_SPECTRUM				0x620
 #define		SSEN					(1 << 0)
diff --git a/drivers/hwmon/Kconfig b/drivers/hwmon/Kconfig
index bc196f4..4af0da9 100644
--- a/drivers/hwmon/Kconfig
+++ b/drivers/hwmon/Kconfig
@@ -1053,7 +1053,7 @@
 
 config SENSORS_NTC_THERMISTOR
 	tristate "NTC thermistor support"
-	depends on (!OF && !IIO) || (OF && IIO)
+	depends on !OF || IIO=n || IIO
 	help
 	  This driver supports NTC thermistors sensor reading and its
 	  interpretation. The driver can also monitor the temperature and
diff --git a/drivers/hwmon/ntc_thermistor.c b/drivers/hwmon/ntc_thermistor.c
index 8a17f01..e76feb86 100644
--- a/drivers/hwmon/ntc_thermistor.c
+++ b/drivers/hwmon/ntc_thermistor.c
@@ -44,6 +44,7 @@
 	unsigned int	ohm;
 };
 
+/* Order matters, ntc_match references the entries by index */
 static const struct platform_device_id ntc_thermistor_id[] = {
 	{ "ncp15wb473", TYPE_NCPXXWB473 },
 	{ "ncp18wb473", TYPE_NCPXXWB473 },
@@ -141,7 +142,7 @@
 	char name[PLATFORM_NAME_SIZE];
 };
 
-#ifdef CONFIG_OF
+#if defined(CONFIG_OF) && IS_ENABLED(CONFIG_IIO)
 static int ntc_adc_iio_read(struct ntc_thermistor_platform_data *pdata)
 {
 	struct iio_channel *channel = pdata->chan;
@@ -163,15 +164,15 @@
 
 static const struct of_device_id ntc_match[] = {
 	{ .compatible = "ntc,ncp15wb473",
-		.data = &ntc_thermistor_id[TYPE_NCPXXWB473] },
+		.data = &ntc_thermistor_id[0] },
 	{ .compatible = "ntc,ncp18wb473",
-		.data = &ntc_thermistor_id[TYPE_NCPXXWB473] },
+		.data = &ntc_thermistor_id[1] },
 	{ .compatible = "ntc,ncp21wb473",
-		.data = &ntc_thermistor_id[TYPE_NCPXXWB473] },
+		.data = &ntc_thermistor_id[2] },
 	{ .compatible = "ntc,ncp03wb473",
-		.data = &ntc_thermistor_id[TYPE_NCPXXWB473] },
+		.data = &ntc_thermistor_id[3] },
 	{ .compatible = "ntc,ncp15wl333",
-		.data = &ntc_thermistor_id[TYPE_NCPXXWL333] },
+		.data = &ntc_thermistor_id[4] },
 	{ },
 };
 MODULE_DEVICE_TABLE(of, ntc_match);
@@ -223,6 +224,8 @@
 	return NULL;
 }
 
+#define ntc_match	NULL
+
 static void ntc_iio_channel_release(struct ntc_thermistor_platform_data *pdata)
 { }
 #endif
diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c
index 1b6dbe15..199c789 100644
--- a/drivers/infiniband/hw/mlx4/main.c
+++ b/drivers/infiniband/hw/mlx4/main.c
@@ -48,6 +48,7 @@
 
 #include <linux/mlx4/driver.h>
 #include <linux/mlx4/cmd.h>
+#include <linux/mlx4/qp.h>
 
 #include "mlx4_ib.h"
 #include "user.h"
@@ -1614,6 +1615,53 @@
 }
 #endif
 
+#define MLX4_IB_INVALID_MAC	((u64)-1)
+static void mlx4_ib_update_qps(struct mlx4_ib_dev *ibdev,
+			       struct net_device *dev,
+			       int port)
+{
+	u64 new_smac = 0;
+	u64 release_mac = MLX4_IB_INVALID_MAC;
+	struct mlx4_ib_qp *qp;
+
+	read_lock(&dev_base_lock);
+	new_smac = mlx4_mac_to_u64(dev->dev_addr);
+	read_unlock(&dev_base_lock);
+
+	mutex_lock(&ibdev->qp1_proxy_lock[port - 1]);
+	qp = ibdev->qp1_proxy[port - 1];
+	if (qp) {
+		int new_smac_index;
+		u64 old_smac = qp->pri.smac;
+		struct mlx4_update_qp_params update_params;
+
+		if (new_smac == old_smac)
+			goto unlock;
+
+		new_smac_index = mlx4_register_mac(ibdev->dev, port, new_smac);
+
+		if (new_smac_index < 0)
+			goto unlock;
+
+		update_params.smac_index = new_smac_index;
+		if (mlx4_update_qp(ibdev->dev, &qp->mqp, MLX4_UPDATE_QP_SMAC,
+				   &update_params)) {
+			release_mac = new_smac;
+			goto unlock;
+		}
+
+		qp->pri.smac = new_smac;
+		qp->pri.smac_index = new_smac_index;
+
+		release_mac = old_smac;
+	}
+
+unlock:
+	mutex_unlock(&ibdev->qp1_proxy_lock[port - 1]);
+	if (release_mac != MLX4_IB_INVALID_MAC)
+		mlx4_unregister_mac(ibdev->dev, port, release_mac);
+}
+
 static void mlx4_ib_get_dev_addr(struct net_device *dev,
 				 struct mlx4_ib_dev *ibdev, u8 port)
 {
@@ -1689,9 +1737,13 @@
 	return 0;
 }
 
-static void mlx4_ib_scan_netdevs(struct mlx4_ib_dev *ibdev)
+static void mlx4_ib_scan_netdevs(struct mlx4_ib_dev *ibdev,
+				 struct net_device *dev,
+				 unsigned long event)
+
 {
 	struct mlx4_ib_iboe *iboe;
+	int update_qps_port = -1;
 	int port;
 
 	iboe = &ibdev->iboe;
@@ -1719,6 +1771,11 @@
 		}
 		curr_master = iboe->masters[port - 1];
 
+		if (dev == iboe->netdevs[port - 1] &&
+		    (event == NETDEV_CHANGEADDR || event == NETDEV_REGISTER ||
+		     event == NETDEV_UP || event == NETDEV_CHANGE))
+			update_qps_port = port;
+
 		if (curr_netdev) {
 			port_state = (netif_running(curr_netdev) && netif_carrier_ok(curr_netdev)) ?
 						IB_PORT_ACTIVE : IB_PORT_DOWN;
@@ -1752,6 +1809,9 @@
 	}
 
 	spin_unlock(&iboe->lock);
+
+	if (update_qps_port > 0)
+		mlx4_ib_update_qps(ibdev, dev, update_qps_port);
 }
 
 static int mlx4_ib_netdev_event(struct notifier_block *this,
@@ -1764,7 +1824,7 @@
 		return NOTIFY_DONE;
 
 	ibdev = container_of(this, struct mlx4_ib_dev, iboe.nb);
-	mlx4_ib_scan_netdevs(ibdev);
+	mlx4_ib_scan_netdevs(ibdev, dev, event);
 
 	return NOTIFY_DONE;
 }
@@ -2043,6 +2103,7 @@
 		goto err_map;
 
 	for (i = 0; i < ibdev->num_ports; ++i) {
+		mutex_init(&ibdev->qp1_proxy_lock[i]);
 		if (mlx4_ib_port_link_layer(&ibdev->ib_dev, i + 1) ==
 						IB_LINK_LAYER_ETHERNET) {
 			err = mlx4_counter_alloc(ibdev->dev, &ibdev->counters[i]);
@@ -2126,7 +2187,7 @@
 		for (i = 1 ; i <= ibdev->num_ports ; ++i)
 			reset_gid_table(ibdev, i);
 		rtnl_lock();
-		mlx4_ib_scan_netdevs(ibdev);
+		mlx4_ib_scan_netdevs(ibdev, NULL, 0);
 		rtnl_unlock();
 		mlx4_ib_init_gid_table(ibdev);
 	}
diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h
index f589522..66b0b7d 100644
--- a/drivers/infiniband/hw/mlx4/mlx4_ib.h
+++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h
@@ -522,6 +522,9 @@
 	int steer_qpn_count;
 	int steer_qpn_base;
 	int steering_support;
+	struct mlx4_ib_qp      *qp1_proxy[MLX4_MAX_PORTS];
+	/* lock when destroying qp1_proxy and getting netdev events */
+	struct mutex		qp1_proxy_lock[MLX4_MAX_PORTS];
 };
 
 struct ib_event_work {
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index 41308af..dc57482 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1132,6 +1132,12 @@
 	if (is_qp0(dev, mqp))
 		mlx4_CLOSE_PORT(dev->dev, mqp->port);
 
+	if (dev->qp1_proxy[mqp->port - 1] == mqp) {
+		mutex_lock(&dev->qp1_proxy_lock[mqp->port - 1]);
+		dev->qp1_proxy[mqp->port - 1] = NULL;
+		mutex_unlock(&dev->qp1_proxy_lock[mqp->port - 1]);
+	}
+
 	pd = get_pd(mqp);
 	destroy_qp_common(dev, mqp, !!pd->ibpd.uobject);
 
@@ -1646,6 +1652,8 @@
 				err = handle_eth_ud_smac_index(dev, qp, (u8 *)attr->smac, context);
 				if (err)
 					return -EINVAL;
+				if (qp->mlx4_ib_qp_type == MLX4_IB_QPT_PROXY_GSI)
+					dev->qp1_proxy[qp->port - 1] = qp;
 			}
 		}
 	}
diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig
index df56e4c..d260605 100644
--- a/drivers/iommu/Kconfig
+++ b/drivers/iommu/Kconfig
@@ -178,13 +178,13 @@
 
 config EXYNOS_IOMMU
 	bool "Exynos IOMMU Support"
-	depends on ARCH_EXYNOS && EXYNOS_DEV_SYSMMU
+	depends on ARCH_EXYNOS
 	select IOMMU_API
 	help
-	  Support for the IOMMU(System MMU) of Samsung Exynos application
-	  processor family. This enables H/W multimedia accellerators to see
-	  non-linear physical memory chunks as a linear memory in their
-	  address spaces
+	  Support for the IOMMU (System MMU) of Samsung Exynos application
+	  processor family. This enables H/W multimedia accelerators to see
+	  non-linear physical memory chunks as linear memory in their
+	  address space.
 
 	  If unsure, say N here.
 
@@ -193,9 +193,9 @@
 	depends on EXYNOS_IOMMU
 	help
 	  Select this to see the detailed log message that shows what
-	  happens in the IOMMU driver
+	  happens in the IOMMU driver.
 
-	  Say N unless you need kernel log message for IOMMU debugging
+	  Say N unless you need kernel log message for IOMMU debugging.
 
 config SHMOBILE_IPMMU
 	bool
@@ -272,6 +272,18 @@
 	default 256 if SHMOBILE_IOMMU_ADDRSIZE_64MB
 	default 128 if SHMOBILE_IOMMU_ADDRSIZE_32MB
 
+config IPMMU_VMSA
+	bool "Renesas VMSA-compatible IPMMU"
+	depends on ARM_LPAE
+	depends on ARCH_SHMOBILE || COMPILE_TEST
+	select IOMMU_API
+	select ARM_DMA_USE_IOMMU
+	help
+	  Support for the Renesas VMSA-compatible IPMMU Renesas found in the
+	  R-Mobile APE6 and R-Car H2/M2 SoCs.
+
+	  If unsure, say N.
+
 config SPAPR_TCE_IOMMU
 	bool "sPAPR TCE IOMMU Support"
 	depends on PPC_POWERNV || PPC_PSERIES
diff --git a/drivers/iommu/Makefile b/drivers/iommu/Makefile
index 5d58bf1..8893bad 100644
--- a/drivers/iommu/Makefile
+++ b/drivers/iommu/Makefile
@@ -7,6 +7,7 @@
 obj-$(CONFIG_ARM_SMMU) += arm-smmu.o
 obj-$(CONFIG_DMAR_TABLE) += dmar.o
 obj-$(CONFIG_INTEL_IOMMU) += iova.o intel-iommu.o
+obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o
 obj-$(CONFIG_IRQ_REMAP) += intel_irq_remapping.o irq_remapping.o
 obj-$(CONFIG_OMAP_IOMMU) += omap-iommu.o
 obj-$(CONFIG_OMAP_IOMMU) += omap-iommu2.o
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index 57068e8..4aec6a2 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -3499,8 +3499,6 @@
 {
 	struct iommu_dev_data *dev_data;
 	struct pci_dev *dev = NULL;
-	struct amd_iommu *iommu;
-	u16 devid;
 	int ret;
 
 	ret = alloc_passthrough_domain();
@@ -3514,12 +3512,6 @@
 		dev_data = get_dev_data(&dev->dev);
 		dev_data->passthrough = true;
 
-		devid = get_device_id(&dev->dev);
-
-		iommu = amd_iommu_rlookup_table[devid];
-		if (!iommu)
-			continue;
-
 		attach_device(&dev->dev, pt_domain);
 	}
 
diff --git a/drivers/iommu/amd_iommu_v2.c b/drivers/iommu/amd_iommu_v2.c
index 203b2e6..d4daa05 100644
--- a/drivers/iommu/amd_iommu_v2.c
+++ b/drivers/iommu/amd_iommu_v2.c
@@ -45,6 +45,8 @@
 struct pasid_state {
 	struct list_head list;			/* For global state-list */
 	atomic_t count;				/* Reference count */
+	atomic_t mmu_notifier_count;		/* Counting nested mmu_notifier
+						   calls */
 	struct task_struct *task;		/* Task bound to this PASID */
 	struct mm_struct *mm;			/* mm_struct for the faults */
 	struct mmu_notifier mn;                 /* mmu_otifier handle */
@@ -56,6 +58,8 @@
 };
 
 struct device_state {
+	struct list_head list;
+	u16 devid;
 	atomic_t count;
 	struct pci_dev *pdev;
 	struct pasid_state **states;
@@ -81,13 +85,9 @@
 	u16 flags;
 };
 
-static struct device_state **state_table;
+static LIST_HEAD(state_list);
 static spinlock_t state_lock;
 
-/* List and lock for all pasid_states */
-static LIST_HEAD(pasid_state_list);
-static DEFINE_SPINLOCK(ps_lock);
-
 static struct workqueue_struct *iommu_wq;
 
 /*
@@ -99,7 +99,6 @@
 
 static void free_pasid_states(struct device_state *dev_state);
 static void unbind_pasid(struct device_state *dev_state, int pasid);
-static int task_exit(struct notifier_block *nb, unsigned long e, void *data);
 
 static u16 device_id(struct pci_dev *pdev)
 {
@@ -111,13 +110,25 @@
 	return devid;
 }
 
+static struct device_state *__get_device_state(u16 devid)
+{
+	struct device_state *dev_state;
+
+	list_for_each_entry(dev_state, &state_list, list) {
+		if (dev_state->devid == devid)
+			return dev_state;
+	}
+
+	return NULL;
+}
+
 static struct device_state *get_device_state(u16 devid)
 {
 	struct device_state *dev_state;
 	unsigned long flags;
 
 	spin_lock_irqsave(&state_lock, flags);
-	dev_state = state_table[devid];
+	dev_state = __get_device_state(devid);
 	if (dev_state != NULL)
 		atomic_inc(&dev_state->count);
 	spin_unlock_irqrestore(&state_lock, flags);
@@ -158,29 +169,6 @@
 	free_device_state(dev_state);
 }
 
-static struct notifier_block profile_nb = {
-	.notifier_call = task_exit,
-};
-
-static void link_pasid_state(struct pasid_state *pasid_state)
-{
-	spin_lock(&ps_lock);
-	list_add_tail(&pasid_state->list, &pasid_state_list);
-	spin_unlock(&ps_lock);
-}
-
-static void __unlink_pasid_state(struct pasid_state *pasid_state)
-{
-	list_del(&pasid_state->list);
-}
-
-static void unlink_pasid_state(struct pasid_state *pasid_state)
-{
-	spin_lock(&ps_lock);
-	__unlink_pasid_state(pasid_state);
-	spin_unlock(&ps_lock);
-}
-
 /* Must be called under dev_state->lock */
 static struct pasid_state **__get_pasid_state_ptr(struct device_state *dev_state,
 						  int pasid, bool alloc)
@@ -337,7 +325,6 @@
 	if (pasid_state == NULL)
 		return;
 
-	unlink_pasid_state(pasid_state);
 	__unbind_pasid(pasid_state);
 	put_pasid_state_wait(pasid_state); /* Reference taken in this function */
 }
@@ -379,7 +366,12 @@
 			continue;
 
 		put_pasid_state(pasid_state);
-		unbind_pasid(dev_state, i);
+
+		/*
+		 * This will call the mn_release function and
+		 * unbind the PASID
+		 */
+		mmu_notifier_unregister(&pasid_state->mn, pasid_state->mm);
 	}
 
 	if (dev_state->pasid_levels == 2)
@@ -443,8 +435,11 @@
 	pasid_state = mn_to_state(mn);
 	dev_state   = pasid_state->device_state;
 
-	amd_iommu_domain_set_gcr3(dev_state->domain, pasid_state->pasid,
-				  __pa(empty_page_table));
+	if (atomic_add_return(1, &pasid_state->mmu_notifier_count) == 1) {
+		amd_iommu_domain_set_gcr3(dev_state->domain,
+					  pasid_state->pasid,
+					  __pa(empty_page_table));
+	}
 }
 
 static void mn_invalidate_range_end(struct mmu_notifier *mn,
@@ -457,11 +452,31 @@
 	pasid_state = mn_to_state(mn);
 	dev_state   = pasid_state->device_state;
 
-	amd_iommu_domain_set_gcr3(dev_state->domain, pasid_state->pasid,
-				  __pa(pasid_state->mm->pgd));
+	if (atomic_dec_and_test(&pasid_state->mmu_notifier_count)) {
+		amd_iommu_domain_set_gcr3(dev_state->domain,
+					  pasid_state->pasid,
+					  __pa(pasid_state->mm->pgd));
+	}
+}
+
+static void mn_release(struct mmu_notifier *mn, struct mm_struct *mm)
+{
+	struct pasid_state *pasid_state;
+	struct device_state *dev_state;
+
+	might_sleep();
+
+	pasid_state = mn_to_state(mn);
+	dev_state   = pasid_state->device_state;
+
+	if (pasid_state->device_state->inv_ctx_cb)
+		dev_state->inv_ctx_cb(dev_state->pdev, pasid_state->pasid);
+
+	unbind_pasid(dev_state, pasid_state->pasid);
 }
 
 static struct mmu_notifier_ops iommu_mn = {
+	.release		= mn_release,
 	.clear_flush_young      = mn_clear_flush_young,
 	.change_pte             = mn_change_pte,
 	.invalidate_page        = mn_invalidate_page,
@@ -606,53 +621,6 @@
 	.notifier_call = ppr_notifier,
 };
 
-static int task_exit(struct notifier_block *nb, unsigned long e, void *data)
-{
-	struct pasid_state *pasid_state;
-	struct task_struct *task;
-
-	task = data;
-
-	/*
-	 * Using this notifier is a hack - but there is no other choice
-	 * at the moment. What I really want is a sleeping notifier that
-	 * is called when an MM goes down. But such a notifier doesn't
-	 * exist yet. The notifier needs to sleep because it has to make
-	 * sure that the device does not use the PASID and the address
-	 * space anymore before it is destroyed. This includes waiting
-	 * for pending PRI requests to pass the workqueue. The
-	 * MMU-Notifiers would be a good fit, but they use RCU and so
-	 * they are not allowed to sleep. Lets see how we can solve this
-	 * in a more intelligent way in the future.
-	 */
-again:
-	spin_lock(&ps_lock);
-	list_for_each_entry(pasid_state, &pasid_state_list, list) {
-		struct device_state *dev_state;
-		int pasid;
-
-		if (pasid_state->task != task)
-			continue;
-
-		/* Drop Lock and unbind */
-		spin_unlock(&ps_lock);
-
-		dev_state = pasid_state->device_state;
-		pasid     = pasid_state->pasid;
-
-		if (pasid_state->device_state->inv_ctx_cb)
-			dev_state->inv_ctx_cb(dev_state->pdev, pasid);
-
-		unbind_pasid(dev_state, pasid);
-
-		/* Task may be in the list multiple times */
-		goto again;
-	}
-	spin_unlock(&ps_lock);
-
-	return NOTIFY_OK;
-}
-
 int amd_iommu_bind_pasid(struct pci_dev *pdev, int pasid,
 			 struct task_struct *task)
 {
@@ -682,6 +650,7 @@
 		goto out;
 
 	atomic_set(&pasid_state->count, 1);
+	atomic_set(&pasid_state->mmu_notifier_count, 0);
 	init_waitqueue_head(&pasid_state->wq);
 	spin_lock_init(&pasid_state->lock);
 
@@ -705,8 +674,6 @@
 	if (ret)
 		goto out_clear_state;
 
-	link_pasid_state(pasid_state);
-
 	return 0;
 
 out_clear_state:
@@ -727,6 +694,7 @@
 
 void amd_iommu_unbind_pasid(struct pci_dev *pdev, int pasid)
 {
+	struct pasid_state *pasid_state;
 	struct device_state *dev_state;
 	u16 devid;
 
@@ -743,7 +711,17 @@
 	if (pasid < 0 || pasid >= dev_state->max_pasids)
 		goto out;
 
-	unbind_pasid(dev_state, pasid);
+	pasid_state = get_pasid_state(dev_state, pasid);
+	if (pasid_state == NULL)
+		goto out;
+	/*
+	 * Drop reference taken here. We are safe because we still hold
+	 * the reference taken in the amd_iommu_bind_pasid function.
+	 */
+	put_pasid_state(pasid_state);
+
+	/* This will call the mn_release function and unbind the PASID */
+	mmu_notifier_unregister(&pasid_state->mn, pasid_state->mm);
 
 out:
 	put_device_state(dev_state);
@@ -773,7 +751,8 @@
 
 	spin_lock_init(&dev_state->lock);
 	init_waitqueue_head(&dev_state->wq);
-	dev_state->pdev = pdev;
+	dev_state->pdev  = pdev;
+	dev_state->devid = devid;
 
 	tmp = pasids;
 	for (dev_state->pasid_levels = 0; (tmp - 1) & ~0x1ff; tmp >>= 9)
@@ -803,13 +782,13 @@
 
 	spin_lock_irqsave(&state_lock, flags);
 
-	if (state_table[devid] != NULL) {
+	if (__get_device_state(devid) != NULL) {
 		spin_unlock_irqrestore(&state_lock, flags);
 		ret = -EBUSY;
 		goto out_free_domain;
 	}
 
-	state_table[devid] = dev_state;
+	list_add_tail(&dev_state->list, &state_list);
 
 	spin_unlock_irqrestore(&state_lock, flags);
 
@@ -841,13 +820,13 @@
 
 	spin_lock_irqsave(&state_lock, flags);
 
-	dev_state = state_table[devid];
+	dev_state = __get_device_state(devid);
 	if (dev_state == NULL) {
 		spin_unlock_irqrestore(&state_lock, flags);
 		return;
 	}
 
-	state_table[devid] = NULL;
+	list_del(&dev_state->list);
 
 	spin_unlock_irqrestore(&state_lock, flags);
 
@@ -874,7 +853,7 @@
 	spin_lock_irqsave(&state_lock, flags);
 
 	ret = -EINVAL;
-	dev_state = state_table[devid];
+	dev_state = __get_device_state(devid);
 	if (dev_state == NULL)
 		goto out_unlock;
 
@@ -905,7 +884,7 @@
 	spin_lock_irqsave(&state_lock, flags);
 
 	ret = -EINVAL;
-	dev_state = state_table[devid];
+	dev_state = __get_device_state(devid);
 	if (dev_state == NULL)
 		goto out_unlock;
 
@@ -922,7 +901,6 @@
 
 static int __init amd_iommu_v2_init(void)
 {
-	size_t state_table_size;
 	int ret;
 
 	pr_info("AMD IOMMUv2 driver by Joerg Roedel <joerg.roedel@amd.com>\n");
@@ -938,16 +916,10 @@
 
 	spin_lock_init(&state_lock);
 
-	state_table_size = MAX_DEVICES * sizeof(struct device_state *);
-	state_table = (void *)__get_free_pages(GFP_KERNEL | __GFP_ZERO,
-					       get_order(state_table_size));
-	if (state_table == NULL)
-		return -ENOMEM;
-
 	ret = -ENOMEM;
 	iommu_wq = create_workqueue("amd_iommu_v2");
 	if (iommu_wq == NULL)
-		goto out_free;
+		goto out;
 
 	ret = -ENOMEM;
 	empty_page_table = (u64 *)get_zeroed_page(GFP_KERNEL);
@@ -955,29 +927,24 @@
 		goto out_destroy_wq;
 
 	amd_iommu_register_ppr_notifier(&ppr_nb);
-	profile_event_register(PROFILE_TASK_EXIT, &profile_nb);
 
 	return 0;
 
 out_destroy_wq:
 	destroy_workqueue(iommu_wq);
 
-out_free:
-	free_pages((unsigned long)state_table, get_order(state_table_size));
-
+out:
 	return ret;
 }
 
 static void __exit amd_iommu_v2_exit(void)
 {
 	struct device_state *dev_state;
-	size_t state_table_size;
 	int i;
 
 	if (!amd_iommu_v2_supported())
 		return;
 
-	profile_event_unregister(PROFILE_TASK_EXIT, &profile_nb);
 	amd_iommu_unregister_ppr_notifier(&ppr_nb);
 
 	flush_workqueue(iommu_wq);
@@ -1000,9 +967,6 @@
 
 	destroy_workqueue(iommu_wq);
 
-	state_table_size = MAX_DEVICES * sizeof(struct device_state *);
-	free_pages((unsigned long)state_table, get_order(state_table_size));
-
 	free_page((unsigned long)empty_page_table);
 }
 
diff --git a/drivers/iommu/arm-smmu.c b/drivers/iommu/arm-smmu.c
index 647c3c7..1599354 100644
--- a/drivers/iommu/arm-smmu.c
+++ b/drivers/iommu/arm-smmu.c
@@ -1167,7 +1167,7 @@
 	for (i = 0; i < master->num_streamids; ++i) {
 		u32 idx, s2cr;
 		idx = master->smrs ? master->smrs[i].idx : master->streamids[i];
-		s2cr = (S2CR_TYPE_TRANS << S2CR_TYPE_SHIFT) |
+		s2cr = S2CR_TYPE_TRANS |
 		       (smmu_domain->root_cfg.cbndx << S2CR_CBNDX_SHIFT);
 		writel_relaxed(s2cr, gr0_base + ARM_SMMU_GR0_S2CR(idx));
 	}
@@ -1804,7 +1804,7 @@
 	 * allocation (PTRS_PER_PGD).
 	 */
 #ifdef CONFIG_64BIT
-	smmu->s1_output_size = min(39UL, size);
+	smmu->s1_output_size = min((unsigned long)VA_BITS, size);
 #else
 	smmu->s1_output_size = min(32UL, size);
 #endif
diff --git a/drivers/iommu/exynos-iommu.c b/drivers/iommu/exynos-iommu.c
index 0740189..99054d2 100644
--- a/drivers/iommu/exynos-iommu.c
+++ b/drivers/iommu/exynos-iommu.c
@@ -29,7 +29,8 @@
 #include <asm/cacheflush.h>
 #include <asm/pgtable.h>
 
-#include <mach/sysmmu.h>
+typedef u32 sysmmu_iova_t;
+typedef u32 sysmmu_pte_t;
 
 /* We does not consider super section mapping (16MB) */
 #define SECT_ORDER 20
@@ -44,28 +45,44 @@
 #define LPAGE_MASK (~(LPAGE_SIZE - 1))
 #define SPAGE_MASK (~(SPAGE_SIZE - 1))
 
-#define lv1ent_fault(sent) (((*(sent) & 3) == 0) || ((*(sent) & 3) == 3))
-#define lv1ent_page(sent) ((*(sent) & 3) == 1)
+#define lv1ent_fault(sent) ((*(sent) == ZERO_LV2LINK) || \
+			   ((*(sent) & 3) == 0) || ((*(sent) & 3) == 3))
+#define lv1ent_zero(sent) (*(sent) == ZERO_LV2LINK)
+#define lv1ent_page_zero(sent) ((*(sent) & 3) == 1)
+#define lv1ent_page(sent) ((*(sent) != ZERO_LV2LINK) && \
+			  ((*(sent) & 3) == 1))
 #define lv1ent_section(sent) ((*(sent) & 3) == 2)
 
 #define lv2ent_fault(pent) ((*(pent) & 3) == 0)
 #define lv2ent_small(pent) ((*(pent) & 2) == 2)
 #define lv2ent_large(pent) ((*(pent) & 3) == 1)
 
-#define section_phys(sent) (*(sent) & SECT_MASK)
-#define section_offs(iova) ((iova) & 0xFFFFF)
-#define lpage_phys(pent) (*(pent) & LPAGE_MASK)
-#define lpage_offs(iova) ((iova) & 0xFFFF)
-#define spage_phys(pent) (*(pent) & SPAGE_MASK)
-#define spage_offs(iova) ((iova) & 0xFFF)
+static u32 sysmmu_page_offset(sysmmu_iova_t iova, u32 size)
+{
+	return iova & (size - 1);
+}
 
-#define lv1ent_offset(iova) ((iova) >> SECT_ORDER)
-#define lv2ent_offset(iova) (((iova) & 0xFF000) >> SPAGE_ORDER)
+#define section_phys(sent) (*(sent) & SECT_MASK)
+#define section_offs(iova) sysmmu_page_offset((iova), SECT_SIZE)
+#define lpage_phys(pent) (*(pent) & LPAGE_MASK)
+#define lpage_offs(iova) sysmmu_page_offset((iova), LPAGE_SIZE)
+#define spage_phys(pent) (*(pent) & SPAGE_MASK)
+#define spage_offs(iova) sysmmu_page_offset((iova), SPAGE_SIZE)
 
 #define NUM_LV1ENTRIES 4096
-#define NUM_LV2ENTRIES 256
+#define NUM_LV2ENTRIES (SECT_SIZE / SPAGE_SIZE)
 
-#define LV2TABLE_SIZE (NUM_LV2ENTRIES * sizeof(long))
+static u32 lv1ent_offset(sysmmu_iova_t iova)
+{
+	return iova >> SECT_ORDER;
+}
+
+static u32 lv2ent_offset(sysmmu_iova_t iova)
+{
+	return (iova >> SPAGE_ORDER) & (NUM_LV2ENTRIES - 1);
+}
+
+#define LV2TABLE_SIZE (NUM_LV2ENTRIES * sizeof(sysmmu_pte_t))
 
 #define SPAGES_PER_LPAGE (LPAGE_SIZE / SPAGE_SIZE)
 
@@ -80,6 +97,13 @@
 #define CTRL_BLOCK	0x7
 #define CTRL_DISABLE	0x0
 
+#define CFG_LRU		0x1
+#define CFG_QOS(n)	((n & 0xF) << 7)
+#define CFG_MASK	0x0150FFFF /* Selecting bit 0-15, 20, 22 and 24 */
+#define CFG_ACGEN	(1 << 24) /* System MMU 3.3 only */
+#define CFG_SYSSEL	(1 << 22) /* System MMU 3.2 only */
+#define CFG_FLPDCACHE	(1 << 20) /* System MMU 3.2+ only */
+
 #define REG_MMU_CTRL		0x000
 #define REG_MMU_CFG		0x004
 #define REG_MMU_STATUS		0x008
@@ -96,19 +120,32 @@
 
 #define REG_MMU_VERSION		0x034
 
+#define MMU_MAJ_VER(val)	((val) >> 7)
+#define MMU_MIN_VER(val)	((val) & 0x7F)
+#define MMU_RAW_VER(reg)	(((reg) >> 21) & ((1 << 11) - 1)) /* 11 bits */
+
+#define MAKE_MMU_VER(maj, min)	((((maj) & 0xF) << 7) | ((min) & 0x7F))
+
 #define REG_PB0_SADDR		0x04C
 #define REG_PB0_EADDR		0x050
 #define REG_PB1_SADDR		0x054
 #define REG_PB1_EADDR		0x058
 
-static unsigned long *section_entry(unsigned long *pgtable, unsigned long iova)
+#define has_sysmmu(dev)		(dev->archdata.iommu != NULL)
+
+static struct kmem_cache *lv2table_kmem_cache;
+static sysmmu_pte_t *zero_lv2_table;
+#define ZERO_LV2LINK mk_lv1ent_page(virt_to_phys(zero_lv2_table))
+
+static sysmmu_pte_t *section_entry(sysmmu_pte_t *pgtable, sysmmu_iova_t iova)
 {
 	return pgtable + lv1ent_offset(iova);
 }
 
-static unsigned long *page_entry(unsigned long *sent, unsigned long iova)
+static sysmmu_pte_t *page_entry(sysmmu_pte_t *sent, sysmmu_iova_t iova)
 {
-	return (unsigned long *)__va(lv2table_base(sent)) + lv2ent_offset(iova);
+	return (sysmmu_pte_t *)phys_to_virt(
+				lv2table_base(sent)) + lv2ent_offset(iova);
 }
 
 enum exynos_sysmmu_inttype {
@@ -124,16 +161,6 @@
 	SYSMMU_FAULTS_NUM
 };
 
-/*
- * @itype: type of fault.
- * @pgtable_base: the physical address of page table base. This is 0 if @itype
- *                is SYSMMU_BUSERROR.
- * @fault_addr: the device (virtual) address that the System MMU tried to
- *             translated. This is 0 if @itype is SYSMMU_BUSERROR.
- */
-typedef int (*sysmmu_fault_handler_t)(enum exynos_sysmmu_inttype itype,
-			unsigned long pgtable_base, unsigned long fault_addr);
-
 static unsigned short fault_reg_offset[SYSMMU_FAULTS_NUM] = {
 	REG_PAGE_FAULT_ADDR,
 	REG_AR_FAULT_ADDR,
@@ -157,27 +184,34 @@
 	"UNKNOWN FAULT"
 };
 
+/* attached to dev.archdata.iommu of the master device */
+struct exynos_iommu_owner {
+	struct list_head client; /* entry of exynos_iommu_domain.clients */
+	struct device *dev;
+	struct device *sysmmu;
+	struct iommu_domain *domain;
+	void *vmm_data;         /* IO virtual memory manager's data */
+	spinlock_t lock;        /* Lock to preserve consistency of System MMU */
+};
+
 struct exynos_iommu_domain {
 	struct list_head clients; /* list of sysmmu_drvdata.node */
-	unsigned long *pgtable; /* lv1 page table, 16KB */
+	sysmmu_pte_t *pgtable; /* lv1 page table, 16KB */
 	short *lv2entcnt; /* free lv2 entry counter for each section */
 	spinlock_t lock; /* lock for this structure */
 	spinlock_t pgtablelock; /* lock for modifying page table @ pgtable */
 };
 
 struct sysmmu_drvdata {
-	struct list_head node; /* entry of exynos_iommu_domain.clients */
 	struct device *sysmmu;	/* System MMU's device descriptor */
-	struct device *dev;	/* Owner of system MMU */
-	char *dbgname;
-	int nsfrs;
-	void __iomem **sfrbases;
-	struct clk *clk[2];
+	struct device *master;	/* Owner of system MMU */
+	void __iomem *sfrbase;
+	struct clk *clk;
+	struct clk *clk_master;
 	int activations;
-	rwlock_t lock;
+	spinlock_t lock;
 	struct iommu_domain *domain;
-	sysmmu_fault_handler_t fault_handler;
-	unsigned long pgtable;
+	phys_addr_t pgtable;
 };
 
 static bool set_sysmmu_active(struct sysmmu_drvdata *data)
@@ -204,6 +238,11 @@
 	__raw_writel(CTRL_ENABLE, sfrbase + REG_MMU_CTRL);
 }
 
+static unsigned int __raw_sysmmu_version(struct sysmmu_drvdata *data)
+{
+	return MMU_RAW_VER(__raw_readl(data->sfrbase + REG_MMU_VERSION));
+}
+
 static bool sysmmu_block(void __iomem *sfrbase)
 {
 	int i = 120;
@@ -226,434 +265,428 @@
 }
 
 static void __sysmmu_tlb_invalidate_entry(void __iomem *sfrbase,
-						unsigned long iova)
+				sysmmu_iova_t iova, unsigned int num_inv)
 {
-	__raw_writel((iova & SPAGE_MASK) | 1, sfrbase + REG_MMU_FLUSH_ENTRY);
+	unsigned int i;
+
+	for (i = 0; i < num_inv; i++) {
+		__raw_writel((iova & SPAGE_MASK) | 1,
+				sfrbase + REG_MMU_FLUSH_ENTRY);
+		iova += SPAGE_SIZE;
+	}
 }
 
 static void __sysmmu_set_ptbase(void __iomem *sfrbase,
-				       unsigned long pgd)
+				       phys_addr_t pgd)
 {
-	__raw_writel(0x1, sfrbase + REG_MMU_CFG); /* 16KB LV1, LRU */
 	__raw_writel(pgd, sfrbase + REG_PT_BASE_ADDR);
 
 	__sysmmu_tlb_invalidate(sfrbase);
 }
 
-static void __sysmmu_set_prefbuf(void __iomem *sfrbase, unsigned long base,
-						unsigned long size, int idx)
+static void show_fault_information(const char *name,
+		enum exynos_sysmmu_inttype itype,
+		phys_addr_t pgtable_base, sysmmu_iova_t fault_addr)
 {
-	__raw_writel(base, sfrbase + REG_PB0_SADDR + idx * 8);
-	__raw_writel(size - 1 + base,  sfrbase + REG_PB0_EADDR + idx * 8);
-}
-
-static void __set_fault_handler(struct sysmmu_drvdata *data,
-					sysmmu_fault_handler_t handler)
-{
-	unsigned long flags;
-
-	write_lock_irqsave(&data->lock, flags);
-	data->fault_handler = handler;
-	write_unlock_irqrestore(&data->lock, flags);
-}
-
-void exynos_sysmmu_set_fault_handler(struct device *dev,
-					sysmmu_fault_handler_t handler)
-{
-	struct sysmmu_drvdata *data = dev_get_drvdata(dev->archdata.iommu);
-
-	__set_fault_handler(data, handler);
-}
-
-static int default_fault_handler(enum exynos_sysmmu_inttype itype,
-		     unsigned long pgtable_base, unsigned long fault_addr)
-{
-	unsigned long *ent;
+	sysmmu_pte_t *ent;
 
 	if ((itype >= SYSMMU_FAULTS_NUM) || (itype < SYSMMU_PAGEFAULT))
 		itype = SYSMMU_FAULT_UNKNOWN;
 
-	pr_err("%s occurred at 0x%lx(Page table base: 0x%lx)\n",
-			sysmmu_fault_name[itype], fault_addr, pgtable_base);
+	pr_err("%s occurred at %#x by %s(Page table base: %pa)\n",
+		sysmmu_fault_name[itype], fault_addr, name, &pgtable_base);
 
-	ent = section_entry(__va(pgtable_base), fault_addr);
-	pr_err("\tLv1 entry: 0x%lx\n", *ent);
+	ent = section_entry(phys_to_virt(pgtable_base), fault_addr);
+	pr_err("\tLv1 entry: %#x\n", *ent);
 
 	if (lv1ent_page(ent)) {
 		ent = page_entry(ent, fault_addr);
-		pr_err("\t Lv2 entry: 0x%lx\n", *ent);
+		pr_err("\t Lv2 entry: %#x\n", *ent);
 	}
-
-	pr_err("Generating Kernel OOPS... because it is unrecoverable.\n");
-
-	BUG();
-
-	return 0;
 }
 
 static irqreturn_t exynos_sysmmu_irq(int irq, void *dev_id)
 {
 	/* SYSMMU is in blocked when interrupt occurred. */
 	struct sysmmu_drvdata *data = dev_id;
-	struct resource *irqres;
-	struct platform_device *pdev;
 	enum exynos_sysmmu_inttype itype;
-	unsigned long addr = -1;
-
-	int i, ret = -ENOSYS;
-
-	read_lock(&data->lock);
+	sysmmu_iova_t addr = -1;
+	int ret = -ENOSYS;
 
 	WARN_ON(!is_sysmmu_active(data));
 
-	pdev = to_platform_device(data->sysmmu);
-	for (i = 0; i < (pdev->num_resources / 2); i++) {
-		irqres = platform_get_resource(pdev, IORESOURCE_IRQ, i);
-		if (irqres && ((int)irqres->start == irq))
-			break;
-	}
+	spin_lock(&data->lock);
 
-	if (i == pdev->num_resources) {
+	if (!IS_ERR(data->clk_master))
+		clk_enable(data->clk_master);
+
+	itype = (enum exynos_sysmmu_inttype)
+		__ffs(__raw_readl(data->sfrbase + REG_INT_STATUS));
+	if (WARN_ON(!((itype >= 0) && (itype < SYSMMU_FAULT_UNKNOWN))))
 		itype = SYSMMU_FAULT_UNKNOWN;
-	} else {
-		itype = (enum exynos_sysmmu_inttype)
-			__ffs(__raw_readl(data->sfrbases[i] + REG_INT_STATUS));
-		if (WARN_ON(!((itype >= 0) && (itype < SYSMMU_FAULT_UNKNOWN))))
-			itype = SYSMMU_FAULT_UNKNOWN;
-		else
-			addr = __raw_readl(
-				data->sfrbases[i] + fault_reg_offset[itype]);
-	}
-
-	if (data->domain)
-		ret = report_iommu_fault(data->domain, data->dev,
-				addr, itype);
-
-	if ((ret == -ENOSYS) && data->fault_handler) {
-		unsigned long base = data->pgtable;
-		if (itype != SYSMMU_FAULT_UNKNOWN)
-			base = __raw_readl(
-					data->sfrbases[i] + REG_PT_BASE_ADDR);
-		ret = data->fault_handler(itype, base, addr);
-	}
-
-	if (!ret && (itype != SYSMMU_FAULT_UNKNOWN))
-		__raw_writel(1 << itype, data->sfrbases[i] + REG_INT_CLEAR);
 	else
-		dev_dbg(data->sysmmu, "(%s) %s is not handled.\n",
-				data->dbgname, sysmmu_fault_name[itype]);
+		addr = __raw_readl(data->sfrbase + fault_reg_offset[itype]);
 
-	if (itype != SYSMMU_FAULT_UNKNOWN)
-		sysmmu_unblock(data->sfrbases[i]);
+	if (itype == SYSMMU_FAULT_UNKNOWN) {
+		pr_err("%s: Fault is not occurred by System MMU '%s'!\n",
+			__func__, dev_name(data->sysmmu));
+		pr_err("%s: Please check if IRQ is correctly configured.\n",
+			__func__);
+		BUG();
+	} else {
+		unsigned int base =
+				__raw_readl(data->sfrbase + REG_PT_BASE_ADDR);
+		show_fault_information(dev_name(data->sysmmu),
+					itype, base, addr);
+		if (data->domain)
+			ret = report_iommu_fault(data->domain,
+					data->master, addr, itype);
+	}
 
-	read_unlock(&data->lock);
+	/* fault is not recovered by fault handler */
+	BUG_ON(ret != 0);
+
+	__raw_writel(1 << itype, data->sfrbase + REG_INT_CLEAR);
+
+	sysmmu_unblock(data->sfrbase);
+
+	if (!IS_ERR(data->clk_master))
+		clk_disable(data->clk_master);
+
+	spin_unlock(&data->lock);
 
 	return IRQ_HANDLED;
 }
 
-static bool __exynos_sysmmu_disable(struct sysmmu_drvdata *data)
+static void __sysmmu_disable_nocount(struct sysmmu_drvdata *data)
 {
+	if (!IS_ERR(data->clk_master))
+		clk_enable(data->clk_master);
+
+	__raw_writel(CTRL_DISABLE, data->sfrbase + REG_MMU_CTRL);
+	__raw_writel(0, data->sfrbase + REG_MMU_CFG);
+
+	clk_disable(data->clk);
+	if (!IS_ERR(data->clk_master))
+		clk_disable(data->clk_master);
+}
+
+static bool __sysmmu_disable(struct sysmmu_drvdata *data)
+{
+	bool disabled;
 	unsigned long flags;
-	bool disabled = false;
-	int i;
 
-	write_lock_irqsave(&data->lock, flags);
+	spin_lock_irqsave(&data->lock, flags);
 
-	if (!set_sysmmu_inactive(data))
-		goto finish;
+	disabled = set_sysmmu_inactive(data);
 
-	for (i = 0; i < data->nsfrs; i++)
-		__raw_writel(CTRL_DISABLE, data->sfrbases[i] + REG_MMU_CTRL);
+	if (disabled) {
+		data->pgtable = 0;
+		data->domain = NULL;
 
-	if (data->clk[1])
-		clk_disable(data->clk[1]);
-	if (data->clk[0])
-		clk_disable(data->clk[0]);
+		__sysmmu_disable_nocount(data);
 
-	disabled = true;
-	data->pgtable = 0;
-	data->domain = NULL;
-finish:
-	write_unlock_irqrestore(&data->lock, flags);
+		dev_dbg(data->sysmmu, "Disabled\n");
+	} else  {
+		dev_dbg(data->sysmmu, "%d times left to disable\n",
+					data->activations);
+	}
 
-	if (disabled)
-		dev_dbg(data->sysmmu, "(%s) Disabled\n", data->dbgname);
-	else
-		dev_dbg(data->sysmmu, "(%s) %d times left to be disabled\n",
-					data->dbgname, data->activations);
+	spin_unlock_irqrestore(&data->lock, flags);
 
 	return disabled;
 }
 
+static void __sysmmu_init_config(struct sysmmu_drvdata *data)
+{
+	unsigned int cfg = CFG_LRU | CFG_QOS(15);
+	unsigned int ver;
+
+	ver = __raw_sysmmu_version(data);
+	if (MMU_MAJ_VER(ver) == 3) {
+		if (MMU_MIN_VER(ver) >= 2) {
+			cfg |= CFG_FLPDCACHE;
+			if (MMU_MIN_VER(ver) == 3) {
+				cfg |= CFG_ACGEN;
+				cfg &= ~CFG_LRU;
+			} else {
+				cfg |= CFG_SYSSEL;
+			}
+		}
+	}
+
+	__raw_writel(cfg, data->sfrbase + REG_MMU_CFG);
+}
+
+static void __sysmmu_enable_nocount(struct sysmmu_drvdata *data)
+{
+	if (!IS_ERR(data->clk_master))
+		clk_enable(data->clk_master);
+	clk_enable(data->clk);
+
+	__raw_writel(CTRL_BLOCK, data->sfrbase + REG_MMU_CTRL);
+
+	__sysmmu_init_config(data);
+
+	__sysmmu_set_ptbase(data->sfrbase, data->pgtable);
+
+	__raw_writel(CTRL_ENABLE, data->sfrbase + REG_MMU_CTRL);
+
+	if (!IS_ERR(data->clk_master))
+		clk_disable(data->clk_master);
+}
+
+static int __sysmmu_enable(struct sysmmu_drvdata *data,
+			phys_addr_t pgtable, struct iommu_domain *domain)
+{
+	int ret = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&data->lock, flags);
+	if (set_sysmmu_active(data)) {
+		data->pgtable = pgtable;
+		data->domain = domain;
+
+		__sysmmu_enable_nocount(data);
+
+		dev_dbg(data->sysmmu, "Enabled\n");
+	} else {
+		ret = (pgtable == data->pgtable) ? 1 : -EBUSY;
+
+		dev_dbg(data->sysmmu, "already enabled\n");
+	}
+
+	if (WARN_ON(ret < 0))
+		set_sysmmu_inactive(data); /* decrement count */
+
+	spin_unlock_irqrestore(&data->lock, flags);
+
+	return ret;
+}
+
 /* __exynos_sysmmu_enable: Enables System MMU
  *
  * returns -error if an error occurred and System MMU is not enabled,
  * 0 if the System MMU has been just enabled and 1 if System MMU was already
  * enabled before.
  */
-static int __exynos_sysmmu_enable(struct sysmmu_drvdata *data,
-			unsigned long pgtable, struct iommu_domain *domain)
+static int __exynos_sysmmu_enable(struct device *dev, phys_addr_t pgtable,
+				  struct iommu_domain *domain)
 {
-	int i, ret = 0;
+	int ret = 0;
 	unsigned long flags;
+	struct exynos_iommu_owner *owner = dev->archdata.iommu;
+	struct sysmmu_drvdata *data;
 
-	write_lock_irqsave(&data->lock, flags);
+	BUG_ON(!has_sysmmu(dev));
 
-	if (!set_sysmmu_active(data)) {
-		if (WARN_ON(pgtable != data->pgtable)) {
-			ret = -EBUSY;
-			set_sysmmu_inactive(data);
-		} else {
-			ret = 1;
-		}
+	spin_lock_irqsave(&owner->lock, flags);
 
-		dev_dbg(data->sysmmu, "(%s) Already enabled\n", data->dbgname);
-		goto finish;
-	}
+	data = dev_get_drvdata(owner->sysmmu);
 
-	if (data->clk[0])
-		clk_enable(data->clk[0]);
-	if (data->clk[1])
-		clk_enable(data->clk[1]);
+	ret = __sysmmu_enable(data, pgtable, domain);
+	if (ret >= 0)
+		data->master = dev;
 
-	data->pgtable = pgtable;
-
-	for (i = 0; i < data->nsfrs; i++) {
-		__sysmmu_set_ptbase(data->sfrbases[i], pgtable);
-
-		if ((readl(data->sfrbases[i] + REG_MMU_VERSION) >> 28) == 3) {
-			/* System MMU version is 3.x */
-			__raw_writel((1 << 12) | (2 << 28),
-					data->sfrbases[i] + REG_MMU_CFG);
-			__sysmmu_set_prefbuf(data->sfrbases[i], 0, -1, 0);
-			__sysmmu_set_prefbuf(data->sfrbases[i], 0, -1, 1);
-		}
-
-		__raw_writel(CTRL_ENABLE, data->sfrbases[i] + REG_MMU_CTRL);
-	}
-
-	data->domain = domain;
-
-	dev_dbg(data->sysmmu, "(%s) Enabled\n", data->dbgname);
-finish:
-	write_unlock_irqrestore(&data->lock, flags);
+	spin_unlock_irqrestore(&owner->lock, flags);
 
 	return ret;
 }
 
-int exynos_sysmmu_enable(struct device *dev, unsigned long pgtable)
+int exynos_sysmmu_enable(struct device *dev, phys_addr_t pgtable)
 {
-	struct sysmmu_drvdata *data = dev_get_drvdata(dev->archdata.iommu);
-	int ret;
-
 	BUG_ON(!memblock_is_memory(pgtable));
 
-	ret = pm_runtime_get_sync(data->sysmmu);
-	if (ret < 0) {
-		dev_dbg(data->sysmmu, "(%s) Failed to enable\n", data->dbgname);
-		return ret;
-	}
-
-	ret = __exynos_sysmmu_enable(data, pgtable, NULL);
-	if (WARN_ON(ret < 0)) {
-		pm_runtime_put(data->sysmmu);
-		dev_err(data->sysmmu,
-			"(%s) Already enabled with page table %#lx\n",
-			data->dbgname, data->pgtable);
-	} else {
-		data->dev = dev;
-	}
-
-	return ret;
+	return __exynos_sysmmu_enable(dev, pgtable, NULL);
 }
 
 static bool exynos_sysmmu_disable(struct device *dev)
 {
-	struct sysmmu_drvdata *data = dev_get_drvdata(dev->archdata.iommu);
-	bool disabled;
+	unsigned long flags;
+	bool disabled = true;
+	struct exynos_iommu_owner *owner = dev->archdata.iommu;
+	struct sysmmu_drvdata *data;
 
-	disabled = __exynos_sysmmu_disable(data);
-	pm_runtime_put(data->sysmmu);
+	BUG_ON(!has_sysmmu(dev));
+
+	spin_lock_irqsave(&owner->lock, flags);
+
+	data = dev_get_drvdata(owner->sysmmu);
+
+	disabled = __sysmmu_disable(data);
+	if (disabled)
+		data->master = NULL;
+
+	spin_unlock_irqrestore(&owner->lock, flags);
 
 	return disabled;
 }
 
-static void sysmmu_tlb_invalidate_entry(struct device *dev, unsigned long iova)
+static void __sysmmu_tlb_invalidate_flpdcache(struct sysmmu_drvdata *data,
+					      sysmmu_iova_t iova)
+{
+	if (__raw_sysmmu_version(data) == MAKE_MMU_VER(3, 3))
+		__raw_writel(iova | 0x1, data->sfrbase + REG_MMU_FLUSH_ENTRY);
+}
+
+static void sysmmu_tlb_invalidate_flpdcache(struct device *dev,
+					    sysmmu_iova_t iova)
 {
 	unsigned long flags;
-	struct sysmmu_drvdata *data = dev_get_drvdata(dev->archdata.iommu);
+	struct exynos_iommu_owner *owner = dev->archdata.iommu;
+	struct sysmmu_drvdata *data = dev_get_drvdata(owner->sysmmu);
 
-	read_lock_irqsave(&data->lock, flags);
+	if (!IS_ERR(data->clk_master))
+		clk_enable(data->clk_master);
 
+	spin_lock_irqsave(&data->lock, flags);
+	if (is_sysmmu_active(data))
+		__sysmmu_tlb_invalidate_flpdcache(data, iova);
+	spin_unlock_irqrestore(&data->lock, flags);
+
+	if (!IS_ERR(data->clk_master))
+		clk_disable(data->clk_master);
+}
+
+static void sysmmu_tlb_invalidate_entry(struct device *dev, sysmmu_iova_t iova,
+					size_t size)
+{
+	struct exynos_iommu_owner *owner = dev->archdata.iommu;
+	unsigned long flags;
+	struct sysmmu_drvdata *data;
+
+	data = dev_get_drvdata(owner->sysmmu);
+
+	spin_lock_irqsave(&data->lock, flags);
 	if (is_sysmmu_active(data)) {
-		int i;
-		for (i = 0; i < data->nsfrs; i++) {
-			if (sysmmu_block(data->sfrbases[i])) {
-				__sysmmu_tlb_invalidate_entry(
-						data->sfrbases[i], iova);
-				sysmmu_unblock(data->sfrbases[i]);
-			}
-		}
-	} else {
-		dev_dbg(data->sysmmu,
-			"(%s) Disabled. Skipping invalidating TLB.\n",
-			data->dbgname);
-	}
+		unsigned int num_inv = 1;
 
-	read_unlock_irqrestore(&data->lock, flags);
+		if (!IS_ERR(data->clk_master))
+			clk_enable(data->clk_master);
+
+		/*
+		 * L2TLB invalidation required
+		 * 4KB page: 1 invalidation
+		 * 64KB page: 16 invalidation
+		 * 1MB page: 64 invalidation
+		 * because it is set-associative TLB
+		 * with 8-way and 64 sets.
+		 * 1MB page can be cached in one of all sets.
+		 * 64KB page can be one of 16 consecutive sets.
+		 */
+		if (MMU_MAJ_VER(__raw_sysmmu_version(data)) == 2)
+			num_inv = min_t(unsigned int, size / PAGE_SIZE, 64);
+
+		if (sysmmu_block(data->sfrbase)) {
+			__sysmmu_tlb_invalidate_entry(
+				data->sfrbase, iova, num_inv);
+			sysmmu_unblock(data->sfrbase);
+		}
+		if (!IS_ERR(data->clk_master))
+			clk_disable(data->clk_master);
+	} else {
+		dev_dbg(dev, "disabled. Skipping TLB invalidation @ %#x\n",
+			iova);
+	}
+	spin_unlock_irqrestore(&data->lock, flags);
 }
 
 void exynos_sysmmu_tlb_invalidate(struct device *dev)
 {
+	struct exynos_iommu_owner *owner = dev->archdata.iommu;
 	unsigned long flags;
-	struct sysmmu_drvdata *data = dev_get_drvdata(dev->archdata.iommu);
-
-	read_lock_irqsave(&data->lock, flags);
-
-	if (is_sysmmu_active(data)) {
-		int i;
-		for (i = 0; i < data->nsfrs; i++) {
-			if (sysmmu_block(data->sfrbases[i])) {
-				__sysmmu_tlb_invalidate(data->sfrbases[i]);
-				sysmmu_unblock(data->sfrbases[i]);
-			}
-		}
-	} else {
-		dev_dbg(data->sysmmu,
-			"(%s) Disabled. Skipping invalidating TLB.\n",
-			data->dbgname);
-	}
-
-	read_unlock_irqrestore(&data->lock, flags);
-}
-
-static int exynos_sysmmu_probe(struct platform_device *pdev)
-{
-	int i, ret;
-	struct device *dev;
 	struct sysmmu_drvdata *data;
 
-	dev = &pdev->dev;
+	data = dev_get_drvdata(owner->sysmmu);
 
-	data = kzalloc(sizeof(*data), GFP_KERNEL);
-	if (!data) {
-		dev_dbg(dev, "Not enough memory\n");
-		ret = -ENOMEM;
-		goto err_alloc;
+	spin_lock_irqsave(&data->lock, flags);
+	if (is_sysmmu_active(data)) {
+		if (!IS_ERR(data->clk_master))
+			clk_enable(data->clk_master);
+		if (sysmmu_block(data->sfrbase)) {
+			__sysmmu_tlb_invalidate(data->sfrbase);
+			sysmmu_unblock(data->sfrbase);
+		}
+		if (!IS_ERR(data->clk_master))
+			clk_disable(data->clk_master);
+	} else {
+		dev_dbg(dev, "disabled. Skipping TLB invalidation\n");
+	}
+	spin_unlock_irqrestore(&data->lock, flags);
+}
+
+static int __init exynos_sysmmu_probe(struct platform_device *pdev)
+{
+	int irq, ret;
+	struct device *dev = &pdev->dev;
+	struct sysmmu_drvdata *data;
+	struct resource *res;
+
+	data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL);
+	if (!data)
+		return -ENOMEM;
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	data->sfrbase = devm_ioremap_resource(dev, res);
+	if (IS_ERR(data->sfrbase))
+		return PTR_ERR(data->sfrbase);
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq <= 0) {
+		dev_err(dev, "Unable to find IRQ resource\n");
+		return irq;
 	}
 
-	ret = dev_set_drvdata(dev, data);
+	ret = devm_request_irq(dev, irq, exynos_sysmmu_irq, 0,
+				dev_name(dev), data);
 	if (ret) {
-		dev_dbg(dev, "Unabled to initialize driver data\n");
-		goto err_init;
+		dev_err(dev, "Unabled to register handler of irq %d\n", irq);
+		return ret;
 	}
 
-	data->nsfrs = pdev->num_resources / 2;
-	data->sfrbases = kmalloc(sizeof(*data->sfrbases) * data->nsfrs,
-								GFP_KERNEL);
-	if (data->sfrbases == NULL) {
-		dev_dbg(dev, "Not enough memory\n");
-		ret = -ENOMEM;
-		goto err_init;
-	}
-
-	for (i = 0; i < data->nsfrs; i++) {
-		struct resource *res;
-		res = platform_get_resource(pdev, IORESOURCE_MEM, i);
-		if (!res) {
-			dev_dbg(dev, "Unable to find IOMEM region\n");
-			ret = -ENOENT;
-			goto err_res;
-		}
-
-		data->sfrbases[i] = ioremap(res->start, resource_size(res));
-		if (!data->sfrbases[i]) {
-			dev_dbg(dev, "Unable to map IOMEM @ PA:%#x\n",
-							res->start);
-			ret = -ENOENT;
-			goto err_res;
-		}
-	}
-
-	for (i = 0; i < data->nsfrs; i++) {
-		ret = platform_get_irq(pdev, i);
-		if (ret <= 0) {
-			dev_dbg(dev, "Unable to find IRQ resource\n");
-			goto err_irq;
-		}
-
-		ret = request_irq(ret, exynos_sysmmu_irq, 0,
-					dev_name(dev), data);
+	data->clk = devm_clk_get(dev, "sysmmu");
+	if (IS_ERR(data->clk)) {
+		dev_err(dev, "Failed to get clock!\n");
+		return PTR_ERR(data->clk);
+	} else  {
+		ret = clk_prepare(data->clk);
 		if (ret) {
-			dev_dbg(dev, "Unabled to register interrupt handler\n");
-			goto err_irq;
+			dev_err(dev, "Failed to prepare clk\n");
+			return ret;
 		}
 	}
 
-	if (dev_get_platdata(dev)) {
-		char *deli, *beg;
-		struct sysmmu_platform_data *platdata = dev_get_platdata(dev);
-
-		beg = platdata->clockname;
-
-		for (deli = beg; (*deli != '\0') && (*deli != ','); deli++)
-			/* NOTHING */;
-
-		if (*deli == '\0')
-			deli = NULL;
-		else
-			*deli = '\0';
-
-		data->clk[0] = clk_get(dev, beg);
-		if (IS_ERR(data->clk[0])) {
-			data->clk[0] = NULL;
-			dev_dbg(dev, "No clock descriptor registered\n");
+	data->clk_master = devm_clk_get(dev, "master");
+	if (!IS_ERR(data->clk_master)) {
+		ret = clk_prepare(data->clk_master);
+		if (ret) {
+			clk_unprepare(data->clk);
+			dev_err(dev, "Failed to prepare master's clk\n");
+			return ret;
 		}
-
-		if (data->clk[0] && deli) {
-			*deli = ',';
-			data->clk[1] = clk_get(dev, deli + 1);
-			if (IS_ERR(data->clk[1]))
-				data->clk[1] = NULL;
-		}
-
-		data->dbgname = platdata->dbgname;
 	}
 
 	data->sysmmu = dev;
-	rwlock_init(&data->lock);
-	INIT_LIST_HEAD(&data->node);
+	spin_lock_init(&data->lock);
 
-	__set_fault_handler(data, &default_fault_handler);
+	platform_set_drvdata(pdev, data);
 
-	if (dev->parent)
-		pm_runtime_enable(dev);
+	pm_runtime_enable(dev);
 
-	dev_dbg(dev, "(%s) Initialized\n", data->dbgname);
 	return 0;
-err_irq:
-	while (i-- > 0) {
-		int irq;
-
-		irq = platform_get_irq(pdev, i);
-		free_irq(irq, data);
-	}
-err_res:
-	while (data->nsfrs-- > 0)
-		iounmap(data->sfrbases[data->nsfrs]);
-	kfree(data->sfrbases);
-err_init:
-	kfree(data);
-err_alloc:
-	dev_err(dev, "Failed to initialize\n");
-	return ret;
 }
 
-static struct platform_driver exynos_sysmmu_driver = {
-	.probe		= exynos_sysmmu_probe,
-	.driver		= {
+static const struct of_device_id sysmmu_of_match[] __initconst = {
+	{ .compatible	= "samsung,exynos-sysmmu", },
+	{ },
+};
+
+static struct platform_driver exynos_sysmmu_driver __refdata = {
+	.probe	= exynos_sysmmu_probe,
+	.driver	= {
 		.owner		= THIS_MODULE,
 		.name		= "exynos-sysmmu",
+		.of_match_table	= sysmmu_of_match,
 	}
 };
 
@@ -667,21 +700,32 @@
 static int exynos_iommu_domain_init(struct iommu_domain *domain)
 {
 	struct exynos_iommu_domain *priv;
+	int i;
 
 	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
 	if (!priv)
 		return -ENOMEM;
 
-	priv->pgtable = (unsigned long *)__get_free_pages(
-						GFP_KERNEL | __GFP_ZERO, 2);
+	priv->pgtable = (sysmmu_pte_t *)__get_free_pages(GFP_KERNEL, 2);
 	if (!priv->pgtable)
 		goto err_pgtable;
 
-	priv->lv2entcnt = (short *)__get_free_pages(
-						GFP_KERNEL | __GFP_ZERO, 1);
+	priv->lv2entcnt = (short *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, 1);
 	if (!priv->lv2entcnt)
 		goto err_counter;
 
+	/* w/a of System MMU v3.3 to prevent caching 1MiB mapping */
+	for (i = 0; i < NUM_LV1ENTRIES; i += 8) {
+		priv->pgtable[i + 0] = ZERO_LV2LINK;
+		priv->pgtable[i + 1] = ZERO_LV2LINK;
+		priv->pgtable[i + 2] = ZERO_LV2LINK;
+		priv->pgtable[i + 3] = ZERO_LV2LINK;
+		priv->pgtable[i + 4] = ZERO_LV2LINK;
+		priv->pgtable[i + 5] = ZERO_LV2LINK;
+		priv->pgtable[i + 6] = ZERO_LV2LINK;
+		priv->pgtable[i + 7] = ZERO_LV2LINK;
+	}
+
 	pgtable_flush(priv->pgtable, priv->pgtable + NUM_LV1ENTRIES);
 
 	spin_lock_init(&priv->lock);
@@ -705,7 +749,7 @@
 static void exynos_iommu_domain_destroy(struct iommu_domain *domain)
 {
 	struct exynos_iommu_domain *priv = domain->priv;
-	struct sysmmu_drvdata *data;
+	struct exynos_iommu_owner *owner;
 	unsigned long flags;
 	int i;
 
@@ -713,16 +757,20 @@
 
 	spin_lock_irqsave(&priv->lock, flags);
 
-	list_for_each_entry(data, &priv->clients, node) {
-		while (!exynos_sysmmu_disable(data->dev))
+	list_for_each_entry(owner, &priv->clients, client) {
+		while (!exynos_sysmmu_disable(owner->dev))
 			; /* until System MMU is actually disabled */
 	}
 
+	while (!list_empty(&priv->clients))
+		list_del_init(priv->clients.next);
+
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 	for (i = 0; i < NUM_LV1ENTRIES; i++)
 		if (lv1ent_page(priv->pgtable + i))
-			kfree(__va(lv2table_base(priv->pgtable + i)));
+			kmem_cache_free(lv2table_kmem_cache,
+				phys_to_virt(lv2table_base(priv->pgtable + i)));
 
 	free_pages((unsigned long)priv->pgtable, 2);
 	free_pages((unsigned long)priv->lv2entcnt, 1);
@@ -733,114 +781,134 @@
 static int exynos_iommu_attach_device(struct iommu_domain *domain,
 				   struct device *dev)
 {
-	struct sysmmu_drvdata *data = dev_get_drvdata(dev->archdata.iommu);
+	struct exynos_iommu_owner *owner = dev->archdata.iommu;
 	struct exynos_iommu_domain *priv = domain->priv;
+	phys_addr_t pagetable = virt_to_phys(priv->pgtable);
 	unsigned long flags;
 	int ret;
 
-	ret = pm_runtime_get_sync(data->sysmmu);
-	if (ret < 0)
-		return ret;
-
-	ret = 0;
-
 	spin_lock_irqsave(&priv->lock, flags);
 
-	ret = __exynos_sysmmu_enable(data, __pa(priv->pgtable), domain);
-
+	ret = __exynos_sysmmu_enable(dev, pagetable, domain);
 	if (ret == 0) {
-		/* 'data->node' must not be appeared in priv->clients */
-		BUG_ON(!list_empty(&data->node));
-		data->dev = dev;
-		list_add_tail(&data->node, &priv->clients);
+		list_add_tail(&owner->client, &priv->clients);
+		owner->domain = domain;
 	}
 
 	spin_unlock_irqrestore(&priv->lock, flags);
 
 	if (ret < 0) {
-		dev_err(dev, "%s: Failed to attach IOMMU with pgtable %#lx\n",
-				__func__, __pa(priv->pgtable));
-		pm_runtime_put(data->sysmmu);
-	} else if (ret > 0) {
-		dev_dbg(dev, "%s: IOMMU with pgtable 0x%lx already attached\n",
-					__func__, __pa(priv->pgtable));
-	} else {
-		dev_dbg(dev, "%s: Attached new IOMMU with pgtable 0x%lx\n",
-					__func__, __pa(priv->pgtable));
+		dev_err(dev, "%s: Failed to attach IOMMU with pgtable %pa\n",
+					__func__, &pagetable);
+		return ret;
 	}
 
+	dev_dbg(dev, "%s: Attached IOMMU with pgtable %pa %s\n",
+		__func__, &pagetable, (ret == 0) ? "" : ", again");
+
 	return ret;
 }
 
 static void exynos_iommu_detach_device(struct iommu_domain *domain,
 				    struct device *dev)
 {
-	struct sysmmu_drvdata *data = dev_get_drvdata(dev->archdata.iommu);
+	struct exynos_iommu_owner *owner;
 	struct exynos_iommu_domain *priv = domain->priv;
-	struct list_head *pos;
+	phys_addr_t pagetable = virt_to_phys(priv->pgtable);
 	unsigned long flags;
-	bool found = false;
 
 	spin_lock_irqsave(&priv->lock, flags);
 
-	list_for_each(pos, &priv->clients) {
-		if (list_entry(pos, struct sysmmu_drvdata, node) == data) {
-			found = true;
+	list_for_each_entry(owner, &priv->clients, client) {
+		if (owner == dev->archdata.iommu) {
+			if (exynos_sysmmu_disable(dev)) {
+				list_del_init(&owner->client);
+				owner->domain = NULL;
+			}
 			break;
 		}
 	}
 
-	if (!found)
-		goto finish;
-
-	if (__exynos_sysmmu_disable(data)) {
-		dev_dbg(dev, "%s: Detached IOMMU with pgtable %#lx\n",
-					__func__, __pa(priv->pgtable));
-		list_del_init(&data->node);
-
-	} else {
-		dev_dbg(dev, "%s: Detaching IOMMU with pgtable %#lx delayed",
-					__func__, __pa(priv->pgtable));
-	}
-
-finish:
 	spin_unlock_irqrestore(&priv->lock, flags);
 
-	if (found)
-		pm_runtime_put(data->sysmmu);
+	if (owner == dev->archdata.iommu)
+		dev_dbg(dev, "%s: Detached IOMMU with pgtable %pa\n",
+					__func__, &pagetable);
+	else
+		dev_err(dev, "%s: No IOMMU is attached\n", __func__);
 }
 
-static unsigned long *alloc_lv2entry(unsigned long *sent, unsigned long iova,
-					short *pgcounter)
+static sysmmu_pte_t *alloc_lv2entry(struct exynos_iommu_domain *priv,
+		sysmmu_pte_t *sent, sysmmu_iova_t iova, short *pgcounter)
 {
+	if (lv1ent_section(sent)) {
+		WARN(1, "Trying mapping on %#08x mapped with 1MiB page", iova);
+		return ERR_PTR(-EADDRINUSE);
+	}
+
 	if (lv1ent_fault(sent)) {
-		unsigned long *pent;
+		sysmmu_pte_t *pent;
+		bool need_flush_flpd_cache = lv1ent_zero(sent);
 
-		pent = kzalloc(LV2TABLE_SIZE, GFP_ATOMIC);
-		BUG_ON((unsigned long)pent & (LV2TABLE_SIZE - 1));
+		pent = kmem_cache_zalloc(lv2table_kmem_cache, GFP_ATOMIC);
+		BUG_ON((unsigned int)pent & (LV2TABLE_SIZE - 1));
 		if (!pent)
-			return NULL;
+			return ERR_PTR(-ENOMEM);
 
-		*sent = mk_lv1ent_page(__pa(pent));
+		*sent = mk_lv1ent_page(virt_to_phys(pent));
 		*pgcounter = NUM_LV2ENTRIES;
 		pgtable_flush(pent, pent + NUM_LV2ENTRIES);
 		pgtable_flush(sent, sent + 1);
+
+		/*
+		 * If pretched SLPD is a fault SLPD in zero_l2_table, FLPD cache
+		 * may caches the address of zero_l2_table. This function
+		 * replaces the zero_l2_table with new L2 page table to write
+		 * valid mappings.
+		 * Accessing the valid area may cause page fault since FLPD
+		 * cache may still caches zero_l2_table for the valid area
+		 * instead of new L2 page table that have the mapping
+		 * information of the valid area
+		 * Thus any replacement of zero_l2_table with other valid L2
+		 * page table must involve FLPD cache invalidation for System
+		 * MMU v3.3.
+		 * FLPD cache invalidation is performed with TLB invalidation
+		 * by VPN without blocking. It is safe to invalidate TLB without
+		 * blocking because the target address of TLB invalidation is
+		 * not currently mapped.
+		 */
+		if (need_flush_flpd_cache) {
+			struct exynos_iommu_owner *owner;
+
+			spin_lock(&priv->lock);
+			list_for_each_entry(owner, &priv->clients, client)
+				sysmmu_tlb_invalidate_flpdcache(
+							owner->dev, iova);
+			spin_unlock(&priv->lock);
+		}
 	}
 
 	return page_entry(sent, iova);
 }
 
-static int lv1set_section(unsigned long *sent, phys_addr_t paddr, short *pgcnt)
+static int lv1set_section(struct exynos_iommu_domain *priv,
+			  sysmmu_pte_t *sent, sysmmu_iova_t iova,
+			  phys_addr_t paddr, short *pgcnt)
 {
-	if (lv1ent_section(sent))
+	if (lv1ent_section(sent)) {
+		WARN(1, "Trying mapping on 1MiB@%#08x that is mapped",
+			iova);
 		return -EADDRINUSE;
+	}
 
 	if (lv1ent_page(sent)) {
-		if (*pgcnt != NUM_LV2ENTRIES)
+		if (*pgcnt != NUM_LV2ENTRIES) {
+			WARN(1, "Trying mapping on 1MiB@%#08x that is mapped",
+				iova);
 			return -EADDRINUSE;
+		}
 
-		kfree(page_entry(sent, 0));
-
+		kmem_cache_free(lv2table_kmem_cache, page_entry(sent, 0));
 		*pgcnt = 0;
 	}
 
@@ -848,14 +916,26 @@
 
 	pgtable_flush(sent, sent + 1);
 
+	spin_lock(&priv->lock);
+	if (lv1ent_page_zero(sent)) {
+		struct exynos_iommu_owner *owner;
+		/*
+		 * Flushing FLPD cache in System MMU v3.3 that may cache a FLPD
+		 * entry by speculative prefetch of SLPD which has no mapping.
+		 */
+		list_for_each_entry(owner, &priv->clients, client)
+			sysmmu_tlb_invalidate_flpdcache(owner->dev, iova);
+	}
+	spin_unlock(&priv->lock);
+
 	return 0;
 }
 
-static int lv2set_page(unsigned long *pent, phys_addr_t paddr, size_t size,
+static int lv2set_page(sysmmu_pte_t *pent, phys_addr_t paddr, size_t size,
 								short *pgcnt)
 {
 	if (size == SPAGE_SIZE) {
-		if (!lv2ent_fault(pent))
+		if (WARN_ON(!lv2ent_fault(pent)))
 			return -EADDRINUSE;
 
 		*pent = mk_lv2ent_spage(paddr);
@@ -863,9 +943,11 @@
 		*pgcnt -= 1;
 	} else { /* size == LPAGE_SIZE */
 		int i;
+
 		for (i = 0; i < SPAGES_PER_LPAGE; i++, pent++) {
-			if (!lv2ent_fault(pent)) {
-				memset(pent, 0, sizeof(*pent) * i);
+			if (WARN_ON(!lv2ent_fault(pent))) {
+				if (i > 0)
+					memset(pent - i, 0, sizeof(*pent) * i);
 				return -EADDRINUSE;
 			}
 
@@ -878,11 +960,38 @@
 	return 0;
 }
 
-static int exynos_iommu_map(struct iommu_domain *domain, unsigned long iova,
+/*
+ * *CAUTION* to the I/O virtual memory managers that support exynos-iommu:
+ *
+ * System MMU v3.x have an advanced logic to improve address translation
+ * performance with caching more page table entries by a page table walk.
+ * However, the logic has a bug that caching fault page table entries and System
+ * MMU reports page fault if the cached fault entry is hit even though the fault
+ * entry is updated to a valid entry after the entry is cached.
+ * To prevent caching fault page table entries which may be updated to valid
+ * entries later, the virtual memory manager should care about the w/a about the
+ * problem. The followings describe w/a.
+ *
+ * Any two consecutive I/O virtual address regions must have a hole of 128KiB
+ * in maximum to prevent misbehavior of System MMU 3.x. (w/a of h/w bug)
+ *
+ * Precisely, any start address of I/O virtual region must be aligned by
+ * the following sizes for System MMU v3.1 and v3.2.
+ * System MMU v3.1: 128KiB
+ * System MMU v3.2: 256KiB
+ *
+ * Because System MMU v3.3 caches page table entries more aggressively, it needs
+ * more w/a.
+ * - Any two consecutive I/O virtual regions must be have a hole of larger size
+ *   than or equal size to 128KiB.
+ * - Start address of an I/O virtual region must be aligned by 128KiB.
+ */
+static int exynos_iommu_map(struct iommu_domain *domain, unsigned long l_iova,
 			 phys_addr_t paddr, size_t size, int prot)
 {
 	struct exynos_iommu_domain *priv = domain->priv;
-	unsigned long *entry;
+	sysmmu_pte_t *entry;
+	sysmmu_iova_t iova = (sysmmu_iova_t)l_iova;
 	unsigned long flags;
 	int ret = -ENOMEM;
 
@@ -893,38 +1002,52 @@
 	entry = section_entry(priv->pgtable, iova);
 
 	if (size == SECT_SIZE) {
-		ret = lv1set_section(entry, paddr,
+		ret = lv1set_section(priv, entry, iova, paddr,
 					&priv->lv2entcnt[lv1ent_offset(iova)]);
 	} else {
-		unsigned long *pent;
+		sysmmu_pte_t *pent;
 
-		pent = alloc_lv2entry(entry, iova,
+		pent = alloc_lv2entry(priv, entry, iova,
 					&priv->lv2entcnt[lv1ent_offset(iova)]);
 
-		if (!pent)
-			ret = -ENOMEM;
+		if (IS_ERR(pent))
+			ret = PTR_ERR(pent);
 		else
 			ret = lv2set_page(pent, paddr, size,
 					&priv->lv2entcnt[lv1ent_offset(iova)]);
 	}
 
-	if (ret) {
-		pr_debug("%s: Failed to map iova 0x%lx/0x%x bytes\n",
-							__func__, iova, size);
-	}
+	if (ret)
+		pr_err("%s: Failed(%d) to map %#zx bytes @ %#x\n",
+			__func__, ret, size, iova);
 
 	spin_unlock_irqrestore(&priv->pgtablelock, flags);
 
 	return ret;
 }
 
+static void exynos_iommu_tlb_invalidate_entry(struct exynos_iommu_domain *priv,
+						sysmmu_iova_t iova, size_t size)
+{
+	struct exynos_iommu_owner *owner;
+	unsigned long flags;
+
+	spin_lock_irqsave(&priv->lock, flags);
+
+	list_for_each_entry(owner, &priv->clients, client)
+		sysmmu_tlb_invalidate_entry(owner->dev, iova, size);
+
+	spin_unlock_irqrestore(&priv->lock, flags);
+}
+
 static size_t exynos_iommu_unmap(struct iommu_domain *domain,
-					       unsigned long iova, size_t size)
+					unsigned long l_iova, size_t size)
 {
 	struct exynos_iommu_domain *priv = domain->priv;
-	struct sysmmu_drvdata *data;
+	sysmmu_iova_t iova = (sysmmu_iova_t)l_iova;
+	sysmmu_pte_t *ent;
+	size_t err_pgsize;
 	unsigned long flags;
-	unsigned long *ent;
 
 	BUG_ON(priv->pgtable == NULL);
 
@@ -933,9 +1056,12 @@
 	ent = section_entry(priv->pgtable, iova);
 
 	if (lv1ent_section(ent)) {
-		BUG_ON(size < SECT_SIZE);
+		if (WARN_ON(size < SECT_SIZE)) {
+			err_pgsize = SECT_SIZE;
+			goto err;
+		}
 
-		*ent = 0;
+		*ent = ZERO_LV2LINK; /* w/a for h/w bug in Sysmem MMU v3.3 */
 		pgtable_flush(ent, ent + 1);
 		size = SECT_SIZE;
 		goto done;
@@ -959,34 +1085,42 @@
 	if (lv2ent_small(ent)) {
 		*ent = 0;
 		size = SPAGE_SIZE;
+		pgtable_flush(ent, ent + 1);
 		priv->lv2entcnt[lv1ent_offset(iova)] += 1;
 		goto done;
 	}
 
 	/* lv1ent_large(ent) == true here */
-	BUG_ON(size < LPAGE_SIZE);
+	if (WARN_ON(size < LPAGE_SIZE)) {
+		err_pgsize = LPAGE_SIZE;
+		goto err;
+	}
 
 	memset(ent, 0, sizeof(*ent) * SPAGES_PER_LPAGE);
+	pgtable_flush(ent, ent + SPAGES_PER_LPAGE);
 
 	size = LPAGE_SIZE;
 	priv->lv2entcnt[lv1ent_offset(iova)] += SPAGES_PER_LPAGE;
 done:
 	spin_unlock_irqrestore(&priv->pgtablelock, flags);
 
-	spin_lock_irqsave(&priv->lock, flags);
-	list_for_each_entry(data, &priv->clients, node)
-		sysmmu_tlb_invalidate_entry(data->dev, iova);
-	spin_unlock_irqrestore(&priv->lock, flags);
-
+	exynos_iommu_tlb_invalidate_entry(priv, iova, size);
 
 	return size;
+err:
+	spin_unlock_irqrestore(&priv->pgtablelock, flags);
+
+	pr_err("%s: Failed: size(%#zx) @ %#x is smaller than page size %#zx\n",
+		__func__, size, iova, err_pgsize);
+
+	return 0;
 }
 
 static phys_addr_t exynos_iommu_iova_to_phys(struct iommu_domain *domain,
 					  dma_addr_t iova)
 {
 	struct exynos_iommu_domain *priv = domain->priv;
-	unsigned long *entry;
+	sysmmu_pte_t *entry;
 	unsigned long flags;
 	phys_addr_t phys = 0;
 
@@ -1010,14 +1144,42 @@
 	return phys;
 }
 
+static int exynos_iommu_add_device(struct device *dev)
+{
+	struct iommu_group *group;
+	int ret;
+
+	group = iommu_group_get(dev);
+
+	if (!group) {
+		group = iommu_group_alloc();
+		if (IS_ERR(group)) {
+			dev_err(dev, "Failed to allocate IOMMU group\n");
+			return PTR_ERR(group);
+		}
+	}
+
+	ret = iommu_group_add_device(group, dev);
+	iommu_group_put(group);
+
+	return ret;
+}
+
+static void exynos_iommu_remove_device(struct device *dev)
+{
+	iommu_group_remove_device(dev);
+}
+
 static struct iommu_ops exynos_iommu_ops = {
-	.domain_init = &exynos_iommu_domain_init,
-	.domain_destroy = &exynos_iommu_domain_destroy,
-	.attach_dev = &exynos_iommu_attach_device,
-	.detach_dev = &exynos_iommu_detach_device,
-	.map = &exynos_iommu_map,
-	.unmap = &exynos_iommu_unmap,
-	.iova_to_phys = &exynos_iommu_iova_to_phys,
+	.domain_init = exynos_iommu_domain_init,
+	.domain_destroy = exynos_iommu_domain_destroy,
+	.attach_dev = exynos_iommu_attach_device,
+	.detach_dev = exynos_iommu_detach_device,
+	.map = exynos_iommu_map,
+	.unmap = exynos_iommu_unmap,
+	.iova_to_phys = exynos_iommu_iova_to_phys,
+	.add_device = exynos_iommu_add_device,
+	.remove_device = exynos_iommu_remove_device,
 	.pgsize_bitmap = SECT_SIZE | LPAGE_SIZE | SPAGE_SIZE,
 };
 
@@ -1025,11 +1187,41 @@
 {
 	int ret;
 
+	lv2table_kmem_cache = kmem_cache_create("exynos-iommu-lv2table",
+				LV2TABLE_SIZE, LV2TABLE_SIZE, 0, NULL);
+	if (!lv2table_kmem_cache) {
+		pr_err("%s: Failed to create kmem cache\n", __func__);
+		return -ENOMEM;
+	}
+
 	ret = platform_driver_register(&exynos_sysmmu_driver);
+	if (ret) {
+		pr_err("%s: Failed to register driver\n", __func__);
+		goto err_reg_driver;
+	}
 
-	if (ret == 0)
-		bus_set_iommu(&platform_bus_type, &exynos_iommu_ops);
+	zero_lv2_table = kmem_cache_zalloc(lv2table_kmem_cache, GFP_KERNEL);
+	if (zero_lv2_table == NULL) {
+		pr_err("%s: Failed to allocate zero level2 page table\n",
+			__func__);
+		ret = -ENOMEM;
+		goto err_zero_lv2;
+	}
 
+	ret = bus_set_iommu(&platform_bus_type, &exynos_iommu_ops);
+	if (ret) {
+		pr_err("%s: Failed to register exynos-iommu driver.\n",
+								__func__);
+		goto err_set_iommu;
+	}
+
+	return 0;
+err_set_iommu:
+	kmem_cache_free(lv2table_kmem_cache, zero_lv2_table);
+err_zero_lv2:
+	platform_driver_unregister(&exynos_sysmmu_driver);
+err_reg_driver:
+	kmem_cache_destroy(lv2table_kmem_cache);
 	return ret;
 }
 subsys_initcall(exynos_iommu_init);
diff --git a/drivers/iommu/ipmmu-vmsa.c b/drivers/iommu/ipmmu-vmsa.c
new file mode 100644
index 0000000..53cde08
--- /dev/null
+++ b/drivers/iommu/ipmmu-vmsa.c
@@ -0,0 +1,1255 @@
+/*
+ * IPMMU VMSA
+ *
+ * Copyright (C) 2014 Renesas Electronics Corporation
+ *
+ * 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; version 2 of the License.
+ */
+
+#include <linux/delay.h>
+#include <linux/dma-mapping.h>
+#include <linux/err.h>
+#include <linux/export.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/iommu.h>
+#include <linux/module.h>
+#include <linux/platform_data/ipmmu-vmsa.h>
+#include <linux/platform_device.h>
+#include <linux/sizes.h>
+#include <linux/slab.h>
+
+#include <asm/dma-iommu.h>
+#include <asm/pgalloc.h>
+
+struct ipmmu_vmsa_device {
+	struct device *dev;
+	void __iomem *base;
+	struct list_head list;
+
+	const struct ipmmu_vmsa_platform_data *pdata;
+	unsigned int num_utlbs;
+
+	struct dma_iommu_mapping *mapping;
+};
+
+struct ipmmu_vmsa_domain {
+	struct ipmmu_vmsa_device *mmu;
+	struct iommu_domain *io_domain;
+
+	unsigned int context_id;
+	spinlock_t lock;			/* Protects mappings */
+	pgd_t *pgd;
+};
+
+struct ipmmu_vmsa_archdata {
+	struct ipmmu_vmsa_device *mmu;
+	unsigned int utlb;
+};
+
+static DEFINE_SPINLOCK(ipmmu_devices_lock);
+static LIST_HEAD(ipmmu_devices);
+
+#define TLB_LOOP_TIMEOUT		100	/* 100us */
+
+/* -----------------------------------------------------------------------------
+ * Registers Definition
+ */
+
+#define IM_CTX_SIZE			0x40
+
+#define IMCTR				0x0000
+#define IMCTR_TRE			(1 << 17)
+#define IMCTR_AFE			(1 << 16)
+#define IMCTR_RTSEL_MASK		(3 << 4)
+#define IMCTR_RTSEL_SHIFT		4
+#define IMCTR_TREN			(1 << 3)
+#define IMCTR_INTEN			(1 << 2)
+#define IMCTR_FLUSH			(1 << 1)
+#define IMCTR_MMUEN			(1 << 0)
+
+#define IMCAAR				0x0004
+
+#define IMTTBCR				0x0008
+#define IMTTBCR_EAE			(1 << 31)
+#define IMTTBCR_PMB			(1 << 30)
+#define IMTTBCR_SH1_NON_SHAREABLE	(0 << 28)
+#define IMTTBCR_SH1_OUTER_SHAREABLE	(2 << 28)
+#define IMTTBCR_SH1_INNER_SHAREABLE	(3 << 28)
+#define IMTTBCR_SH1_MASK		(3 << 28)
+#define IMTTBCR_ORGN1_NC		(0 << 26)
+#define IMTTBCR_ORGN1_WB_WA		(1 << 26)
+#define IMTTBCR_ORGN1_WT		(2 << 26)
+#define IMTTBCR_ORGN1_WB		(3 << 26)
+#define IMTTBCR_ORGN1_MASK		(3 << 26)
+#define IMTTBCR_IRGN1_NC		(0 << 24)
+#define IMTTBCR_IRGN1_WB_WA		(1 << 24)
+#define IMTTBCR_IRGN1_WT		(2 << 24)
+#define IMTTBCR_IRGN1_WB		(3 << 24)
+#define IMTTBCR_IRGN1_MASK		(3 << 24)
+#define IMTTBCR_TSZ1_MASK		(7 << 16)
+#define IMTTBCR_TSZ1_SHIFT		16
+#define IMTTBCR_SH0_NON_SHAREABLE	(0 << 12)
+#define IMTTBCR_SH0_OUTER_SHAREABLE	(2 << 12)
+#define IMTTBCR_SH0_INNER_SHAREABLE	(3 << 12)
+#define IMTTBCR_SH0_MASK		(3 << 12)
+#define IMTTBCR_ORGN0_NC		(0 << 10)
+#define IMTTBCR_ORGN0_WB_WA		(1 << 10)
+#define IMTTBCR_ORGN0_WT		(2 << 10)
+#define IMTTBCR_ORGN0_WB		(3 << 10)
+#define IMTTBCR_ORGN0_MASK		(3 << 10)
+#define IMTTBCR_IRGN0_NC		(0 << 8)
+#define IMTTBCR_IRGN0_WB_WA		(1 << 8)
+#define IMTTBCR_IRGN0_WT		(2 << 8)
+#define IMTTBCR_IRGN0_WB		(3 << 8)
+#define IMTTBCR_IRGN0_MASK		(3 << 8)
+#define IMTTBCR_SL0_LVL_2		(0 << 4)
+#define IMTTBCR_SL0_LVL_1		(1 << 4)
+#define IMTTBCR_TSZ0_MASK		(7 << 0)
+#define IMTTBCR_TSZ0_SHIFT		O
+
+#define IMBUSCR				0x000c
+#define IMBUSCR_DVM			(1 << 2)
+#define IMBUSCR_BUSSEL_SYS		(0 << 0)
+#define IMBUSCR_BUSSEL_CCI		(1 << 0)
+#define IMBUSCR_BUSSEL_IMCAAR		(2 << 0)
+#define IMBUSCR_BUSSEL_CCI_IMCAAR	(3 << 0)
+#define IMBUSCR_BUSSEL_MASK		(3 << 0)
+
+#define IMTTLBR0			0x0010
+#define IMTTUBR0			0x0014
+#define IMTTLBR1			0x0018
+#define IMTTUBR1			0x001c
+
+#define IMSTR				0x0020
+#define IMSTR_ERRLVL_MASK		(3 << 12)
+#define IMSTR_ERRLVL_SHIFT		12
+#define IMSTR_ERRCODE_TLB_FORMAT	(1 << 8)
+#define IMSTR_ERRCODE_ACCESS_PERM	(4 << 8)
+#define IMSTR_ERRCODE_SECURE_ACCESS	(5 << 8)
+#define IMSTR_ERRCODE_MASK		(7 << 8)
+#define IMSTR_MHIT			(1 << 4)
+#define IMSTR_ABORT			(1 << 2)
+#define IMSTR_PF			(1 << 1)
+#define IMSTR_TF			(1 << 0)
+
+#define IMMAIR0				0x0028
+#define IMMAIR1				0x002c
+#define IMMAIR_ATTR_MASK		0xff
+#define IMMAIR_ATTR_DEVICE		0x04
+#define IMMAIR_ATTR_NC			0x44
+#define IMMAIR_ATTR_WBRWA		0xff
+#define IMMAIR_ATTR_SHIFT(n)		((n) << 3)
+#define IMMAIR_ATTR_IDX_NC		0
+#define IMMAIR_ATTR_IDX_WBRWA		1
+#define IMMAIR_ATTR_IDX_DEV		2
+
+#define IMEAR				0x0030
+
+#define IMPCTR				0x0200
+#define IMPSTR				0x0208
+#define IMPEAR				0x020c
+#define IMPMBA(n)			(0x0280 + ((n) * 4))
+#define IMPMBD(n)			(0x02c0 + ((n) * 4))
+
+#define IMUCTR(n)			(0x0300 + ((n) * 16))
+#define IMUCTR_FIXADDEN			(1 << 31)
+#define IMUCTR_FIXADD_MASK		(0xff << 16)
+#define IMUCTR_FIXADD_SHIFT		16
+#define IMUCTR_TTSEL_MMU(n)		((n) << 4)
+#define IMUCTR_TTSEL_PMB		(8 << 4)
+#define IMUCTR_TTSEL_MASK		(15 << 4)
+#define IMUCTR_FLUSH			(1 << 1)
+#define IMUCTR_MMUEN			(1 << 0)
+
+#define IMUASID(n)			(0x0308 + ((n) * 16))
+#define IMUASID_ASID8_MASK		(0xff << 8)
+#define IMUASID_ASID8_SHIFT		8
+#define IMUASID_ASID0_MASK		(0xff << 0)
+#define IMUASID_ASID0_SHIFT		0
+
+/* -----------------------------------------------------------------------------
+ * Page Table Bits
+ */
+
+/*
+ * VMSA states in section B3.6.3 "Control of Secure or Non-secure memory access,
+ * Long-descriptor format" that the NStable bit being set in a table descriptor
+ * will result in the NStable and NS bits of all child entries being ignored and
+ * considered as being set. The IPMMU seems not to comply with this, as it
+ * generates a secure access page fault if any of the NStable and NS bits isn't
+ * set when running in non-secure mode.
+ */
+#ifndef PMD_NSTABLE
+#define PMD_NSTABLE			(_AT(pmdval_t, 1) << 63)
+#endif
+
+#define ARM_VMSA_PTE_XN			(((pteval_t)3) << 53)
+#define ARM_VMSA_PTE_CONT		(((pteval_t)1) << 52)
+#define ARM_VMSA_PTE_AF			(((pteval_t)1) << 10)
+#define ARM_VMSA_PTE_SH_NS		(((pteval_t)0) << 8)
+#define ARM_VMSA_PTE_SH_OS		(((pteval_t)2) << 8)
+#define ARM_VMSA_PTE_SH_IS		(((pteval_t)3) << 8)
+#define ARM_VMSA_PTE_SH_MASK		(((pteval_t)3) << 8)
+#define ARM_VMSA_PTE_NS			(((pteval_t)1) << 5)
+#define ARM_VMSA_PTE_PAGE		(((pteval_t)3) << 0)
+
+/* Stage-1 PTE */
+#define ARM_VMSA_PTE_nG			(((pteval_t)1) << 11)
+#define ARM_VMSA_PTE_AP_UNPRIV		(((pteval_t)1) << 6)
+#define ARM_VMSA_PTE_AP_RDONLY		(((pteval_t)2) << 6)
+#define ARM_VMSA_PTE_AP_MASK		(((pteval_t)3) << 6)
+#define ARM_VMSA_PTE_ATTRINDX_MASK	(((pteval_t)3) << 2)
+#define ARM_VMSA_PTE_ATTRINDX_SHIFT	2
+
+#define ARM_VMSA_PTE_ATTRS_MASK \
+	(ARM_VMSA_PTE_XN | ARM_VMSA_PTE_CONT | ARM_VMSA_PTE_nG | \
+	 ARM_VMSA_PTE_AF | ARM_VMSA_PTE_SH_MASK | ARM_VMSA_PTE_AP_MASK | \
+	 ARM_VMSA_PTE_NS | ARM_VMSA_PTE_ATTRINDX_MASK)
+
+#define ARM_VMSA_PTE_CONT_ENTRIES	16
+#define ARM_VMSA_PTE_CONT_SIZE		(PAGE_SIZE * ARM_VMSA_PTE_CONT_ENTRIES)
+
+#define IPMMU_PTRS_PER_PTE		512
+#define IPMMU_PTRS_PER_PMD		512
+#define IPMMU_PTRS_PER_PGD		4
+
+/* -----------------------------------------------------------------------------
+ * Read/Write Access
+ */
+
+static u32 ipmmu_read(struct ipmmu_vmsa_device *mmu, unsigned int offset)
+{
+	return ioread32(mmu->base + offset);
+}
+
+static void ipmmu_write(struct ipmmu_vmsa_device *mmu, unsigned int offset,
+			u32 data)
+{
+	iowrite32(data, mmu->base + offset);
+}
+
+static u32 ipmmu_ctx_read(struct ipmmu_vmsa_domain *domain, unsigned int reg)
+{
+	return ipmmu_read(domain->mmu, domain->context_id * IM_CTX_SIZE + reg);
+}
+
+static void ipmmu_ctx_write(struct ipmmu_vmsa_domain *domain, unsigned int reg,
+			    u32 data)
+{
+	ipmmu_write(domain->mmu, domain->context_id * IM_CTX_SIZE + reg, data);
+}
+
+/* -----------------------------------------------------------------------------
+ * TLB and microTLB Management
+ */
+
+/* Wait for any pending TLB invalidations to complete */
+static void ipmmu_tlb_sync(struct ipmmu_vmsa_domain *domain)
+{
+	unsigned int count = 0;
+
+	while (ipmmu_ctx_read(domain, IMCTR) & IMCTR_FLUSH) {
+		cpu_relax();
+		if (++count == TLB_LOOP_TIMEOUT) {
+			dev_err_ratelimited(domain->mmu->dev,
+			"TLB sync timed out -- MMU may be deadlocked\n");
+			return;
+		}
+		udelay(1);
+	}
+}
+
+static void ipmmu_tlb_invalidate(struct ipmmu_vmsa_domain *domain)
+{
+	u32 reg;
+
+	reg = ipmmu_ctx_read(domain, IMCTR);
+	reg |= IMCTR_FLUSH;
+	ipmmu_ctx_write(domain, IMCTR, reg);
+
+	ipmmu_tlb_sync(domain);
+}
+
+/*
+ * Enable MMU translation for the microTLB.
+ */
+static void ipmmu_utlb_enable(struct ipmmu_vmsa_domain *domain,
+			      unsigned int utlb)
+{
+	struct ipmmu_vmsa_device *mmu = domain->mmu;
+
+	/*
+	 * TODO: Reference-count the microTLB as several bus masters can be
+	 * connected to the same microTLB.
+	 */
+
+	/* TODO: What should we set the ASID to ? */
+	ipmmu_write(mmu, IMUASID(utlb), 0);
+	/* TODO: Do we need to flush the microTLB ? */
+	ipmmu_write(mmu, IMUCTR(utlb),
+		    IMUCTR_TTSEL_MMU(domain->context_id) | IMUCTR_FLUSH |
+		    IMUCTR_MMUEN);
+}
+
+/*
+ * Disable MMU translation for the microTLB.
+ */
+static void ipmmu_utlb_disable(struct ipmmu_vmsa_domain *domain,
+			       unsigned int utlb)
+{
+	struct ipmmu_vmsa_device *mmu = domain->mmu;
+
+	ipmmu_write(mmu, IMUCTR(utlb), 0);
+}
+
+static void ipmmu_flush_pgtable(struct ipmmu_vmsa_device *mmu, void *addr,
+				size_t size)
+{
+	unsigned long offset = (unsigned long)addr & ~PAGE_MASK;
+
+	/*
+	 * TODO: Add support for coherent walk through CCI with DVM and remove
+	 * cache handling.
+	 */
+	dma_map_page(mmu->dev, virt_to_page(addr), offset, size, DMA_TO_DEVICE);
+}
+
+/* -----------------------------------------------------------------------------
+ * Domain/Context Management
+ */
+
+static int ipmmu_domain_init_context(struct ipmmu_vmsa_domain *domain)
+{
+	phys_addr_t ttbr;
+	u32 reg;
+
+	/*
+	 * TODO: When adding support for multiple contexts, find an unused
+	 * context.
+	 */
+	domain->context_id = 0;
+
+	/* TTBR0 */
+	ipmmu_flush_pgtable(domain->mmu, domain->pgd,
+			    IPMMU_PTRS_PER_PGD * sizeof(*domain->pgd));
+	ttbr = __pa(domain->pgd);
+	ipmmu_ctx_write(domain, IMTTLBR0, ttbr);
+	ipmmu_ctx_write(domain, IMTTUBR0, ttbr >> 32);
+
+	/*
+	 * TTBCR
+	 * We use long descriptors with inner-shareable WBWA tables and allocate
+	 * the whole 32-bit VA space to TTBR0.
+	 */
+	ipmmu_ctx_write(domain, IMTTBCR, IMTTBCR_EAE |
+			IMTTBCR_SH0_INNER_SHAREABLE | IMTTBCR_ORGN0_WB_WA |
+			IMTTBCR_IRGN0_WB_WA | IMTTBCR_SL0_LVL_1);
+
+	/*
+	 * MAIR0
+	 * We need three attributes only, non-cacheable, write-back read/write
+	 * allocate and device memory.
+	 */
+	reg = (IMMAIR_ATTR_NC << IMMAIR_ATTR_SHIFT(IMMAIR_ATTR_IDX_NC))
+	    | (IMMAIR_ATTR_WBRWA << IMMAIR_ATTR_SHIFT(IMMAIR_ATTR_IDX_WBRWA))
+	    | (IMMAIR_ATTR_DEVICE << IMMAIR_ATTR_SHIFT(IMMAIR_ATTR_IDX_DEV));
+	ipmmu_ctx_write(domain, IMMAIR0, reg);
+
+	/* IMBUSCR */
+	ipmmu_ctx_write(domain, IMBUSCR,
+			ipmmu_ctx_read(domain, IMBUSCR) &
+			~(IMBUSCR_DVM | IMBUSCR_BUSSEL_MASK));
+
+	/*
+	 * IMSTR
+	 * Clear all interrupt flags.
+	 */
+	ipmmu_ctx_write(domain, IMSTR, ipmmu_ctx_read(domain, IMSTR));
+
+	/*
+	 * IMCTR
+	 * Enable the MMU and interrupt generation. The long-descriptor
+	 * translation table format doesn't use TEX remapping. Don't enable AF
+	 * software management as we have no use for it. Flush the TLB as
+	 * required when modifying the context registers.
+	 */
+	ipmmu_ctx_write(domain, IMCTR, IMCTR_INTEN | IMCTR_FLUSH | IMCTR_MMUEN);
+
+	return 0;
+}
+
+static void ipmmu_domain_destroy_context(struct ipmmu_vmsa_domain *domain)
+{
+	/*
+	 * Disable the context. Flush the TLB as required when modifying the
+	 * context registers.
+	 *
+	 * TODO: Is TLB flush really needed ?
+	 */
+	ipmmu_ctx_write(domain, IMCTR, IMCTR_FLUSH);
+	ipmmu_tlb_sync(domain);
+}
+
+/* -----------------------------------------------------------------------------
+ * Fault Handling
+ */
+
+static irqreturn_t ipmmu_domain_irq(struct ipmmu_vmsa_domain *domain)
+{
+	const u32 err_mask = IMSTR_MHIT | IMSTR_ABORT | IMSTR_PF | IMSTR_TF;
+	struct ipmmu_vmsa_device *mmu = domain->mmu;
+	u32 status;
+	u32 iova;
+
+	status = ipmmu_ctx_read(domain, IMSTR);
+	if (!(status & err_mask))
+		return IRQ_NONE;
+
+	iova = ipmmu_ctx_read(domain, IMEAR);
+
+	/*
+	 * Clear the error status flags. Unlike traditional interrupt flag
+	 * registers that must be cleared by writing 1, this status register
+	 * seems to require 0. The error address register must be read before,
+	 * otherwise its value will be 0.
+	 */
+	ipmmu_ctx_write(domain, IMSTR, 0);
+
+	/* Log fatal errors. */
+	if (status & IMSTR_MHIT)
+		dev_err_ratelimited(mmu->dev, "Multiple TLB hits @0x%08x\n",
+				    iova);
+	if (status & IMSTR_ABORT)
+		dev_err_ratelimited(mmu->dev, "Page Table Walk Abort @0x%08x\n",
+				    iova);
+
+	if (!(status & (IMSTR_PF | IMSTR_TF)))
+		return IRQ_NONE;
+
+	/*
+	 * Try to handle page faults and translation faults.
+	 *
+	 * TODO: We need to look up the faulty device based on the I/O VA. Use
+	 * the IOMMU device for now.
+	 */
+	if (!report_iommu_fault(domain->io_domain, mmu->dev, iova, 0))
+		return IRQ_HANDLED;
+
+	dev_err_ratelimited(mmu->dev,
+			    "Unhandled fault: status 0x%08x iova 0x%08x\n",
+			    status, iova);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t ipmmu_irq(int irq, void *dev)
+{
+	struct ipmmu_vmsa_device *mmu = dev;
+	struct iommu_domain *io_domain;
+	struct ipmmu_vmsa_domain *domain;
+
+	if (!mmu->mapping)
+		return IRQ_NONE;
+
+	io_domain = mmu->mapping->domain;
+	domain = io_domain->priv;
+
+	return ipmmu_domain_irq(domain);
+}
+
+/* -----------------------------------------------------------------------------
+ * Page Table Management
+ */
+
+#define pud_pgtable(pud) pfn_to_page(__phys_to_pfn(pud_val(pud) & PHYS_MASK))
+
+static void ipmmu_free_ptes(pmd_t *pmd)
+{
+	pgtable_t table = pmd_pgtable(*pmd);
+	__free_page(table);
+}
+
+static void ipmmu_free_pmds(pud_t *pud)
+{
+	pmd_t *pmd = pmd_offset(pud, 0);
+	pgtable_t table;
+	unsigned int i;
+
+	for (i = 0; i < IPMMU_PTRS_PER_PMD; ++i) {
+		if (!pmd_table(*pmd))
+			continue;
+
+		ipmmu_free_ptes(pmd);
+		pmd++;
+	}
+
+	table = pud_pgtable(*pud);
+	__free_page(table);
+}
+
+static void ipmmu_free_pgtables(struct ipmmu_vmsa_domain *domain)
+{
+	pgd_t *pgd, *pgd_base = domain->pgd;
+	unsigned int i;
+
+	/*
+	 * Recursively free the page tables for this domain. We don't care about
+	 * speculative TLB filling, because the TLB will be nuked next time this
+	 * context bank is re-allocated and no devices currently map to these
+	 * tables.
+	 */
+	pgd = pgd_base;
+	for (i = 0; i < IPMMU_PTRS_PER_PGD; ++i) {
+		if (pgd_none(*pgd))
+			continue;
+		ipmmu_free_pmds((pud_t *)pgd);
+		pgd++;
+	}
+
+	kfree(pgd_base);
+}
+
+/*
+ * We can't use the (pgd|pud|pmd|pte)_populate or the set_(pgd|pud|pmd|pte)
+ * functions as they would flush the CPU TLB.
+ */
+
+static pte_t *ipmmu_alloc_pte(struct ipmmu_vmsa_device *mmu, pmd_t *pmd,
+			      unsigned long iova)
+{
+	pte_t *pte;
+
+	if (!pmd_none(*pmd))
+		return pte_offset_kernel(pmd, iova);
+
+	pte = (pte_t *)get_zeroed_page(GFP_ATOMIC);
+	if (!pte)
+		return NULL;
+
+	ipmmu_flush_pgtable(mmu, pte, PAGE_SIZE);
+	*pmd = __pmd(__pa(pte) | PMD_NSTABLE | PMD_TYPE_TABLE);
+	ipmmu_flush_pgtable(mmu, pmd, sizeof(*pmd));
+
+	return pte + pte_index(iova);
+}
+
+static pmd_t *ipmmu_alloc_pmd(struct ipmmu_vmsa_device *mmu, pgd_t *pgd,
+			      unsigned long iova)
+{
+	pud_t *pud = (pud_t *)pgd;
+	pmd_t *pmd;
+
+	if (!pud_none(*pud))
+		return pmd_offset(pud, iova);
+
+	pmd = (pmd_t *)get_zeroed_page(GFP_ATOMIC);
+	if (!pmd)
+		return NULL;
+
+	ipmmu_flush_pgtable(mmu, pmd, PAGE_SIZE);
+	*pud = __pud(__pa(pmd) | PMD_NSTABLE | PMD_TYPE_TABLE);
+	ipmmu_flush_pgtable(mmu, pud, sizeof(*pud));
+
+	return pmd + pmd_index(iova);
+}
+
+static u64 ipmmu_page_prot(unsigned int prot, u64 type)
+{
+	u64 pgprot = ARM_VMSA_PTE_XN | ARM_VMSA_PTE_nG | ARM_VMSA_PTE_AF
+		   | ARM_VMSA_PTE_SH_IS | ARM_VMSA_PTE_AP_UNPRIV
+		   | ARM_VMSA_PTE_NS | type;
+
+	if (!(prot & IOMMU_WRITE) && (prot & IOMMU_READ))
+		pgprot |= ARM_VMSA_PTE_AP_RDONLY;
+
+	if (prot & IOMMU_CACHE)
+		pgprot |= IMMAIR_ATTR_IDX_WBRWA << ARM_VMSA_PTE_ATTRINDX_SHIFT;
+
+	if (prot & IOMMU_EXEC)
+		pgprot &= ~ARM_VMSA_PTE_XN;
+	else if (!(prot & (IOMMU_READ | IOMMU_WRITE)))
+		/* If no access create a faulting entry to avoid TLB fills. */
+		pgprot &= ~ARM_VMSA_PTE_PAGE;
+
+	return pgprot;
+}
+
+static int ipmmu_alloc_init_pte(struct ipmmu_vmsa_device *mmu, pmd_t *pmd,
+				unsigned long iova, unsigned long pfn,
+				size_t size, int prot)
+{
+	pteval_t pteval = ipmmu_page_prot(prot, ARM_VMSA_PTE_PAGE);
+	unsigned int num_ptes = 1;
+	pte_t *pte, *start;
+	unsigned int i;
+
+	pte = ipmmu_alloc_pte(mmu, pmd, iova);
+	if (!pte)
+		return -ENOMEM;
+
+	start = pte;
+
+	/*
+	 * Install the page table entries. We can be called both for a single
+	 * page or for a block of 16 physically contiguous pages. In the latter
+	 * case set the PTE contiguous hint.
+	 */
+	if (size == SZ_64K) {
+		pteval |= ARM_VMSA_PTE_CONT;
+		num_ptes = ARM_VMSA_PTE_CONT_ENTRIES;
+	}
+
+	for (i = num_ptes; i; --i)
+		*pte++ = pfn_pte(pfn++, __pgprot(pteval));
+
+	ipmmu_flush_pgtable(mmu, start, sizeof(*pte) * num_ptes);
+
+	return 0;
+}
+
+static int ipmmu_alloc_init_pmd(struct ipmmu_vmsa_device *mmu, pmd_t *pmd,
+				unsigned long iova, unsigned long pfn,
+				int prot)
+{
+	pmdval_t pmdval = ipmmu_page_prot(prot, PMD_TYPE_SECT);
+
+	*pmd = pfn_pmd(pfn, __pgprot(pmdval));
+	ipmmu_flush_pgtable(mmu, pmd, sizeof(*pmd));
+
+	return 0;
+}
+
+static int ipmmu_create_mapping(struct ipmmu_vmsa_domain *domain,
+				unsigned long iova, phys_addr_t paddr,
+				size_t size, int prot)
+{
+	struct ipmmu_vmsa_device *mmu = domain->mmu;
+	pgd_t *pgd = domain->pgd;
+	unsigned long flags;
+	unsigned long pfn;
+	pmd_t *pmd;
+	int ret;
+
+	if (!pgd)
+		return -EINVAL;
+
+	if (size & ~PAGE_MASK)
+		return -EINVAL;
+
+	if (paddr & ~((1ULL << 40) - 1))
+		return -ERANGE;
+
+	pfn = __phys_to_pfn(paddr);
+	pgd += pgd_index(iova);
+
+	/* Update the page tables. */
+	spin_lock_irqsave(&domain->lock, flags);
+
+	pmd = ipmmu_alloc_pmd(mmu, pgd, iova);
+	if (!pmd) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	switch (size) {
+	case SZ_2M:
+		ret = ipmmu_alloc_init_pmd(mmu, pmd, iova, pfn, prot);
+		break;
+	case SZ_64K:
+	case SZ_4K:
+		ret = ipmmu_alloc_init_pte(mmu, pmd, iova, pfn, size, prot);
+		break;
+	default:
+		ret = -EINVAL;
+		break;
+	}
+
+done:
+	spin_unlock_irqrestore(&domain->lock, flags);
+
+	if (!ret)
+		ipmmu_tlb_invalidate(domain);
+
+	return ret;
+}
+
+static void ipmmu_clear_pud(struct ipmmu_vmsa_device *mmu, pud_t *pud)
+{
+	/* Free the page table. */
+	pgtable_t table = pud_pgtable(*pud);
+	__free_page(table);
+
+	/* Clear the PUD. */
+	*pud = __pud(0);
+	ipmmu_flush_pgtable(mmu, pud, sizeof(*pud));
+}
+
+static void ipmmu_clear_pmd(struct ipmmu_vmsa_device *mmu, pud_t *pud,
+			    pmd_t *pmd)
+{
+	unsigned int i;
+
+	/* Free the page table. */
+	if (pmd_table(*pmd)) {
+		pgtable_t table = pmd_pgtable(*pmd);
+		__free_page(table);
+	}
+
+	/* Clear the PMD. */
+	*pmd = __pmd(0);
+	ipmmu_flush_pgtable(mmu, pmd, sizeof(*pmd));
+
+	/* Check whether the PUD is still needed. */
+	pmd = pmd_offset(pud, 0);
+	for (i = 0; i < IPMMU_PTRS_PER_PMD; ++i) {
+		if (!pmd_none(pmd[i]))
+			return;
+	}
+
+	/* Clear the parent PUD. */
+	ipmmu_clear_pud(mmu, pud);
+}
+
+static void ipmmu_clear_pte(struct ipmmu_vmsa_device *mmu, pud_t *pud,
+			    pmd_t *pmd, pte_t *pte, unsigned int num_ptes)
+{
+	unsigned int i;
+
+	/* Clear the PTE. */
+	for (i = num_ptes; i; --i)
+		pte[i-1] = __pte(0);
+
+	ipmmu_flush_pgtable(mmu, pte, sizeof(*pte) * num_ptes);
+
+	/* Check whether the PMD is still needed. */
+	pte = pte_offset_kernel(pmd, 0);
+	for (i = 0; i < IPMMU_PTRS_PER_PTE; ++i) {
+		if (!pte_none(pte[i]))
+			return;
+	}
+
+	/* Clear the parent PMD. */
+	ipmmu_clear_pmd(mmu, pud, pmd);
+}
+
+static int ipmmu_split_pmd(struct ipmmu_vmsa_device *mmu, pmd_t *pmd)
+{
+	pte_t *pte, *start;
+	pteval_t pteval;
+	unsigned long pfn;
+	unsigned int i;
+
+	pte = (pte_t *)get_zeroed_page(GFP_ATOMIC);
+	if (!pte)
+		return -ENOMEM;
+
+	/* Copy the PMD attributes. */
+	pteval = (pmd_val(*pmd) & ARM_VMSA_PTE_ATTRS_MASK)
+	       | ARM_VMSA_PTE_CONT | ARM_VMSA_PTE_PAGE;
+
+	pfn = pmd_pfn(*pmd);
+	start = pte;
+
+	for (i = IPMMU_PTRS_PER_PTE; i; --i)
+		*pte++ = pfn_pte(pfn++, __pgprot(pteval));
+
+	ipmmu_flush_pgtable(mmu, start, PAGE_SIZE);
+	*pmd = __pmd(__pa(start) | PMD_NSTABLE | PMD_TYPE_TABLE);
+	ipmmu_flush_pgtable(mmu, pmd, sizeof(*pmd));
+
+	return 0;
+}
+
+static void ipmmu_split_pte(struct ipmmu_vmsa_device *mmu, pte_t *pte)
+{
+	unsigned int i;
+
+	for (i = ARM_VMSA_PTE_CONT_ENTRIES; i; --i)
+		pte[i-1] = __pte(pte_val(*pte) & ~ARM_VMSA_PTE_CONT);
+
+	ipmmu_flush_pgtable(mmu, pte, sizeof(*pte) * ARM_VMSA_PTE_CONT_ENTRIES);
+}
+
+static int ipmmu_clear_mapping(struct ipmmu_vmsa_domain *domain,
+			       unsigned long iova, size_t size)
+{
+	struct ipmmu_vmsa_device *mmu = domain->mmu;
+	unsigned long flags;
+	pgd_t *pgd = domain->pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	int ret = 0;
+
+	if (!pgd)
+		return -EINVAL;
+
+	if (size & ~PAGE_MASK)
+		return -EINVAL;
+
+	pgd += pgd_index(iova);
+	pud = (pud_t *)pgd;
+
+	spin_lock_irqsave(&domain->lock, flags);
+
+	/* If there's no PUD or PMD we're done. */
+	if (pud_none(*pud))
+		goto done;
+
+	pmd = pmd_offset(pud, iova);
+	if (pmd_none(*pmd))
+		goto done;
+
+	/*
+	 * When freeing a 2MB block just clear the PMD. In the unlikely case the
+	 * block is mapped as individual pages this will free the corresponding
+	 * PTE page table.
+	 */
+	if (size == SZ_2M) {
+		ipmmu_clear_pmd(mmu, pud, pmd);
+		goto done;
+	}
+
+	/*
+	 * If the PMD has been mapped as a section remap it as pages to allow
+	 * freeing individual pages.
+	 */
+	if (pmd_sect(*pmd))
+		ipmmu_split_pmd(mmu, pmd);
+
+	pte = pte_offset_kernel(pmd, iova);
+
+	/*
+	 * When freeing a 64kB block just clear the PTE entries. We don't have
+	 * to care about the contiguous hint of the surrounding entries.
+	 */
+	if (size == SZ_64K) {
+		ipmmu_clear_pte(mmu, pud, pmd, pte, ARM_VMSA_PTE_CONT_ENTRIES);
+		goto done;
+	}
+
+	/*
+	 * If the PTE has been mapped with the contiguous hint set remap it and
+	 * its surrounding PTEs to allow unmapping a single page.
+	 */
+	if (pte_val(*pte) & ARM_VMSA_PTE_CONT)
+		ipmmu_split_pte(mmu, pte);
+
+	/* Clear the PTE. */
+	ipmmu_clear_pte(mmu, pud, pmd, pte, 1);
+
+done:
+	spin_unlock_irqrestore(&domain->lock, flags);
+
+	if (ret)
+		ipmmu_tlb_invalidate(domain);
+
+	return 0;
+}
+
+/* -----------------------------------------------------------------------------
+ * IOMMU Operations
+ */
+
+static int ipmmu_domain_init(struct iommu_domain *io_domain)
+{
+	struct ipmmu_vmsa_domain *domain;
+
+	domain = kzalloc(sizeof(*domain), GFP_KERNEL);
+	if (!domain)
+		return -ENOMEM;
+
+	spin_lock_init(&domain->lock);
+
+	domain->pgd = kzalloc(IPMMU_PTRS_PER_PGD * sizeof(pgd_t), GFP_KERNEL);
+	if (!domain->pgd) {
+		kfree(domain);
+		return -ENOMEM;
+	}
+
+	io_domain->priv = domain;
+	domain->io_domain = io_domain;
+
+	return 0;
+}
+
+static void ipmmu_domain_destroy(struct iommu_domain *io_domain)
+{
+	struct ipmmu_vmsa_domain *domain = io_domain->priv;
+
+	/*
+	 * Free the domain resources. We assume that all devices have already
+	 * been detached.
+	 */
+	ipmmu_domain_destroy_context(domain);
+	ipmmu_free_pgtables(domain);
+	kfree(domain);
+}
+
+static int ipmmu_attach_device(struct iommu_domain *io_domain,
+			       struct device *dev)
+{
+	struct ipmmu_vmsa_archdata *archdata = dev->archdata.iommu;
+	struct ipmmu_vmsa_device *mmu = archdata->mmu;
+	struct ipmmu_vmsa_domain *domain = io_domain->priv;
+	unsigned long flags;
+	int ret = 0;
+
+	if (!mmu) {
+		dev_err(dev, "Cannot attach to IPMMU\n");
+		return -ENXIO;
+	}
+
+	spin_lock_irqsave(&domain->lock, flags);
+
+	if (!domain->mmu) {
+		/* The domain hasn't been used yet, initialize it. */
+		domain->mmu = mmu;
+		ret = ipmmu_domain_init_context(domain);
+	} else if (domain->mmu != mmu) {
+		/*
+		 * Something is wrong, we can't attach two devices using
+		 * different IOMMUs to the same domain.
+		 */
+		dev_err(dev, "Can't attach IPMMU %s to domain on IPMMU %s\n",
+			dev_name(mmu->dev), dev_name(domain->mmu->dev));
+		ret = -EINVAL;
+	}
+
+	spin_unlock_irqrestore(&domain->lock, flags);
+
+	if (ret < 0)
+		return ret;
+
+	ipmmu_utlb_enable(domain, archdata->utlb);
+
+	return 0;
+}
+
+static void ipmmu_detach_device(struct iommu_domain *io_domain,
+				struct device *dev)
+{
+	struct ipmmu_vmsa_archdata *archdata = dev->archdata.iommu;
+	struct ipmmu_vmsa_domain *domain = io_domain->priv;
+
+	ipmmu_utlb_disable(domain, archdata->utlb);
+
+	/*
+	 * TODO: Optimize by disabling the context when no device is attached.
+	 */
+}
+
+static int ipmmu_map(struct iommu_domain *io_domain, unsigned long iova,
+		     phys_addr_t paddr, size_t size, int prot)
+{
+	struct ipmmu_vmsa_domain *domain = io_domain->priv;
+
+	if (!domain)
+		return -ENODEV;
+
+	return ipmmu_create_mapping(domain, iova, paddr, size, prot);
+}
+
+static size_t ipmmu_unmap(struct iommu_domain *io_domain, unsigned long iova,
+			  size_t size)
+{
+	struct ipmmu_vmsa_domain *domain = io_domain->priv;
+	int ret;
+
+	ret = ipmmu_clear_mapping(domain, iova, size);
+	return ret ? 0 : size;
+}
+
+static phys_addr_t ipmmu_iova_to_phys(struct iommu_domain *io_domain,
+				      dma_addr_t iova)
+{
+	struct ipmmu_vmsa_domain *domain = io_domain->priv;
+	pgd_t pgd;
+	pud_t pud;
+	pmd_t pmd;
+	pte_t pte;
+
+	/* TODO: Is locking needed ? */
+
+	if (!domain->pgd)
+		return 0;
+
+	pgd = *(domain->pgd + pgd_index(iova));
+	if (pgd_none(pgd))
+		return 0;
+
+	pud = *pud_offset(&pgd, iova);
+	if (pud_none(pud))
+		return 0;
+
+	pmd = *pmd_offset(&pud, iova);
+	if (pmd_none(pmd))
+		return 0;
+
+	if (pmd_sect(pmd))
+		return __pfn_to_phys(pmd_pfn(pmd)) | (iova & ~PMD_MASK);
+
+	pte = *(pmd_page_vaddr(pmd) + pte_index(iova));
+	if (pte_none(pte))
+		return 0;
+
+	return __pfn_to_phys(pte_pfn(pte)) | (iova & ~PAGE_MASK);
+}
+
+static int ipmmu_find_utlb(struct ipmmu_vmsa_device *mmu, struct device *dev)
+{
+	const struct ipmmu_vmsa_master *master = mmu->pdata->masters;
+	const char *devname = dev_name(dev);
+	unsigned int i;
+
+	for (i = 0; i < mmu->pdata->num_masters; ++i, ++master) {
+		if (strcmp(master->name, devname) == 0)
+			return master->utlb;
+	}
+
+	return -1;
+}
+
+static int ipmmu_add_device(struct device *dev)
+{
+	struct ipmmu_vmsa_archdata *archdata;
+	struct ipmmu_vmsa_device *mmu;
+	struct iommu_group *group;
+	int utlb = -1;
+	int ret;
+
+	if (dev->archdata.iommu) {
+		dev_warn(dev, "IOMMU driver already assigned to device %s\n",
+			 dev_name(dev));
+		return -EINVAL;
+	}
+
+	/* Find the master corresponding to the device. */
+	spin_lock(&ipmmu_devices_lock);
+
+	list_for_each_entry(mmu, &ipmmu_devices, list) {
+		utlb = ipmmu_find_utlb(mmu, dev);
+		if (utlb >= 0) {
+			/*
+			 * TODO Take a reference to the MMU to protect
+			 * against device removal.
+			 */
+			break;
+		}
+	}
+
+	spin_unlock(&ipmmu_devices_lock);
+
+	if (utlb < 0)
+		return -ENODEV;
+
+	if (utlb >= mmu->num_utlbs)
+		return -EINVAL;
+
+	/* Create a device group and add the device to it. */
+	group = iommu_group_alloc();
+	if (IS_ERR(group)) {
+		dev_err(dev, "Failed to allocate IOMMU group\n");
+		return PTR_ERR(group);
+	}
+
+	ret = iommu_group_add_device(group, dev);
+	iommu_group_put(group);
+
+	if (ret < 0) {
+		dev_err(dev, "Failed to add device to IPMMU group\n");
+		return ret;
+	}
+
+	archdata = kzalloc(sizeof(*archdata), GFP_KERNEL);
+	if (!archdata) {
+		ret = -ENOMEM;
+		goto error;
+	}
+
+	archdata->mmu = mmu;
+	archdata->utlb = utlb;
+	dev->archdata.iommu = archdata;
+
+	/*
+	 * Create the ARM mapping, used by the ARM DMA mapping core to allocate
+	 * VAs. This will allocate a corresponding IOMMU domain.
+	 *
+	 * TODO:
+	 * - Create one mapping per context (TLB).
+	 * - Make the mapping size configurable ? We currently use a 2GB mapping
+	 *   at a 1GB offset to ensure that NULL VAs will fault.
+	 */
+	if (!mmu->mapping) {
+		struct dma_iommu_mapping *mapping;
+
+		mapping = arm_iommu_create_mapping(&platform_bus_type,
+						   SZ_1G, SZ_2G);
+		if (IS_ERR(mapping)) {
+			dev_err(mmu->dev, "failed to create ARM IOMMU mapping\n");
+			return PTR_ERR(mapping);
+		}
+
+		mmu->mapping = mapping;
+	}
+
+	/* Attach the ARM VA mapping to the device. */
+	ret = arm_iommu_attach_device(dev, mmu->mapping);
+	if (ret < 0) {
+		dev_err(dev, "Failed to attach device to VA mapping\n");
+		goto error;
+	}
+
+	return 0;
+
+error:
+	kfree(dev->archdata.iommu);
+	dev->archdata.iommu = NULL;
+	iommu_group_remove_device(dev);
+	return ret;
+}
+
+static void ipmmu_remove_device(struct device *dev)
+{
+	arm_iommu_detach_device(dev);
+	iommu_group_remove_device(dev);
+	kfree(dev->archdata.iommu);
+	dev->archdata.iommu = NULL;
+}
+
+static struct iommu_ops ipmmu_ops = {
+	.domain_init = ipmmu_domain_init,
+	.domain_destroy = ipmmu_domain_destroy,
+	.attach_dev = ipmmu_attach_device,
+	.detach_dev = ipmmu_detach_device,
+	.map = ipmmu_map,
+	.unmap = ipmmu_unmap,
+	.iova_to_phys = ipmmu_iova_to_phys,
+	.add_device = ipmmu_add_device,
+	.remove_device = ipmmu_remove_device,
+	.pgsize_bitmap = SZ_2M | SZ_64K | SZ_4K,
+};
+
+/* -----------------------------------------------------------------------------
+ * Probe/remove and init
+ */
+
+static void ipmmu_device_reset(struct ipmmu_vmsa_device *mmu)
+{
+	unsigned int i;
+
+	/* Disable all contexts. */
+	for (i = 0; i < 4; ++i)
+		ipmmu_write(mmu, i * IM_CTX_SIZE + IMCTR, 0);
+}
+
+static int ipmmu_probe(struct platform_device *pdev)
+{
+	struct ipmmu_vmsa_device *mmu;
+	struct resource *res;
+	int irq;
+	int ret;
+
+	if (!pdev->dev.platform_data) {
+		dev_err(&pdev->dev, "missing platform data\n");
+		return -EINVAL;
+	}
+
+	mmu = devm_kzalloc(&pdev->dev, sizeof(*mmu), GFP_KERNEL);
+	if (!mmu) {
+		dev_err(&pdev->dev, "cannot allocate device data\n");
+		return -ENOMEM;
+	}
+
+	mmu->dev = &pdev->dev;
+	mmu->pdata = pdev->dev.platform_data;
+	mmu->num_utlbs = 32;
+
+	/* Map I/O memory and request IRQ. */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	mmu->base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(mmu->base))
+		return PTR_ERR(mmu->base);
+
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0) {
+		dev_err(&pdev->dev, "no IRQ found\n");
+		return irq;
+	}
+
+	ret = devm_request_irq(&pdev->dev, irq, ipmmu_irq, 0,
+			       dev_name(&pdev->dev), mmu);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "failed to request IRQ %d\n", irq);
+		return irq;
+	}
+
+	ipmmu_device_reset(mmu);
+
+	/*
+	 * We can't create the ARM mapping here as it requires the bus to have
+	 * an IOMMU, which only happens when bus_set_iommu() is called in
+	 * ipmmu_init() after the probe function returns.
+	 */
+
+	spin_lock(&ipmmu_devices_lock);
+	list_add(&mmu->list, &ipmmu_devices);
+	spin_unlock(&ipmmu_devices_lock);
+
+	platform_set_drvdata(pdev, mmu);
+
+	return 0;
+}
+
+static int ipmmu_remove(struct platform_device *pdev)
+{
+	struct ipmmu_vmsa_device *mmu = platform_get_drvdata(pdev);
+
+	spin_lock(&ipmmu_devices_lock);
+	list_del(&mmu->list);
+	spin_unlock(&ipmmu_devices_lock);
+
+	arm_iommu_release_mapping(mmu->mapping);
+
+	ipmmu_device_reset(mmu);
+
+	return 0;
+}
+
+static struct platform_driver ipmmu_driver = {
+	.driver = {
+		.owner = THIS_MODULE,
+		.name = "ipmmu-vmsa",
+	},
+	.probe = ipmmu_probe,
+	.remove	= ipmmu_remove,
+};
+
+static int __init ipmmu_init(void)
+{
+	int ret;
+
+	ret = platform_driver_register(&ipmmu_driver);
+	if (ret < 0)
+		return ret;
+
+	if (!iommu_present(&platform_bus_type))
+		bus_set_iommu(&platform_bus_type, &ipmmu_ops);
+
+	return 0;
+}
+
+static void __exit ipmmu_exit(void)
+{
+	return platform_driver_unregister(&ipmmu_driver);
+}
+
+subsys_initcall(ipmmu_init);
+module_exit(ipmmu_exit);
+
+MODULE_DESCRIPTION("IOMMU API for Renesas VMSA-compatible IPMMU");
+MODULE_AUTHOR("Laurent Pinchart <laurent.pinchart@ideasonboard.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/iommu/msm_iommu_dev.c b/drivers/iommu/msm_iommu_dev.c
index 08ba497..61def7cb 100644
--- a/drivers/iommu/msm_iommu_dev.c
+++ b/drivers/iommu/msm_iommu_dev.c
@@ -127,13 +127,12 @@
 
 static int msm_iommu_probe(struct platform_device *pdev)
 {
-	struct resource *r, *r2;
+	struct resource *r;
 	struct clk *iommu_clk;
 	struct clk *iommu_pclk;
 	struct msm_iommu_drvdata *drvdata;
 	struct msm_iommu_dev *iommu_dev = pdev->dev.platform_data;
 	void __iomem *regs_base;
-	resource_size_t	len;
 	int ret, irq, par;
 
 	if (pdev->id == -1) {
@@ -178,35 +177,16 @@
 		iommu_clk = NULL;
 
 	r = platform_get_resource_byname(pdev, IORESOURCE_MEM, "physbase");
-
-	if (!r) {
-		ret = -ENODEV;
+	regs_base = devm_ioremap_resource(&pdev->dev, r);
+	if (IS_ERR(regs_base)) {
+		ret = PTR_ERR(regs_base);
 		goto fail_clk;
 	}
 
-	len = resource_size(r);
-
-	r2 = request_mem_region(r->start, len, r->name);
-	if (!r2) {
-		pr_err("Could not request memory region: start=%p, len=%d\n",
-							(void *) r->start, len);
-		ret = -EBUSY;
-		goto fail_clk;
-	}
-
-	regs_base = ioremap(r2->start, len);
-
-	if (!regs_base) {
-		pr_err("Could not ioremap: start=%p, len=%d\n",
-			 (void *) r2->start, len);
-		ret = -EBUSY;
-		goto fail_mem;
-	}
-
 	irq = platform_get_irq_byname(pdev, "secure_irq");
 	if (irq < 0) {
 		ret = -ENODEV;
-		goto fail_io;
+		goto fail_clk;
 	}
 
 	msm_iommu_reset(regs_base, iommu_dev->ncb);
@@ -222,14 +202,14 @@
 	if (!par) {
 		pr_err("%s: Invalid PAR value detected\n", iommu_dev->name);
 		ret = -ENODEV;
-		goto fail_io;
+		goto fail_clk;
 	}
 
 	ret = request_irq(irq, msm_iommu_fault_handler, 0,
 			"msm_iommu_secure_irpt_handler", drvdata);
 	if (ret) {
 		pr_err("Request IRQ %d failed with ret=%d\n", irq, ret);
-		goto fail_io;
+		goto fail_clk;
 	}
 
 
@@ -250,10 +230,6 @@
 	clk_disable(iommu_pclk);
 
 	return 0;
-fail_io:
-	iounmap(regs_base);
-fail_mem:
-	release_mem_region(r->start, len);
 fail_clk:
 	if (iommu_clk) {
 		clk_disable(iommu_clk);
diff --git a/drivers/iommu/omap-iommu.c b/drivers/iommu/omap-iommu.c
index 7fcbfc4..895af06 100644
--- a/drivers/iommu/omap-iommu.c
+++ b/drivers/iommu/omap-iommu.c
@@ -34,6 +34,9 @@
 #include "omap-iopgtable.h"
 #include "omap-iommu.h"
 
+#define to_iommu(dev)							\
+	((struct omap_iommu *)platform_get_drvdata(to_platform_device(dev)))
+
 #define for_each_iotlb_cr(obj, n, __i, cr)				\
 	for (__i = 0;							\
 	     (__i < (n)) && (cr = __iotlb_read_cr((obj), __i), true);	\
@@ -391,6 +394,7 @@
 				__func__, start, da, bytes);
 			iotlb_load_cr(obj, &cr);
 			iommu_write_reg(obj, 1, MMU_FLUSH_ENTRY);
+			break;
 		}
 	}
 	pm_runtime_put_sync(obj->dev);
@@ -1037,19 +1041,18 @@
 	clean_dcache_area(iopte, IOPTE_TABLE_SIZE);
 }
 
-static u32 iotlb_init_entry(struct iotlb_entry *e, u32 da, u32 pa,
-				   u32 flags)
+static u32 iotlb_init_entry(struct iotlb_entry *e, u32 da, u32 pa, int pgsz)
 {
 	memset(e, 0, sizeof(*e));
 
 	e->da		= da;
 	e->pa		= pa;
-	e->valid	= 1;
+	e->valid	= MMU_CAM_V;
 	/* FIXME: add OMAP1 support */
-	e->pgsz		= flags & MMU_CAM_PGSZ_MASK;
-	e->endian	= flags & MMU_RAM_ENDIAN_MASK;
-	e->elsz		= flags & MMU_RAM_ELSZ_MASK;
-	e->mixed	= flags & MMU_RAM_MIXED_MASK;
+	e->pgsz		= pgsz;
+	e->endian	= MMU_RAM_ENDIAN_LITTLE;
+	e->elsz		= MMU_RAM_ELSZ_8;
+	e->mixed	= 0;
 
 	return iopgsz_to_bytes(e->pgsz);
 }
@@ -1062,9 +1065,8 @@
 	struct device *dev = oiommu->dev;
 	struct iotlb_entry e;
 	int omap_pgsz;
-	u32 ret, flags;
+	u32 ret;
 
-	/* we only support mapping a single iommu page for now */
 	omap_pgsz = bytes_to_iopgsz(bytes);
 	if (omap_pgsz < 0) {
 		dev_err(dev, "invalid size to map: %d\n", bytes);
@@ -1073,9 +1075,7 @@
 
 	dev_dbg(dev, "mapping da 0x%lx to pa 0x%x size 0x%x\n", da, pa, bytes);
 
-	flags = omap_pgsz | prot;
-
-	iotlb_init_entry(&e, da, pa, flags);
+	iotlb_init_entry(&e, da, pa, omap_pgsz);
 
 	ret = omap_iopgtable_store_entry(oiommu, &e);
 	if (ret)
@@ -1248,12 +1248,6 @@
 	return ret;
 }
 
-static int omap_iommu_domain_has_cap(struct iommu_domain *domain,
-				    unsigned long cap)
-{
-	return 0;
-}
-
 static int omap_iommu_add_device(struct device *dev)
 {
 	struct omap_iommu_arch_data *arch_data;
@@ -1305,7 +1299,6 @@
 	.map		= omap_iommu_map,
 	.unmap		= omap_iommu_unmap,
 	.iova_to_phys	= omap_iommu_iova_to_phys,
-	.domain_has_cap	= omap_iommu_domain_has_cap,
 	.add_device	= omap_iommu_add_device,
 	.remove_device	= omap_iommu_remove_device,
 	.pgsize_bitmap	= OMAP_IOMMU_PGSIZES,
diff --git a/drivers/iommu/omap-iopgtable.h b/drivers/iommu/omap-iopgtable.h
index b6f9a51..f891683 100644
--- a/drivers/iommu/omap-iopgtable.h
+++ b/drivers/iommu/omap-iopgtable.h
@@ -93,6 +93,3 @@
 /* to find an entry in the second-level page table. */
 #define iopte_index(da)		(((da) >> IOPTE_SHIFT) & (PTRS_PER_IOPTE - 1))
 #define iopte_offset(iopgd, da)	(iopgd_page_vaddr(iopgd) + iopte_index(da))
-
-#define to_iommu(dev)							\
-	(platform_get_drvdata(to_platform_device(dev)))
diff --git a/drivers/iommu/shmobile-ipmmu.c b/drivers/iommu/shmobile-ipmmu.c
index e3bc2e1..bd97ade 100644
--- a/drivers/iommu/shmobile-ipmmu.c
+++ b/drivers/iommu/shmobile-ipmmu.c
@@ -94,11 +94,6 @@
 	struct resource *res;
 	struct shmobile_ipmmu_platform_data *pdata = pdev->dev.platform_data;
 
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (!res) {
-		dev_err(&pdev->dev, "cannot get platform resources\n");
-		return -ENOENT;
-	}
 	ipmmu = devm_kzalloc(&pdev->dev, sizeof(*ipmmu), GFP_KERNEL);
 	if (!ipmmu) {
 		dev_err(&pdev->dev, "cannot allocate device data\n");
@@ -106,19 +101,18 @@
 	}
 	spin_lock_init(&ipmmu->flush_lock);
 	ipmmu->dev = &pdev->dev;
-	ipmmu->ipmmu_base = devm_ioremap_nocache(&pdev->dev, res->start,
-						resource_size(res));
-	if (!ipmmu->ipmmu_base) {
-		dev_err(&pdev->dev, "ioremap_nocache failed\n");
-		return -ENOMEM;
-	}
+
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	ipmmu->ipmmu_base = devm_ioremap_resource(&pdev->dev, res);
+	if (IS_ERR(ipmmu->ipmmu_base))
+		return PTR_ERR(ipmmu->ipmmu_base);
+
 	ipmmu->dev_names = pdata->dev_names;
 	ipmmu->num_dev_names = pdata->num_dev_names;
 	platform_set_drvdata(pdev, ipmmu);
 	ipmmu_reg_write(ipmmu, IMCTR1, 0x0); /* disable TLB */
 	ipmmu_reg_write(ipmmu, IMCTR2, 0x0); /* disable PMB */
-	ipmmu_iommu_init(ipmmu);
-	return 0;
+	return ipmmu_iommu_init(ipmmu);
 }
 
 static struct platform_driver ipmmu_driver = {
diff --git a/drivers/net/bonding/bond_alb.c b/drivers/net/bonding/bond_alb.c
index 9f69e81..93580a4 100644
--- a/drivers/net/bonding/bond_alb.c
+++ b/drivers/net/bonding/bond_alb.c
@@ -82,7 +82,8 @@
 }
 
 /* Forward declaration */
-static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[]);
+static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[],
+				      bool strict_match);
 static void rlb_purge_src_ip(struct bonding *bond, struct arp_pkt *arp);
 static void rlb_src_unlink(struct bonding *bond, u32 index);
 static void rlb_src_link(struct bonding *bond, u32 ip_src_hash,
@@ -459,7 +460,7 @@
 
 	bond->alb_info.rlb_promisc_timeout_counter = 0;
 
-	alb_send_learning_packets(bond->curr_active_slave, addr);
+	alb_send_learning_packets(bond->curr_active_slave, addr, true);
 }
 
 /* slave being removed should not be active at this point
@@ -995,7 +996,7 @@
 /*********************** tlb/rlb shared functions *********************/
 
 static void alb_send_lp_vid(struct slave *slave, u8 mac_addr[],
-			    u16 vid)
+			    __be16 vlan_proto, u16 vid)
 {
 	struct learning_pkt pkt;
 	struct sk_buff *skb;
@@ -1021,7 +1022,7 @@
 	skb->dev = slave->dev;
 
 	if (vid) {
-		skb = vlan_put_tag(skb, htons(ETH_P_8021Q), vid);
+		skb = vlan_put_tag(skb, vlan_proto, vid);
 		if (!skb) {
 			pr_err("%s: Error: failed to insert VLAN tag\n",
 			       slave->bond->dev->name);
@@ -1032,22 +1033,32 @@
 	dev_queue_xmit(skb);
 }
 
-
-static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[])
+static void alb_send_learning_packets(struct slave *slave, u8 mac_addr[],
+				      bool strict_match)
 {
 	struct bonding *bond = bond_get_bond_by_slave(slave);
 	struct net_device *upper;
 	struct list_head *iter;
 
 	/* send untagged */
-	alb_send_lp_vid(slave, mac_addr, 0);
+	alb_send_lp_vid(slave, mac_addr, 0, 0);
 
 	/* loop through vlans and send one packet for each */
 	rcu_read_lock();
 	netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
-		if (upper->priv_flags & IFF_802_1Q_VLAN)
-			alb_send_lp_vid(slave, mac_addr,
-					vlan_dev_vlan_id(upper));
+		if (is_vlan_dev(upper) && vlan_get_encap_level(upper) == 0) {
+			if (strict_match &&
+			    ether_addr_equal_64bits(mac_addr,
+						    upper->dev_addr)) {
+				alb_send_lp_vid(slave, mac_addr,
+						vlan_dev_vlan_proto(upper),
+						vlan_dev_vlan_id(upper));
+			} else if (!strict_match) {
+				alb_send_lp_vid(slave, upper->dev_addr,
+						vlan_dev_vlan_proto(upper),
+						vlan_dev_vlan_id(upper));
+			}
+		}
 	}
 	rcu_read_unlock();
 }
@@ -1107,7 +1118,7 @@
 
 	/* fasten the change in the switch */
 	if (SLAVE_IS_OK(slave1)) {
-		alb_send_learning_packets(slave1, slave1->dev->dev_addr);
+		alb_send_learning_packets(slave1, slave1->dev->dev_addr, false);
 		if (bond->alb_info.rlb_enabled) {
 			/* inform the clients that the mac address
 			 * has changed
@@ -1119,7 +1130,7 @@
 	}
 
 	if (SLAVE_IS_OK(slave2)) {
-		alb_send_learning_packets(slave2, slave2->dev->dev_addr);
+		alb_send_learning_packets(slave2, slave2->dev->dev_addr, false);
 		if (bond->alb_info.rlb_enabled) {
 			/* inform the clients that the mac address
 			 * has changed
@@ -1490,6 +1501,8 @@
 
 	/* send learning packets */
 	if (bond_info->lp_counter >= BOND_ALB_LP_TICKS(bond)) {
+		bool strict_match;
+
 		/* change of curr_active_slave involves swapping of mac addresses.
 		 * in order to avoid this swapping from happening while
 		 * sending the learning packets, the curr_slave_lock must be held for
@@ -1497,8 +1510,15 @@
 		 */
 		read_lock(&bond->curr_slave_lock);
 
-		bond_for_each_slave_rcu(bond, slave, iter)
-			alb_send_learning_packets(slave, slave->dev->dev_addr);
+		bond_for_each_slave_rcu(bond, slave, iter) {
+			/* If updating current_active, use all currently
+			 * user mac addreses (!strict_match).  Otherwise, only
+			 * use mac of the slave device.
+			 */
+			strict_match = (slave != bond->curr_active_slave);
+			alb_send_learning_packets(slave, slave->dev->dev_addr,
+						  strict_match);
+		}
 
 		read_unlock(&bond->curr_slave_lock);
 
@@ -1721,7 +1741,8 @@
 	} else {
 		/* set the new_slave to the bond mac address */
 		alb_set_slave_mac_addr(new_slave, bond->dev->dev_addr);
-		alb_send_learning_packets(new_slave, bond->dev->dev_addr);
+		alb_send_learning_packets(new_slave, bond->dev->dev_addr,
+					  false);
 	}
 
 	write_lock_bh(&bond->curr_slave_lock);
@@ -1764,7 +1785,8 @@
 		alb_set_slave_mac_addr(bond->curr_active_slave, bond_dev->dev_addr);
 
 		read_lock(&bond->lock);
-		alb_send_learning_packets(bond->curr_active_slave, bond_dev->dev_addr);
+		alb_send_learning_packets(bond->curr_active_slave,
+					  bond_dev->dev_addr, false);
 		if (bond->alb_info.rlb_enabled) {
 			/* inform clients mac address has changed */
 			rlb_req_update_slave_clients(bond, bond->curr_active_slave);
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 69aff72..d3a6789 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2126,10 +2126,10 @@
  */
 static void bond_arp_send(struct net_device *slave_dev, int arp_op,
 			  __be32 dest_ip, __be32 src_ip,
-			  struct bond_vlan_tag *inner,
-			  struct bond_vlan_tag *outer)
+			  struct bond_vlan_tag *tags)
 {
 	struct sk_buff *skb;
+	int i;
 
 	pr_debug("arp %d on slave %s: dst %pI4 src %pI4\n",
 		 arp_op, slave_dev->name, &dest_ip, &src_ip);
@@ -2141,21 +2141,26 @@
 		net_err_ratelimited("ARP packet allocation failed\n");
 		return;
 	}
-	if (outer->vlan_id) {
-		if (inner->vlan_id) {
-			pr_debug("inner tag: proto %X vid %X\n",
-				 ntohs(inner->vlan_proto), inner->vlan_id);
-			skb = __vlan_put_tag(skb, inner->vlan_proto,
-					     inner->vlan_id);
-			if (!skb) {
-				net_err_ratelimited("failed to insert inner VLAN tag\n");
-				return;
-			}
-		}
 
-		pr_debug("outer reg: proto %X vid %X\n",
-			 ntohs(outer->vlan_proto), outer->vlan_id);
-		skb = vlan_put_tag(skb, outer->vlan_proto, outer->vlan_id);
+	/* Go through all the tags backwards and add them to the packet */
+	for (i = BOND_MAX_VLAN_ENCAP - 1; i > 0; i--) {
+		if (!tags[i].vlan_id)
+			continue;
+
+		pr_debug("inner tag: proto %X vid %X\n",
+			 ntohs(tags[i].vlan_proto), tags[i].vlan_id);
+		skb = __vlan_put_tag(skb, tags[i].vlan_proto,
+				     tags[i].vlan_id);
+		if (!skb) {
+			net_err_ratelimited("failed to insert inner VLAN tag\n");
+			return;
+		}
+	}
+	/* Set the outer tag */
+	if (tags[0].vlan_id) {
+		pr_debug("outer tag: proto %X vid %X\n",
+			 ntohs(tags[0].vlan_proto), tags[0].vlan_id);
+		skb = vlan_put_tag(skb, tags[0].vlan_proto, tags[0].vlan_id);
 		if (!skb) {
 			net_err_ratelimited("failed to insert outer VLAN tag\n");
 			return;
@@ -2164,22 +2169,52 @@
 	arp_xmit(skb);
 }
 
+/* Validate the device path between the @start_dev and the @end_dev.
+ * The path is valid if the @end_dev is reachable through device
+ * stacking.
+ * When the path is validated, collect any vlan information in the
+ * path.
+ */
+static bool bond_verify_device_path(struct net_device *start_dev,
+				    struct net_device *end_dev,
+				    struct bond_vlan_tag *tags)
+{
+	struct net_device *upper;
+	struct list_head  *iter;
+	int  idx;
+
+	if (start_dev == end_dev)
+		return true;
+
+	netdev_for_each_upper_dev_rcu(start_dev, upper, iter) {
+		if (bond_verify_device_path(upper, end_dev, tags)) {
+			if (is_vlan_dev(upper)) {
+				idx = vlan_get_encap_level(upper);
+				if (idx >= BOND_MAX_VLAN_ENCAP)
+					return false;
+
+				tags[idx].vlan_proto =
+						    vlan_dev_vlan_proto(upper);
+				tags[idx].vlan_id = vlan_dev_vlan_id(upper);
+			}
+			return true;
+		}
+	}
+
+	return false;
+}
 
 static void bond_arp_send_all(struct bonding *bond, struct slave *slave)
 {
-	struct net_device *upper, *vlan_upper;
-	struct list_head *iter, *vlan_iter;
 	struct rtable *rt;
-	struct bond_vlan_tag inner, outer;
+	struct bond_vlan_tag tags[BOND_MAX_VLAN_ENCAP];
 	__be32 *targets = bond->params.arp_targets, addr;
 	int i;
+	bool ret;
 
 	for (i = 0; i < BOND_MAX_ARP_TARGETS && targets[i]; i++) {
 		pr_debug("basa: target %pI4\n", &targets[i]);
-		inner.vlan_proto = 0;
-		inner.vlan_id = 0;
-		outer.vlan_proto = 0;
-		outer.vlan_id = 0;
+		memset(tags, 0, sizeof(tags));
 
 		/* Find out through which dev should the packet go */
 		rt = ip_route_output(dev_net(bond->dev), targets[i], 0,
@@ -2192,7 +2227,8 @@
 				net_warn_ratelimited("%s: no route to arp_ip_target %pI4 and arp_validate is set\n",
 						     bond->dev->name,
 						     &targets[i]);
-			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i], 0, &inner, &outer);
+			bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
+				      0, tags);
 			continue;
 		}
 
@@ -2201,52 +2237,12 @@
 			goto found;
 
 		rcu_read_lock();
-		/* first we search only for vlan devices. for every vlan
-		 * found we verify its upper dev list, searching for the
-		 * rt->dst.dev. If found we save the tag of the vlan and
-		 * proceed to send the packet.
-		 */
-		netdev_for_each_all_upper_dev_rcu(bond->dev, vlan_upper,
-						  vlan_iter) {
-			if (!is_vlan_dev(vlan_upper))
-				continue;
-
-			if (vlan_upper == rt->dst.dev) {
-				outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
-				outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
-				rcu_read_unlock();
-				goto found;
-			}
-			netdev_for_each_all_upper_dev_rcu(vlan_upper, upper,
-							  iter) {
-				if (upper == rt->dst.dev) {
-					/* If the upper dev is a vlan dev too,
-					 *  set the vlan tag to inner tag.
-					 */
-					if (is_vlan_dev(upper)) {
-						inner.vlan_proto = vlan_dev_vlan_proto(upper);
-						inner.vlan_id = vlan_dev_vlan_id(upper);
-					}
-					outer.vlan_proto = vlan_dev_vlan_proto(vlan_upper);
-					outer.vlan_id = vlan_dev_vlan_id(vlan_upper);
-					rcu_read_unlock();
-					goto found;
-				}
-			}
-		}
-
-		/* if the device we're looking for is not on top of any of
-		 * our upper vlans, then just search for any dev that
-		 * matches, and in case it's a vlan - save the id
-		 */
-		netdev_for_each_all_upper_dev_rcu(bond->dev, upper, iter) {
-			if (upper == rt->dst.dev) {
-				rcu_read_unlock();
-				goto found;
-			}
-		}
+		ret = bond_verify_device_path(bond->dev, rt->dst.dev, tags);
 		rcu_read_unlock();
 
+		if (ret)
+			goto found;
+
 		/* Not our device - skip */
 		pr_debug("%s: no path to arp_ip_target %pI4 via rt.dev %s\n",
 			 bond->dev->name, &targets[i],
@@ -2259,7 +2255,7 @@
 		addr = bond_confirm_addr(rt->dst.dev, targets[i], 0);
 		ip_rt_put(rt);
 		bond_arp_send(slave->dev, ARPOP_REQUEST, targets[i],
-			      addr, &inner, &outer);
+			      addr, tags);
 	}
 }
 
diff --git a/drivers/net/bonding/bond_options.c b/drivers/net/bonding/bond_options.c
index 724e30f..8320702 100644
--- a/drivers/net/bonding/bond_options.c
+++ b/drivers/net/bonding/bond_options.c
@@ -125,6 +125,7 @@
 static const struct bond_opt_value bond_intmax_tbl[] = {
 	{ "off",     0,       BOND_VALFLAG_DEFAULT},
 	{ "maxval",  INT_MAX, BOND_VALFLAG_MAX},
+	{ NULL,      -1,      0}
 };
 
 static const struct bond_opt_value bond_lacp_rate_tbl[] = {
diff --git a/drivers/net/bonding/bonding.h b/drivers/net/bonding/bonding.h
index b8bdd0a..00bea32 100644
--- a/drivers/net/bonding/bonding.h
+++ b/drivers/net/bonding/bonding.h
@@ -36,6 +36,7 @@
 
 #define bond_version DRV_DESCRIPTION ": v" DRV_VERSION " (" DRV_RELDATE ")\n"
 
+#define BOND_MAX_VLAN_ENCAP	2
 #define BOND_MAX_ARP_TARGETS	16
 
 #define BOND_DEFAULT_MIIMON	100
diff --git a/drivers/net/can/c_can/Kconfig b/drivers/net/can/c_can/Kconfig
index 8ab7103..61ffc12 100644
--- a/drivers/net/can/c_can/Kconfig
+++ b/drivers/net/can/c_can/Kconfig
@@ -14,13 +14,6 @@
 	  SPEAr1310 and SPEAr320 evaluation boards & TI (www.ti.com)
 	  boards like am335x, dm814x, dm813x and dm811x.
 
-config CAN_C_CAN_STRICT_FRAME_ORDERING
-	bool "Force a strict RX CAN frame order (may cause frame loss)"
-	---help---
-	  The RX split buffer prevents packet reordering but can cause packet
-	  loss. Only enable this option when you accept to lose CAN frames
-	  in favour of getting the received CAN frames in the correct order.
-
 config CAN_C_CAN_PCI
 	tristate "Generic PCI Bus based C_CAN/D_CAN driver"
 	depends on PCI
diff --git a/drivers/net/can/c_can/c_can.c b/drivers/net/can/c_can/c_can.c
index a2ca820..95e04e2 100644
--- a/drivers/net/can/c_can/c_can.c
+++ b/drivers/net/can/c_can/c_can.c
@@ -732,26 +732,12 @@
 static inline void c_can_rx_object_get(struct net_device *dev,
 				       struct c_can_priv *priv, u32 obj)
 {
-#ifdef CONFIG_CAN_C_CAN_STRICT_FRAME_ORDERING
-	if (obj < C_CAN_MSG_RX_LOW_LAST)
-		c_can_object_get(dev, IF_RX, obj, IF_COMM_RCV_LOW);
-	else
-#endif
 		c_can_object_get(dev, IF_RX, obj, priv->comm_rcv_high);
 }
 
 static inline void c_can_rx_finalize(struct net_device *dev,
 				     struct c_can_priv *priv, u32 obj)
 {
-#ifdef CONFIG_CAN_C_CAN_STRICT_FRAME_ORDERING
-	if (obj < C_CAN_MSG_RX_LOW_LAST)
-		priv->rxmasked |= BIT(obj - 1);
-	else if (obj == C_CAN_MSG_RX_LOW_LAST) {
-		priv->rxmasked = 0;
-		/* activate all lower message objects */
-		c_can_activate_all_lower_rx_msg_obj(dev, IF_RX);
-	}
-#endif
 	if (priv->type != BOSCH_D_CAN)
 		c_can_object_get(dev, IF_RX, obj, IF_COMM_CLR_NEWDAT);
 }
@@ -799,9 +785,6 @@
 {
 	u32 pend = priv->read_reg(priv, C_CAN_NEWDAT1_REG);
 
-#ifdef CONFIG_CAN_C_CAN_STRICT_FRAME_ORDERING
-	pend &= ~priv->rxmasked;
-#endif
 	return pend;
 }
 
@@ -814,25 +797,6 @@
  * has arrived. To work-around this issue, we keep two groups of message
  * objects whose partitioning is defined by C_CAN_MSG_OBJ_RX_SPLIT.
  *
- * If CONFIG_CAN_C_CAN_STRICT_FRAME_ORDERING = y
- *
- * To ensure in-order frame reception we use the following
- * approach while re-activating a message object to receive further
- * frames:
- * - if the current message object number is lower than
- *   C_CAN_MSG_RX_LOW_LAST, do not clear the NEWDAT bit while clearing
- *   the INTPND bit.
- * - if the current message object number is equal to
- *   C_CAN_MSG_RX_LOW_LAST then clear the NEWDAT bit of all lower
- *   receive message objects.
- * - if the current message object number is greater than
- *   C_CAN_MSG_RX_LOW_LAST then clear the NEWDAT bit of
- *   only this message object.
- *
- * This can cause packet loss!
- *
- * If CONFIG_CAN_C_CAN_STRICT_FRAME_ORDERING = n
- *
  * We clear the newdat bit right away.
  *
  * This can result in packet reordering when the readout is slow.
diff --git a/drivers/net/can/sja1000/peak_pci.c b/drivers/net/can/sja1000/peak_pci.c
index c540e3d..564933a 100644
--- a/drivers/net/can/sja1000/peak_pci.c
+++ b/drivers/net/can/sja1000/peak_pci.c
@@ -551,7 +551,7 @@
 {
 	struct sja1000_priv *priv;
 	struct peak_pci_chan *chan;
-	struct net_device *dev;
+	struct net_device *dev, *prev_dev;
 	void __iomem *cfg_base, *reg_base;
 	u16 sub_sys_id, icr;
 	int i, err, channels;
@@ -688,11 +688,13 @@
 	writew(0x0, cfg_base + PITA_ICR + 2);
 
 	chan = NULL;
-	for (dev = pci_get_drvdata(pdev); dev; dev = chan->prev_dev) {
-		unregister_sja1000dev(dev);
-		free_sja1000dev(dev);
+	for (dev = pci_get_drvdata(pdev); dev; dev = prev_dev) {
 		priv = netdev_priv(dev);
 		chan = priv->priv;
+		prev_dev = chan->prev_dev;
+
+		unregister_sja1000dev(dev);
+		free_sja1000dev(dev);
 	}
 
 	/* free any PCIeC resources too */
@@ -726,10 +728,12 @@
 
 	/* Loop over all registered devices */
 	while (1) {
+		struct net_device *prev_dev = chan->prev_dev;
+
 		dev_info(&pdev->dev, "removing device %s\n", dev->name);
 		unregister_sja1000dev(dev);
 		free_sja1000dev(dev);
-		dev = chan->prev_dev;
+		dev = prev_dev;
 
 		if (!dev) {
 			/* do that only for first channel */
diff --git a/drivers/net/ethernet/Kconfig b/drivers/net/ethernet/Kconfig
index 39b26fe..d740101 100644
--- a/drivers/net/ethernet/Kconfig
+++ b/drivers/net/ethernet/Kconfig
@@ -35,6 +35,18 @@
 source "drivers/net/ethernet/chelsio/Kconfig"
 source "drivers/net/ethernet/cirrus/Kconfig"
 source "drivers/net/ethernet/cisco/Kconfig"
+
+config CX_ECAT
+	tristate "Beckhoff CX5020 EtherCAT master support"
+	depends on PCI
+	---help---
+	  Driver for EtherCAT master module located on CCAT FPGA
+	  that can be found on Beckhoff CX5020, and possibly other of CX
+	  Beckhoff CX series industrial PCs.
+
+	  To compile this driver as a module, choose M here. The module
+	  will be called ec_bhf.
+
 source "drivers/net/ethernet/davicom/Kconfig"
 
 config DNET
diff --git a/drivers/net/ethernet/Makefile b/drivers/net/ethernet/Makefile
index 545d0b3..35190e3 100644
--- a/drivers/net/ethernet/Makefile
+++ b/drivers/net/ethernet/Makefile
@@ -21,6 +21,7 @@
 obj-$(CONFIG_NET_VENDOR_CHELSIO) += chelsio/
 obj-$(CONFIG_NET_VENDOR_CIRRUS) += cirrus/
 obj-$(CONFIG_NET_VENDOR_CISCO) += cisco/
+obj-$(CONFIG_CX_ECAT) += ec_bhf.o
 obj-$(CONFIG_DM9000) += davicom/
 obj-$(CONFIG_DNET) += dnet.o
 obj-$(CONFIG_NET_VENDOR_DEC) += dec/
diff --git a/drivers/net/ethernet/altera/Makefile b/drivers/net/ethernet/altera/Makefile
index d4a187e..3eff2fd 100644
--- a/drivers/net/ethernet/altera/Makefile
+++ b/drivers/net/ethernet/altera/Makefile
@@ -5,3 +5,4 @@
 obj-$(CONFIG_ALTERA_TSE) += altera_tse.o
 altera_tse-objs := altera_tse_main.o altera_tse_ethtool.o \
 altera_msgdma.o altera_sgdma.o altera_utils.o
+ccflags-y += -D__CHECK_ENDIAN__
diff --git a/drivers/net/ethernet/altera/altera_msgdma.c b/drivers/net/ethernet/altera/altera_msgdma.c
index 4d1f2fd..0fb986b 100644
--- a/drivers/net/ethernet/altera/altera_msgdma.c
+++ b/drivers/net/ethernet/altera/altera_msgdma.c
@@ -37,18 +37,16 @@
 void msgdma_reset(struct altera_tse_private *priv)
 {
 	int counter;
-	struct msgdma_csr *txcsr =
-		(struct msgdma_csr *)priv->tx_dma_csr;
-	struct msgdma_csr *rxcsr =
-		(struct msgdma_csr *)priv->rx_dma_csr;
 
 	/* Reset Rx mSGDMA */
-	iowrite32(MSGDMA_CSR_STAT_MASK, &rxcsr->status);
-	iowrite32(MSGDMA_CSR_CTL_RESET, &rxcsr->control);
+	csrwr32(MSGDMA_CSR_STAT_MASK, priv->rx_dma_csr,
+		msgdma_csroffs(status));
+	csrwr32(MSGDMA_CSR_CTL_RESET, priv->rx_dma_csr,
+		msgdma_csroffs(control));
 
 	counter = 0;
 	while (counter++ < ALTERA_TSE_SW_RESET_WATCHDOG_CNTR) {
-		if (tse_bit_is_clear(&rxcsr->status,
+		if (tse_bit_is_clear(priv->rx_dma_csr, msgdma_csroffs(status),
 				     MSGDMA_CSR_STAT_RESETTING))
 			break;
 		udelay(1);
@@ -59,15 +57,18 @@
 			   "TSE Rx mSGDMA resetting bit never cleared!\n");
 
 	/* clear all status bits */
-	iowrite32(MSGDMA_CSR_STAT_MASK, &rxcsr->status);
+	csrwr32(MSGDMA_CSR_STAT_MASK, priv->rx_dma_csr, msgdma_csroffs(status));
 
 	/* Reset Tx mSGDMA */
-	iowrite32(MSGDMA_CSR_STAT_MASK, &txcsr->status);
-	iowrite32(MSGDMA_CSR_CTL_RESET, &txcsr->control);
+	csrwr32(MSGDMA_CSR_STAT_MASK, priv->tx_dma_csr,
+		msgdma_csroffs(status));
+
+	csrwr32(MSGDMA_CSR_CTL_RESET, priv->tx_dma_csr,
+		msgdma_csroffs(control));
 
 	counter = 0;
 	while (counter++ < ALTERA_TSE_SW_RESET_WATCHDOG_CNTR) {
-		if (tse_bit_is_clear(&txcsr->status,
+		if (tse_bit_is_clear(priv->tx_dma_csr, msgdma_csroffs(status),
 				     MSGDMA_CSR_STAT_RESETTING))
 			break;
 		udelay(1);
@@ -78,58 +79,58 @@
 			   "TSE Tx mSGDMA resetting bit never cleared!\n");
 
 	/* clear all status bits */
-	iowrite32(MSGDMA_CSR_STAT_MASK, &txcsr->status);
+	csrwr32(MSGDMA_CSR_STAT_MASK, priv->tx_dma_csr, msgdma_csroffs(status));
 }
 
 void msgdma_disable_rxirq(struct altera_tse_private *priv)
 {
-	struct msgdma_csr *csr = priv->rx_dma_csr;
-	tse_clear_bit(&csr->control, MSGDMA_CSR_CTL_GLOBAL_INTR);
+	tse_clear_bit(priv->rx_dma_csr, msgdma_csroffs(control),
+		      MSGDMA_CSR_CTL_GLOBAL_INTR);
 }
 
 void msgdma_enable_rxirq(struct altera_tse_private *priv)
 {
-	struct msgdma_csr *csr = priv->rx_dma_csr;
-	tse_set_bit(&csr->control, MSGDMA_CSR_CTL_GLOBAL_INTR);
+	tse_set_bit(priv->rx_dma_csr, msgdma_csroffs(control),
+		    MSGDMA_CSR_CTL_GLOBAL_INTR);
 }
 
 void msgdma_disable_txirq(struct altera_tse_private *priv)
 {
-	struct msgdma_csr *csr = priv->tx_dma_csr;
-	tse_clear_bit(&csr->control, MSGDMA_CSR_CTL_GLOBAL_INTR);
+	tse_clear_bit(priv->tx_dma_csr, msgdma_csroffs(control),
+		      MSGDMA_CSR_CTL_GLOBAL_INTR);
 }
 
 void msgdma_enable_txirq(struct altera_tse_private *priv)
 {
-	struct msgdma_csr *csr = priv->tx_dma_csr;
-	tse_set_bit(&csr->control, MSGDMA_CSR_CTL_GLOBAL_INTR);
+	tse_set_bit(priv->tx_dma_csr, msgdma_csroffs(control),
+		    MSGDMA_CSR_CTL_GLOBAL_INTR);
 }
 
 void msgdma_clear_rxirq(struct altera_tse_private *priv)
 {
-	struct msgdma_csr *csr = priv->rx_dma_csr;
-	iowrite32(MSGDMA_CSR_STAT_IRQ, &csr->status);
+	csrwr32(MSGDMA_CSR_STAT_IRQ, priv->rx_dma_csr, msgdma_csroffs(status));
 }
 
 void msgdma_clear_txirq(struct altera_tse_private *priv)
 {
-	struct msgdma_csr *csr = priv->tx_dma_csr;
-	iowrite32(MSGDMA_CSR_STAT_IRQ, &csr->status);
+	csrwr32(MSGDMA_CSR_STAT_IRQ, priv->tx_dma_csr, msgdma_csroffs(status));
 }
 
 /* return 0 to indicate transmit is pending */
 int msgdma_tx_buffer(struct altera_tse_private *priv, struct tse_buffer *buffer)
 {
-	struct msgdma_extended_desc *desc = priv->tx_dma_desc;
-
-	iowrite32(lower_32_bits(buffer->dma_addr), &desc->read_addr_lo);
-	iowrite32(upper_32_bits(buffer->dma_addr), &desc->read_addr_hi);
-	iowrite32(0, &desc->write_addr_lo);
-	iowrite32(0, &desc->write_addr_hi);
-	iowrite32(buffer->len, &desc->len);
-	iowrite32(0, &desc->burst_seq_num);
-	iowrite32(MSGDMA_DESC_TX_STRIDE, &desc->stride);
-	iowrite32(MSGDMA_DESC_CTL_TX_SINGLE, &desc->control);
+	csrwr32(lower_32_bits(buffer->dma_addr), priv->tx_dma_desc,
+		msgdma_descroffs(read_addr_lo));
+	csrwr32(upper_32_bits(buffer->dma_addr), priv->tx_dma_desc,
+		msgdma_descroffs(read_addr_hi));
+	csrwr32(0, priv->tx_dma_desc, msgdma_descroffs(write_addr_lo));
+	csrwr32(0, priv->tx_dma_desc, msgdma_descroffs(write_addr_hi));
+	csrwr32(buffer->len, priv->tx_dma_desc, msgdma_descroffs(len));
+	csrwr32(0, priv->tx_dma_desc, msgdma_descroffs(burst_seq_num));
+	csrwr32(MSGDMA_DESC_TX_STRIDE, priv->tx_dma_desc,
+		msgdma_descroffs(stride));
+	csrwr32(MSGDMA_DESC_CTL_TX_SINGLE, priv->tx_dma_desc,
+		msgdma_descroffs(control));
 	return 0;
 }
 
@@ -138,17 +139,16 @@
 	u32 ready = 0;
 	u32 inuse;
 	u32 status;
-	struct msgdma_csr *txcsr =
-		(struct msgdma_csr *)priv->tx_dma_csr;
 
 	/* Get number of sent descriptors */
-	inuse = ioread32(&txcsr->rw_fill_level) & 0xffff;
+	inuse = csrrd32(priv->tx_dma_csr, msgdma_csroffs(rw_fill_level))
+			& 0xffff;
 
 	if (inuse) { /* Tx FIFO is not empty */
 		ready = priv->tx_prod - priv->tx_cons - inuse - 1;
 	} else {
 		/* Check for buffered last packet */
-		status = ioread32(&txcsr->status);
+		status = csrrd32(priv->tx_dma_csr, msgdma_csroffs(status));
 		if (status & MSGDMA_CSR_STAT_BUSY)
 			ready = priv->tx_prod - priv->tx_cons - 1;
 		else
@@ -162,7 +162,6 @@
 void msgdma_add_rx_desc(struct altera_tse_private *priv,
 			struct tse_buffer *rxbuffer)
 {
-	struct msgdma_extended_desc *desc = priv->rx_dma_desc;
 	u32 len = priv->rx_dma_buf_sz;
 	dma_addr_t dma_addr = rxbuffer->dma_addr;
 	u32 control = (MSGDMA_DESC_CTL_END_ON_EOP
@@ -172,14 +171,16 @@
 			| MSGDMA_DESC_CTL_TR_ERR_IRQ
 			| MSGDMA_DESC_CTL_GO);
 
-	iowrite32(0, &desc->read_addr_lo);
-	iowrite32(0, &desc->read_addr_hi);
-	iowrite32(lower_32_bits(dma_addr), &desc->write_addr_lo);
-	iowrite32(upper_32_bits(dma_addr), &desc->write_addr_hi);
-	iowrite32(len, &desc->len);
-	iowrite32(0, &desc->burst_seq_num);
-	iowrite32(0x00010001, &desc->stride);
-	iowrite32(control, &desc->control);
+	csrwr32(0, priv->rx_dma_desc, msgdma_descroffs(read_addr_lo));
+	csrwr32(0, priv->rx_dma_desc, msgdma_descroffs(read_addr_hi));
+	csrwr32(lower_32_bits(dma_addr), priv->rx_dma_desc,
+		msgdma_descroffs(write_addr_lo));
+	csrwr32(upper_32_bits(dma_addr), priv->rx_dma_desc,
+		msgdma_descroffs(write_addr_hi));
+	csrwr32(len, priv->rx_dma_desc, msgdma_descroffs(len));
+	csrwr32(0, priv->rx_dma_desc, msgdma_descroffs(burst_seq_num));
+	csrwr32(0x00010001, priv->rx_dma_desc, msgdma_descroffs(stride));
+	csrwr32(control, priv->rx_dma_desc, msgdma_descroffs(control));
 }
 
 /* status is returned on upper 16 bits,
@@ -190,14 +191,13 @@
 	u32 rxstatus = 0;
 	u32 pktlength;
 	u32 pktstatus;
-	struct msgdma_csr *rxcsr =
-		(struct msgdma_csr *)priv->rx_dma_csr;
-	struct msgdma_response *rxresp =
-		(struct msgdma_response *)priv->rx_dma_resp;
 
-	if (ioread32(&rxcsr->resp_fill_level) & 0xffff) {
-		pktlength = ioread32(&rxresp->bytes_transferred);
-		pktstatus = ioread32(&rxresp->status);
+	if (csrrd32(priv->rx_dma_csr, msgdma_csroffs(resp_fill_level))
+	    & 0xffff) {
+		pktlength = csrrd32(priv->rx_dma_resp,
+				    msgdma_respoffs(bytes_transferred));
+		pktstatus = csrrd32(priv->rx_dma_resp,
+				    msgdma_respoffs(status));
 		rxstatus = pktstatus;
 		rxstatus = rxstatus << 16;
 		rxstatus |= (pktlength & 0xffff);
diff --git a/drivers/net/ethernet/altera/altera_msgdmahw.h b/drivers/net/ethernet/altera/altera_msgdmahw.h
index d7b59ba..e335626 100644
--- a/drivers/net/ethernet/altera/altera_msgdmahw.h
+++ b/drivers/net/ethernet/altera/altera_msgdmahw.h
@@ -17,15 +17,6 @@
 #ifndef __ALTERA_MSGDMAHW_H__
 #define __ALTERA_MSGDMAHW_H__
 
-/* mSGDMA standard descriptor format
- */
-struct msgdma_desc {
-	u32 read_addr;	/* data buffer source address */
-	u32 write_addr;	/* data buffer destination address */
-	u32 len;	/* the number of bytes to transfer per descriptor */
-	u32 control;	/* characteristics of the transfer */
-};
-
 /* mSGDMA extended descriptor format
  */
 struct msgdma_extended_desc {
@@ -159,6 +150,10 @@
 	u32 status;
 };
 
+#define msgdma_respoffs(a) (offsetof(struct msgdma_response, a))
+#define msgdma_csroffs(a) (offsetof(struct msgdma_csr, a))
+#define msgdma_descroffs(a) (offsetof(struct msgdma_extended_desc, a))
+
 /* mSGDMA response register bit definitions
  */
 #define MSGDMA_RESP_EARLY_TERM	BIT(8)
diff --git a/drivers/net/ethernet/altera/altera_sgdma.c b/drivers/net/ethernet/altera/altera_sgdma.c
index 9ce8630..99cc56f 100644
--- a/drivers/net/ethernet/altera/altera_sgdma.c
+++ b/drivers/net/ethernet/altera/altera_sgdma.c
@@ -20,8 +20,8 @@
 #include "altera_sgdmahw.h"
 #include "altera_sgdma.h"
 
-static void sgdma_setup_descrip(struct sgdma_descrip *desc,
-				struct sgdma_descrip *ndesc,
+static void sgdma_setup_descrip(struct sgdma_descrip __iomem *desc,
+				struct sgdma_descrip __iomem *ndesc,
 				dma_addr_t ndesc_phys,
 				dma_addr_t raddr,
 				dma_addr_t waddr,
@@ -31,17 +31,17 @@
 				int wfixed);
 
 static int sgdma_async_write(struct altera_tse_private *priv,
-			      struct sgdma_descrip *desc);
+			      struct sgdma_descrip __iomem *desc);
 
 static int sgdma_async_read(struct altera_tse_private *priv);
 
 static dma_addr_t
 sgdma_txphysaddr(struct altera_tse_private *priv,
-		 struct sgdma_descrip *desc);
+		 struct sgdma_descrip __iomem *desc);
 
 static dma_addr_t
 sgdma_rxphysaddr(struct altera_tse_private *priv,
-		 struct sgdma_descrip *desc);
+		 struct sgdma_descrip __iomem *desc);
 
 static int sgdma_txbusy(struct altera_tse_private *priv);
 
@@ -79,7 +79,8 @@
 	priv->rxdescphys = (dma_addr_t) 0;
 	priv->txdescphys = (dma_addr_t) 0;
 
-	priv->rxdescphys = dma_map_single(priv->device, priv->rx_dma_desc,
+	priv->rxdescphys = dma_map_single(priv->device,
+					  (void __force *)priv->rx_dma_desc,
 					  priv->rxdescmem, DMA_BIDIRECTIONAL);
 
 	if (dma_mapping_error(priv->device, priv->rxdescphys)) {
@@ -88,7 +89,8 @@
 		return -EINVAL;
 	}
 
-	priv->txdescphys = dma_map_single(priv->device, priv->tx_dma_desc,
+	priv->txdescphys = dma_map_single(priv->device,
+					  (void __force *)priv->tx_dma_desc,
 					  priv->txdescmem, DMA_TO_DEVICE);
 
 	if (dma_mapping_error(priv->device, priv->txdescphys)) {
@@ -98,8 +100,8 @@
 	}
 
 	/* Initialize descriptor memory to all 0's, sync memory to cache */
-	memset(priv->tx_dma_desc, 0, priv->txdescmem);
-	memset(priv->rx_dma_desc, 0, priv->rxdescmem);
+	memset_io(priv->tx_dma_desc, 0, priv->txdescmem);
+	memset_io(priv->rx_dma_desc, 0, priv->rxdescmem);
 
 	dma_sync_single_for_device(priv->device, priv->txdescphys,
 				   priv->txdescmem, DMA_TO_DEVICE);
@@ -126,22 +128,15 @@
  */
 void sgdma_reset(struct altera_tse_private *priv)
 {
-	u32 *ptxdescripmem = (u32 *)priv->tx_dma_desc;
-	u32 txdescriplen   = priv->txdescmem;
-	u32 *prxdescripmem = (u32 *)priv->rx_dma_desc;
-	u32 rxdescriplen   = priv->rxdescmem;
-	struct sgdma_csr *ptxsgdma = (struct sgdma_csr *)priv->tx_dma_csr;
-	struct sgdma_csr *prxsgdma = (struct sgdma_csr *)priv->rx_dma_csr;
-
 	/* Initialize descriptor memory to 0 */
-	memset(ptxdescripmem, 0, txdescriplen);
-	memset(prxdescripmem, 0, rxdescriplen);
+	memset_io(priv->tx_dma_desc, 0, priv->txdescmem);
+	memset_io(priv->rx_dma_desc, 0, priv->rxdescmem);
 
-	iowrite32(SGDMA_CTRLREG_RESET, &ptxsgdma->control);
-	iowrite32(0, &ptxsgdma->control);
+	csrwr32(SGDMA_CTRLREG_RESET, priv->tx_dma_csr, sgdma_csroffs(control));
+	csrwr32(0, priv->tx_dma_csr, sgdma_csroffs(control));
 
-	iowrite32(SGDMA_CTRLREG_RESET, &prxsgdma->control);
-	iowrite32(0, &prxsgdma->control);
+	csrwr32(SGDMA_CTRLREG_RESET, priv->rx_dma_csr, sgdma_csroffs(control));
+	csrwr32(0, priv->rx_dma_csr, sgdma_csroffs(control));
 }
 
 /* For SGDMA, interrupts remain enabled after initially enabling,
@@ -167,14 +162,14 @@
 
 void sgdma_clear_rxirq(struct altera_tse_private *priv)
 {
-	struct sgdma_csr *csr = (struct sgdma_csr *)priv->rx_dma_csr;
-	tse_set_bit(&csr->control, SGDMA_CTRLREG_CLRINT);
+	tse_set_bit(priv->rx_dma_csr, sgdma_csroffs(control),
+		    SGDMA_CTRLREG_CLRINT);
 }
 
 void sgdma_clear_txirq(struct altera_tse_private *priv)
 {
-	struct sgdma_csr *csr = (struct sgdma_csr *)priv->tx_dma_csr;
-	tse_set_bit(&csr->control, SGDMA_CTRLREG_CLRINT);
+	tse_set_bit(priv->tx_dma_csr, sgdma_csroffs(control),
+		    SGDMA_CTRLREG_CLRINT);
 }
 
 /* transmits buffer through SGDMA. Returns number of buffers
@@ -184,12 +179,11 @@
  */
 int sgdma_tx_buffer(struct altera_tse_private *priv, struct tse_buffer *buffer)
 {
-	int pktstx = 0;
-	struct sgdma_descrip *descbase =
-		(struct sgdma_descrip *)priv->tx_dma_desc;
+	struct sgdma_descrip __iomem *descbase =
+		(struct sgdma_descrip __iomem *)priv->tx_dma_desc;
 
-	struct sgdma_descrip *cdesc = &descbase[0];
-	struct sgdma_descrip *ndesc = &descbase[1];
+	struct sgdma_descrip __iomem *cdesc = &descbase[0];
+	struct sgdma_descrip __iomem *ndesc = &descbase[1];
 
 	/* wait 'til the tx sgdma is ready for the next transmit request */
 	if (sgdma_txbusy(priv))
@@ -205,7 +199,7 @@
 			    0,				/* read fixed */
 			    SGDMA_CONTROL_WR_FIXED);	/* Generate SOP */
 
-	pktstx = sgdma_async_write(priv, cdesc);
+	sgdma_async_write(priv, cdesc);
 
 	/* enqueue the request to the pending transmit queue */
 	queue_tx(priv, buffer);
@@ -219,10 +213,10 @@
 u32 sgdma_tx_completions(struct altera_tse_private *priv)
 {
 	u32 ready = 0;
-	struct sgdma_descrip *desc = (struct sgdma_descrip *)priv->tx_dma_desc;
 
 	if (!sgdma_txbusy(priv) &&
-	    ((desc->control & SGDMA_CONTROL_HW_OWNED) == 0) &&
+	    ((csrrd8(priv->tx_dma_desc, sgdma_descroffs(control))
+	     & SGDMA_CONTROL_HW_OWNED) == 0) &&
 	    (dequeue_tx(priv))) {
 		ready = 1;
 	}
@@ -246,32 +240,31 @@
  */
 u32 sgdma_rx_status(struct altera_tse_private *priv)
 {
-	struct sgdma_csr *csr = (struct sgdma_csr *)priv->rx_dma_csr;
-	struct sgdma_descrip *base = (struct sgdma_descrip *)priv->rx_dma_desc;
-	struct sgdma_descrip *desc = NULL;
-	int pktsrx;
-	unsigned int rxstatus = 0;
-	unsigned int pktlength = 0;
-	unsigned int pktstatus = 0;
+	struct sgdma_descrip __iomem *base =
+		(struct sgdma_descrip __iomem *)priv->rx_dma_desc;
+	struct sgdma_descrip __iomem *desc = NULL;
 	struct tse_buffer *rxbuffer = NULL;
+	unsigned int rxstatus = 0;
 
-	u32 sts = ioread32(&csr->status);
+	u32 sts = csrrd32(priv->rx_dma_csr, sgdma_csroffs(status));
 
 	desc = &base[0];
 	if (sts & SGDMA_STSREG_EOP) {
+		unsigned int pktlength = 0;
+		unsigned int pktstatus = 0;
 		dma_sync_single_for_cpu(priv->device,
 					priv->rxdescphys,
 					priv->sgdmadesclen,
 					DMA_FROM_DEVICE);
 
-		pktlength = desc->bytes_xferred;
-		pktstatus = desc->status & 0x3f;
-		rxstatus = pktstatus;
+		pktlength = csrrd16(desc, sgdma_descroffs(bytes_xferred));
+		pktstatus = csrrd8(desc, sgdma_descroffs(status));
+		rxstatus = pktstatus & ~SGDMA_STATUS_EOP;
 		rxstatus = rxstatus << 16;
 		rxstatus |= (pktlength & 0xffff);
 
 		if (rxstatus) {
-			desc->status = 0;
+			csrwr8(0, desc, sgdma_descroffs(status));
 
 			rxbuffer = dequeue_rx(priv);
 			if (rxbuffer == NULL)
@@ -279,12 +272,12 @@
 					    "sgdma rx and rx queue empty!\n");
 
 			/* Clear control */
-			iowrite32(0, &csr->control);
+			csrwr32(0, priv->rx_dma_csr, sgdma_csroffs(control));
 			/* clear status */
-			iowrite32(0xf, &csr->status);
+			csrwr32(0xf, priv->rx_dma_csr, sgdma_csroffs(status));
 
 			/* kick the rx sgdma after reaping this descriptor */
-			pktsrx = sgdma_async_read(priv);
+			sgdma_async_read(priv);
 
 		} else {
 			/* If the SGDMA indicated an end of packet on recv,
@@ -298,10 +291,11 @@
 			 */
 			netdev_err(priv->dev,
 				   "SGDMA RX Error Info: %x, %x, %x\n",
-				   sts, desc->status, rxstatus);
+				   sts, csrrd8(desc, sgdma_descroffs(status)),
+				   rxstatus);
 		}
 	} else if (sts == 0) {
-		pktsrx = sgdma_async_read(priv);
+		sgdma_async_read(priv);
 	}
 
 	return rxstatus;
@@ -309,8 +303,8 @@
 
 
 /* Private functions */
-static void sgdma_setup_descrip(struct sgdma_descrip *desc,
-				struct sgdma_descrip *ndesc,
+static void sgdma_setup_descrip(struct sgdma_descrip __iomem *desc,
+				struct sgdma_descrip __iomem *ndesc,
 				dma_addr_t ndesc_phys,
 				dma_addr_t raddr,
 				dma_addr_t waddr,
@@ -320,27 +314,30 @@
 				int wfixed)
 {
 	/* Clear the next descriptor as not owned by hardware */
-	u32 ctrl = ndesc->control;
-	ctrl &= ~SGDMA_CONTROL_HW_OWNED;
-	ndesc->control = ctrl;
 
-	ctrl = 0;
+	u32 ctrl = csrrd8(ndesc, sgdma_descroffs(control));
+	ctrl &= ~SGDMA_CONTROL_HW_OWNED;
+	csrwr8(ctrl, ndesc, sgdma_descroffs(control));
+
 	ctrl = SGDMA_CONTROL_HW_OWNED;
 	ctrl |= generate_eop;
 	ctrl |= rfixed;
 	ctrl |= wfixed;
 
 	/* Channel is implicitly zero, initialized to 0 by default */
+	csrwr32(lower_32_bits(raddr), desc, sgdma_descroffs(raddr));
+	csrwr32(lower_32_bits(waddr), desc, sgdma_descroffs(waddr));
 
-	desc->raddr = raddr;
-	desc->waddr = waddr;
-	desc->next = lower_32_bits(ndesc_phys);
-	desc->control = ctrl;
-	desc->status = 0;
-	desc->rburst = 0;
-	desc->wburst = 0;
-	desc->bytes = length;
-	desc->bytes_xferred = 0;
+	csrwr32(0, desc, sgdma_descroffs(pad1));
+	csrwr32(0, desc, sgdma_descroffs(pad2));
+	csrwr32(lower_32_bits(ndesc_phys), desc, sgdma_descroffs(next));
+
+	csrwr8(ctrl, desc, sgdma_descroffs(control));
+	csrwr8(0, desc, sgdma_descroffs(status));
+	csrwr8(0, desc, sgdma_descroffs(wburst));
+	csrwr8(0, desc, sgdma_descroffs(rburst));
+	csrwr16(length, desc, sgdma_descroffs(bytes));
+	csrwr16(0, desc, sgdma_descroffs(bytes_xferred));
 }
 
 /* If hardware is busy, don't restart async read.
@@ -351,12 +348,11 @@
  */
 static int sgdma_async_read(struct altera_tse_private *priv)
 {
-	struct sgdma_csr *csr = (struct sgdma_csr *)priv->rx_dma_csr;
-	struct sgdma_descrip *descbase =
-		(struct sgdma_descrip *)priv->rx_dma_desc;
+	struct sgdma_descrip __iomem *descbase =
+		(struct sgdma_descrip __iomem *)priv->rx_dma_desc;
 
-	struct sgdma_descrip *cdesc = &descbase[0];
-	struct sgdma_descrip *ndesc = &descbase[1];
+	struct sgdma_descrip __iomem *cdesc = &descbase[0];
+	struct sgdma_descrip __iomem *ndesc = &descbase[1];
 
 	struct tse_buffer *rxbuffer = NULL;
 
@@ -382,11 +378,13 @@
 					   priv->sgdmadesclen,
 					   DMA_TO_DEVICE);
 
-		iowrite32(lower_32_bits(sgdma_rxphysaddr(priv, cdesc)),
-			  &csr->next_descrip);
+		csrwr32(lower_32_bits(sgdma_rxphysaddr(priv, cdesc)),
+			priv->rx_dma_csr,
+			sgdma_csroffs(next_descrip));
 
-		iowrite32((priv->rxctrlreg | SGDMA_CTRLREG_START),
-			  &csr->control);
+		csrwr32((priv->rxctrlreg | SGDMA_CTRLREG_START),
+			priv->rx_dma_csr,
+			sgdma_csroffs(control));
 
 		return 1;
 	}
@@ -395,32 +393,32 @@
 }
 
 static int sgdma_async_write(struct altera_tse_private *priv,
-			     struct sgdma_descrip *desc)
+			     struct sgdma_descrip __iomem *desc)
 {
-	struct sgdma_csr *csr = (struct sgdma_csr *)priv->tx_dma_csr;
-
 	if (sgdma_txbusy(priv))
 		return 0;
 
 	/* clear control and status */
-	iowrite32(0, &csr->control);
-	iowrite32(0x1f, &csr->status);
+	csrwr32(0, priv->tx_dma_csr, sgdma_csroffs(control));
+	csrwr32(0x1f, priv->tx_dma_csr, sgdma_csroffs(status));
 
 	dma_sync_single_for_device(priv->device, priv->txdescphys,
 				   priv->sgdmadesclen, DMA_TO_DEVICE);
 
-	iowrite32(lower_32_bits(sgdma_txphysaddr(priv, desc)),
-		  &csr->next_descrip);
+	csrwr32(lower_32_bits(sgdma_txphysaddr(priv, desc)),
+		priv->tx_dma_csr,
+		sgdma_csroffs(next_descrip));
 
-	iowrite32((priv->txctrlreg | SGDMA_CTRLREG_START),
-		  &csr->control);
+	csrwr32((priv->txctrlreg | SGDMA_CTRLREG_START),
+		priv->tx_dma_csr,
+		sgdma_csroffs(control));
 
 	return 1;
 }
 
 static dma_addr_t
 sgdma_txphysaddr(struct altera_tse_private *priv,
-		 struct sgdma_descrip *desc)
+		 struct sgdma_descrip __iomem *desc)
 {
 	dma_addr_t paddr = priv->txdescmem_busaddr;
 	uintptr_t offs = (uintptr_t)desc - (uintptr_t)priv->tx_dma_desc;
@@ -429,7 +427,7 @@
 
 static dma_addr_t
 sgdma_rxphysaddr(struct altera_tse_private *priv,
-		 struct sgdma_descrip *desc)
+		 struct sgdma_descrip __iomem *desc)
 {
 	dma_addr_t paddr = priv->rxdescmem_busaddr;
 	uintptr_t offs = (uintptr_t)desc - (uintptr_t)priv->rx_dma_desc;
@@ -518,8 +516,8 @@
  */
 static int sgdma_rxbusy(struct altera_tse_private *priv)
 {
-	struct sgdma_csr *csr = (struct sgdma_csr *)priv->rx_dma_csr;
-	return ioread32(&csr->status) & SGDMA_STSREG_BUSY;
+	return csrrd32(priv->rx_dma_csr, sgdma_csroffs(status))
+		       & SGDMA_STSREG_BUSY;
 }
 
 /* waits for the tx sgdma to finish it's current operation, returns 0
@@ -528,13 +526,14 @@
 static int sgdma_txbusy(struct altera_tse_private *priv)
 {
 	int delay = 0;
-	struct sgdma_csr *csr = (struct sgdma_csr *)priv->tx_dma_csr;
 
 	/* if DMA is busy, wait for current transactino to finish */
-	while ((ioread32(&csr->status) & SGDMA_STSREG_BUSY) && (delay++ < 100))
+	while ((csrrd32(priv->tx_dma_csr, sgdma_csroffs(status))
+		& SGDMA_STSREG_BUSY) && (delay++ < 100))
 		udelay(1);
 
-	if (ioread32(&csr->status) & SGDMA_STSREG_BUSY) {
+	if (csrrd32(priv->tx_dma_csr, sgdma_csroffs(status))
+	    & SGDMA_STSREG_BUSY) {
 		netdev_err(priv->dev, "timeout waiting for tx dma\n");
 		return 1;
 	}
diff --git a/drivers/net/ethernet/altera/altera_sgdmahw.h b/drivers/net/ethernet/altera/altera_sgdmahw.h
index ba3334f..85bc33b 100644
--- a/drivers/net/ethernet/altera/altera_sgdmahw.h
+++ b/drivers/net/ethernet/altera/altera_sgdmahw.h
@@ -19,16 +19,16 @@
 
 /* SGDMA descriptor structure */
 struct sgdma_descrip {
-	unsigned int	raddr; /* address of data to be read */
-	unsigned int	pad1;
-	unsigned int	waddr;
-	unsigned int    pad2;
-	unsigned int	next;
-	unsigned int	pad3;
-	unsigned short  bytes;
-	unsigned char   rburst;
-	unsigned char	wburst;
-	unsigned short	bytes_xferred;	/* 16 bits, bytes xferred */
+	u32	raddr; /* address of data to be read */
+	u32	pad1;
+	u32	waddr;
+	u32	pad2;
+	u32	next;
+	u32	pad3;
+	u16	bytes;
+	u8	rburst;
+	u8	wburst;
+	u16	bytes_xferred;	/* 16 bits, bytes xferred */
 
 	/* bit 0: error
 	 * bit 1: length error
@@ -39,7 +39,7 @@
 	 * bit 6: reserved
 	 * bit 7: status eop for recv case
 	 */
-	unsigned char	status;
+	u8	status;
 
 	/* bit 0: eop
 	 * bit 1: read_fixed
@@ -47,7 +47,7 @@
 	 * bits 3,4,5,6: Channel (always 0)
 	 * bit 7: hardware owned
 	 */
-	unsigned char	control;
+	u8	control;
 } __packed;
 
 
@@ -101,6 +101,8 @@
 	u32	pad3[3];
 };
 
+#define sgdma_csroffs(a) (offsetof(struct sgdma_csr, a))
+#define sgdma_descroffs(a) (offsetof(struct sgdma_descrip, a))
 
 #define SGDMA_STSREG_ERR	BIT(0) /* Error */
 #define SGDMA_STSREG_EOP	BIT(1) /* EOP */
diff --git a/drivers/net/ethernet/altera/altera_tse.h b/drivers/net/ethernet/altera/altera_tse.h
index 465c4aa..2adb24d 100644
--- a/drivers/net/ethernet/altera/altera_tse.h
+++ b/drivers/net/ethernet/altera/altera_tse.h
@@ -357,6 +357,8 @@
 	u32 reserved5[42];
 };
 
+#define tse_csroffs(a) (offsetof(struct altera_tse_mac, a))
+
 /* Transmit and Receive Command Registers Bit Definitions
  */
 #define ALTERA_TSE_TX_CMD_STAT_OMIT_CRC		BIT(17)
@@ -487,4 +489,49 @@
  */
 void altera_tse_set_ethtool_ops(struct net_device *);
 
+static inline
+u32 csrrd32(void __iomem *mac, size_t offs)
+{
+	void __iomem *paddr = (void __iomem *)((uintptr_t)mac + offs);
+	return readl(paddr);
+}
+
+static inline
+u16 csrrd16(void __iomem *mac, size_t offs)
+{
+	void __iomem *paddr = (void __iomem *)((uintptr_t)mac + offs);
+	return readw(paddr);
+}
+
+static inline
+u8 csrrd8(void __iomem *mac, size_t offs)
+{
+	void __iomem *paddr = (void __iomem *)((uintptr_t)mac + offs);
+	return readb(paddr);
+}
+
+static inline
+void csrwr32(u32 val, void __iomem *mac, size_t offs)
+{
+	void __iomem *paddr = (void __iomem *)((uintptr_t)mac + offs);
+
+	writel(val, paddr);
+}
+
+static inline
+void csrwr16(u16 val, void __iomem *mac, size_t offs)
+{
+	void __iomem *paddr = (void __iomem *)((uintptr_t)mac + offs);
+
+	writew(val, paddr);
+}
+
+static inline
+void csrwr8(u8 val, void __iomem *mac, size_t offs)
+{
+	void __iomem *paddr = (void __iomem *)((uintptr_t)mac + offs);
+
+	writeb(val, paddr);
+}
+
 #endif /* __ALTERA_TSE_H__ */
diff --git a/drivers/net/ethernet/altera/altera_tse_ethtool.c b/drivers/net/ethernet/altera/altera_tse_ethtool.c
index 76133ca..54c25ef 100644
--- a/drivers/net/ethernet/altera/altera_tse_ethtool.c
+++ b/drivers/net/ethernet/altera/altera_tse_ethtool.c
@@ -96,54 +96,89 @@
 			   u64 *buf)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
-	struct altera_tse_mac *mac = priv->mac_dev;
 	u64 ext;
 
-	buf[0] = ioread32(&mac->frames_transmitted_ok);
-	buf[1] = ioread32(&mac->frames_received_ok);
-	buf[2] = ioread32(&mac->frames_check_sequence_errors);
-	buf[3] = ioread32(&mac->alignment_errors);
+	buf[0] = csrrd32(priv->mac_dev,
+			 tse_csroffs(frames_transmitted_ok));
+	buf[1] = csrrd32(priv->mac_dev,
+			 tse_csroffs(frames_received_ok));
+	buf[2] = csrrd32(priv->mac_dev,
+			 tse_csroffs(frames_check_sequence_errors));
+	buf[3] = csrrd32(priv->mac_dev,
+			 tse_csroffs(alignment_errors));
 
 	/* Extended aOctetsTransmittedOK counter */
-	ext = (u64) ioread32(&mac->msb_octets_transmitted_ok) << 32;
-	ext |= ioread32(&mac->octets_transmitted_ok);
+	ext = (u64) csrrd32(priv->mac_dev,
+			    tse_csroffs(msb_octets_transmitted_ok)) << 32;
+
+	ext |= csrrd32(priv->mac_dev,
+		       tse_csroffs(octets_transmitted_ok));
 	buf[4] = ext;
 
 	/* Extended aOctetsReceivedOK counter */
-	ext = (u64) ioread32(&mac->msb_octets_received_ok) << 32;
-	ext |= ioread32(&mac->octets_received_ok);
+	ext = (u64) csrrd32(priv->mac_dev,
+			    tse_csroffs(msb_octets_received_ok)) << 32;
+
+	ext |= csrrd32(priv->mac_dev,
+		       tse_csroffs(octets_received_ok));
 	buf[5] = ext;
 
-	buf[6] = ioread32(&mac->tx_pause_mac_ctrl_frames);
-	buf[7] = ioread32(&mac->rx_pause_mac_ctrl_frames);
-	buf[8] = ioread32(&mac->if_in_errors);
-	buf[9] = ioread32(&mac->if_out_errors);
-	buf[10] = ioread32(&mac->if_in_ucast_pkts);
-	buf[11] = ioread32(&mac->if_in_multicast_pkts);
-	buf[12] = ioread32(&mac->if_in_broadcast_pkts);
-	buf[13] = ioread32(&mac->if_out_discards);
-	buf[14] = ioread32(&mac->if_out_ucast_pkts);
-	buf[15] = ioread32(&mac->if_out_multicast_pkts);
-	buf[16] = ioread32(&mac->if_out_broadcast_pkts);
-	buf[17] = ioread32(&mac->ether_stats_drop_events);
+	buf[6] = csrrd32(priv->mac_dev,
+			 tse_csroffs(tx_pause_mac_ctrl_frames));
+	buf[7] = csrrd32(priv->mac_dev,
+			 tse_csroffs(rx_pause_mac_ctrl_frames));
+	buf[8] = csrrd32(priv->mac_dev,
+			 tse_csroffs(if_in_errors));
+	buf[9] = csrrd32(priv->mac_dev,
+			 tse_csroffs(if_out_errors));
+	buf[10] = csrrd32(priv->mac_dev,
+			  tse_csroffs(if_in_ucast_pkts));
+	buf[11] = csrrd32(priv->mac_dev,
+			  tse_csroffs(if_in_multicast_pkts));
+	buf[12] = csrrd32(priv->mac_dev,
+			  tse_csroffs(if_in_broadcast_pkts));
+	buf[13] = csrrd32(priv->mac_dev,
+			  tse_csroffs(if_out_discards));
+	buf[14] = csrrd32(priv->mac_dev,
+			  tse_csroffs(if_out_ucast_pkts));
+	buf[15] = csrrd32(priv->mac_dev,
+			  tse_csroffs(if_out_multicast_pkts));
+	buf[16] = csrrd32(priv->mac_dev,
+			  tse_csroffs(if_out_broadcast_pkts));
+	buf[17] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_drop_events));
 
 	/* Extended etherStatsOctets counter */
-	ext = (u64) ioread32(&mac->msb_ether_stats_octets) << 32;
-	ext |= ioread32(&mac->ether_stats_octets);
+	ext = (u64) csrrd32(priv->mac_dev,
+			    tse_csroffs(msb_ether_stats_octets)) << 32;
+	ext |= csrrd32(priv->mac_dev,
+		       tse_csroffs(ether_stats_octets));
 	buf[18] = ext;
 
-	buf[19] = ioread32(&mac->ether_stats_pkts);
-	buf[20] = ioread32(&mac->ether_stats_undersize_pkts);
-	buf[21] = ioread32(&mac->ether_stats_oversize_pkts);
-	buf[22] = ioread32(&mac->ether_stats_pkts_64_octets);
-	buf[23] = ioread32(&mac->ether_stats_pkts_65to127_octets);
-	buf[24] = ioread32(&mac->ether_stats_pkts_128to255_octets);
-	buf[25] = ioread32(&mac->ether_stats_pkts_256to511_octets);
-	buf[26] = ioread32(&mac->ether_stats_pkts_512to1023_octets);
-	buf[27] = ioread32(&mac->ether_stats_pkts_1024to1518_octets);
-	buf[28] = ioread32(&mac->ether_stats_pkts_1519tox_octets);
-	buf[29] = ioread32(&mac->ether_stats_jabbers);
-	buf[30] = ioread32(&mac->ether_stats_fragments);
+	buf[19] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_pkts));
+	buf[20] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_undersize_pkts));
+	buf[21] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_oversize_pkts));
+	buf[22] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_pkts_64_octets));
+	buf[23] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_pkts_65to127_octets));
+	buf[24] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_pkts_128to255_octets));
+	buf[25] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_pkts_256to511_octets));
+	buf[26] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_pkts_512to1023_octets));
+	buf[27] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_pkts_1024to1518_octets));
+	buf[28] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_pkts_1519tox_octets));
+	buf[29] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_jabbers));
+	buf[30] = csrrd32(priv->mac_dev,
+			  tse_csroffs(ether_stats_fragments));
 }
 
 static int tse_sset_count(struct net_device *dev, int sset)
@@ -178,7 +213,6 @@
 {
 	int i;
 	struct altera_tse_private *priv = netdev_priv(dev);
-	u32 *tse_mac_regs = (u32 *)priv->mac_dev;
 	u32 *buf = regbuf;
 
 	/* Set version to a known value, so ethtool knows
@@ -196,7 +230,7 @@
 	regs->version = 1;
 
 	for (i = 0; i < TSE_NUM_REGS; i++)
-		buf[i] = ioread32(&tse_mac_regs[i]);
+		buf[i] = csrrd32(priv->mac_dev, i * 4);
 }
 
 static int tse_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c
index e44a4ae..7330681 100644
--- a/drivers/net/ethernet/altera/altera_tse_main.c
+++ b/drivers/net/ethernet/altera/altera_tse_main.c
@@ -100,29 +100,30 @@
  */
 static int altera_tse_mdio_read(struct mii_bus *bus, int mii_id, int regnum)
 {
-	struct altera_tse_mac *mac = (struct altera_tse_mac *)bus->priv;
-	unsigned int *mdio_regs = (unsigned int *)&mac->mdio_phy0;
-	u32 data;
+	struct net_device *ndev = bus->priv;
+	struct altera_tse_private *priv = netdev_priv(ndev);
 
 	/* set MDIO address */
-	iowrite32((mii_id & 0x1f), &mac->mdio_phy0_addr);
+	csrwr32((mii_id & 0x1f), priv->mac_dev,
+		tse_csroffs(mdio_phy0_addr));
 
 	/* get the data */
-	data = ioread32(&mdio_regs[regnum]) & 0xffff;
-	return data;
+	return csrrd32(priv->mac_dev,
+		       tse_csroffs(mdio_phy0) + regnum * 4) & 0xffff;
 }
 
 static int altera_tse_mdio_write(struct mii_bus *bus, int mii_id, int regnum,
 				 u16 value)
 {
-	struct altera_tse_mac *mac = (struct altera_tse_mac *)bus->priv;
-	unsigned int *mdio_regs = (unsigned int *)&mac->mdio_phy0;
+	struct net_device *ndev = bus->priv;
+	struct altera_tse_private *priv = netdev_priv(ndev);
 
 	/* set MDIO address */
-	iowrite32((mii_id & 0x1f), &mac->mdio_phy0_addr);
+	csrwr32((mii_id & 0x1f), priv->mac_dev,
+		tse_csroffs(mdio_phy0_addr));
 
 	/* write the data */
-	iowrite32((u32) value, &mdio_regs[regnum]);
+	csrwr32(value, priv->mac_dev, tse_csroffs(mdio_phy0) + regnum * 4);
 	return 0;
 }
 
@@ -168,7 +169,7 @@
 	for (i = 0; i < PHY_MAX_ADDR; i++)
 		mdio->irq[i] = PHY_POLL;
 
-	mdio->priv = priv->mac_dev;
+	mdio->priv = dev;
 	mdio->parent = priv->device;
 
 	ret = of_mdiobus_register(mdio, mdio_node);
@@ -563,7 +564,6 @@
 	unsigned int nopaged_len = skb_headlen(skb);
 	enum netdev_tx ret = NETDEV_TX_OK;
 	dma_addr_t dma_addr;
-	int txcomplete = 0;
 
 	spin_lock_bh(&priv->tx_lock);
 
@@ -599,7 +599,7 @@
 	dma_sync_single_for_device(priv->device, buffer->dma_addr,
 				   buffer->len, DMA_TO_DEVICE);
 
-	txcomplete = priv->dmaops->tx_buffer(priv, buffer);
+	priv->dmaops->tx_buffer(priv, buffer);
 
 	skb_tx_timestamp(skb);
 
@@ -698,7 +698,6 @@
 	struct altera_tse_private *priv = netdev_priv(dev);
 	struct phy_device *phydev = NULL;
 	char phy_id_fmt[MII_BUS_ID_SIZE + 3];
-	int ret;
 
 	if (priv->phy_addr != POLL_PHY) {
 		snprintf(phy_id_fmt, MII_BUS_ID_SIZE + 3, PHY_ID_FMT,
@@ -712,6 +711,7 @@
 			netdev_err(dev, "Could not attach to PHY\n");
 
 	} else {
+		int ret;
 		phydev = phy_find_first(priv->mdio);
 		if (phydev == NULL) {
 			netdev_err(dev, "No PHY found\n");
@@ -791,7 +791,6 @@
 
 static void tse_update_mac_addr(struct altera_tse_private *priv, u8 *addr)
 {
-	struct altera_tse_mac *mac = priv->mac_dev;
 	u32 msb;
 	u32 lsb;
 
@@ -799,8 +798,8 @@
 	lsb = ((addr[5] << 8) | addr[4]) & 0xffff;
 
 	/* Set primary MAC address */
-	iowrite32(msb, &mac->mac_addr_0);
-	iowrite32(lsb, &mac->mac_addr_1);
+	csrwr32(msb, priv->mac_dev, tse_csroffs(mac_addr_0));
+	csrwr32(lsb, priv->mac_dev, tse_csroffs(mac_addr_1));
 }
 
 /* MAC software reset.
@@ -811,26 +810,26 @@
  */
 static int reset_mac(struct altera_tse_private *priv)
 {
-	void __iomem *cmd_cfg_reg = &priv->mac_dev->command_config;
 	int counter;
 	u32 dat;
 
-	dat = ioread32(cmd_cfg_reg);
+	dat = csrrd32(priv->mac_dev, tse_csroffs(command_config));
 	dat &= ~(MAC_CMDCFG_TX_ENA | MAC_CMDCFG_RX_ENA);
 	dat |= MAC_CMDCFG_SW_RESET | MAC_CMDCFG_CNT_RESET;
-	iowrite32(dat, cmd_cfg_reg);
+	csrwr32(dat, priv->mac_dev, tse_csroffs(command_config));
 
 	counter = 0;
 	while (counter++ < ALTERA_TSE_SW_RESET_WATCHDOG_CNTR) {
-		if (tse_bit_is_clear(cmd_cfg_reg, MAC_CMDCFG_SW_RESET))
+		if (tse_bit_is_clear(priv->mac_dev, tse_csroffs(command_config),
+				     MAC_CMDCFG_SW_RESET))
 			break;
 		udelay(1);
 	}
 
 	if (counter >= ALTERA_TSE_SW_RESET_WATCHDOG_CNTR) {
-		dat = ioread32(cmd_cfg_reg);
+		dat = csrrd32(priv->mac_dev, tse_csroffs(command_config));
 		dat &= ~MAC_CMDCFG_SW_RESET;
-		iowrite32(dat, cmd_cfg_reg);
+		csrwr32(dat, priv->mac_dev, tse_csroffs(command_config));
 		return -1;
 	}
 	return 0;
@@ -840,41 +839,57 @@
 */
 static int init_mac(struct altera_tse_private *priv)
 {
-	struct altera_tse_mac *mac = priv->mac_dev;
 	unsigned int cmd = 0;
 	u32 frm_length;
 
 	/* Setup Rx FIFO */
-	iowrite32(priv->rx_fifo_depth - ALTERA_TSE_RX_SECTION_EMPTY,
-		  &mac->rx_section_empty);
-	iowrite32(ALTERA_TSE_RX_SECTION_FULL, &mac->rx_section_full);
-	iowrite32(ALTERA_TSE_RX_ALMOST_EMPTY, &mac->rx_almost_empty);
-	iowrite32(ALTERA_TSE_RX_ALMOST_FULL, &mac->rx_almost_full);
+	csrwr32(priv->rx_fifo_depth - ALTERA_TSE_RX_SECTION_EMPTY,
+		priv->mac_dev, tse_csroffs(rx_section_empty));
+
+	csrwr32(ALTERA_TSE_RX_SECTION_FULL, priv->mac_dev,
+		tse_csroffs(rx_section_full));
+
+	csrwr32(ALTERA_TSE_RX_ALMOST_EMPTY, priv->mac_dev,
+		tse_csroffs(rx_almost_empty));
+
+	csrwr32(ALTERA_TSE_RX_ALMOST_FULL, priv->mac_dev,
+		tse_csroffs(rx_almost_full));
 
 	/* Setup Tx FIFO */
-	iowrite32(priv->tx_fifo_depth - ALTERA_TSE_TX_SECTION_EMPTY,
-		  &mac->tx_section_empty);
-	iowrite32(ALTERA_TSE_TX_SECTION_FULL, &mac->tx_section_full);
-	iowrite32(ALTERA_TSE_TX_ALMOST_EMPTY, &mac->tx_almost_empty);
-	iowrite32(ALTERA_TSE_TX_ALMOST_FULL, &mac->tx_almost_full);
+	csrwr32(priv->tx_fifo_depth - ALTERA_TSE_TX_SECTION_EMPTY,
+		priv->mac_dev, tse_csroffs(tx_section_empty));
+
+	csrwr32(ALTERA_TSE_TX_SECTION_FULL, priv->mac_dev,
+		tse_csroffs(tx_section_full));
+
+	csrwr32(ALTERA_TSE_TX_ALMOST_EMPTY, priv->mac_dev,
+		tse_csroffs(tx_almost_empty));
+
+	csrwr32(ALTERA_TSE_TX_ALMOST_FULL, priv->mac_dev,
+		tse_csroffs(tx_almost_full));
 
 	/* MAC Address Configuration */
 	tse_update_mac_addr(priv, priv->dev->dev_addr);
 
 	/* MAC Function Configuration */
 	frm_length = ETH_HLEN + priv->dev->mtu + ETH_FCS_LEN;
-	iowrite32(frm_length, &mac->frm_length);
-	iowrite32(ALTERA_TSE_TX_IPG_LENGTH, &mac->tx_ipg_length);
+	csrwr32(frm_length, priv->mac_dev, tse_csroffs(frm_length));
+
+	csrwr32(ALTERA_TSE_TX_IPG_LENGTH, priv->mac_dev,
+		tse_csroffs(tx_ipg_length));
 
 	/* Disable RX/TX shift 16 for alignment of all received frames on 16-bit
 	 * start address
 	 */
-	tse_set_bit(&mac->rx_cmd_stat, ALTERA_TSE_RX_CMD_STAT_RX_SHIFT16);
-	tse_clear_bit(&mac->tx_cmd_stat, ALTERA_TSE_TX_CMD_STAT_TX_SHIFT16 |
-					 ALTERA_TSE_TX_CMD_STAT_OMIT_CRC);
+	tse_set_bit(priv->mac_dev, tse_csroffs(rx_cmd_stat),
+		    ALTERA_TSE_RX_CMD_STAT_RX_SHIFT16);
+
+	tse_clear_bit(priv->mac_dev, tse_csroffs(tx_cmd_stat),
+		      ALTERA_TSE_TX_CMD_STAT_TX_SHIFT16 |
+		      ALTERA_TSE_TX_CMD_STAT_OMIT_CRC);
 
 	/* Set the MAC options */
-	cmd = ioread32(&mac->command_config);
+	cmd = csrrd32(priv->mac_dev, tse_csroffs(command_config));
 	cmd &= ~MAC_CMDCFG_PAD_EN;	/* No padding Removal on Receive */
 	cmd &= ~MAC_CMDCFG_CRC_FWD;	/* CRC Removal */
 	cmd |= MAC_CMDCFG_RX_ERR_DISC;	/* Automatically discard frames
@@ -889,9 +904,10 @@
 	cmd &= ~MAC_CMDCFG_ETH_SPEED;
 	cmd &= ~MAC_CMDCFG_ENA_10;
 
-	iowrite32(cmd, &mac->command_config);
+	csrwr32(cmd, priv->mac_dev, tse_csroffs(command_config));
 
-	iowrite32(ALTERA_TSE_PAUSE_QUANTA, &mac->pause_quanta);
+	csrwr32(ALTERA_TSE_PAUSE_QUANTA, priv->mac_dev,
+		tse_csroffs(pause_quanta));
 
 	if (netif_msg_hw(priv))
 		dev_dbg(priv->device,
@@ -904,15 +920,14 @@
  */
 static void tse_set_mac(struct altera_tse_private *priv, bool enable)
 {
-	struct altera_tse_mac *mac = priv->mac_dev;
-	u32 value = ioread32(&mac->command_config);
+	u32 value = csrrd32(priv->mac_dev, tse_csroffs(command_config));
 
 	if (enable)
 		value |= MAC_CMDCFG_TX_ENA | MAC_CMDCFG_RX_ENA;
 	else
 		value &= ~(MAC_CMDCFG_TX_ENA | MAC_CMDCFG_RX_ENA);
 
-	iowrite32(value, &mac->command_config);
+	csrwr32(value, priv->mac_dev, tse_csroffs(command_config));
 }
 
 /* Change the MTU
@@ -942,13 +957,12 @@
 static void altera_tse_set_mcfilter(struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
-	struct altera_tse_mac *mac = priv->mac_dev;
 	int i;
 	struct netdev_hw_addr *ha;
 
 	/* clear the hash filter */
 	for (i = 0; i < 64; i++)
-		iowrite32(0, &(mac->hash_table[i]));
+		csrwr32(0, priv->mac_dev, tse_csroffs(hash_table) + i * 4);
 
 	netdev_for_each_mc_addr(ha, dev) {
 		unsigned int hash = 0;
@@ -964,7 +978,7 @@
 
 			hash = (hash << 1) | xor_bit;
 		}
-		iowrite32(1, &(mac->hash_table[hash]));
+		csrwr32(1, priv->mac_dev, tse_csroffs(hash_table) + hash * 4);
 	}
 }
 
@@ -972,12 +986,11 @@
 static void altera_tse_set_mcfilterall(struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
-	struct altera_tse_mac *mac = priv->mac_dev;
 	int i;
 
 	/* set the hash filter */
 	for (i = 0; i < 64; i++)
-		iowrite32(1, &(mac->hash_table[i]));
+		csrwr32(1, priv->mac_dev, tse_csroffs(hash_table) + i * 4);
 }
 
 /* Set or clear the multicast filter for this adaptor
@@ -985,12 +998,12 @@
 static void tse_set_rx_mode_hashfilter(struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
-	struct altera_tse_mac *mac = priv->mac_dev;
 
 	spin_lock(&priv->mac_cfg_lock);
 
 	if (dev->flags & IFF_PROMISC)
-		tse_set_bit(&mac->command_config, MAC_CMDCFG_PROMIS_EN);
+		tse_set_bit(priv->mac_dev, tse_csroffs(command_config),
+			    MAC_CMDCFG_PROMIS_EN);
 
 	if (dev->flags & IFF_ALLMULTI)
 		altera_tse_set_mcfilterall(dev);
@@ -1005,15 +1018,16 @@
 static void tse_set_rx_mode(struct net_device *dev)
 {
 	struct altera_tse_private *priv = netdev_priv(dev);
-	struct altera_tse_mac *mac = priv->mac_dev;
 
 	spin_lock(&priv->mac_cfg_lock);
 
 	if ((dev->flags & IFF_PROMISC) || (dev->flags & IFF_ALLMULTI) ||
 	    !netdev_mc_empty(dev) || !netdev_uc_empty(dev))
-		tse_set_bit(&mac->command_config, MAC_CMDCFG_PROMIS_EN);
+		tse_set_bit(priv->mac_dev, tse_csroffs(command_config),
+			    MAC_CMDCFG_PROMIS_EN);
 	else
-		tse_clear_bit(&mac->command_config, MAC_CMDCFG_PROMIS_EN);
+		tse_clear_bit(priv->mac_dev, tse_csroffs(command_config),
+			      MAC_CMDCFG_PROMIS_EN);
 
 	spin_unlock(&priv->mac_cfg_lock);
 }
@@ -1362,6 +1376,11 @@
 		of_property_read_bool(pdev->dev.of_node,
 				      "altr,has-hash-multicast-filter");
 
+	/* Set hash filter to not set for now until the
+	 * multicast filter receive issue is debugged
+	 */
+	priv->hash_filter = 0;
+
 	/* get supplemental address settings for this instance */
 	priv->added_unicast =
 		of_property_read_bool(pdev->dev.of_node,
@@ -1493,7 +1512,7 @@
 	return 0;
 }
 
-struct altera_dmaops altera_dtype_sgdma = {
+static const struct altera_dmaops altera_dtype_sgdma = {
 	.altera_dtype = ALTERA_DTYPE_SGDMA,
 	.dmamask = 32,
 	.reset_dma = sgdma_reset,
@@ -1512,7 +1531,7 @@
 	.start_rxdma = sgdma_start_rxdma,
 };
 
-struct altera_dmaops altera_dtype_msgdma = {
+static const struct altera_dmaops altera_dtype_msgdma = {
 	.altera_dtype = ALTERA_DTYPE_MSGDMA,
 	.dmamask = 64,
 	.reset_dma = msgdma_reset,
diff --git a/drivers/net/ethernet/altera/altera_utils.c b/drivers/net/ethernet/altera/altera_utils.c
index 70fa13f..d7eeb17 100644
--- a/drivers/net/ethernet/altera/altera_utils.c
+++ b/drivers/net/ethernet/altera/altera_utils.c
@@ -17,28 +17,28 @@
 #include "altera_tse.h"
 #include "altera_utils.h"
 
-void tse_set_bit(void __iomem *ioaddr, u32 bit_mask)
+void tse_set_bit(void __iomem *ioaddr, size_t offs, u32 bit_mask)
 {
-	u32 value = ioread32(ioaddr);
+	u32 value = csrrd32(ioaddr, offs);
 	value |= bit_mask;
-	iowrite32(value, ioaddr);
+	csrwr32(value, ioaddr, offs);
 }
 
-void tse_clear_bit(void __iomem *ioaddr, u32 bit_mask)
+void tse_clear_bit(void __iomem *ioaddr, size_t offs, u32 bit_mask)
 {
-	u32 value = ioread32(ioaddr);
+	u32 value = csrrd32(ioaddr, offs);
 	value &= ~bit_mask;
-	iowrite32(value, ioaddr);
+	csrwr32(value, ioaddr, offs);
 }
 
-int tse_bit_is_set(void __iomem *ioaddr, u32 bit_mask)
+int tse_bit_is_set(void __iomem *ioaddr, size_t offs, u32 bit_mask)
 {
-	u32 value = ioread32(ioaddr);
+	u32 value = csrrd32(ioaddr, offs);
 	return (value & bit_mask) ? 1 : 0;
 }
 
-int tse_bit_is_clear(void __iomem *ioaddr, u32 bit_mask)
+int tse_bit_is_clear(void __iomem *ioaddr, size_t offs, u32 bit_mask)
 {
-	u32 value = ioread32(ioaddr);
+	u32 value = csrrd32(ioaddr, offs);
 	return (value & bit_mask) ? 0 : 1;
 }
diff --git a/drivers/net/ethernet/altera/altera_utils.h b/drivers/net/ethernet/altera/altera_utils.h
index ce1db36..baf100c 100644
--- a/drivers/net/ethernet/altera/altera_utils.h
+++ b/drivers/net/ethernet/altera/altera_utils.h
@@ -19,9 +19,9 @@
 #ifndef __ALTERA_UTILS_H__
 #define __ALTERA_UTILS_H__
 
-void tse_set_bit(void __iomem *ioaddr, u32 bit_mask);
-void tse_clear_bit(void __iomem *ioaddr, u32 bit_mask);
-int tse_bit_is_set(void __iomem *ioaddr, u32 bit_mask);
-int tse_bit_is_clear(void __iomem *ioaddr, u32 bit_mask);
+void tse_set_bit(void __iomem *ioaddr, size_t offs, u32 bit_mask);
+void tse_clear_bit(void __iomem *ioaddr, size_t offs, u32 bit_mask);
+int tse_bit_is_set(void __iomem *ioaddr, size_t offs, u32 bit_mask);
+int tse_bit_is_clear(void __iomem *ioaddr, size_t offs, u32 bit_mask);
 
 #endif /* __ALTERA_UTILS_H__*/
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index b260913..3b0d431 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -10051,8 +10051,8 @@
 #define BCM_5710_UNDI_FW_MF_MAJOR	(0x07)
 #define BCM_5710_UNDI_FW_MF_MINOR	(0x08)
 #define BCM_5710_UNDI_FW_MF_VERS	(0x05)
-#define BNX2X_PREV_UNDI_MF_PORT(p)	(0x1a150c + ((p) << 4))
-#define BNX2X_PREV_UNDI_MF_FUNC(f)	(0x1a184c + ((f) << 4))
+#define BNX2X_PREV_UNDI_MF_PORT(p) (BAR_TSTRORM_INTMEM + 0x150c + ((p) << 4))
+#define BNX2X_PREV_UNDI_MF_FUNC(f) (BAR_TSTRORM_INTMEM + 0x184c + ((f) << 4))
 static bool bnx2x_prev_unload_undi_fw_supports_mf(struct bnx2x *bp)
 {
 	u8 major, minor, version;
@@ -10352,6 +10352,7 @@
 	/* Reset should be performed after BRB is emptied */
 	if (reset_reg & MISC_REGISTERS_RESET_REG_1_RST_BRB1) {
 		u32 timer_count = 1000;
+		bool need_write = true;
 
 		/* Close the MAC Rx to prevent BRB from filling up */
 		bnx2x_prev_unload_close_mac(bp, &mac_vals);
@@ -10398,7 +10399,10 @@
 			 * cleaning methods - might be redundant but harmless.
 			 */
 			if (bnx2x_prev_unload_undi_fw_supports_mf(bp)) {
-				bnx2x_prev_unload_undi_mf(bp);
+				if (need_write) {
+					bnx2x_prev_unload_undi_mf(bp);
+					need_write = false;
+				}
 			} else if (prev_undi) {
 				/* If UNDI resides in memory,
 				 * manually increment it
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
index 81cc2d9..b8078d5 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
@@ -2695,7 +2695,7 @@
 		bnx2x_unlock_vf_pf_channel(bp, vf, CHANNEL_TLV_PF_SET_MAC);
 	}
 
-	return 0;
+	return rc;
 }
 
 int bnx2x_set_vf_vlan(struct net_device *dev, int vfidx, u16 vlan, u8 qos)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
index 0c067e8..784c715 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
@@ -747,7 +747,7 @@
 out:
 	bnx2x_vfpf_finalize(bp, &req->first_tlv);
 
-	return 0;
+	return rc;
 }
 
 /* request pf to config rss table for vf queues*/
diff --git a/drivers/net/ethernet/ec_bhf.c b/drivers/net/ethernet/ec_bhf.c
new file mode 100644
index 0000000..4884205
--- /dev/null
+++ b/drivers/net/ethernet/ec_bhf.c
@@ -0,0 +1,706 @@
+ /*
+ * drivers/net/ethernet/beckhoff/ec_bhf.c
+ *
+ * Copyright (C) 2014 Darek Marcinkiewicz <reksio@newterm.pl>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+/* This is a driver for EtherCAT master module present on CCAT FPGA.
+ * Those can be found on Bechhoff CX50xx industrial PCs.
+ */
+
+#if 0
+#define DEBUG
+#endif
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/pci.h>
+#include <linux/init.h>
+
+#include <linux/netdevice.h>
+#include <linux/etherdevice.h>
+#include <linux/ip.h>
+#include <linux/skbuff.h>
+#include <linux/hrtimer.h>
+#include <linux/interrupt.h>
+#include <linux/stat.h>
+
+#define TIMER_INTERVAL_NSEC	20000
+
+#define INFO_BLOCK_SIZE		0x10
+#define INFO_BLOCK_TYPE		0x0
+#define INFO_BLOCK_REV		0x2
+#define INFO_BLOCK_BLK_CNT	0x4
+#define INFO_BLOCK_TX_CHAN	0x4
+#define INFO_BLOCK_RX_CHAN	0x5
+#define INFO_BLOCK_OFFSET	0x8
+
+#define EC_MII_OFFSET		0x4
+#define EC_FIFO_OFFSET		0x8
+#define EC_MAC_OFFSET		0xc
+
+#define MAC_FRAME_ERR_CNT	0x0
+#define MAC_RX_ERR_CNT		0x1
+#define MAC_CRC_ERR_CNT		0x2
+#define MAC_LNK_LST_ERR_CNT	0x3
+#define MAC_TX_FRAME_CNT	0x10
+#define MAC_RX_FRAME_CNT	0x14
+#define MAC_TX_FIFO_LVL		0x20
+#define MAC_DROPPED_FRMS	0x28
+#define MAC_CONNECTED_CCAT_FLAG	0x78
+
+#define MII_MAC_ADDR		0x8
+#define MII_MAC_FILT_FLAG	0xe
+#define MII_LINK_STATUS		0xf
+
+#define FIFO_TX_REG		0x0
+#define FIFO_TX_RESET		0x8
+#define FIFO_RX_REG		0x10
+#define FIFO_RX_ADDR_VALID	(1u << 31)
+#define FIFO_RX_RESET		0x18
+
+#define DMA_CHAN_OFFSET		0x1000
+#define DMA_CHAN_SIZE		0x8
+
+#define DMA_WINDOW_SIZE_MASK	0xfffffffc
+
+static struct pci_device_id ids[] = {
+	{ PCI_DEVICE(0x15ec, 0x5000), },
+	{ 0, }
+};
+MODULE_DEVICE_TABLE(pci, ids);
+
+struct rx_header {
+#define RXHDR_NEXT_ADDR_MASK	0xffffffu
+#define RXHDR_NEXT_VALID	(1u << 31)
+	__le32 next;
+#define RXHDR_NEXT_RECV_FLAG	0x1
+	__le32 recv;
+#define RXHDR_LEN_MASK		0xfffu
+	__le16 len;
+	__le16 port;
+	__le32 reserved;
+	u8 timestamp[8];
+} __packed;
+
+#define PKT_PAYLOAD_SIZE	0x7e8
+struct rx_desc {
+	struct rx_header header;
+	u8 data[PKT_PAYLOAD_SIZE];
+} __packed;
+
+struct tx_header {
+	__le16 len;
+#define TX_HDR_PORT_0		0x1
+#define TX_HDR_PORT_1		0x2
+	u8 port;
+	u8 ts_enable;
+#define TX_HDR_SENT		0x1
+	__le32 sent;
+	u8 timestamp[8];
+} __packed;
+
+struct tx_desc {
+	struct tx_header header;
+	u8 data[PKT_PAYLOAD_SIZE];
+} __packed;
+
+#define FIFO_SIZE		64
+
+static long polling_frequency = TIMER_INTERVAL_NSEC;
+
+struct bhf_dma {
+	u8 *buf;
+	size_t len;
+	dma_addr_t buf_phys;
+
+	u8 *alloc;
+	size_t alloc_len;
+	dma_addr_t alloc_phys;
+};
+
+struct ec_bhf_priv {
+	struct net_device *net_dev;
+
+	struct pci_dev *dev;
+
+	void * __iomem io;
+	void * __iomem dma_io;
+
+	struct hrtimer hrtimer;
+
+	int tx_dma_chan;
+	int rx_dma_chan;
+	void * __iomem ec_io;
+	void * __iomem fifo_io;
+	void * __iomem mii_io;
+	void * __iomem mac_io;
+
+	struct bhf_dma rx_buf;
+	struct rx_desc *rx_descs;
+	int rx_dnext;
+	int rx_dcount;
+
+	struct bhf_dma tx_buf;
+	struct tx_desc *tx_descs;
+	int tx_dcount;
+	int tx_dnext;
+
+	u64 stat_rx_bytes;
+	u64 stat_tx_bytes;
+};
+
+#define PRIV_TO_DEV(priv) (&(priv)->dev->dev)
+
+#define ETHERCAT_MASTER_ID	0x14
+
+static void ec_bhf_print_status(struct ec_bhf_priv *priv)
+{
+	struct device *dev = PRIV_TO_DEV(priv);
+
+	dev_dbg(dev, "Frame error counter: %d\n",
+		ioread8(priv->mac_io + MAC_FRAME_ERR_CNT));
+	dev_dbg(dev, "RX error counter: %d\n",
+		ioread8(priv->mac_io + MAC_RX_ERR_CNT));
+	dev_dbg(dev, "CRC error counter: %d\n",
+		ioread8(priv->mac_io + MAC_CRC_ERR_CNT));
+	dev_dbg(dev, "TX frame counter: %d\n",
+		ioread32(priv->mac_io + MAC_TX_FRAME_CNT));
+	dev_dbg(dev, "RX frame counter: %d\n",
+		ioread32(priv->mac_io + MAC_RX_FRAME_CNT));
+	dev_dbg(dev, "TX fifo level: %d\n",
+		ioread8(priv->mac_io + MAC_TX_FIFO_LVL));
+	dev_dbg(dev, "Dropped frames: %d\n",
+		ioread8(priv->mac_io + MAC_DROPPED_FRMS));
+	dev_dbg(dev, "Connected with CCAT slot: %d\n",
+		ioread8(priv->mac_io + MAC_CONNECTED_CCAT_FLAG));
+	dev_dbg(dev, "Link status: %d\n",
+		ioread8(priv->mii_io + MII_LINK_STATUS));
+}
+
+static void ec_bhf_reset(struct ec_bhf_priv *priv)
+{
+	iowrite8(0, priv->mac_io + MAC_FRAME_ERR_CNT);
+	iowrite8(0, priv->mac_io + MAC_RX_ERR_CNT);
+	iowrite8(0, priv->mac_io + MAC_CRC_ERR_CNT);
+	iowrite8(0, priv->mac_io + MAC_LNK_LST_ERR_CNT);
+	iowrite32(0, priv->mac_io + MAC_TX_FRAME_CNT);
+	iowrite32(0, priv->mac_io + MAC_RX_FRAME_CNT);
+	iowrite8(0, priv->mac_io + MAC_DROPPED_FRMS);
+
+	iowrite8(0, priv->fifo_io + FIFO_TX_RESET);
+	iowrite8(0, priv->fifo_io + FIFO_RX_RESET);
+
+	iowrite8(0, priv->mac_io + MAC_TX_FIFO_LVL);
+}
+
+static void ec_bhf_send_packet(struct ec_bhf_priv *priv, struct tx_desc *desc)
+{
+	u32 len = le16_to_cpu(desc->header.len) + sizeof(desc->header);
+	u32 addr = (u8 *)desc - priv->tx_buf.buf;
+
+	iowrite32((ALIGN(len, 8) << 24) | addr, priv->fifo_io + FIFO_TX_REG);
+
+	dev_dbg(PRIV_TO_DEV(priv), "Done sending packet\n");
+}
+
+static int ec_bhf_desc_sent(struct tx_desc *desc)
+{
+	return le32_to_cpu(desc->header.sent) & TX_HDR_SENT;
+}
+
+static void ec_bhf_process_tx(struct ec_bhf_priv *priv)
+{
+	if (unlikely(netif_queue_stopped(priv->net_dev))) {
+		/* Make sure that we perceive changes to tx_dnext. */
+		smp_rmb();
+
+		if (ec_bhf_desc_sent(&priv->tx_descs[priv->tx_dnext]))
+			netif_wake_queue(priv->net_dev);
+	}
+}
+
+static int ec_bhf_pkt_received(struct rx_desc *desc)
+{
+	return le32_to_cpu(desc->header.recv) & RXHDR_NEXT_RECV_FLAG;
+}
+
+static void ec_bhf_add_rx_desc(struct ec_bhf_priv *priv, struct rx_desc *desc)
+{
+	iowrite32(FIFO_RX_ADDR_VALID | ((u8 *)(desc) - priv->rx_buf.buf),
+		  priv->fifo_io + FIFO_RX_REG);
+}
+
+static void ec_bhf_process_rx(struct ec_bhf_priv *priv)
+{
+	struct rx_desc *desc = &priv->rx_descs[priv->rx_dnext];
+	struct device *dev = PRIV_TO_DEV(priv);
+
+	while (ec_bhf_pkt_received(desc)) {
+		int pkt_size = (le16_to_cpu(desc->header.len) &
+			       RXHDR_LEN_MASK) - sizeof(struct rx_header) - 4;
+		u8 *data = desc->data;
+		struct sk_buff *skb;
+
+		skb = netdev_alloc_skb_ip_align(priv->net_dev, pkt_size);
+		dev_dbg(dev, "Received packet, size: %d\n", pkt_size);
+
+		if (skb) {
+			memcpy(skb_put(skb, pkt_size), data, pkt_size);
+			skb->protocol = eth_type_trans(skb, priv->net_dev);
+			dev_dbg(dev, "Protocol type: %x\n", skb->protocol);
+
+			priv->stat_rx_bytes += pkt_size;
+
+			netif_rx(skb);
+		} else {
+			dev_err_ratelimited(dev,
+				"Couldn't allocate a skb_buff for a packet of size %u\n",
+				pkt_size);
+		}
+
+		desc->header.recv = 0;
+
+		ec_bhf_add_rx_desc(priv, desc);
+
+		priv->rx_dnext = (priv->rx_dnext + 1) % priv->rx_dcount;
+		desc = &priv->rx_descs[priv->rx_dnext];
+	}
+
+}
+
+static enum hrtimer_restart ec_bhf_timer_fun(struct hrtimer *timer)
+{
+	struct ec_bhf_priv *priv = container_of(timer, struct ec_bhf_priv,
+						hrtimer);
+	ec_bhf_process_rx(priv);
+	ec_bhf_process_tx(priv);
+
+	if (!netif_running(priv->net_dev))
+		return HRTIMER_NORESTART;
+
+	hrtimer_forward_now(timer, ktime_set(0, polling_frequency));
+	return HRTIMER_RESTART;
+}
+
+static int ec_bhf_setup_offsets(struct ec_bhf_priv *priv)
+{
+	struct device *dev = PRIV_TO_DEV(priv);
+	unsigned block_count, i;
+	void * __iomem ec_info;
+
+	dev_dbg(dev, "Info block:\n");
+	dev_dbg(dev, "Type of function: %x\n", (unsigned)ioread16(priv->io));
+	dev_dbg(dev, "Revision of function: %x\n",
+		(unsigned)ioread16(priv->io + INFO_BLOCK_REV));
+
+	block_count = ioread8(priv->io + INFO_BLOCK_BLK_CNT);
+	dev_dbg(dev, "Number of function blocks: %x\n", block_count);
+
+	for (i = 0; i < block_count; i++) {
+		u16 type = ioread16(priv->io + i * INFO_BLOCK_SIZE +
+				    INFO_BLOCK_TYPE);
+		if (type == ETHERCAT_MASTER_ID)
+			break;
+	}
+	if (i == block_count) {
+		dev_err(dev, "EtherCAT master with DMA block not found\n");
+		return -ENODEV;
+	}
+	dev_dbg(dev, "EtherCAT master with DMA block found at pos: %d\n", i);
+
+	ec_info = priv->io + i * INFO_BLOCK_SIZE;
+	dev_dbg(dev, "EtherCAT master revision: %d\n",
+		ioread16(ec_info + INFO_BLOCK_REV));
+
+	priv->tx_dma_chan = ioread8(ec_info + INFO_BLOCK_TX_CHAN);
+	dev_dbg(dev, "EtherCAT master tx dma channel: %d\n",
+		priv->tx_dma_chan);
+
+	priv->rx_dma_chan = ioread8(ec_info + INFO_BLOCK_RX_CHAN);
+	dev_dbg(dev, "EtherCAT master rx dma channel: %d\n",
+		 priv->rx_dma_chan);
+
+	priv->ec_io = priv->io + ioread32(ec_info + INFO_BLOCK_OFFSET);
+	priv->mii_io = priv->ec_io + ioread32(priv->ec_io + EC_MII_OFFSET);
+	priv->fifo_io = priv->ec_io + ioread32(priv->ec_io + EC_FIFO_OFFSET);
+	priv->mac_io = priv->ec_io + ioread32(priv->ec_io + EC_MAC_OFFSET);
+
+	dev_dbg(dev,
+		"EtherCAT block addres: %p, fifo address: %p, mii address: %p, mac address: %p\n",
+		priv->ec_io, priv->fifo_io, priv->mii_io, priv->mac_io);
+
+	return 0;
+}
+
+static netdev_tx_t ec_bhf_start_xmit(struct sk_buff *skb,
+				     struct net_device *net_dev)
+{
+	struct ec_bhf_priv *priv = netdev_priv(net_dev);
+	struct tx_desc *desc;
+	unsigned len;
+
+	dev_dbg(PRIV_TO_DEV(priv), "Starting xmit\n");
+
+	desc = &priv->tx_descs[priv->tx_dnext];
+
+	skb_copy_and_csum_dev(skb, desc->data);
+	len = skb->len;
+
+	memset(&desc->header, 0, sizeof(desc->header));
+	desc->header.len = cpu_to_le16(len);
+	desc->header.port = TX_HDR_PORT_0;
+
+	ec_bhf_send_packet(priv, desc);
+
+	priv->tx_dnext = (priv->tx_dnext + 1) % priv->tx_dcount;
+
+	if (!ec_bhf_desc_sent(&priv->tx_descs[priv->tx_dnext])) {
+		/* Make sure that update updates to tx_dnext are perceived
+		 * by timer routine.
+		 */
+		smp_wmb();
+
+		netif_stop_queue(net_dev);
+
+		dev_dbg(PRIV_TO_DEV(priv), "Stopping netif queue\n");
+		ec_bhf_print_status(priv);
+	}
+
+	priv->stat_tx_bytes += len;
+
+	dev_kfree_skb(skb);
+
+	return NETDEV_TX_OK;
+}
+
+static int ec_bhf_alloc_dma_mem(struct ec_bhf_priv *priv,
+				struct bhf_dma *buf,
+				int channel,
+				int size)
+{
+	int offset = channel * DMA_CHAN_SIZE + DMA_CHAN_OFFSET;
+	struct device *dev = PRIV_TO_DEV(priv);
+	u32 mask;
+
+	iowrite32(0xffffffff, priv->dma_io + offset);
+
+	mask = ioread32(priv->dma_io + offset);
+	mask &= DMA_WINDOW_SIZE_MASK;
+	dev_dbg(dev, "Read mask %x for channel %d\n", mask, channel);
+
+	/* We want to allocate a chunk of memory that is:
+	 * - aligned to the mask we just read
+	 * - is of size 2^mask bytes (at most)
+	 * In order to ensure that we will allocate buffer of
+	 * 2 * 2^mask bytes.
+	 */
+	buf->len = min_t(int, ~mask + 1, size);
+	buf->alloc_len = 2 * buf->len;
+
+	dev_dbg(dev, "Allocating %d bytes for channel %d",
+		(int)buf->alloc_len, channel);
+	buf->alloc = dma_alloc_coherent(dev, buf->alloc_len, &buf->alloc_phys,
+					GFP_KERNEL);
+	if (buf->alloc == NULL) {
+		dev_info(dev, "Failed to allocate buffer\n");
+		return -ENOMEM;
+	}
+
+	buf->buf_phys = (buf->alloc_phys + buf->len) & mask;
+	buf->buf = buf->alloc + (buf->buf_phys - buf->alloc_phys);
+
+	iowrite32(0, priv->dma_io + offset + 4);
+	iowrite32(buf->buf_phys, priv->dma_io + offset);
+	dev_dbg(dev, "Buffer: %x and read from dev: %x",
+		(unsigned)buf->buf_phys, ioread32(priv->dma_io + offset));
+
+	return 0;
+}
+
+static void ec_bhf_setup_tx_descs(struct ec_bhf_priv *priv)
+{
+	int i = 0;
+
+	priv->tx_dcount = priv->tx_buf.len / sizeof(struct tx_desc);
+	priv->tx_descs = (struct tx_desc *) priv->tx_buf.buf;
+	priv->tx_dnext = 0;
+
+	for (i = 0; i < priv->tx_dcount; i++)
+		priv->tx_descs[i].header.sent = cpu_to_le32(TX_HDR_SENT);
+}
+
+static void ec_bhf_setup_rx_descs(struct ec_bhf_priv *priv)
+{
+	int i;
+
+	priv->rx_dcount = priv->rx_buf.len / sizeof(struct rx_desc);
+	priv->rx_descs = (struct rx_desc *) priv->rx_buf.buf;
+	priv->rx_dnext = 0;
+
+	for (i = 0; i < priv->rx_dcount; i++) {
+		struct rx_desc *desc = &priv->rx_descs[i];
+		u32 next;
+
+		if (i != priv->rx_dcount - 1)
+			next = (u8 *)(desc + 1) - priv->rx_buf.buf;
+		else
+			next = 0;
+		next |= RXHDR_NEXT_VALID;
+		desc->header.next = cpu_to_le32(next);
+		desc->header.recv = 0;
+		ec_bhf_add_rx_desc(priv, desc);
+	}
+}
+
+static int ec_bhf_open(struct net_device *net_dev)
+{
+	struct ec_bhf_priv *priv = netdev_priv(net_dev);
+	struct device *dev = PRIV_TO_DEV(priv);
+	int err = 0;
+
+	dev_info(dev, "Opening device\n");
+
+	ec_bhf_reset(priv);
+
+	err = ec_bhf_alloc_dma_mem(priv, &priv->rx_buf, priv->rx_dma_chan,
+				   FIFO_SIZE * sizeof(struct rx_desc));
+	if (err) {
+		dev_err(dev, "Failed to allocate rx buffer\n");
+		goto out;
+	}
+	ec_bhf_setup_rx_descs(priv);
+
+	dev_info(dev, "RX buffer allocated, address: %x\n",
+		 (unsigned)priv->rx_buf.buf_phys);
+
+	err = ec_bhf_alloc_dma_mem(priv, &priv->tx_buf, priv->tx_dma_chan,
+				   FIFO_SIZE * sizeof(struct tx_desc));
+	if (err) {
+		dev_err(dev, "Failed to allocate tx buffer\n");
+		goto error_rx_free;
+	}
+	dev_dbg(dev, "TX buffer allocated, addres: %x\n",
+		(unsigned)priv->tx_buf.buf_phys);
+
+	iowrite8(0, priv->mii_io + MII_MAC_FILT_FLAG);
+
+	ec_bhf_setup_tx_descs(priv);
+
+	netif_start_queue(net_dev);
+
+	hrtimer_init(&priv->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+	priv->hrtimer.function = ec_bhf_timer_fun;
+	hrtimer_start(&priv->hrtimer, ktime_set(0, polling_frequency),
+		      HRTIMER_MODE_REL);
+
+	dev_info(PRIV_TO_DEV(priv), "Device open\n");
+
+	ec_bhf_print_status(priv);
+
+	return 0;
+
+error_rx_free:
+	dma_free_coherent(dev, priv->rx_buf.alloc_len, priv->rx_buf.alloc,
+			  priv->rx_buf.alloc_len);
+out:
+	return err;
+}
+
+static int ec_bhf_stop(struct net_device *net_dev)
+{
+	struct ec_bhf_priv *priv = netdev_priv(net_dev);
+	struct device *dev = PRIV_TO_DEV(priv);
+
+	hrtimer_cancel(&priv->hrtimer);
+
+	ec_bhf_reset(priv);
+
+	netif_tx_disable(net_dev);
+
+	dma_free_coherent(dev, priv->tx_buf.alloc_len,
+			  priv->tx_buf.alloc, priv->tx_buf.alloc_phys);
+	dma_free_coherent(dev, priv->rx_buf.alloc_len,
+			  priv->rx_buf.alloc, priv->rx_buf.alloc_phys);
+
+	return 0;
+}
+
+static struct rtnl_link_stats64 *
+ec_bhf_get_stats(struct net_device *net_dev,
+		 struct rtnl_link_stats64 *stats)
+{
+	struct ec_bhf_priv *priv = netdev_priv(net_dev);
+
+	stats->rx_errors = ioread8(priv->mac_io + MAC_RX_ERR_CNT) +
+				ioread8(priv->mac_io + MAC_CRC_ERR_CNT) +
+				ioread8(priv->mac_io + MAC_FRAME_ERR_CNT);
+	stats->rx_packets = ioread32(priv->mac_io + MAC_RX_FRAME_CNT);
+	stats->tx_packets = ioread32(priv->mac_io + MAC_TX_FRAME_CNT);
+	stats->rx_dropped = ioread8(priv->mac_io + MAC_DROPPED_FRMS);
+
+	stats->tx_bytes = priv->stat_tx_bytes;
+	stats->rx_bytes = priv->stat_rx_bytes;
+
+	return stats;
+}
+
+static const struct net_device_ops ec_bhf_netdev_ops = {
+	.ndo_start_xmit		= ec_bhf_start_xmit,
+	.ndo_open		= ec_bhf_open,
+	.ndo_stop		= ec_bhf_stop,
+	.ndo_get_stats64	= ec_bhf_get_stats,
+	.ndo_change_mtu		= eth_change_mtu,
+	.ndo_validate_addr	= eth_validate_addr,
+	.ndo_set_mac_address	= eth_mac_addr
+};
+
+static int ec_bhf_probe(struct pci_dev *dev, const struct pci_device_id *id)
+{
+	struct net_device *net_dev;
+	struct ec_bhf_priv *priv;
+	void * __iomem dma_io;
+	void * __iomem io;
+	int err = 0;
+
+	err = pci_enable_device(dev);
+	if (err)
+		return err;
+
+	pci_set_master(dev);
+
+	err = pci_set_dma_mask(dev, DMA_BIT_MASK(32));
+	if (err) {
+		dev_err(&dev->dev,
+			"Required dma mask not supported, failed to initialize device\n");
+		err = -EIO;
+		goto err_disable_dev;
+	}
+
+	err = pci_set_consistent_dma_mask(dev, DMA_BIT_MASK(32));
+	if (err) {
+		dev_err(&dev->dev,
+			"Required dma mask not supported, failed to initialize device\n");
+		goto err_disable_dev;
+	}
+
+	err = pci_request_regions(dev, "ec_bhf");
+	if (err) {
+		dev_err(&dev->dev, "Failed to request pci memory regions\n");
+		goto err_disable_dev;
+	}
+
+	io = pci_iomap(dev, 0, 0);
+	if (!io) {
+		dev_err(&dev->dev, "Failed to map pci card memory bar 0");
+		err = -EIO;
+		goto err_release_regions;
+	}
+
+	dma_io = pci_iomap(dev, 2, 0);
+	if (!dma_io) {
+		dev_err(&dev->dev, "Failed to map pci card memory bar 2");
+		err = -EIO;
+		goto err_unmap;
+	}
+
+	net_dev = alloc_etherdev(sizeof(struct ec_bhf_priv));
+	if (net_dev == 0) {
+		err = -ENOMEM;
+		goto err_unmap_dma_io;
+	}
+
+	pci_set_drvdata(dev, net_dev);
+	SET_NETDEV_DEV(net_dev, &dev->dev);
+
+	net_dev->features = 0;
+	net_dev->flags |= IFF_NOARP;
+
+	net_dev->netdev_ops = &ec_bhf_netdev_ops;
+
+	priv = netdev_priv(net_dev);
+	priv->net_dev = net_dev;
+	priv->io = io;
+	priv->dma_io = dma_io;
+	priv->dev = dev;
+
+	err = ec_bhf_setup_offsets(priv);
+	if (err < 0)
+		goto err_free_net_dev;
+
+	memcpy_fromio(net_dev->dev_addr, priv->mii_io + MII_MAC_ADDR, 6);
+
+	dev_dbg(&dev->dev, "CX5020 Ethercat master address: %pM\n",
+		net_dev->dev_addr);
+
+	err = register_netdev(net_dev);
+	if (err < 0)
+		goto err_free_net_dev;
+
+	return 0;
+
+err_free_net_dev:
+	free_netdev(net_dev);
+err_unmap_dma_io:
+	pci_iounmap(dev, dma_io);
+err_unmap:
+	pci_iounmap(dev, io);
+err_release_regions:
+	pci_release_regions(dev);
+err_disable_dev:
+	pci_clear_master(dev);
+	pci_disable_device(dev);
+
+	return err;
+}
+
+static void ec_bhf_remove(struct pci_dev *dev)
+{
+	struct net_device *net_dev = pci_get_drvdata(dev);
+	struct ec_bhf_priv *priv = netdev_priv(net_dev);
+
+	unregister_netdev(net_dev);
+	free_netdev(net_dev);
+
+	pci_iounmap(dev, priv->dma_io);
+	pci_iounmap(dev, priv->io);
+	pci_release_regions(dev);
+	pci_clear_master(dev);
+	pci_disable_device(dev);
+}
+
+static struct pci_driver pci_driver = {
+	.name		= "ec_bhf",
+	.id_table	= ids,
+	.probe		= ec_bhf_probe,
+	.remove		= ec_bhf_remove,
+};
+
+static int __init ec_bhf_init(void)
+{
+	return pci_register_driver(&pci_driver);
+}
+
+static void __exit ec_bhf_exit(void)
+{
+	pci_unregister_driver(&pci_driver);
+}
+
+module_init(ec_bhf_init);
+module_exit(ec_bhf_exit);
+
+module_param(polling_frequency, long, S_IRUGO);
+MODULE_PARM_DESC(polling_frequency, "Polling timer frequency in ns");
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Dariusz Marcinkiewicz <reksio@newterm.pl>");
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index a186454..dc19bc5 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -4949,6 +4949,12 @@
 	if (status)
 		goto err;
 
+	/* On some BE3 FW versions, after a HW reset,
+	 * interrupts will remain disabled for each function.
+	 * So, explicitly enable interrupts
+	 */
+	be_intr_set(adapter, true);
+
 	/* tell fw we're ready to fire cmds */
 	status = be_cmd_fw_init(adapter);
 	if (status)
diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c
index b0c6050..b78378c 100644
--- a/drivers/net/ethernet/jme.c
+++ b/drivers/net/ethernet/jme.c
@@ -1988,7 +1988,7 @@
 	return idx;
 }
 
-static void
+static int
 jme_fill_tx_map(struct pci_dev *pdev,
 		struct txdesc *txdesc,
 		struct jme_buffer_info *txbi,
@@ -2005,6 +2005,9 @@
 				len,
 				PCI_DMA_TODEVICE);
 
+	if (unlikely(pci_dma_mapping_error(pdev, dmaaddr)))
+		return -EINVAL;
+
 	pci_dma_sync_single_for_device(pdev,
 				       dmaaddr,
 				       len,
@@ -2021,9 +2024,30 @@
 
 	txbi->mapping = dmaaddr;
 	txbi->len = len;
+	return 0;
 }
 
-static void
+static void jme_drop_tx_map(struct jme_adapter *jme, int startidx, int count)
+{
+	struct jme_ring *txring = &(jme->txring[0]);
+	struct jme_buffer_info *txbi = txring->bufinf, *ctxbi;
+	int mask = jme->tx_ring_mask;
+	int j;
+
+	for (j = 0 ; j < count ; j++) {
+		ctxbi = txbi + ((startidx + j + 2) & (mask));
+		pci_unmap_page(jme->pdev,
+				ctxbi->mapping,
+				ctxbi->len,
+				PCI_DMA_TODEVICE);
+
+				ctxbi->mapping = 0;
+				ctxbi->len = 0;
+	}
+
+}
+
+static int
 jme_map_tx_skb(struct jme_adapter *jme, struct sk_buff *skb, int idx)
 {
 	struct jme_ring *txring = &(jme->txring[0]);
@@ -2034,25 +2058,37 @@
 	int mask = jme->tx_ring_mask;
 	const struct skb_frag_struct *frag;
 	u32 len;
+	int ret = 0;
 
 	for (i = 0 ; i < nr_frags ; ++i) {
 		frag = &skb_shinfo(skb)->frags[i];
 		ctxdesc = txdesc + ((idx + i + 2) & (mask));
 		ctxbi = txbi + ((idx + i + 2) & (mask));
 
-		jme_fill_tx_map(jme->pdev, ctxdesc, ctxbi,
+		ret = jme_fill_tx_map(jme->pdev, ctxdesc, ctxbi,
 				skb_frag_page(frag),
 				frag->page_offset, skb_frag_size(frag), hidma);
+		if (ret) {
+			jme_drop_tx_map(jme, idx, i);
+			goto out;
+		}
+
 	}
 
 	len = skb_is_nonlinear(skb) ? skb_headlen(skb) : skb->len;
 	ctxdesc = txdesc + ((idx + 1) & (mask));
 	ctxbi = txbi + ((idx + 1) & (mask));
-	jme_fill_tx_map(jme->pdev, ctxdesc, ctxbi, virt_to_page(skb->data),
+	ret = jme_fill_tx_map(jme->pdev, ctxdesc, ctxbi, virt_to_page(skb->data),
 			offset_in_page(skb->data), len, hidma);
+	if (ret)
+		jme_drop_tx_map(jme, idx, i);
+
+out:
+	return ret;
 
 }
 
+
 static int
 jme_tx_tso(struct sk_buff *skb, __le16 *mss, u8 *flags)
 {
@@ -2131,6 +2167,7 @@
 	struct txdesc *txdesc;
 	struct jme_buffer_info *txbi;
 	u8 flags;
+	int ret = 0;
 
 	txdesc = (struct txdesc *)txring->desc + idx;
 	txbi = txring->bufinf + idx;
@@ -2155,7 +2192,10 @@
 	if (jme_tx_tso(skb, &txdesc->desc1.mss, &flags))
 		jme_tx_csum(jme, skb, &flags);
 	jme_tx_vlan(skb, &txdesc->desc1.vlan, &flags);
-	jme_map_tx_skb(jme, skb, idx);
+	ret = jme_map_tx_skb(jme, skb, idx);
+	if (ret)
+		return ret;
+
 	txdesc->desc1.flags = flags;
 	/*
 	 * Set tx buffer info after telling NIC to send
@@ -2228,7 +2268,8 @@
 		return NETDEV_TX_BUSY;
 	}
 
-	jme_fill_tx_desc(jme, skb, idx);
+	if (jme_fill_tx_desc(jme, skb, idx))
+		return NETDEV_TX_OK;
 
 	jwrite32(jme, JME_TXCS, jme->reg_txcs |
 				TXCS_SELECT_QUEUE0 |
diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c
index 78099ea..92d3249 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c
@@ -1253,12 +1253,12 @@
 	},
 	{
 		.opcode = MLX4_CMD_UPDATE_QP,
-		.has_inbox = false,
+		.has_inbox = true,
 		.has_outbox = false,
 		.out_is_imm = false,
 		.encode_slave_id = false,
 		.verify = NULL,
-		.wrapper = mlx4_CMD_EPERM_wrapper
+		.wrapper = mlx4_UPDATE_QP_wrapper
 	},
 	{
 		.opcode = MLX4_CMD_GET_OP_REQ,
diff --git a/drivers/net/ethernet/mellanox/mlx4/mlx4.h b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
index f9c4651..212cea4 100644
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -1195,6 +1195,12 @@
 			   struct mlx4_cmd_mailbox *outbox,
 			   struct mlx4_cmd_info *cmd);
 
+int mlx4_UPDATE_QP_wrapper(struct mlx4_dev *dev, int slave,
+			   struct mlx4_vhcr *vhcr,
+			   struct mlx4_cmd_mailbox *inbox,
+			   struct mlx4_cmd_mailbox *outbox,
+			   struct mlx4_cmd_info *cmd);
+
 int mlx4_PROMISC_wrapper(struct mlx4_dev *dev, int slave,
 			 struct mlx4_vhcr *vhcr,
 			 struct mlx4_cmd_mailbox *inbox,
diff --git a/drivers/net/ethernet/mellanox/mlx4/qp.c b/drivers/net/ethernet/mellanox/mlx4/qp.c
index 61d64eb..fbd32af 100644
--- a/drivers/net/ethernet/mellanox/mlx4/qp.c
+++ b/drivers/net/ethernet/mellanox/mlx4/qp.c
@@ -389,6 +389,41 @@
 
 EXPORT_SYMBOL_GPL(mlx4_qp_alloc);
 
+#define MLX4_UPDATE_QP_SUPPORTED_ATTRS MLX4_UPDATE_QP_SMAC
+int mlx4_update_qp(struct mlx4_dev *dev, struct mlx4_qp *qp,
+		   enum mlx4_update_qp_attr attr,
+		   struct mlx4_update_qp_params *params)
+{
+	struct mlx4_cmd_mailbox *mailbox;
+	struct mlx4_update_qp_context *cmd;
+	u64 pri_addr_path_mask = 0;
+	int err = 0;
+
+	mailbox = mlx4_alloc_cmd_mailbox(dev);
+	if (IS_ERR(mailbox))
+		return PTR_ERR(mailbox);
+
+	cmd = (struct mlx4_update_qp_context *)mailbox->buf;
+
+	if (!attr || (attr & ~MLX4_UPDATE_QP_SUPPORTED_ATTRS))
+		return -EINVAL;
+
+	if (attr & MLX4_UPDATE_QP_SMAC) {
+		pri_addr_path_mask |= 1ULL << MLX4_UPD_QP_PATH_MASK_MAC_INDEX;
+		cmd->qp_context.pri_path.grh_mylmc = params->smac_index;
+	}
+
+	cmd->primary_addr_path_mask = cpu_to_be64(pri_addr_path_mask);
+
+	err = mlx4_cmd(dev, mailbox->dma, qp->qpn & 0xffffff, 0,
+		       MLX4_CMD_UPDATE_QP, MLX4_CMD_TIME_CLASS_A,
+		       MLX4_CMD_NATIVE);
+
+	mlx4_free_cmd_mailbox(dev, mailbox);
+	return err;
+}
+EXPORT_SYMBOL_GPL(mlx4_update_qp);
+
 void mlx4_qp_remove(struct mlx4_dev *dev, struct mlx4_qp *qp)
 {
 	struct mlx4_qp_table *qp_table = &mlx4_priv(dev)->qp_table;
diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index 1c3fdd4..8f1254a 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -3895,6 +3895,60 @@
 
 }
 
+#define MLX4_UPD_QP_PATH_MASK_SUPPORTED (1ULL << MLX4_UPD_QP_PATH_MASK_MAC_INDEX)
+int mlx4_UPDATE_QP_wrapper(struct mlx4_dev *dev, int slave,
+			   struct mlx4_vhcr *vhcr,
+			   struct mlx4_cmd_mailbox *inbox,
+			   struct mlx4_cmd_mailbox *outbox,
+			   struct mlx4_cmd_info *cmd_info)
+{
+	int err;
+	u32 qpn = vhcr->in_modifier & 0xffffff;
+	struct res_qp *rqp;
+	u64 mac;
+	unsigned port;
+	u64 pri_addr_path_mask;
+	struct mlx4_update_qp_context *cmd;
+	int smac_index;
+
+	cmd = (struct mlx4_update_qp_context *)inbox->buf;
+
+	pri_addr_path_mask = be64_to_cpu(cmd->primary_addr_path_mask);
+	if (cmd->qp_mask || cmd->secondary_addr_path_mask ||
+	    (pri_addr_path_mask & ~MLX4_UPD_QP_PATH_MASK_SUPPORTED))
+		return -EPERM;
+
+	/* Just change the smac for the QP */
+	err = get_res(dev, slave, qpn, RES_QP, &rqp);
+	if (err) {
+		mlx4_err(dev, "Updating qpn 0x%x for slave %d rejected\n", qpn, slave);
+		return err;
+	}
+
+	port = (rqp->sched_queue >> 6 & 1) + 1;
+	smac_index = cmd->qp_context.pri_path.grh_mylmc;
+	err = mac_find_smac_ix_in_slave(dev, slave, port,
+					smac_index, &mac);
+	if (err) {
+		mlx4_err(dev, "Failed to update qpn 0x%x, MAC is invalid. smac_ix: %d\n",
+			 qpn, smac_index);
+		goto err_mac;
+	}
+
+	err = mlx4_cmd(dev, inbox->dma,
+		       vhcr->in_modifier, 0,
+		       MLX4_CMD_UPDATE_QP, MLX4_CMD_TIME_CLASS_A,
+		       MLX4_CMD_NATIVE);
+	if (err) {
+		mlx4_err(dev, "Failed to update qpn on qpn 0x%x, command failed\n", qpn);
+		goto err_mac;
+	}
+
+err_mac:
+	put_res(dev, slave, qpn, RES_QP);
+	return err;
+}
+
 int mlx4_QP_FLOW_STEERING_ATTACH_wrapper(struct mlx4_dev *dev, int slave,
 					 struct mlx4_vhcr *vhcr,
 					 struct mlx4_cmd_mailbox *inbox,
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
index 7b52a88..f785d01 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic.h
@@ -1719,22 +1719,6 @@
 				tx_ring->producer;
 }
 
-static inline int qlcnic_set_real_num_queues(struct qlcnic_adapter *adapter,
-					     struct net_device *netdev)
-{
-	int err;
-
-	netdev->num_tx_queues = adapter->drv_tx_rings;
-	netdev->real_num_tx_queues = adapter->drv_tx_rings;
-
-	err = netif_set_real_num_tx_queues(netdev, adapter->drv_tx_rings);
-	if (err)
-		netdev_err(netdev, "failed to set %d Tx queues\n",
-			   adapter->drv_tx_rings);
-
-	return err;
-}
-
 struct qlcnic_nic_template {
 	int (*config_bridged_mode) (struct qlcnic_adapter *, u32);
 	int (*config_led) (struct qlcnic_adapter *, u32, u32);
diff --git a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
index 0bc9148..7e55e88 100644
--- a/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
+++ b/drivers/net/ethernet/qlogic/qlcnic/qlcnic_main.c
@@ -2206,6 +2206,31 @@
 	ahw->max_uc_count = count;
 }
 
+static int qlcnic_set_real_num_queues(struct qlcnic_adapter *adapter,
+				      u8 tx_queues, u8 rx_queues)
+{
+	struct net_device *netdev = adapter->netdev;
+	int err = 0;
+
+	if (tx_queues) {
+		err = netif_set_real_num_tx_queues(netdev, tx_queues);
+		if (err) {
+			netdev_err(netdev, "failed to set %d Tx queues\n",
+				   tx_queues);
+			return err;
+		}
+	}
+
+	if (rx_queues) {
+		err = netif_set_real_num_rx_queues(netdev, rx_queues);
+		if (err)
+			netdev_err(netdev, "failed to set %d Rx queues\n",
+				   rx_queues);
+	}
+
+	return err;
+}
+
 int
 qlcnic_setup_netdev(struct qlcnic_adapter *adapter, struct net_device *netdev,
 		    int pci_using_dac)
@@ -2269,7 +2294,8 @@
 	netdev->priv_flags |= IFF_UNICAST_FLT;
 	netdev->irq = adapter->msix_entries[0].vector;
 
-	err = qlcnic_set_real_num_queues(adapter, netdev);
+	err = qlcnic_set_real_num_queues(adapter, adapter->drv_tx_rings,
+					 adapter->drv_sds_rings);
 	if (err)
 		return err;
 
@@ -2943,9 +2969,13 @@
 			    tx_ring->tx_stats.xmit_called,
 			    tx_ring->tx_stats.xmit_on,
 			    tx_ring->tx_stats.xmit_off);
+
+		if (tx_ring->crb_intr_mask)
+			netdev_info(netdev, "crb_intr_mask=%d\n",
+				    readl(tx_ring->crb_intr_mask));
+
 		netdev_info(netdev,
-			    "crb_intr_mask=%d, hw_producer=%d, sw_producer=%d sw_consumer=%d, hw_consumer=%d\n",
-			    readl(tx_ring->crb_intr_mask),
+			    "hw_producer=%d, sw_producer=%d sw_consumer=%d, hw_consumer=%d\n",
 			    readl(tx_ring->crb_cmd_producer),
 			    tx_ring->producer, tx_ring->sw_consumer,
 			    le32_to_cpu(*(tx_ring->hw_consumer)));
@@ -3978,12 +4008,21 @@
 int qlcnic_setup_rings(struct qlcnic_adapter *adapter)
 {
 	struct net_device *netdev = adapter->netdev;
+	u8 tx_rings, rx_rings;
 	int err;
 
 	if (test_bit(__QLCNIC_RESETTING, &adapter->state))
 		return -EBUSY;
 
+	tx_rings = adapter->drv_tss_rings;
+	rx_rings = adapter->drv_rss_rings;
+
 	netif_device_detach(netdev);
+
+	err = qlcnic_set_real_num_queues(adapter, tx_rings, rx_rings);
+	if (err)
+		goto done;
+
 	if (netif_running(netdev))
 		__qlcnic_down(adapter, netdev);
 
@@ -4003,7 +4042,17 @@
 		return err;
 	}
 
-	netif_set_real_num_tx_queues(netdev, adapter->drv_tx_rings);
+	/* Check if we need to update real_num_{tx|rx}_queues because
+	 * qlcnic_setup_intr() may change Tx/Rx rings size
+	 */
+	if ((tx_rings != adapter->drv_tx_rings) ||
+	    (rx_rings != adapter->drv_sds_rings)) {
+		err = qlcnic_set_real_num_queues(adapter,
+						 adapter->drv_tx_rings,
+						 adapter->drv_sds_rings);
+		if (err)
+			goto done;
+	}
 
 	if (qlcnic_83xx_check(adapter)) {
 		qlcnic_83xx_initialize_nic(adapter, 1);
diff --git a/drivers/net/ethernet/sfc/nic.c b/drivers/net/ethernet/sfc/nic.c
index 32d969e..89b83e5 100644
--- a/drivers/net/ethernet/sfc/nic.c
+++ b/drivers/net/ethernet/sfc/nic.c
@@ -156,13 +156,15 @@
 	efx->net_dev->rx_cpu_rmap = NULL;
 #endif
 
-	/* Disable MSI/MSI-X interrupts */
-	efx_for_each_channel(channel, efx)
-		free_irq(channel->irq, &efx->msi_context[channel->channel]);
-
-	/* Disable legacy interrupt */
-	if (efx->legacy_irq)
+	if (EFX_INT_MODE_USE_MSI(efx)) {
+		/* Disable MSI/MSI-X interrupts */
+		efx_for_each_channel(channel, efx)
+			free_irq(channel->irq,
+				 &efx->msi_context[channel->channel]);
+	} else {
+		/* Disable legacy interrupt */
 		free_irq(efx->legacy_irq, efx);
+	}
 }
 
 /* Register dump */
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
index d940034..0f4841d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_main.c
@@ -1704,7 +1704,7 @@
 		if (ret) {
 			pr_err("%s: Cannot attach to PHY (error: %d)\n",
 			       __func__, ret);
-			goto phy_error;
+			return ret;
 		}
 	}
 
@@ -1779,8 +1779,6 @@
 dma_desc_error:
 	if (priv->phydev)
 		phy_disconnect(priv->phydev);
-phy_error:
-	clk_disable_unprepare(priv->stmmac_clk);
 
 	return ret;
 }
diff --git a/drivers/net/ethernet/sun/cassini.c b/drivers/net/ethernet/sun/cassini.c
index df8d383..b9ac20f 100644
--- a/drivers/net/ethernet/sun/cassini.c
+++ b/drivers/net/ethernet/sun/cassini.c
@@ -246,7 +246,7 @@
 	int i;
 
 	for (i = 0; i < N_TX_RINGS; i++)
-		spin_lock(&cp->tx_lock[i]);
+		spin_lock_nested(&cp->tx_lock[i], i);
 }
 
 static inline void cas_lock_all(struct cas *cp)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 36aa109..c331b7e 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -1871,18 +1871,13 @@
 		mdio_node = of_find_node_by_phandle(be32_to_cpup(parp));
 		phyid = be32_to_cpup(parp+1);
 		mdio = of_find_device_by_node(mdio_node);
-
-		if (strncmp(mdio->name, "gpio", 4) == 0) {
-			/* GPIO bitbang MDIO driver attached */
-			struct mii_bus *bus = dev_get_drvdata(&mdio->dev);
-
-			snprintf(slave_data->phy_id, sizeof(slave_data->phy_id),
-				 PHY_ID_FMT, bus->id, phyid);
-		} else {
-			/* davinci MDIO driver attached */
-			snprintf(slave_data->phy_id, sizeof(slave_data->phy_id),
-				 PHY_ID_FMT, mdio->name, phyid);
+		of_node_put(mdio_node);
+		if (!mdio) {
+			pr_err("Missing mdio platform device\n");
+			return -EINVAL;
 		}
+		snprintf(slave_data->phy_id, sizeof(slave_data->phy_id),
+			 PHY_ID_FMT, mdio->name, phyid);
 
 		mac_addr = of_get_mac_address(slave_node);
 		if (mac_addr)
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index b0e2865..d53e299 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -458,8 +458,10 @@
 	struct macvlan_dev *vlan = netdev_priv(dev);
 	struct net_device *lowerdev = vlan->lowerdev;
 
-	if (change & IFF_ALLMULTI)
-		dev_set_allmulti(lowerdev, dev->flags & IFF_ALLMULTI ? 1 : -1);
+	if (dev->flags & IFF_UP) {
+		if (change & IFF_ALLMULTI)
+			dev_set_allmulti(lowerdev, dev->flags & IFF_ALLMULTI ? 1 : -1);
+	}
 }
 
 static void macvlan_set_mac_lists(struct net_device *dev)
@@ -515,6 +517,11 @@
 #define MACVLAN_STATE_MASK \
 	((1<<__LINK_STATE_NOCARRIER) | (1<<__LINK_STATE_DORMANT))
 
+static int macvlan_get_nest_level(struct net_device *dev)
+{
+	return ((struct macvlan_dev *)netdev_priv(dev))->nest_level;
+}
+
 static void macvlan_set_lockdep_class_one(struct net_device *dev,
 					  struct netdev_queue *txq,
 					  void *_unused)
@@ -525,8 +532,9 @@
 
 static void macvlan_set_lockdep_class(struct net_device *dev)
 {
-	lockdep_set_class(&dev->addr_list_lock,
-			  &macvlan_netdev_addr_lock_key);
+	lockdep_set_class_and_subclass(&dev->addr_list_lock,
+				       &macvlan_netdev_addr_lock_key,
+				       macvlan_get_nest_level(dev));
 	netdev_for_each_tx_queue(dev, macvlan_set_lockdep_class_one, NULL);
 }
 
@@ -721,6 +729,7 @@
 	.ndo_fdb_add		= macvlan_fdb_add,
 	.ndo_fdb_del		= macvlan_fdb_del,
 	.ndo_fdb_dump		= ndo_dflt_fdb_dump,
+	.ndo_get_lock_subclass  = macvlan_get_nest_level,
 };
 
 void macvlan_common_setup(struct net_device *dev)
@@ -849,6 +858,7 @@
 	vlan->dev      = dev;
 	vlan->port     = port;
 	vlan->set_features = MACVLAN_FEATURES;
+	vlan->nest_level = dev_get_nest_level(lowerdev, netif_is_macvlan) + 1;
 
 	vlan->mode     = MACVLAN_MODE_VEPA;
 	if (data && data[IFLA_MACVLAN_MODE])
diff --git a/drivers/net/phy/mdio-gpio.c b/drivers/net/phy/mdio-gpio.c
index 9c4defd..5f1a225 100644
--- a/drivers/net/phy/mdio-gpio.c
+++ b/drivers/net/phy/mdio-gpio.c
@@ -215,6 +215,10 @@
 	if (pdev->dev.of_node) {
 		pdata = mdio_gpio_of_get_data(pdev);
 		bus_id = of_alias_get_id(pdev->dev.of_node, "mdio-gpio");
+		if (bus_id < 0) {
+			dev_warn(&pdev->dev, "failed to get alias id\n");
+			bus_id = 0;
+		}
 	} else {
 		pdata = dev_get_platdata(&pdev->dev);
 		bus_id = pdev->id;
diff --git a/drivers/net/phy/phy.c b/drivers/net/phy/phy.c
index a972056..3bc079a 100644
--- a/drivers/net/phy/phy.c
+++ b/drivers/net/phy/phy.c
@@ -715,7 +715,7 @@
 	struct delayed_work *dwork = to_delayed_work(work);
 	struct phy_device *phydev =
 			container_of(dwork, struct phy_device, state_queue);
-	int needs_aneg = 0, do_suspend = 0;
+	bool needs_aneg = false, do_suspend = false, do_resume = false;
 	int err = 0;
 
 	mutex_lock(&phydev->lock);
@@ -727,7 +727,7 @@
 	case PHY_PENDING:
 		break;
 	case PHY_UP:
-		needs_aneg = 1;
+		needs_aneg = true;
 
 		phydev->link_timeout = PHY_AN_TIMEOUT;
 
@@ -757,7 +757,7 @@
 			phydev->adjust_link(phydev->attached_dev);
 
 		} else if (0 == phydev->link_timeout--)
-			needs_aneg = 1;
+			needs_aneg = true;
 		break;
 	case PHY_NOLINK:
 		err = phy_read_status(phydev);
@@ -791,7 +791,7 @@
 			netif_carrier_on(phydev->attached_dev);
 		} else {
 			if (0 == phydev->link_timeout--)
-				needs_aneg = 1;
+				needs_aneg = true;
 		}
 
 		phydev->adjust_link(phydev->attached_dev);
@@ -827,7 +827,7 @@
 			phydev->link = 0;
 			netif_carrier_off(phydev->attached_dev);
 			phydev->adjust_link(phydev->attached_dev);
-			do_suspend = 1;
+			do_suspend = true;
 		}
 		break;
 	case PHY_RESUMING:
@@ -876,6 +876,7 @@
 			}
 			phydev->adjust_link(phydev->attached_dev);
 		}
+		do_resume = true;
 		break;
 	}
 
@@ -883,9 +884,10 @@
 
 	if (needs_aneg)
 		err = phy_start_aneg(phydev);
-
-	if (do_suspend)
+	else if (do_suspend)
 		phy_suspend(phydev);
+	else if (do_resume)
+		phy_resume(phydev);
 
 	if (err < 0)
 		phy_error(phydev);
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 0ce6066..4987a1c 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -614,8 +614,8 @@
 	err = phy_init_hw(phydev);
 	if (err)
 		phy_detach(phydev);
-
-	phy_resume(phydev);
+	else
+		phy_resume(phydev);
 
 	return err;
 }
diff --git a/drivers/net/usb/cdc_mbim.c b/drivers/net/usb/cdc_mbim.c
index c9f3281..2e025dd 100644
--- a/drivers/net/usb/cdc_mbim.c
+++ b/drivers/net/usb/cdc_mbim.c
@@ -120,6 +120,16 @@
 	cdc_ncm_unbind(dev, intf);
 }
 
+/* verify that the ethernet protocol is IPv4 or IPv6 */
+static bool is_ip_proto(__be16 proto)
+{
+	switch (proto) {
+	case htons(ETH_P_IP):
+	case htons(ETH_P_IPV6):
+		return true;
+	}
+	return false;
+}
 
 static struct sk_buff *cdc_mbim_tx_fixup(struct usbnet *dev, struct sk_buff *skb, gfp_t flags)
 {
@@ -128,6 +138,7 @@
 	struct cdc_ncm_ctx *ctx = info->ctx;
 	__le32 sign = cpu_to_le32(USB_CDC_MBIM_NDP16_IPS_SIGN);
 	u16 tci = 0;
+	bool is_ip;
 	u8 *c;
 
 	if (!ctx)
@@ -137,25 +148,32 @@
 		if (skb->len <= ETH_HLEN)
 			goto error;
 
+		/* Some applications using e.g. packet sockets will
+		 * bypass the VLAN acceleration and create tagged
+		 * ethernet frames directly.  We primarily look for
+		 * the accelerated out-of-band tag, but fall back if
+		 * required
+		 */
+		skb_reset_mac_header(skb);
+		if (vlan_get_tag(skb, &tci) < 0 && skb->len > VLAN_ETH_HLEN &&
+		    __vlan_get_tag(skb, &tci) == 0) {
+			is_ip = is_ip_proto(vlan_eth_hdr(skb)->h_vlan_encapsulated_proto);
+			skb_pull(skb, VLAN_ETH_HLEN);
+		} else {
+			is_ip = is_ip_proto(eth_hdr(skb)->h_proto);
+			skb_pull(skb, ETH_HLEN);
+		}
+
 		/* mapping VLANs to MBIM sessions:
 		 *   no tag     => IPS session <0>
 		 *   1 - 255    => IPS session <vlanid>
 		 *   256 - 511  => DSS session <vlanid - 256>
 		 *   512 - 4095 => unsupported, drop
 		 */
-		vlan_get_tag(skb, &tci);
-
 		switch (tci & 0x0f00) {
 		case 0x0000: /* VLAN ID 0 - 255 */
-			/* verify that datagram is IPv4 or IPv6 */
-			skb_reset_mac_header(skb);
-			switch (eth_hdr(skb)->h_proto) {
-			case htons(ETH_P_IP):
-			case htons(ETH_P_IPV6):
-				break;
-			default:
+			if (!is_ip)
 				goto error;
-			}
 			c = (u8 *)&sign;
 			c[3] = tci;
 			break;
@@ -169,7 +187,6 @@
 				  "unsupported tci=0x%04x\n", tci);
 			goto error;
 		}
-		skb_pull(skb, ETH_HLEN);
 	}
 
 	spin_lock_bh(&ctx->mtx);
@@ -204,17 +221,23 @@
 		return;
 
 	/* need to send the NA on the VLAN dev, if any */
-	if (tci)
+	rcu_read_lock();
+	if (tci) {
 		netdev = __vlan_find_dev_deep(dev->net, htons(ETH_P_8021Q),
 					      tci);
-	else
+		if (!netdev) {
+			rcu_read_unlock();
+			return;
+		}
+	} else {
 		netdev = dev->net;
-	if (!netdev)
-		return;
+	}
+	dev_hold(netdev);
+	rcu_read_unlock();
 
 	in6_dev = in6_dev_get(netdev);
 	if (!in6_dev)
-		return;
+		goto out;
 	is_router = !!in6_dev->cnf.forwarding;
 	in6_dev_put(in6_dev);
 
@@ -224,6 +247,8 @@
 				 true /* solicited */,
 				 false /* override */,
 				 true /* inc_opt */);
+out:
+	dev_put(netdev);
 }
 
 static bool is_neigh_solicit(u8 *buf, size_t len)
diff --git a/drivers/net/wireless/ath/ath9k/htc_drv_main.c b/drivers/net/wireless/ath/ath9k/htc_drv_main.c
index f46cd02..5627917 100644
--- a/drivers/net/wireless/ath/ath9k/htc_drv_main.c
+++ b/drivers/net/wireless/ath/ath9k/htc_drv_main.c
@@ -95,8 +95,10 @@
 
 	if ((vif->type == NL80211_IFTYPE_AP ||
 	     vif->type == NL80211_IFTYPE_MESH_POINT) &&
-	    bss_conf->enable_beacon)
+	    bss_conf->enable_beacon) {
 		priv->reconfig_beacon = true;
+		priv->rearm_ani = true;
+	}
 
 	if (bss_conf->assoc) {
 		priv->rearm_ani = true;
@@ -257,6 +259,7 @@
 
 	ath9k_htc_ps_wakeup(priv);
 
+	ath9k_htc_stop_ani(priv);
 	del_timer_sync(&priv->tx.cleanup_timer);
 	ath9k_htc_tx_drain(priv);
 
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c b/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
index afb3d15..be198529 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/wl_cfg80211.c
@@ -4948,7 +4948,7 @@
 	if (!err) {
 		/* only set 2G bandwidth using bw_cap command */
 		band_bwcap.band = cpu_to_le32(WLC_BAND_2G);
-		band_bwcap.bw_cap = cpu_to_le32(WLC_BW_40MHZ_BIT);
+		band_bwcap.bw_cap = cpu_to_le32(WLC_BW_CAP_40MHZ);
 		err = brcmf_fil_iovar_data_set(ifp, "bw_cap", &band_bwcap,
 					       sizeof(band_bwcap));
 	} else {
diff --git a/drivers/net/wireless/iwlwifi/mvm/coex.c b/drivers/net/wireless/iwlwifi/mvm/coex.c
index fa858d54..0489314 100644
--- a/drivers/net/wireless/iwlwifi/mvm/coex.c
+++ b/drivers/net/wireless/iwlwifi/mvm/coex.c
@@ -611,14 +611,14 @@
 		bt_cmd->flags |= cpu_to_le32(BT_COEX_SYNC2SCO);
 
 	if (IWL_MVM_BT_COEX_CORUNNING) {
-		bt_cmd->valid_bit_msk = cpu_to_le32(BT_VALID_CORUN_LUT_20 |
-						    BT_VALID_CORUN_LUT_40);
+		bt_cmd->valid_bit_msk |= cpu_to_le32(BT_VALID_CORUN_LUT_20 |
+						     BT_VALID_CORUN_LUT_40);
 		bt_cmd->flags |= cpu_to_le32(BT_COEX_CORUNNING);
 	}
 
 	if (IWL_MVM_BT_COEX_MPLUT) {
 		bt_cmd->flags |= cpu_to_le32(BT_COEX_MPLUT);
-		bt_cmd->valid_bit_msk = cpu_to_le32(BT_VALID_MULTI_PRIO_LUT);
+		bt_cmd->valid_bit_msk |= cpu_to_le32(BT_VALID_MULTI_PRIO_LUT);
 	}
 
 	if (mvm->cfg->bt_shared_single_ant)
diff --git a/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h b/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h
index 9426905..d73a89e 100644
--- a/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h
+++ b/drivers/net/wireless/iwlwifi/mvm/fw-api-scan.h
@@ -183,9 +183,9 @@
  *	this number of packets were received (typically 1)
  * @passive2active: is auto switching from passive to active during scan allowed
  * @rxchain_sel_flags: RXON_RX_CHAIN_*
- * @max_out_time: in usecs, max out of serving channel time
+ * @max_out_time: in TUs, max out of serving channel time
  * @suspend_time: how long to pause scan when returning to service channel:
- *	bits 0-19: beacon interal in usecs (suspend before executing)
+ *	bits 0-19: beacon interal in TUs (suspend before executing)
  *	bits 20-23: reserved
  *	bits 24-31: number of beacons (suspend between channels)
  * @rxon_flags: RXON_FLG_*
@@ -383,8 +383,8 @@
  * @quiet_plcp_th:	quiet channel num of packets threshold
  * @good_CRC_th:	passive to active promotion threshold
  * @rx_chain:		RXON rx chain.
- * @max_out_time:	max uSec to be out of assoceated channel
- * @suspend_time:	pause scan this long when returning to service channel
+ * @max_out_time:	max TUs to be out of assoceated channel
+ * @suspend_time:	pause scan this TUs when returning to service channel
  * @flags:		RXON flags
  * @filter_flags:	RXONfilter
  * @tx_cmd:		tx command for active scan; for 2GHz and for 5GHz.
diff --git a/drivers/net/wireless/iwlwifi/mvm/mac80211.c b/drivers/net/wireless/iwlwifi/mvm/mac80211.c
index f0cebf1..b41dc84 100644
--- a/drivers/net/wireless/iwlwifi/mvm/mac80211.c
+++ b/drivers/net/wireless/iwlwifi/mvm/mac80211.c
@@ -1007,7 +1007,7 @@
 	memcpy(cmd->bssid, vif->bss_conf.bssid, ETH_ALEN);
 	len = roundup(sizeof(*cmd) + cmd->count * ETH_ALEN, 4);
 
-	ret = iwl_mvm_send_cmd_pdu(mvm, MCAST_FILTER_CMD, CMD_SYNC, len, cmd);
+	ret = iwl_mvm_send_cmd_pdu(mvm, MCAST_FILTER_CMD, CMD_ASYNC, len, cmd);
 	if (ret)
 		IWL_ERR(mvm, "mcast filter cmd error. ret=%d\n", ret);
 }
@@ -1023,7 +1023,7 @@
 	if (WARN_ON_ONCE(!mvm->mcast_filter_cmd))
 		return;
 
-	ieee80211_iterate_active_interfaces(
+	ieee80211_iterate_active_interfaces_atomic(
 		mvm->hw, IEEE80211_IFACE_ITER_NORMAL,
 		iwl_mvm_mc_iface_iterator, &iter_data);
 }
@@ -1807,6 +1807,11 @@
 
 	mutex_lock(&mvm->mutex);
 
+	if (!iwl_mvm_is_idle(mvm)) {
+		ret = -EBUSY;
+		goto out;
+	}
+
 	switch (mvm->scan_status) {
 	case IWL_MVM_SCAN_OS:
 		IWL_DEBUG_SCAN(mvm, "Stopping previous scan for sched_scan\n");
diff --git a/drivers/net/wireless/iwlwifi/mvm/mvm.h b/drivers/net/wireless/iwlwifi/mvm/mvm.h
index d564233..f1ec098 100644
--- a/drivers/net/wireless/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/iwlwifi/mvm/mvm.h
@@ -1003,6 +1003,9 @@
 	return mvmvif->low_latency;
 }
 
+/* Assoc status */
+bool iwl_mvm_is_idle(struct iwl_mvm *mvm);
+
 /* Thermal management and CT-kill */
 void iwl_mvm_tt_tx_backoff(struct iwl_mvm *mvm, u32 backoff);
 void iwl_mvm_tt_handler(struct iwl_mvm *mvm);
diff --git a/drivers/net/wireless/iwlwifi/mvm/rs.c b/drivers/net/wireless/iwlwifi/mvm/rs.c
index 9f52c5b..e1c8388 100644
--- a/drivers/net/wireless/iwlwifi/mvm/rs.c
+++ b/drivers/net/wireless/iwlwifi/mvm/rs.c
@@ -1010,7 +1010,7 @@
 		return;
 	}
 
-#ifdef CPTCFG_MAC80211_DEBUGFS
+#ifdef CONFIG_MAC80211_DEBUGFS
 	/* Disable last tx check if we are debugging with fixed rate */
 	if (lq_sta->dbg_fixed_rate) {
 		IWL_DEBUG_RATE(mvm, "Fixed rate. avoid rate scaling\n");
diff --git a/drivers/net/wireless/iwlwifi/mvm/scan.c b/drivers/net/wireless/iwlwifi/mvm/scan.c
index c91dc84..c28de54 100644
--- a/drivers/net/wireless/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/iwlwifi/mvm/scan.c
@@ -277,51 +277,22 @@
 					    IEEE80211_IFACE_ITER_NORMAL,
 					    iwl_mvm_scan_condition_iterator,
 					    &global_bound);
-	/*
-	 * Under low latency traffic passive scan is fragmented meaning
-	 * that dwell on a particular channel will be fragmented. Each fragment
-	 * dwell time is 20ms and fragments period is 105ms. Skipping to next
-	 * channel will be delayed by the same period - 105ms. So suspend_time
-	 * parameter describing both fragments and channels skipping periods is
-	 * set to 105ms. This value is chosen so that overall passive scan
-	 * duration will not be too long. Max_out_time in this case is set to
-	 * 70ms, so for active scanning operating channel will be left for 70ms
-	 * while for passive still for 20ms (fragment dwell).
-	 */
-	if (global_bound) {
-		if (!iwl_mvm_low_latency(mvm)) {
-			params->suspend_time = ieee80211_tu_to_usec(100);
-			params->max_out_time = ieee80211_tu_to_usec(600);
-		} else {
-			params->suspend_time = ieee80211_tu_to_usec(105);
-			/* P2P doesn't support fragmented passive scan, so
-			 * configure max_out_time to be at least longest dwell
-			 * time for passive scan.
-			 */
-			if (vif->type == NL80211_IFTYPE_STATION && !vif->p2p) {
-				params->max_out_time = ieee80211_tu_to_usec(70);
-				params->passive_fragmented = true;
-			} else {
-				u32 passive_dwell;
 
-				/*
-				 * Use band G so that passive channel dwell time
-				 * will be assigned with maximum value.
-				 */
-				band = IEEE80211_BAND_2GHZ;
-				passive_dwell = iwl_mvm_get_passive_dwell(band);
-				params->max_out_time =
-					ieee80211_tu_to_usec(passive_dwell);
-			}
-		}
+	if (!global_bound)
+		goto not_bound;
+
+	params->suspend_time = 100;
+	params->max_out_time = 600;
+
+	if (iwl_mvm_low_latency(mvm)) {
+		params->suspend_time = 250;
+		params->max_out_time = 250;
 	}
 
+not_bound:
+
 	for (band = IEEE80211_BAND_2GHZ; band < IEEE80211_NUM_BANDS; band++) {
-		if (params->passive_fragmented)
-			params->dwell[band].passive = 20;
-		else
-			params->dwell[band].passive =
-				iwl_mvm_get_passive_dwell(band);
+		params->dwell[band].passive = iwl_mvm_get_passive_dwell(band);
 		params->dwell[band].active = iwl_mvm_get_active_dwell(band,
 								      n_ssids);
 	}
@@ -761,7 +732,7 @@
 	int band_2ghz = mvm->nvm_data->bands[IEEE80211_BAND_2GHZ].n_channels;
 	int band_5ghz = mvm->nvm_data->bands[IEEE80211_BAND_5GHZ].n_channels;
 	int head = 0;
-	int tail = band_2ghz + band_5ghz;
+	int tail = band_2ghz + band_5ghz - 1;
 	u32 ssid_bitmap;
 	int cmd_len;
 	int ret;
diff --git a/drivers/net/wireless/iwlwifi/mvm/utils.c b/drivers/net/wireless/iwlwifi/mvm/utils.c
index d619851..2180902 100644
--- a/drivers/net/wireless/iwlwifi/mvm/utils.c
+++ b/drivers/net/wireless/iwlwifi/mvm/utils.c
@@ -644,3 +644,22 @@
 
 	return result;
 }
+
+static void iwl_mvm_idle_iter(void *_data, u8 *mac, struct ieee80211_vif *vif)
+{
+	bool *idle = _data;
+
+	if (!vif->bss_conf.idle)
+		*idle = false;
+}
+
+bool iwl_mvm_is_idle(struct iwl_mvm *mvm)
+{
+	bool idle = true;
+
+	ieee80211_iterate_active_interfaces_atomic(
+			mvm->hw, IEEE80211_IFACE_ITER_NORMAL,
+			iwl_mvm_idle_iter, &idle);
+
+	return idle;
+}
diff --git a/drivers/net/wireless/iwlwifi/pcie/trans.c b/drivers/net/wireless/iwlwifi/pcie/trans.c
index dcfd6d8..2365553 100644
--- a/drivers/net/wireless/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/iwlwifi/pcie/trans.c
@@ -1749,6 +1749,10 @@
 	 * PCI Tx retries from interfering with C3 CPU state */
 	pci_write_config_byte(pdev, PCI_CFG_RETRY_TIMEOUT, 0x00);
 
+	trans->dev = &pdev->dev;
+	trans_pcie->pci_dev = pdev;
+	iwl_disable_interrupts(trans);
+
 	err = pci_enable_msi(pdev);
 	if (err) {
 		dev_err(&pdev->dev, "pci_enable_msi failed(0X%x)\n", err);
@@ -1760,8 +1764,6 @@
 		}
 	}
 
-	trans->dev = &pdev->dev;
-	trans_pcie->pci_dev = pdev;
 	trans->hw_rev = iwl_read32(trans, CSR_HW_REV);
 	trans->hw_id = (pdev->device << 16) + pdev->subsystem_device;
 	snprintf(trans->hw_id_str, sizeof(trans->hw_id_str),
@@ -1787,8 +1789,6 @@
 		goto out_pci_disable_msi;
 	}
 
-	trans_pcie->inta_mask = CSR_INI_SET_MASK;
-
 	if (iwl_pcie_alloc_ict(trans))
 		goto out_free_cmd_pool;
 
@@ -1800,6 +1800,8 @@
 		goto out_free_ict;
 	}
 
+	trans_pcie->inta_mask = CSR_INI_SET_MASK;
+
 	return trans;
 
 out_free_ict:
diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 630a3fc..0d4a285 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -226,7 +226,7 @@
 			      grant_ref_t rx_ring_ref);
 
 /* Check for SKBs from frontend and schedule backend processing */
-void xenvif_check_rx_xenvif(struct xenvif *vif);
+void xenvif_napi_schedule_or_enable_events(struct xenvif *vif);
 
 /* Prevent the device from generating any further traffic. */
 void xenvif_carrier_off(struct xenvif *vif);
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index ef05c5c..20e9def 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -75,32 +75,8 @@
 	work_done = xenvif_tx_action(vif, budget);
 
 	if (work_done < budget) {
-		int more_to_do = 0;
-		unsigned long flags;
-
-		/* It is necessary to disable IRQ before calling
-		 * RING_HAS_UNCONSUMED_REQUESTS. Otherwise we might
-		 * lose event from the frontend.
-		 *
-		 * Consider:
-		 *   RING_HAS_UNCONSUMED_REQUESTS
-		 *   <frontend generates event to trigger napi_schedule>
-		 *   __napi_complete
-		 *
-		 * This handler is still in scheduled state so the
-		 * event has no effect at all. After __napi_complete
-		 * this handler is descheduled and cannot get
-		 * scheduled again. We lose event in this case and the ring
-		 * will be completely stalled.
-		 */
-
-		local_irq_save(flags);
-
-		RING_FINAL_CHECK_FOR_REQUESTS(&vif->tx, more_to_do);
-		if (!more_to_do)
-			__napi_complete(napi);
-
-		local_irq_restore(flags);
+		napi_complete(napi);
+		xenvif_napi_schedule_or_enable_events(vif);
 	}
 
 	return work_done;
@@ -194,7 +170,7 @@
 	enable_irq(vif->tx_irq);
 	if (vif->tx_irq != vif->rx_irq)
 		enable_irq(vif->rx_irq);
-	xenvif_check_rx_xenvif(vif);
+	xenvif_napi_schedule_or_enable_events(vif);
 }
 
 static void xenvif_down(struct xenvif *vif)
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 7666540..7367208 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -104,7 +104,7 @@
 
 /* Find the containing VIF's structure from a pointer in pending_tx_info array
  */
-static inline struct xenvif* ubuf_to_vif(struct ubuf_info *ubuf)
+static inline struct xenvif *ubuf_to_vif(const struct ubuf_info *ubuf)
 {
 	u16 pending_idx = ubuf->desc;
 	struct pending_tx_info *temp =
@@ -323,6 +323,35 @@
 }
 
 /*
+ * Find the grant ref for a given frag in a chain of struct ubuf_info's
+ * skb: the skb itself
+ * i: the frag's number
+ * ubuf: a pointer to an element in the chain. It should not be NULL
+ *
+ * Returns a pointer to the element in the chain where the page were found. If
+ * not found, returns NULL.
+ * See the definition of callback_struct in common.h for more details about
+ * the chain.
+ */
+static const struct ubuf_info *xenvif_find_gref(const struct sk_buff *const skb,
+						const int i,
+						const struct ubuf_info *ubuf)
+{
+	struct xenvif *foreign_vif = ubuf_to_vif(ubuf);
+
+	do {
+		u16 pending_idx = ubuf->desc;
+
+		if (skb_shinfo(skb)->frags[i].page.p ==
+		    foreign_vif->mmap_pages[pending_idx])
+			break;
+		ubuf = (struct ubuf_info *) ubuf->ctx;
+	} while (ubuf);
+
+	return ubuf;
+}
+
+/*
  * Prepare an SKB to be transmitted to the frontend.
  *
  * This function is responsible for allocating grant operations, meta
@@ -346,9 +375,8 @@
 	int head = 1;
 	int old_meta_prod;
 	int gso_type;
-	struct ubuf_info *ubuf = skb_shinfo(skb)->destructor_arg;
-	grant_ref_t foreign_grefs[MAX_SKB_FRAGS];
-	struct xenvif *foreign_vif = NULL;
+	const struct ubuf_info *ubuf = skb_shinfo(skb)->destructor_arg;
+	const struct ubuf_info *const head_ubuf = ubuf;
 
 	old_meta_prod = npo->meta_prod;
 
@@ -386,19 +414,6 @@
 	npo->copy_off = 0;
 	npo->copy_gref = req->gref;
 
-	if ((skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) &&
-		 (ubuf->callback == &xenvif_zerocopy_callback)) {
-		int i = 0;
-		foreign_vif = ubuf_to_vif(ubuf);
-
-		do {
-			u16 pending_idx = ubuf->desc;
-			foreign_grefs[i++] =
-				foreign_vif->pending_tx_info[pending_idx].req.gref;
-			ubuf = (struct ubuf_info *) ubuf->ctx;
-		} while (ubuf);
-	}
-
 	data = skb->data;
 	while (data < skb_tail_pointer(skb)) {
 		unsigned int offset = offset_in_page(data);
@@ -415,13 +430,60 @@
 	}
 
 	for (i = 0; i < nr_frags; i++) {
+		/* This variable also signals whether foreign_gref has a real
+		 * value or not.
+		 */
+		struct xenvif *foreign_vif = NULL;
+		grant_ref_t foreign_gref;
+
+		if ((skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) &&
+			(ubuf->callback == &xenvif_zerocopy_callback)) {
+			const struct ubuf_info *const startpoint = ubuf;
+
+			/* Ideally ubuf points to the chain element which
+			 * belongs to this frag. Or if frags were removed from
+			 * the beginning, then shortly before it.
+			 */
+			ubuf = xenvif_find_gref(skb, i, ubuf);
+
+			/* Try again from the beginning of the list, if we
+			 * haven't tried from there. This only makes sense in
+			 * the unlikely event of reordering the original frags.
+			 * For injected local pages it's an unnecessary second
+			 * run.
+			 */
+			if (unlikely(!ubuf) && startpoint != head_ubuf)
+				ubuf = xenvif_find_gref(skb, i, head_ubuf);
+
+			if (likely(ubuf)) {
+				u16 pending_idx = ubuf->desc;
+
+				foreign_vif = ubuf_to_vif(ubuf);
+				foreign_gref = foreign_vif->pending_tx_info[pending_idx].req.gref;
+				/* Just a safety measure. If this was the last
+				 * element on the list, the for loop will
+				 * iterate again if a local page were added to
+				 * the end. Using head_ubuf here prevents the
+				 * second search on the chain. Or the original
+				 * frags changed order, but that's less likely.
+				 * In any way, ubuf shouldn't be NULL.
+				 */
+				ubuf = ubuf->ctx ?
+					(struct ubuf_info *) ubuf->ctx :
+					head_ubuf;
+			} else
+				/* This frag was a local page, added to the
+				 * array after the skb left netback.
+				 */
+				ubuf = head_ubuf;
+		}
 		xenvif_gop_frag_copy(vif, skb, npo,
 				     skb_frag_page(&skb_shinfo(skb)->frags[i]),
 				     skb_frag_size(&skb_shinfo(skb)->frags[i]),
 				     skb_shinfo(skb)->frags[i].page_offset,
 				     &head,
 				     foreign_vif,
-				     foreign_grefs[i]);
+				     foreign_vif ? foreign_gref : UINT_MAX);
 	}
 
 	return npo->meta_prod - old_meta_prod;
@@ -654,7 +716,7 @@
 		notify_remote_via_irq(vif->rx_irq);
 }
 
-void xenvif_check_rx_xenvif(struct xenvif *vif)
+void xenvif_napi_schedule_or_enable_events(struct xenvif *vif)
 {
 	int more_to_do;
 
@@ -688,7 +750,7 @@
 {
 	struct xenvif *vif = (struct xenvif *)data;
 	tx_add_credit(vif);
-	xenvif_check_rx_xenvif(vif);
+	xenvif_napi_schedule_or_enable_events(vif);
 }
 
 static void xenvif_tx_err(struct xenvif *vif,
diff --git a/drivers/ptp/Kconfig b/drivers/ptp/Kconfig
index 6963bdf..6aea373 100644
--- a/drivers/ptp/Kconfig
+++ b/drivers/ptp/Kconfig
@@ -6,6 +6,7 @@
 
 config PTP_1588_CLOCK
 	tristate "PTP clock support"
+	depends on NET
 	select PPS
 	select NET_PTP_CLASSIFY
 	help
@@ -74,7 +75,7 @@
 config PTP_1588_CLOCK_PCH
 	tristate "Intel PCH EG20T as PTP clock"
 	depends on X86 || COMPILE_TEST
-	depends on HAS_IOMEM
+	depends on HAS_IOMEM && NET
 	select PTP_1588_CLOCK
 	help
 	  This driver adds support for using the PCH EG20T as a PTP
diff --git a/drivers/scsi/scsi_transport_sas.c b/drivers/scsi/scsi_transport_sas.c
index 1b68142..c341f85 100644
--- a/drivers/scsi/scsi_transport_sas.c
+++ b/drivers/scsi/scsi_transport_sas.c
@@ -1621,8 +1621,6 @@
 	list_del(&rphy->list);
 	mutex_unlock(&sas_host->lock);
 
-	sas_bsg_remove(shost, rphy);
-
 	transport_destroy_device(dev);
 
 	put_device(dev);
@@ -1681,6 +1679,7 @@
 	}
 
 	sas_rphy_unlink(rphy);
+	sas_bsg_remove(NULL, rphy);
 	transport_remove_device(dev);
 	device_del(dev);
 }
diff --git a/fs/afs/cmservice.c b/fs/afs/cmservice.c
index 1c8c6cc..4b0eff6 100644
--- a/fs/afs/cmservice.c
+++ b/fs/afs/cmservice.c
@@ -130,6 +130,15 @@
 {
 	_enter("");
 
+	/* Break the callbacks here so that we do it after the final ACK is
+	 * received.  The step number here must match the final number in
+	 * afs_deliver_cb_callback().
+	 */
+	if (call->unmarshall == 6) {
+		ASSERT(call->server && call->count && call->request);
+		afs_break_callbacks(call->server, call->count, call->request);
+	}
+
 	afs_put_server(call->server);
 	call->server = NULL;
 	kfree(call->buffer);
@@ -272,6 +281,16 @@
 		_debug("trailer");
 		if (skb->len != 0)
 			return -EBADMSG;
+
+		/* Record that the message was unmarshalled successfully so
+		 * that the call destructor can know do the callback breaking
+		 * work, even if the final ACK isn't received.
+		 *
+		 * If the step number changes, then afs_cm_destructor() must be
+		 * updated also.
+		 */
+		call->unmarshall++;
+	case 6:
 		break;
 	}
 
diff --git a/fs/afs/internal.h b/fs/afs/internal.h
index be75b50..590b55f 100644
--- a/fs/afs/internal.h
+++ b/fs/afs/internal.h
@@ -75,7 +75,7 @@
 	const struct afs_call_type *type;	/* type of call */
 	const struct afs_wait_mode *wait_mode;	/* completion wait mode */
 	wait_queue_head_t	waitq;		/* processes awaiting completion */
-	work_func_t		async_workfn;
+	void (*async_workfn)(struct afs_call *call); /* asynchronous work function */
 	struct work_struct	async_work;	/* asynchronous work processor */
 	struct work_struct	work;		/* actual work processor */
 	struct sk_buff_head	rx_queue;	/* received packets */
diff --git a/fs/afs/rxrpc.c b/fs/afs/rxrpc.c
index ef943df..03a3beb1 100644
--- a/fs/afs/rxrpc.c
+++ b/fs/afs/rxrpc.c
@@ -25,7 +25,7 @@
 static int afs_wait_for_call_to_complete(struct afs_call *);
 static void afs_wake_up_async_call(struct afs_call *);
 static int afs_dont_wait_for_call_to_complete(struct afs_call *);
-static void afs_process_async_call(struct work_struct *);
+static void afs_process_async_call(struct afs_call *);
 static void afs_rx_interceptor(struct sock *, unsigned long, struct sk_buff *);
 static int afs_deliver_cm_op_id(struct afs_call *, struct sk_buff *, bool);
 
@@ -58,6 +58,13 @@
 static struct sk_buff_head afs_incoming_calls;
 static DECLARE_WORK(afs_collect_incoming_call_work, afs_collect_incoming_call);
 
+static void afs_async_workfn(struct work_struct *work)
+{
+	struct afs_call *call = container_of(work, struct afs_call, async_work);
+
+	call->async_workfn(call);
+}
+
 /*
  * open an RxRPC socket and bind it to be a server for callback notifications
  * - the socket is left in blocking mode and non-blocking ops use MSG_DONTWAIT
@@ -184,6 +191,28 @@
 }
 
 /*
+ * End a call but do not free it
+ */
+static void afs_end_call_nofree(struct afs_call *call)
+{
+	if (call->rxcall) {
+		rxrpc_kernel_end_call(call->rxcall);
+		call->rxcall = NULL;
+	}
+	if (call->type->destructor)
+		call->type->destructor(call);
+}
+
+/*
+ * End a call and free it
+ */
+static void afs_end_call(struct afs_call *call)
+{
+	afs_end_call_nofree(call);
+	afs_free_call(call);
+}
+
+/*
  * allocate a call with flat request and reply buffers
  */
 struct afs_call *afs_alloc_flat_call(const struct afs_call_type *type,
@@ -326,7 +355,8 @@
 	       atomic_read(&afs_outstanding_calls));
 
 	call->wait_mode = wait_mode;
-	INIT_WORK(&call->async_work, afs_process_async_call);
+	call->async_workfn = afs_process_async_call;
+	INIT_WORK(&call->async_work, afs_async_workfn);
 
 	memset(&srx, 0, sizeof(srx));
 	srx.srx_family = AF_RXRPC;
@@ -383,11 +413,8 @@
 	rxrpc_kernel_abort_call(rxcall, RX_USER_ABORT);
 	while ((skb = skb_dequeue(&call->rx_queue)))
 		afs_free_skb(skb);
-	rxrpc_kernel_end_call(rxcall);
-	call->rxcall = NULL;
 error_kill_call:
-	call->type->destructor(call);
-	afs_free_call(call);
+	afs_end_call(call);
 	_leave(" = %d", ret);
 	return ret;
 }
@@ -509,12 +536,8 @@
 	if (call->state >= AFS_CALL_COMPLETE) {
 		while ((skb = skb_dequeue(&call->rx_queue)))
 			afs_free_skb(skb);
-		if (call->incoming) {
-			rxrpc_kernel_end_call(call->rxcall);
-			call->rxcall = NULL;
-			call->type->destructor(call);
-			afs_free_call(call);
-		}
+		if (call->incoming)
+			afs_end_call(call);
 	}
 
 	_leave("");
@@ -564,10 +587,7 @@
 	}
 
 	_debug("call complete");
-	rxrpc_kernel_end_call(call->rxcall);
-	call->rxcall = NULL;
-	call->type->destructor(call);
-	afs_free_call(call);
+	afs_end_call(call);
 	_leave(" = %d", ret);
 	return ret;
 }
@@ -603,11 +623,8 @@
 /*
  * delete an asynchronous call
  */
-static void afs_delete_async_call(struct work_struct *work)
+static void afs_delete_async_call(struct afs_call *call)
 {
-	struct afs_call *call =
-		container_of(work, struct afs_call, async_work);
-
 	_enter("");
 
 	afs_free_call(call);
@@ -620,11 +637,8 @@
  * - on a multiple-thread workqueue this work item may try to run on several
  *   CPUs at the same time
  */
-static void afs_process_async_call(struct work_struct *work)
+static void afs_process_async_call(struct afs_call *call)
 {
-	struct afs_call *call =
-		container_of(work, struct afs_call, async_work);
-
 	_enter("");
 
 	if (!skb_queue_empty(&call->rx_queue))
@@ -637,10 +651,7 @@
 		call->reply = NULL;
 
 		/* kill the call */
-		rxrpc_kernel_end_call(call->rxcall);
-		call->rxcall = NULL;
-		if (call->type->destructor)
-			call->type->destructor(call);
+		afs_end_call_nofree(call);
 
 		/* we can't just delete the call because the work item may be
 		 * queued */
@@ -663,13 +674,6 @@
 	call->reply_size += len;
 }
 
-static void afs_async_workfn(struct work_struct *work)
-{
-	struct afs_call *call = container_of(work, struct afs_call, async_work);
-
-	call->async_workfn(work);
-}
-
 /*
  * accept the backlog of incoming calls
  */
@@ -790,10 +794,7 @@
 		_debug("oom");
 		rxrpc_kernel_abort_call(call->rxcall, RX_USER_ABORT);
 	default:
-		rxrpc_kernel_end_call(call->rxcall);
-		call->rxcall = NULL;
-		call->type->destructor(call);
-		afs_free_call(call);
+		afs_end_call(call);
 		_leave(" [error]");
 		return;
 	}
@@ -823,17 +824,16 @@
 	call->state = AFS_CALL_AWAIT_ACK;
 	n = rxrpc_kernel_send_data(call->rxcall, &msg, len);
 	if (n >= 0) {
+		/* Success */
 		_leave(" [replied]");
 		return;
 	}
+
 	if (n == -ENOMEM) {
 		_debug("oom");
 		rxrpc_kernel_abort_call(call->rxcall, RX_USER_ABORT);
 	}
-	rxrpc_kernel_end_call(call->rxcall);
-	call->rxcall = NULL;
-	call->type->destructor(call);
-	afs_free_call(call);
+	afs_end_call(call);
 	_leave(" [error]");
 }
 
diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c
index b6f4601..f66c66b 100644
--- a/fs/nfsd/nfs4acl.c
+++ b/fs/nfsd/nfs4acl.c
@@ -590,7 +590,7 @@
 		add_to_mask(state, &state->groups->aces[i].perms);
 	}
 
-	if (!state->users->n && !state->groups->n) {
+	if (state->users->n || state->groups->n) {
 		pace++;
 		pace->e_tag = ACL_MASK;
 		low_mode_from_nfs4(state->mask.allow, &pace->e_perm, flags);
diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c
index 32b699b..9a77a5a 100644
--- a/fs/nfsd/nfs4state.c
+++ b/fs/nfsd/nfs4state.c
@@ -3717,9 +3717,16 @@
 static __be32
 nfsd4_free_lock_stateid(struct nfs4_ol_stateid *stp)
 {
-	if (check_for_locks(stp->st_file, lockowner(stp->st_stateowner)))
+	struct nfs4_lockowner *lo = lockowner(stp->st_stateowner);
+
+	if (check_for_locks(stp->st_file, lo))
 		return nfserr_locks_held;
-	release_lock_stateid(stp);
+	/*
+	 * Currently there's a 1-1 lock stateid<->lockowner
+	 * correspondance, and we have to delete the lockowner when we
+	 * delete the lock stateid:
+	 */
+	unhash_lockowner(lo);
 	return nfs_ok;
 }
 
@@ -4159,6 +4166,10 @@
 
 	if (!same_owner_str(&lo->lo_owner, owner, clid))
 		return false;
+	if (list_empty(&lo->lo_owner.so_stateids)) {
+		WARN_ON_ONCE(1);
+		return false;
+	}
 	lst = list_first_entry(&lo->lo_owner.so_stateids,
 			       struct nfs4_ol_stateid, st_perstateowner);
 	return lst->st_file->fi_inode == inode;
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index af3f7aa..ee1f884 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -472,11 +472,15 @@
 
 void dlm_destroy_master_caches(void)
 {
-	if (dlm_lockname_cache)
+	if (dlm_lockname_cache) {
 		kmem_cache_destroy(dlm_lockname_cache);
+		dlm_lockname_cache = NULL;
+	}
 
-	if (dlm_lockres_cache)
+	if (dlm_lockres_cache) {
 		kmem_cache_destroy(dlm_lockres_cache);
+		dlm_lockres_cache = NULL;
+	}
 }
 
 static void dlm_lockres_release(struct kref *kref)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 8300fb8..72cb0dd 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -429,6 +429,7 @@
 typedef void (*dma_async_tx_callback)(void *dma_async_param);
 
 struct dmaengine_unmap_data {
+	u8 map_cnt;
 	u8 to_cnt;
 	u8 from_cnt;
 	u8 bidi_cnt;
diff --git a/include/linux/if_macvlan.h b/include/linux/if_macvlan.h
index 7c8b20b1..a9a53b1 100644
--- a/include/linux/if_macvlan.h
+++ b/include/linux/if_macvlan.h
@@ -56,6 +56,7 @@
 	int			numqueues;
 	netdev_features_t	tap_features;
 	int			minor;
+	int			nest_level;
 };
 
 static inline void macvlan_count_rx(const struct macvlan_dev *vlan,
diff --git a/include/linux/if_vlan.h b/include/linux/if_vlan.h
index 13bbbde..b2acc4a 100644
--- a/include/linux/if_vlan.h
+++ b/include/linux/if_vlan.h
@@ -73,7 +73,7 @@
 /* found in socket.c */
 extern void vlan_ioctl_set(int (*hook)(struct net *, void __user *));
 
-static inline int is_vlan_dev(struct net_device *dev)
+static inline bool is_vlan_dev(struct net_device *dev)
 {
         return dev->priv_flags & IFF_802_1Q_VLAN;
 }
@@ -159,6 +159,7 @@
 #ifdef CONFIG_NET_POLL_CONTROLLER
 	struct netpoll				*netpoll;
 #endif
+	unsigned int				nest_level;
 };
 
 static inline struct vlan_dev_priv *vlan_dev_priv(const struct net_device *dev)
@@ -197,6 +198,12 @@
 				 const struct net_device *by_dev);
 
 extern bool vlan_uses_dev(const struct net_device *dev);
+
+static inline int vlan_get_encap_level(struct net_device *dev)
+{
+	BUG_ON(!is_vlan_dev(dev));
+	return vlan_dev_priv(dev)->nest_level;
+}
 #else
 static inline struct net_device *
 __vlan_find_dev_deep(struct net_device *real_dev,
@@ -263,6 +270,11 @@
 {
 	return false;
 }
+static inline int vlan_get_encap_level(struct net_device *dev)
+{
+	BUG();
+	return 0;
+}
 #endif
 
 static inline bool vlan_hw_offload_capable(netdev_features_t features,
@@ -483,4 +495,5 @@
 		 */
 		skb->protocol = htons(ETH_P_802_2);
 }
+
 #endif /* !(_LINUX_IF_VLAN_H_) */
diff --git a/include/linux/mlx4/qp.h b/include/linux/mlx4/qp.h
index b66e761..7040dc9 100644
--- a/include/linux/mlx4/qp.h
+++ b/include/linux/mlx4/qp.h
@@ -421,6 +421,17 @@
 	__be32			byte_count;
 };
 
+enum mlx4_update_qp_attr {
+	MLX4_UPDATE_QP_SMAC		= 1 << 0,
+};
+
+struct mlx4_update_qp_params {
+	u8	smac_index;
+};
+
+int mlx4_update_qp(struct mlx4_dev *dev, struct mlx4_qp *qp,
+		   enum mlx4_update_qp_attr attr,
+		   struct mlx4_update_qp_params *params);
 int mlx4_qp_modify(struct mlx4_dev *dev, struct mlx4_mtt *mtt,
 		   enum mlx4_qp_state cur_state, enum mlx4_qp_state new_state,
 		   struct mlx4_qp_context *context, enum mlx4_qp_optpar optpar,
diff --git a/include/linux/net.h b/include/linux/net.h
index 94734a6..17d8339 100644
--- a/include/linux/net.h
+++ b/include/linux/net.h
@@ -248,24 +248,17 @@
 bool __net_get_random_once(void *buf, int nbytes, bool *done,
 			   struct static_key *done_key);
 
-#ifdef HAVE_JUMP_LABEL
-#define ___NET_RANDOM_STATIC_KEY_INIT ((struct static_key) \
-		{ .enabled = ATOMIC_INIT(0), .entries = (void *)1 })
-#else /* !HAVE_JUMP_LABEL */
-#define ___NET_RANDOM_STATIC_KEY_INIT STATIC_KEY_INIT_FALSE
-#endif /* HAVE_JUMP_LABEL */
-
 #define net_get_random_once(buf, nbytes)				\
 	({								\
 		bool ___ret = false;					\
 		static bool ___done = false;				\
-		static struct static_key ___done_key =			\
-			___NET_RANDOM_STATIC_KEY_INIT;			\
-		if (!static_key_true(&___done_key))			\
+		static struct static_key ___once_key =			\
+			STATIC_KEY_INIT_TRUE;				\
+		if (static_key_true(&___once_key))			\
 			___ret = __net_get_random_once(buf,		\
 						       nbytes,		\
 						       &___done,	\
-						       &___done_key);	\
+						       &___once_key);	\
 		___ret;							\
 	})
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 7ed3a3a..b42d07b 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1144,6 +1144,7 @@
 	netdev_tx_t		(*ndo_dfwd_start_xmit) (struct sk_buff *skb,
 							struct net_device *dev,
 							void *priv);
+	int			(*ndo_get_lock_subclass)(struct net_device *dev);
 };
 
 /**
@@ -2950,7 +2951,12 @@
 
 static inline void netif_addr_lock_nested(struct net_device *dev)
 {
-	spin_lock_nested(&dev->addr_list_lock, SINGLE_DEPTH_NESTING);
+	int subclass = SINGLE_DEPTH_NESTING;
+
+	if (dev->netdev_ops->ndo_get_lock_subclass)
+		subclass = dev->netdev_ops->ndo_get_lock_subclass(dev);
+
+	spin_lock_nested(&dev->addr_list_lock, subclass);
 }
 
 static inline void netif_addr_lock_bh(struct net_device *dev)
@@ -3050,10 +3056,19 @@
 extern int		bpf_jit_enable;
 
 bool netdev_has_upper_dev(struct net_device *dev, struct net_device *upper_dev);
+struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev,
+						     struct list_head **iter);
 struct net_device *netdev_all_upper_get_next_dev_rcu(struct net_device *dev,
 						     struct list_head **iter);
 
 /* iterate through upper list, must be called under RCU read lock */
+#define netdev_for_each_upper_dev_rcu(dev, updev, iter) \
+	for (iter = &(dev)->adj_list.upper, \
+	     updev = netdev_upper_get_next_dev_rcu(dev, &(iter)); \
+	     updev; \
+	     updev = netdev_upper_get_next_dev_rcu(dev, &(iter)))
+
+/* iterate through upper list, must be called under RCU read lock */
 #define netdev_for_each_all_upper_dev_rcu(dev, updev, iter) \
 	for (iter = &(dev)->all_adj_list.upper, \
 	     updev = netdev_all_upper_get_next_dev_rcu(dev, &(iter)); \
@@ -3077,6 +3092,14 @@
 	     priv; \
 	     priv = netdev_lower_get_next_private_rcu(dev, &(iter)))
 
+void *netdev_lower_get_next(struct net_device *dev,
+				struct list_head **iter);
+#define netdev_for_each_lower_dev(dev, ldev, iter) \
+	for (iter = &(dev)->adj_list.lower, \
+	     ldev = netdev_lower_get_next(dev, &(iter)); \
+	     ldev; \
+	     ldev = netdev_lower_get_next(dev, &(iter)))
+
 void *netdev_adjacent_get_private(struct list_head *adj_list);
 void *netdev_lower_get_first_private_rcu(struct net_device *dev);
 struct net_device *netdev_master_upper_dev_get(struct net_device *dev);
@@ -3092,6 +3115,8 @@
 void netdev_adjacent_rename_links(struct net_device *dev, char *oldname);
 void *netdev_lower_dev_get_private(struct net_device *dev,
 				   struct net_device *lower_dev);
+int dev_get_nest_level(struct net_device *dev,
+		       bool (*type_check)(struct net_device *dev));
 int skb_checksum_help(struct sk_buff *skb);
 struct sk_buff *__skb_gso_segment(struct sk_buff *skb,
 				  netdev_features_t features, bool tx_path);
@@ -3180,12 +3205,7 @@
 void netif_stacked_transfer_operstate(const struct net_device *rootdev,
 					struct net_device *dev);
 
-netdev_features_t netif_skb_dev_features(struct sk_buff *skb,
-					 const struct net_device *dev);
-static inline netdev_features_t netif_skb_features(struct sk_buff *skb)
-{
-	return netif_skb_dev_features(skb, skb->dev);
-}
+netdev_features_t netif_skb_features(struct sk_buff *skb);
 
 static inline bool net_gso_ok(netdev_features_t features, int gso_type)
 {
diff --git a/include/linux/of_mdio.h b/include/linux/of_mdio.h
index 6fe8464..881a7c3 100644
--- a/include/linux/of_mdio.h
+++ b/include/linux/of_mdio.h
@@ -31,7 +31,12 @@
 #else /* CONFIG_OF */
 static inline int of_mdiobus_register(struct mii_bus *mdio, struct device_node *np)
 {
-	return -ENOSYS;
+	/*
+	 * Fall back to the non-DT function to register a bus.
+	 * This way, we don't have to keep compat bits around in drivers.
+	 */
+
+	return mdiobus_register(mdio);
 }
 
 static inline struct phy_device *of_phy_find_device(struct device_node *phy_np)
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 3356abc..3ef6ea1 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -402,6 +402,8 @@
 
 	struct ring_buffer		*rb;
 	struct list_head		rb_entry;
+	unsigned long			rcu_batches;
+	int				rcu_pending;
 
 	/* poll related */
 	wait_queue_head_t		waitq;
diff --git a/include/linux/platform_data/ipmmu-vmsa.h b/include/linux/platform_data/ipmmu-vmsa.h
new file mode 100644
index 0000000..5275b3a
--- /dev/null
+++ b/include/linux/platform_data/ipmmu-vmsa.h
@@ -0,0 +1,24 @@
+/*
+ * IPMMU VMSA Platform Data
+ *
+ * Copyright (C) 2014 Renesas Electronics Corporation
+ *
+ * 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; version 2 of the License.
+ */
+
+#ifndef __IPMMU_VMSA_H__
+#define __IPMMU_VMSA_H__
+
+struct ipmmu_vmsa_master {
+	const char *name;
+	unsigned int utlb;
+};
+
+struct ipmmu_vmsa_platform_data {
+	const struct ipmmu_vmsa_master *masters;
+	unsigned int num_masters;
+};
+
+#endif /* __IPMMU_VMSA_H__ */
diff --git a/include/linux/rtnetlink.h b/include/linux/rtnetlink.h
index 8e3e66a..953937e 100644
--- a/include/linux/rtnetlink.h
+++ b/include/linux/rtnetlink.h
@@ -4,6 +4,7 @@
 
 #include <linux/mutex.h>
 #include <linux/netdevice.h>
+#include <linux/wait.h>
 #include <uapi/linux/rtnetlink.h>
 
 extern int rtnetlink_send(struct sk_buff *skb, struct net *net, u32 pid, u32 group, int echo);
@@ -22,6 +23,10 @@
 extern void rtnl_unlock(void);
 extern int rtnl_trylock(void);
 extern int rtnl_is_locked(void);
+
+extern wait_queue_head_t netdev_unregistering_wq;
+extern struct mutex net_mutex;
+
 #ifdef CONFIG_PROVE_LOCKING
 extern int lockdep_rtnl_is_held(void);
 #else
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 25f54c7..221b2bd 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -220,7 +220,7 @@
 #define TASK_PARKED		512
 #define TASK_STATE_MAX		1024
 
-#define TASK_STATE_TO_CHAR_STR "RSDTtZXxKWP"
+#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWP"
 
 extern char ___assert_task_state[1 - 2*!!(
 		sizeof(TASK_STATE_TO_CHAR_STR)-1 != ilog2(TASK_STATE_MAX)+1)];
@@ -1153,9 +1153,12 @@
 	 *
 	 * @dl_boosted tells if we are boosted due to DI. If so we are
 	 * outside bandwidth enforcement mechanism (but only until we
-	 * exit the critical section).
+	 * exit the critical section);
+	 *
+	 * @dl_yielded tells if task gave up the cpu before consuming
+	 * all its available runtime during the last job.
 	 */
-	int dl_throttled, dl_new, dl_boosted;
+	int dl_throttled, dl_new, dl_boosted, dl_yielded;
 
 	/*
 	 * Bandwidth enforcement timer. Each -deadline task has its
diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h
index f3539a1..f856e5a 100644
--- a/include/net/cfg80211.h
+++ b/include/net/cfg80211.h
@@ -3669,6 +3669,18 @@
 void cfg80211_sched_scan_stopped(struct wiphy *wiphy);
 
 /**
+ * cfg80211_sched_scan_stopped_rtnl - notify that the scheduled scan has stopped
+ *
+ * @wiphy: the wiphy on which the scheduled scan stopped
+ *
+ * The driver can call this function to inform cfg80211 that the
+ * scheduled scan had to be stopped, for whatever reason.  The driver
+ * is then called back via the sched_scan_stop operation when done.
+ * This function should be called with rtnl locked.
+ */
+void cfg80211_sched_scan_stopped_rtnl(struct wiphy *wiphy);
+
+/**
  * cfg80211_inform_bss_width_frame - inform cfg80211 of a received BSS frame
  *
  * @wiphy: the wiphy reporting the BSS
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index 6c4f5ea..216cecc 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -127,6 +127,7 @@
 void rt6_ifdown(struct net *net, struct net_device *dev);
 void rt6_mtu_change(struct net_device *dev, unsigned int mtu);
 void rt6_remove_prefsrc(struct inet6_ifaddr *ifp);
+void rt6_clean_tohost(struct net *net, struct in6_addr *gateway);
 
 
 /*
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index 80f500a..b2704fd0 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -20,6 +20,11 @@
 	int		range[2];
 };
 
+struct ping_group_range {
+	seqlock_t	lock;
+	kgid_t		range[2];
+};
+
 struct netns_ipv4 {
 #ifdef CONFIG_SYSCTL
 	struct ctl_table_header	*forw_hdr;
@@ -66,13 +71,13 @@
 	int sysctl_icmp_ratemask;
 	int sysctl_icmp_errors_use_inbound_ifaddr;
 
-	struct local_ports sysctl_local_ports;
+	struct local_ports ip_local_ports;
 
 	int sysctl_tcp_ecn;
 	int sysctl_ip_no_pmtu_disc;
 	int sysctl_ip_fwd_use_pmtu;
 
-	kgid_t sysctl_ping_group_range[2];
+	struct ping_group_range ping_group_range;
 
 	atomic_t dev_addr_genid;
 
diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h
index 1ba9d62..194c1ea 100644
--- a/include/uapi/linux/nl80211.h
+++ b/include/uapi/linux/nl80211.h
@@ -3856,6 +3856,8 @@
  * @NL80211_FEATURE_CELL_BASE_REG_HINTS: This driver has been tested
  *	to work properly to suppport receiving regulatory hints from
  *	cellular base stations.
+ * @NL80211_FEATURE_P2P_DEVICE_NEEDS_CHANNEL: (no longer available, only
+ *	here to reserve the value for API/ABI compatibility)
  * @NL80211_FEATURE_SAE: This driver supports simultaneous authentication of
  *	equals (SAE) with user space SME (NL80211_CMD_AUTHENTICATE) in station
  *	mode
@@ -3897,7 +3899,7 @@
 	NL80211_FEATURE_HT_IBSS				= 1 << 1,
 	NL80211_FEATURE_INACTIVITY_TIMER		= 1 << 2,
 	NL80211_FEATURE_CELL_BASE_REG_HINTS		= 1 << 3,
-	/* bit 4 is reserved - don't use */
+	NL80211_FEATURE_P2P_DEVICE_NEEDS_CHANNEL	= 1 << 4,
 	NL80211_FEATURE_SAE				= 1 << 5,
 	NL80211_FEATURE_LOW_PRIORITY_SCAN		= 1 << 6,
 	NL80211_FEATURE_SCAN_FLUSH			= 1 << 7,
diff --git a/kernel/events/core.c b/kernel/events/core.c
index f83a71a..440eefc 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -1443,6 +1443,11 @@
 		cpuctx->exclusive = 0;
 }
 
+struct remove_event {
+	struct perf_event *event;
+	bool detach_group;
+};
+
 /*
  * Cross CPU call to remove a performance event
  *
@@ -1451,12 +1456,15 @@
  */
 static int __perf_remove_from_context(void *info)
 {
-	struct perf_event *event = info;
+	struct remove_event *re = info;
+	struct perf_event *event = re->event;
 	struct perf_event_context *ctx = event->ctx;
 	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 
 	raw_spin_lock(&ctx->lock);
 	event_sched_out(event, cpuctx, ctx);
+	if (re->detach_group)
+		perf_group_detach(event);
 	list_del_event(event, ctx);
 	if (!ctx->nr_events && cpuctx->task_ctx == ctx) {
 		ctx->is_active = 0;
@@ -1481,10 +1489,14 @@
  * When called from perf_event_exit_task, it's OK because the
  * context has been detached from its task.
  */
-static void perf_remove_from_context(struct perf_event *event)
+static void perf_remove_from_context(struct perf_event *event, bool detach_group)
 {
 	struct perf_event_context *ctx = event->ctx;
 	struct task_struct *task = ctx->task;
+	struct remove_event re = {
+		.event = event,
+		.detach_group = detach_group,
+	};
 
 	lockdep_assert_held(&ctx->mutex);
 
@@ -1493,12 +1505,12 @@
 		 * Per cpu events are removed via an smp call and
 		 * the removal is always successful.
 		 */
-		cpu_function_call(event->cpu, __perf_remove_from_context, event);
+		cpu_function_call(event->cpu, __perf_remove_from_context, &re);
 		return;
 	}
 
 retry:
-	if (!task_function_call(task, __perf_remove_from_context, event))
+	if (!task_function_call(task, __perf_remove_from_context, &re))
 		return;
 
 	raw_spin_lock_irq(&ctx->lock);
@@ -1515,6 +1527,8 @@
 	 * Since the task isn't running, its safe to remove the event, us
 	 * holding the ctx->lock ensures the task won't get scheduled in.
 	 */
+	if (detach_group)
+		perf_group_detach(event);
 	list_del_event(event, ctx);
 	raw_spin_unlock_irq(&ctx->lock);
 }
@@ -3178,7 +3192,8 @@
 }
 
 static void ring_buffer_put(struct ring_buffer *rb);
-static void ring_buffer_detach(struct perf_event *event, struct ring_buffer *rb);
+static void ring_buffer_attach(struct perf_event *event,
+			       struct ring_buffer *rb);
 
 static void unaccount_event_cpu(struct perf_event *event, int cpu)
 {
@@ -3238,8 +3253,6 @@
 	unaccount_event(event);
 
 	if (event->rb) {
-		struct ring_buffer *rb;
-
 		/*
 		 * Can happen when we close an event with re-directed output.
 		 *
@@ -3247,12 +3260,7 @@
 		 * over us; possibly making our ring_buffer_put() the last.
 		 */
 		mutex_lock(&event->mmap_mutex);
-		rb = event->rb;
-		if (rb) {
-			rcu_assign_pointer(event->rb, NULL);
-			ring_buffer_detach(event, rb);
-			ring_buffer_put(rb); /* could be last */
-		}
+		ring_buffer_attach(event, NULL);
 		mutex_unlock(&event->mmap_mutex);
 	}
 
@@ -3281,10 +3289,7 @@
 	 *     to trigger the AB-BA case.
 	 */
 	mutex_lock_nested(&ctx->mutex, SINGLE_DEPTH_NESTING);
-	raw_spin_lock_irq(&ctx->lock);
-	perf_group_detach(event);
-	raw_spin_unlock_irq(&ctx->lock);
-	perf_remove_from_context(event);
+	perf_remove_from_context(event, true);
 	mutex_unlock(&ctx->mutex);
 
 	free_event(event);
@@ -3839,28 +3844,47 @@
 static void ring_buffer_attach(struct perf_event *event,
 			       struct ring_buffer *rb)
 {
+	struct ring_buffer *old_rb = NULL;
 	unsigned long flags;
 
-	if (!list_empty(&event->rb_entry))
-		return;
+	if (event->rb) {
+		/*
+		 * Should be impossible, we set this when removing
+		 * event->rb_entry and wait/clear when adding event->rb_entry.
+		 */
+		WARN_ON_ONCE(event->rcu_pending);
 
-	spin_lock_irqsave(&rb->event_lock, flags);
-	if (list_empty(&event->rb_entry))
-		list_add(&event->rb_entry, &rb->event_list);
-	spin_unlock_irqrestore(&rb->event_lock, flags);
-}
+		old_rb = event->rb;
+		event->rcu_batches = get_state_synchronize_rcu();
+		event->rcu_pending = 1;
 
-static void ring_buffer_detach(struct perf_event *event, struct ring_buffer *rb)
-{
-	unsigned long flags;
+		spin_lock_irqsave(&old_rb->event_lock, flags);
+		list_del_rcu(&event->rb_entry);
+		spin_unlock_irqrestore(&old_rb->event_lock, flags);
+	}
 
-	if (list_empty(&event->rb_entry))
-		return;
+	if (event->rcu_pending && rb) {
+		cond_synchronize_rcu(event->rcu_batches);
+		event->rcu_pending = 0;
+	}
 
-	spin_lock_irqsave(&rb->event_lock, flags);
-	list_del_init(&event->rb_entry);
-	wake_up_all(&event->waitq);
-	spin_unlock_irqrestore(&rb->event_lock, flags);
+	if (rb) {
+		spin_lock_irqsave(&rb->event_lock, flags);
+		list_add_rcu(&event->rb_entry, &rb->event_list);
+		spin_unlock_irqrestore(&rb->event_lock, flags);
+	}
+
+	rcu_assign_pointer(event->rb, rb);
+
+	if (old_rb) {
+		ring_buffer_put(old_rb);
+		/*
+		 * Since we detached before setting the new rb, so that we
+		 * could attach the new rb, we could have missed a wakeup.
+		 * Provide it now.
+		 */
+		wake_up_all(&event->waitq);
+	}
 }
 
 static void ring_buffer_wakeup(struct perf_event *event)
@@ -3929,7 +3953,7 @@
 {
 	struct perf_event *event = vma->vm_file->private_data;
 
-	struct ring_buffer *rb = event->rb;
+	struct ring_buffer *rb = ring_buffer_get(event);
 	struct user_struct *mmap_user = rb->mmap_user;
 	int mmap_locked = rb->mmap_locked;
 	unsigned long size = perf_data_size(rb);
@@ -3937,18 +3961,14 @@
 	atomic_dec(&rb->mmap_count);
 
 	if (!atomic_dec_and_mutex_lock(&event->mmap_count, &event->mmap_mutex))
-		return;
+		goto out_put;
 
-	/* Detach current event from the buffer. */
-	rcu_assign_pointer(event->rb, NULL);
-	ring_buffer_detach(event, rb);
+	ring_buffer_attach(event, NULL);
 	mutex_unlock(&event->mmap_mutex);
 
 	/* If there's still other mmap()s of this buffer, we're done. */
-	if (atomic_read(&rb->mmap_count)) {
-		ring_buffer_put(rb); /* can't be last */
-		return;
-	}
+	if (atomic_read(&rb->mmap_count))
+		goto out_put;
 
 	/*
 	 * No other mmap()s, detach from all other events that might redirect
@@ -3978,11 +3998,9 @@
 		 * still restart the iteration to make sure we're not now
 		 * iterating the wrong list.
 		 */
-		if (event->rb == rb) {
-			rcu_assign_pointer(event->rb, NULL);
-			ring_buffer_detach(event, rb);
-			ring_buffer_put(rb); /* can't be last, we still have one */
-		}
+		if (event->rb == rb)
+			ring_buffer_attach(event, NULL);
+
 		mutex_unlock(&event->mmap_mutex);
 		put_event(event);
 
@@ -4007,6 +4025,7 @@
 	vma->vm_mm->pinned_vm -= mmap_locked;
 	free_uid(mmap_user);
 
+out_put:
 	ring_buffer_put(rb); /* could be last */
 }
 
@@ -4124,7 +4143,6 @@
 	vma->vm_mm->pinned_vm += extra;
 
 	ring_buffer_attach(event, rb);
-	rcu_assign_pointer(event->rb, rb);
 
 	perf_event_init_userpage(event);
 	perf_event_update_userpage(event);
@@ -5408,6 +5426,9 @@
 
 	/* Recursion avoidance in each contexts */
 	int				recursion[PERF_NR_CONTEXTS];
+
+	/* Keeps track of cpu being initialized/exited */
+	bool				online;
 };
 
 static DEFINE_PER_CPU(struct swevent_htable, swevent_htable);
@@ -5654,8 +5675,14 @@
 	hwc->state = !(flags & PERF_EF_START);
 
 	head = find_swevent_head(swhash, event);
-	if (WARN_ON_ONCE(!head))
+	if (!head) {
+		/*
+		 * We can race with cpu hotplug code. Do not
+		 * WARN if the cpu just got unplugged.
+		 */
+		WARN_ON_ONCE(swhash->online);
 		return -EINVAL;
+	}
 
 	hlist_add_head_rcu(&event->hlist_entry, head);
 
@@ -6914,7 +6941,7 @@
 static int
 perf_event_set_output(struct perf_event *event, struct perf_event *output_event)
 {
-	struct ring_buffer *rb = NULL, *old_rb = NULL;
+	struct ring_buffer *rb = NULL;
 	int ret = -EINVAL;
 
 	if (!output_event)
@@ -6942,8 +6969,6 @@
 	if (atomic_read(&event->mmap_count))
 		goto unlock;
 
-	old_rb = event->rb;
-
 	if (output_event) {
 		/* get the rb we want to redirect to */
 		rb = ring_buffer_get(output_event);
@@ -6951,23 +6976,7 @@
 			goto unlock;
 	}
 
-	if (old_rb)
-		ring_buffer_detach(event, old_rb);
-
-	if (rb)
-		ring_buffer_attach(event, rb);
-
-	rcu_assign_pointer(event->rb, rb);
-
-	if (old_rb) {
-		ring_buffer_put(old_rb);
-		/*
-		 * Since we detached before setting the new rb, so that we
-		 * could attach the new rb, we could have missed a wakeup.
-		 * Provide it now.
-		 */
-		wake_up_all(&event->waitq);
-	}
+	ring_buffer_attach(event, rb);
 
 	ret = 0;
 unlock:
@@ -7018,6 +7027,9 @@
 	if (attr.freq) {
 		if (attr.sample_freq > sysctl_perf_event_sample_rate)
 			return -EINVAL;
+	} else {
+		if (attr.sample_period & (1ULL << 63))
+			return -EINVAL;
 	}
 
 	/*
@@ -7165,7 +7177,7 @@
 		struct perf_event_context *gctx = group_leader->ctx;
 
 		mutex_lock(&gctx->mutex);
-		perf_remove_from_context(group_leader);
+		perf_remove_from_context(group_leader, false);
 
 		/*
 		 * Removing from the context ends up with disabled
@@ -7175,7 +7187,7 @@
 		perf_event__state_init(group_leader);
 		list_for_each_entry(sibling, &group_leader->sibling_list,
 				    group_entry) {
-			perf_remove_from_context(sibling);
+			perf_remove_from_context(sibling, false);
 			perf_event__state_init(sibling);
 			put_ctx(gctx);
 		}
@@ -7305,7 +7317,7 @@
 	mutex_lock(&src_ctx->mutex);
 	list_for_each_entry_safe(event, tmp, &src_ctx->event_list,
 				 event_entry) {
-		perf_remove_from_context(event);
+		perf_remove_from_context(event, false);
 		unaccount_event_cpu(event, src_cpu);
 		put_ctx(src_ctx);
 		list_add(&event->migrate_entry, &events);
@@ -7367,13 +7379,7 @@
 			 struct perf_event_context *child_ctx,
 			 struct task_struct *child)
 {
-	if (child_event->parent) {
-		raw_spin_lock_irq(&child_ctx->lock);
-		perf_group_detach(child_event);
-		raw_spin_unlock_irq(&child_ctx->lock);
-	}
-
-	perf_remove_from_context(child_event);
+	perf_remove_from_context(child_event, !!child_event->parent);
 
 	/*
 	 * It can happen that the parent exits first, and has events
@@ -7724,6 +7730,8 @@
 	 * swapped under us.
 	 */
 	parent_ctx = perf_pin_task_context(parent, ctxn);
+	if (!parent_ctx)
+		return 0;
 
 	/*
 	 * No need to check if parent_ctx != NULL here; since we saw
@@ -7835,6 +7843,7 @@
 	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
 	mutex_lock(&swhash->hlist_mutex);
+	swhash->online = true;
 	if (swhash->hlist_refcount > 0) {
 		struct swevent_hlist *hlist;
 
@@ -7857,14 +7866,14 @@
 
 static void __perf_event_exit_context(void *__info)
 {
+	struct remove_event re = { .detach_group = false };
 	struct perf_event_context *ctx = __info;
-	struct perf_event *event;
 
 	perf_pmu_rotate_stop(ctx->pmu);
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(event, &ctx->event_list, event_entry)
-		__perf_remove_from_context(event);
+	list_for_each_entry_rcu(re.event, &ctx->event_list, event_entry)
+		__perf_remove_from_context(&re);
 	rcu_read_unlock();
 }
 
@@ -7892,6 +7901,7 @@
 	perf_event_exit_cpu_context(cpu);
 
 	mutex_lock(&swhash->hlist_mutex);
+	swhash->online = false;
 	swevent_hlist_release(swhash);
 	mutex_unlock(&swhash->hlist_mutex);
 }
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index d9d8ece..204d3d2 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2592,8 +2592,14 @@
 	if (likely(prev->sched_class == class &&
 		   rq->nr_running == rq->cfs.h_nr_running)) {
 		p = fair_sched_class.pick_next_task(rq, prev);
-		if (likely(p && p != RETRY_TASK))
-			return p;
+		if (unlikely(p == RETRY_TASK))
+			goto again;
+
+		/* assumes fair_sched_class->next == idle_sched_class */
+		if (unlikely(!p))
+			p = idle_sched_class.pick_next_task(rq, prev);
+
+		return p;
 	}
 
 again:
@@ -3124,6 +3130,7 @@
 	dl_se->dl_bw = to_ratio(dl_se->dl_period, dl_se->dl_runtime);
 	dl_se->dl_throttled = 0;
 	dl_se->dl_new = 1;
+	dl_se->dl_yielded = 0;
 }
 
 static void __setscheduler_params(struct task_struct *p,
@@ -3639,6 +3646,7 @@
  * sys_sched_setattr - same as above, but with extended sched_attr
  * @pid: the pid in question.
  * @uattr: structure containing the extended parameters.
+ * @flags: for future extension.
  */
 SYSCALL_DEFINE3(sched_setattr, pid_t, pid, struct sched_attr __user *, uattr,
 			       unsigned int, flags)
@@ -3783,6 +3791,7 @@
  * @pid: the pid in question.
  * @uattr: structure containing the extended parameters.
  * @size: sizeof(attr) for fwd/bwd comp.
+ * @flags: for future extension.
  */
 SYSCALL_DEFINE4(sched_getattr, pid_t, pid, struct sched_attr __user *, uattr,
 		unsigned int, size, unsigned int, flags)
@@ -6017,6 +6026,8 @@
 					,
 		.last_balance		= jiffies,
 		.balance_interval	= sd_weight,
+		.max_newidle_lb_cost	= 0,
+		.next_decay_max_lb_cost	= jiffies,
 	};
 	SD_INIT_NAME(sd, NUMA);
 	sd->private = &tl->data;
diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
index 5b9bb42..ab001b5 100644
--- a/kernel/sched/cpudeadline.c
+++ b/kernel/sched/cpudeadline.c
@@ -210,7 +210,5 @@
  */
 void cpudl_cleanup(struct cpudl *cp)
 {
-	/*
-	 * nothing to do for the moment
-	 */
+	free_cpumask_var(cp->free_cpus);
 }
diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c
index 8b836b3..3031bac 100644
--- a/kernel/sched/cpupri.c
+++ b/kernel/sched/cpupri.c
@@ -70,8 +70,7 @@
 	int idx = 0;
 	int task_pri = convert_prio(p->prio);
 
-	if (task_pri >= MAX_RT_PRIO)
-		return 0;
+	BUG_ON(task_pri >= CPUPRI_NR_PRIORITIES);
 
 	for (idx = 0; idx < task_pri; idx++) {
 		struct cpupri_vec *vec  = &cp->pri_to_cpu[idx];
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index a95097c..72fdf06 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -332,50 +332,50 @@
  * softirq as those do not count in task exec_runtime any more.
  */
 static void irqtime_account_process_tick(struct task_struct *p, int user_tick,
-						struct rq *rq)
+					 struct rq *rq, int ticks)
 {
-	cputime_t one_jiffy_scaled = cputime_to_scaled(cputime_one_jiffy);
+	cputime_t scaled = cputime_to_scaled(cputime_one_jiffy);
+	u64 cputime = (__force u64) cputime_one_jiffy;
 	u64 *cpustat = kcpustat_this_cpu->cpustat;
 
 	if (steal_account_process_tick())
 		return;
 
+	cputime *= ticks;
+	scaled *= ticks;
+
 	if (irqtime_account_hi_update()) {
-		cpustat[CPUTIME_IRQ] += (__force u64) cputime_one_jiffy;
+		cpustat[CPUTIME_IRQ] += cputime;
 	} else if (irqtime_account_si_update()) {
-		cpustat[CPUTIME_SOFTIRQ] += (__force u64) cputime_one_jiffy;
+		cpustat[CPUTIME_SOFTIRQ] += cputime;
 	} else if (this_cpu_ksoftirqd() == p) {
 		/*
 		 * ksoftirqd time do not get accounted in cpu_softirq_time.
 		 * So, we have to handle it separately here.
 		 * Also, p->stime needs to be updated for ksoftirqd.
 		 */
-		__account_system_time(p, cputime_one_jiffy, one_jiffy_scaled,
-					CPUTIME_SOFTIRQ);
+		__account_system_time(p, cputime, scaled, CPUTIME_SOFTIRQ);
 	} else if (user_tick) {
-		account_user_time(p, cputime_one_jiffy, one_jiffy_scaled);
+		account_user_time(p, cputime, scaled);
 	} else if (p == rq->idle) {
-		account_idle_time(cputime_one_jiffy);
+		account_idle_time(cputime);
 	} else if (p->flags & PF_VCPU) { /* System time or guest time */
-		account_guest_time(p, cputime_one_jiffy, one_jiffy_scaled);
+		account_guest_time(p, cputime, scaled);
 	} else {
-		__account_system_time(p, cputime_one_jiffy, one_jiffy_scaled,
-					CPUTIME_SYSTEM);
+		__account_system_time(p, cputime, scaled,	CPUTIME_SYSTEM);
 	}
 }
 
 static void irqtime_account_idle_ticks(int ticks)
 {
-	int i;
 	struct rq *rq = this_rq();
 
-	for (i = 0; i < ticks; i++)
-		irqtime_account_process_tick(current, 0, rq);
+	irqtime_account_process_tick(current, 0, rq, ticks);
 }
 #else /* CONFIG_IRQ_TIME_ACCOUNTING */
 static inline void irqtime_account_idle_ticks(int ticks) {}
 static inline void irqtime_account_process_tick(struct task_struct *p, int user_tick,
-						struct rq *rq) {}
+						struct rq *rq, int nr_ticks) {}
 #endif /* CONFIG_IRQ_TIME_ACCOUNTING */
 
 /*
@@ -464,7 +464,7 @@
 		return;
 
 	if (sched_clock_irqtime) {
-		irqtime_account_process_tick(p, user_tick, rq);
+		irqtime_account_process_tick(p, user_tick, rq, 1);
 		return;
 	}
 
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index b080957..800e99b 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -528,6 +528,7 @@
 	sched_clock_tick();
 	update_rq_clock(rq);
 	dl_se->dl_throttled = 0;
+	dl_se->dl_yielded = 0;
 	if (p->on_rq) {
 		enqueue_task_dl(rq, p, ENQUEUE_REPLENISH);
 		if (task_has_dl_policy(rq->curr))
@@ -893,10 +894,10 @@
 	 * We make the task go to sleep until its current deadline by
 	 * forcing its runtime to zero. This way, update_curr_dl() stops
 	 * it and the bandwidth timer will wake it up and will give it
-	 * new scheduling parameters (thanks to dl_new=1).
+	 * new scheduling parameters (thanks to dl_yielded=1).
 	 */
 	if (p->dl.runtime > 0) {
-		rq->curr->dl.dl_new = 1;
+		rq->curr->dl.dl_yielded = 1;
 		p->dl.runtime = 0;
 	}
 	update_curr_dl(rq);
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 7570dd9..0fdb96d 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6653,6 +6653,7 @@
 	int this_cpu = this_rq->cpu;
 
 	idle_enter_fair(this_rq);
+
 	/*
 	 * We must set idle_stamp _before_ calling idle_balance(), such that we
 	 * measure the duration of idle_balance() as idle time.
@@ -6705,14 +6706,16 @@
 
 	raw_spin_lock(&this_rq->lock);
 
+	if (curr_cost > this_rq->max_idle_balance_cost)
+		this_rq->max_idle_balance_cost = curr_cost;
+
 	/*
-	 * While browsing the domains, we released the rq lock.
-	 * A task could have be enqueued in the meantime
+	 * While browsing the domains, we released the rq lock, a task could
+	 * have been enqueued in the meantime. Since we're not going idle,
+	 * pretend we pulled a task.
 	 */
-	if (this_rq->cfs.h_nr_running && !pulled_task) {
+	if (this_rq->cfs.h_nr_running && !pulled_task)
 		pulled_task = 1;
-		goto out;
-	}
 
 	if (pulled_task || time_after(jiffies, this_rq->next_balance)) {
 		/*
@@ -6722,9 +6725,6 @@
 		this_rq->next_balance = next_balance;
 	}
 
-	if (curr_cost > this_rq->max_idle_balance_cost)
-		this_rq->max_idle_balance_cost = curr_cost;
-
 out:
 	/* Is there a task of a high priority class? */
 	if (this_rq->nr_running != this_rq->cfs.h_nr_running &&
diff --git a/mm/filemap.c b/mm/filemap.c
index 000a220..088358c 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -257,9 +257,11 @@
 {
 	int ret = 0;
 	/* Check for outstanding write errors */
-	if (test_and_clear_bit(AS_ENOSPC, &mapping->flags))
+	if (test_bit(AS_ENOSPC, &mapping->flags) &&
+	    test_and_clear_bit(AS_ENOSPC, &mapping->flags))
 		ret = -ENOSPC;
-	if (test_and_clear_bit(AS_EIO, &mapping->flags))
+	if (test_bit(AS_EIO, &mapping->flags) &&
+	    test_and_clear_bit(AS_EIO, &mapping->flags))
 		ret = -EIO;
 	return ret;
 }
diff --git a/mm/madvise.c b/mm/madvise.c
index 539eeb9..a402f8f 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -195,7 +195,7 @@
 	for (; start < end; start += PAGE_SIZE) {
 		index = ((start - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
 
-		page = find_get_page(mapping, index);
+		page = find_get_entry(mapping, index);
 		if (!radix_tree_exceptional_entry(page)) {
 			if (page)
 				page_cache_release(page);
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index c47dffd..5177c6d 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1077,9 +1077,18 @@
 
 	rcu_read_lock();
 	do {
-		memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
-		if (unlikely(!memcg))
+		/*
+		 * Page cache insertions can happen withou an
+		 * actual mm context, e.g. during disk probing
+		 * on boot, loopback IO, acct() writes etc.
+		 */
+		if (unlikely(!mm))
 			memcg = root_mem_cgroup;
+		else {
+			memcg = mem_cgroup_from_task(rcu_dereference(mm->owner));
+			if (unlikely(!memcg))
+				memcg = root_mem_cgroup;
+		}
 	} while (!css_tryget(&memcg->css));
 	rcu_read_unlock();
 	return memcg;
@@ -3958,17 +3967,9 @@
 		return 0;
 	}
 
-	/*
-	 * Page cache insertions can happen without an actual mm
-	 * context, e.g. during disk probing on boot.
-	 */
-	if (unlikely(!mm))
-		memcg = root_mem_cgroup;
-	else {
-		memcg = mem_cgroup_try_charge_mm(mm, gfp_mask, 1, true);
-		if (!memcg)
-			return -ENOMEM;
-	}
+	memcg = mem_cgroup_try_charge_mm(mm, gfp_mask, 1, true);
+	if (!memcg)
+		return -ENOMEM;
 	__mem_cgroup_commit_charge(memcg, page, 1, type, false);
 	return 0;
 }
diff --git a/mm/memory-failure.c b/mm/memory-failure.c
index 35ef28a..9ccef39 100644
--- a/mm/memory-failure.c
+++ b/mm/memory-failure.c
@@ -1081,15 +1081,16 @@
 			return 0;
 		} else if (PageHuge(hpage)) {
 			/*
-			 * Check "just unpoisoned", "filter hit", and
-			 * "race with other subpage."
+			 * Check "filter hit" and "race with other subpage."
 			 */
 			lock_page(hpage);
-			if (!PageHWPoison(hpage)
-			    || (hwpoison_filter(p) && TestClearPageHWPoison(p))
-			    || (p != hpage && TestSetPageHWPoison(hpage))) {
-				atomic_long_sub(nr_pages, &num_poisoned_pages);
-				return 0;
+			if (PageHWPoison(hpage)) {
+				if ((hwpoison_filter(p) && TestClearPageHWPoison(p))
+				    || (p != hpage && TestSetPageHWPoison(hpage))) {
+					atomic_long_sub(nr_pages, &num_poisoned_pages);
+					unlock_page(hpage);
+					return 0;
+				}
 			}
 			set_page_hwpoison_huge_page(hpage);
 			res = dequeue_hwpoisoned_huge_page(hpage);
@@ -1152,6 +1153,8 @@
 	 */
 	if (!PageHWPoison(p)) {
 		printk(KERN_ERR "MCE %#lx: just unpoisoned\n", pfn);
+		atomic_long_sub(nr_pages, &num_poisoned_pages);
+		put_page(hpage);
 		res = 0;
 		goto out;
 	}
diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c
index 175273f..44ebd5c 100644
--- a/net/8021q/vlan.c
+++ b/net/8021q/vlan.c
@@ -169,6 +169,7 @@
 	if (err < 0)
 		goto out_uninit_mvrp;
 
+	vlan->nest_level = dev_get_nest_level(real_dev, is_vlan_dev) + 1;
 	err = register_netdevice(dev);
 	if (err < 0)
 		goto out_uninit_mvrp;
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 733ec28..019efb7 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -493,48 +493,10 @@
 	}
 }
 
-static int vlan_calculate_locking_subclass(struct net_device *real_dev)
-{
-	int subclass = 0;
-
-	while (is_vlan_dev(real_dev)) {
-		subclass++;
-		real_dev = vlan_dev_priv(real_dev)->real_dev;
-	}
-
-	return subclass;
-}
-
-static void vlan_dev_mc_sync(struct net_device *to, struct net_device *from)
-{
-	int err = 0, subclass;
-
-	subclass = vlan_calculate_locking_subclass(to);
-
-	spin_lock_nested(&to->addr_list_lock, subclass);
-	err = __hw_addr_sync(&to->mc, &from->mc, to->addr_len);
-	if (!err)
-		__dev_set_rx_mode(to);
-	spin_unlock(&to->addr_list_lock);
-}
-
-static void vlan_dev_uc_sync(struct net_device *to, struct net_device *from)
-{
-	int err = 0, subclass;
-
-	subclass = vlan_calculate_locking_subclass(to);
-
-	spin_lock_nested(&to->addr_list_lock, subclass);
-	err = __hw_addr_sync(&to->uc, &from->uc, to->addr_len);
-	if (!err)
-		__dev_set_rx_mode(to);
-	spin_unlock(&to->addr_list_lock);
-}
-
 static void vlan_dev_set_rx_mode(struct net_device *vlan_dev)
 {
-	vlan_dev_mc_sync(vlan_dev_priv(vlan_dev)->real_dev, vlan_dev);
-	vlan_dev_uc_sync(vlan_dev_priv(vlan_dev)->real_dev, vlan_dev);
+	dev_mc_sync(vlan_dev_priv(vlan_dev)->real_dev, vlan_dev);
+	dev_uc_sync(vlan_dev_priv(vlan_dev)->real_dev, vlan_dev);
 }
 
 /*
@@ -562,6 +524,11 @@
 	netdev_for_each_tx_queue(dev, vlan_dev_set_lockdep_one, &subclass);
 }
 
+static int vlan_dev_get_lock_subclass(struct net_device *dev)
+{
+	return vlan_dev_priv(dev)->nest_level;
+}
+
 static const struct header_ops vlan_header_ops = {
 	.create	 = vlan_dev_hard_header,
 	.rebuild = vlan_dev_rebuild_header,
@@ -597,7 +564,6 @@
 static int vlan_dev_init(struct net_device *dev)
 {
 	struct net_device *real_dev = vlan_dev_priv(dev)->real_dev;
-	int subclass = 0;
 
 	netif_carrier_off(dev);
 
@@ -646,8 +612,7 @@
 
 	SET_NETDEV_DEVTYPE(dev, &vlan_type);
 
-	subclass = vlan_calculate_locking_subclass(dev);
-	vlan_dev_set_lockdep_class(dev, subclass);
+	vlan_dev_set_lockdep_class(dev, vlan_dev_get_lock_subclass(dev));
 
 	vlan_dev_priv(dev)->vlan_pcpu_stats = netdev_alloc_pcpu_stats(struct vlan_pcpu_stats);
 	if (!vlan_dev_priv(dev)->vlan_pcpu_stats)
@@ -819,6 +784,7 @@
 	.ndo_netpoll_cleanup	= vlan_dev_netpoll_cleanup,
 #endif
 	.ndo_fix_features	= vlan_dev_fix_features,
+	.ndo_get_lock_subclass  = vlan_dev_get_lock_subclass,
 };
 
 void vlan_setup(struct net_device *dev)
diff --git a/net/batman-adv/bat_iv_ogm.c b/net/batman-adv/bat_iv_ogm.c
index b3bd4ec..f04224c 100644
--- a/net/batman-adv/bat_iv_ogm.c
+++ b/net/batman-adv/bat_iv_ogm.c
@@ -1545,6 +1545,8 @@
 	if ((orig_neigh_node) && (!is_single_hop_neigh))
 		batadv_orig_node_free_ref(orig_neigh_node);
 out:
+	if (router_ifinfo)
+		batadv_neigh_ifinfo_free_ref(router_ifinfo);
 	if (router)
 		batadv_neigh_node_free_ref(router);
 	if (router_router)
diff --git a/net/batman-adv/distributed-arp-table.c b/net/batman-adv/distributed-arp-table.c
index b25fd64..aa5d494 100644
--- a/net/batman-adv/distributed-arp-table.c
+++ b/net/batman-adv/distributed-arp-table.c
@@ -940,8 +940,7 @@
 		 * additional DAT answer may trigger kernel warnings about
 		 * a packet coming from the wrong port.
 		 */
-		if (batadv_is_my_client(bat_priv, dat_entry->mac_addr,
-					BATADV_NO_FLAGS)) {
+		if (batadv_is_my_client(bat_priv, dat_entry->mac_addr, vid)) {
 			ret = true;
 			goto out;
 		}
diff --git a/net/batman-adv/fragmentation.c b/net/batman-adv/fragmentation.c
index bcc4bea..f14e54a 100644
--- a/net/batman-adv/fragmentation.c
+++ b/net/batman-adv/fragmentation.c
@@ -418,12 +418,13 @@
 			     struct batadv_neigh_node *neigh_node)
 {
 	struct batadv_priv *bat_priv;
-	struct batadv_hard_iface *primary_if;
+	struct batadv_hard_iface *primary_if = NULL;
 	struct batadv_frag_packet frag_header;
 	struct sk_buff *skb_fragment;
 	unsigned mtu = neigh_node->if_incoming->net_dev->mtu;
 	unsigned header_size = sizeof(frag_header);
 	unsigned max_fragment_size, max_packet_size;
+	bool ret = false;
 
 	/* To avoid merge and refragmentation at next-hops we never send
 	 * fragments larger than BATADV_FRAG_MAX_FRAG_SIZE
@@ -483,7 +484,11 @@
 			   skb->len + ETH_HLEN);
 	batadv_send_skb_packet(skb, neigh_node->if_incoming, neigh_node->addr);
 
-	return true;
+	ret = true;
+
 out_err:
-	return false;
+	if (primary_if)
+		batadv_hardif_free_ref(primary_if);
+
+	return ret;
 }
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index c835e13..90cff58 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -42,8 +42,10 @@
 
 static void batadv_gw_node_free_ref(struct batadv_gw_node *gw_node)
 {
-	if (atomic_dec_and_test(&gw_node->refcount))
+	if (atomic_dec_and_test(&gw_node->refcount)) {
+		batadv_orig_node_free_ref(gw_node->orig_node);
 		kfree_rcu(gw_node, rcu);
+	}
 }
 
 static struct batadv_gw_node *
@@ -406,10 +408,15 @@
 	if (gateway->bandwidth_down == 0)
 		return;
 
-	gw_node = kzalloc(sizeof(*gw_node), GFP_ATOMIC);
-	if (!gw_node)
+	if (!atomic_inc_not_zero(&orig_node->refcount))
 		return;
 
+	gw_node = kzalloc(sizeof(*gw_node), GFP_ATOMIC);
+	if (!gw_node) {
+		batadv_orig_node_free_ref(orig_node);
+		return;
+	}
+
 	INIT_HLIST_NODE(&gw_node->list);
 	gw_node->orig_node = orig_node;
 	atomic_set(&gw_node->refcount, 1);
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index b851cc5..fbda6b5 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -83,7 +83,7 @@
 		return true;
 
 	/* no more parents..stop recursion */
-	if (net_dev->iflink == net_dev->ifindex)
+	if (net_dev->iflink == 0 || net_dev->iflink == net_dev->ifindex)
 		return false;
 
 	/* recurse over the parent device */
diff --git a/net/batman-adv/originator.c b/net/batman-adv/originator.c
index ffd9dfb..6a48451 100644
--- a/net/batman-adv/originator.c
+++ b/net/batman-adv/originator.c
@@ -501,12 +501,17 @@
 static void batadv_orig_ifinfo_free_rcu(struct rcu_head *rcu)
 {
 	struct batadv_orig_ifinfo *orig_ifinfo;
+	struct batadv_neigh_node *router;
 
 	orig_ifinfo = container_of(rcu, struct batadv_orig_ifinfo, rcu);
 
 	if (orig_ifinfo->if_outgoing != BATADV_IF_DEFAULT)
 		batadv_hardif_free_ref_now(orig_ifinfo->if_outgoing);
 
+	/* this is the last reference to this object */
+	router = rcu_dereference_protected(orig_ifinfo->router, true);
+	if (router)
+		batadv_neigh_node_free_ref_now(router);
 	kfree(orig_ifinfo);
 }
 
@@ -702,6 +707,47 @@
 }
 
 /**
+ * batadv_purge_neigh_ifinfo - purge obsolete ifinfo entries from neighbor
+ * @bat_priv: the bat priv with all the soft interface information
+ * @neigh: orig node which is to be checked
+ */
+static void
+batadv_purge_neigh_ifinfo(struct batadv_priv *bat_priv,
+			  struct batadv_neigh_node *neigh)
+{
+	struct batadv_neigh_ifinfo *neigh_ifinfo;
+	struct batadv_hard_iface *if_outgoing;
+	struct hlist_node *node_tmp;
+
+	spin_lock_bh(&neigh->ifinfo_lock);
+
+	/* for all ifinfo objects for this neighinator */
+	hlist_for_each_entry_safe(neigh_ifinfo, node_tmp,
+				  &neigh->ifinfo_list, list) {
+		if_outgoing = neigh_ifinfo->if_outgoing;
+
+		/* always keep the default interface */
+		if (if_outgoing == BATADV_IF_DEFAULT)
+			continue;
+
+		/* don't purge if the interface is not (going) down */
+		if ((if_outgoing->if_status != BATADV_IF_INACTIVE) &&
+		    (if_outgoing->if_status != BATADV_IF_NOT_IN_USE) &&
+		    (if_outgoing->if_status != BATADV_IF_TO_BE_REMOVED))
+			continue;
+
+		batadv_dbg(BATADV_DBG_BATMAN, bat_priv,
+			   "neighbor/ifinfo purge: neighbor %pM, iface: %s\n",
+			   neigh->addr, if_outgoing->net_dev->name);
+
+		hlist_del_rcu(&neigh_ifinfo->list);
+		batadv_neigh_ifinfo_free_ref(neigh_ifinfo);
+	}
+
+	spin_unlock_bh(&neigh->ifinfo_lock);
+}
+
+/**
  * batadv_purge_orig_ifinfo - purge obsolete ifinfo entries from originator
  * @bat_priv: the bat priv with all the soft interface information
  * @orig_node: orig node which is to be checked
@@ -800,6 +846,11 @@
 
 			hlist_del_rcu(&neigh_node->list);
 			batadv_neigh_node_free_ref(neigh_node);
+		} else {
+			/* only necessary if not the whole neighbor is to be
+			 * deleted, but some interface has been removed.
+			 */
+			batadv_purge_neigh_ifinfo(bat_priv, neigh_node);
 		}
 	}
 
@@ -857,7 +908,7 @@
 {
 	struct batadv_neigh_node *best_neigh_node;
 	struct batadv_hard_iface *hard_iface;
-	bool changed;
+	bool changed_ifinfo, changed_neigh;
 
 	if (batadv_has_timed_out(orig_node->last_seen,
 				 2 * BATADV_PURGE_TIMEOUT)) {
@@ -867,10 +918,10 @@
 			   jiffies_to_msecs(orig_node->last_seen));
 		return true;
 	}
-	changed = batadv_purge_orig_ifinfo(bat_priv, orig_node);
-	changed = changed || batadv_purge_orig_neighbors(bat_priv, orig_node);
+	changed_ifinfo = batadv_purge_orig_ifinfo(bat_priv, orig_node);
+	changed_neigh = batadv_purge_orig_neighbors(bat_priv, orig_node);
 
-	if (!changed)
+	if (!changed_ifinfo && !changed_neigh)
 		return false;
 
 	/* first for NULL ... */
@@ -1028,7 +1079,8 @@
 	bat_priv->bat_algo_ops->bat_orig_print(bat_priv, seq, hard_iface);
 
 out:
-	batadv_hardif_free_ref(hard_iface);
+	if (hard_iface)
+		batadv_hardif_free_ref(hard_iface);
 	return 0;
 }
 
diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c
index 80e1b0f..2acf7fa 100644
--- a/net/bridge/br_netfilter.c
+++ b/net/bridge/br_netfilter.c
@@ -859,12 +859,12 @@
 	return NF_STOLEN;
 }
 
-#if IS_ENABLED(CONFIG_NF_CONNTRACK_IPV4)
+#if IS_ENABLED(CONFIG_NF_DEFRAG_IPV4)
 static int br_nf_dev_queue_xmit(struct sk_buff *skb)
 {
 	int ret;
 
-	if (skb->nfct != NULL && skb->protocol == htons(ETH_P_IP) &&
+	if (skb->protocol == htons(ETH_P_IP) &&
 	    skb->len + nf_bridge_mtu_reduction(skb) > skb->dev->mtu &&
 	    !skb_is_gso(skb)) {
 		if (br_parse_ip_options(skb))
diff --git a/net/core/dev.c b/net/core/dev.c
index d2c8a06..9abc503 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -2418,7 +2418,7 @@
  * 2. No high memory really exists on this machine.
  */
 
-static int illegal_highdma(const struct net_device *dev, struct sk_buff *skb)
+static int illegal_highdma(struct net_device *dev, struct sk_buff *skb)
 {
 #ifdef CONFIG_HIGHMEM
 	int i;
@@ -2493,38 +2493,36 @@
 }
 
 static netdev_features_t harmonize_features(struct sk_buff *skb,
-					    const struct net_device *dev,
-					    netdev_features_t features)
+	netdev_features_t features)
 {
 	int tmp;
 
 	if (skb->ip_summed != CHECKSUM_NONE &&
 	    !can_checksum_protocol(features, skb_network_protocol(skb, &tmp))) {
 		features &= ~NETIF_F_ALL_CSUM;
-	} else if (illegal_highdma(dev, skb)) {
+	} else if (illegal_highdma(skb->dev, skb)) {
 		features &= ~NETIF_F_SG;
 	}
 
 	return features;
 }
 
-netdev_features_t netif_skb_dev_features(struct sk_buff *skb,
-					 const struct net_device *dev)
+netdev_features_t netif_skb_features(struct sk_buff *skb)
 {
 	__be16 protocol = skb->protocol;
-	netdev_features_t features = dev->features;
+	netdev_features_t features = skb->dev->features;
 
-	if (skb_shinfo(skb)->gso_segs > dev->gso_max_segs)
+	if (skb_shinfo(skb)->gso_segs > skb->dev->gso_max_segs)
 		features &= ~NETIF_F_GSO_MASK;
 
 	if (protocol == htons(ETH_P_8021Q) || protocol == htons(ETH_P_8021AD)) {
 		struct vlan_ethhdr *veh = (struct vlan_ethhdr *)skb->data;
 		protocol = veh->h_vlan_encapsulated_proto;
 	} else if (!vlan_tx_tag_present(skb)) {
-		return harmonize_features(skb, dev, features);
+		return harmonize_features(skb, features);
 	}
 
-	features &= (dev->vlan_features | NETIF_F_HW_VLAN_CTAG_TX |
+	features &= (skb->dev->vlan_features | NETIF_F_HW_VLAN_CTAG_TX |
 					       NETIF_F_HW_VLAN_STAG_TX);
 
 	if (protocol == htons(ETH_P_8021Q) || protocol == htons(ETH_P_8021AD))
@@ -2532,9 +2530,9 @@
 				NETIF_F_GEN_CSUM | NETIF_F_HW_VLAN_CTAG_TX |
 				NETIF_F_HW_VLAN_STAG_TX;
 
-	return harmonize_features(skb, dev, features);
+	return harmonize_features(skb, features);
 }
-EXPORT_SYMBOL(netif_skb_dev_features);
+EXPORT_SYMBOL(netif_skb_features);
 
 int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev,
 			struct netdev_queue *txq)
@@ -3953,6 +3951,7 @@
 	}
 	NAPI_GRO_CB(skb)->count = 1;
 	NAPI_GRO_CB(skb)->age = jiffies;
+	NAPI_GRO_CB(skb)->last = skb;
 	skb_shinfo(skb)->gso_size = skb_gro_len(skb);
 	skb->next = napi->gro_list;
 	napi->gro_list = skb;
@@ -4543,6 +4542,32 @@
 EXPORT_SYMBOL(netdev_adjacent_get_private);
 
 /**
+ * netdev_upper_get_next_dev_rcu - Get the next dev from upper list
+ * @dev: device
+ * @iter: list_head ** of the current position
+ *
+ * Gets the next device from the dev's upper list, starting from iter
+ * position. The caller must hold RCU read lock.
+ */
+struct net_device *netdev_upper_get_next_dev_rcu(struct net_device *dev,
+						 struct list_head **iter)
+{
+	struct netdev_adjacent *upper;
+
+	WARN_ON_ONCE(!rcu_read_lock_held() && !lockdep_rtnl_is_held());
+
+	upper = list_entry_rcu((*iter)->next, struct netdev_adjacent, list);
+
+	if (&upper->list == &dev->adj_list.upper)
+		return NULL;
+
+	*iter = &upper->list;
+
+	return upper->dev;
+}
+EXPORT_SYMBOL(netdev_upper_get_next_dev_rcu);
+
+/**
  * netdev_all_upper_get_next_dev_rcu - Get the next dev from upper list
  * @dev: device
  * @iter: list_head ** of the current position
@@ -4624,6 +4649,32 @@
 EXPORT_SYMBOL(netdev_lower_get_next_private_rcu);
 
 /**
+ * netdev_lower_get_next - Get the next device from the lower neighbour
+ *                         list
+ * @dev: device
+ * @iter: list_head ** of the current position
+ *
+ * Gets the next netdev_adjacent from the dev's lower neighbour
+ * list, starting from iter position. The caller must hold RTNL lock or
+ * its own locking that guarantees that the neighbour lower
+ * list will remain unchainged.
+ */
+void *netdev_lower_get_next(struct net_device *dev, struct list_head **iter)
+{
+	struct netdev_adjacent *lower;
+
+	lower = list_entry((*iter)->next, struct netdev_adjacent, list);
+
+	if (&lower->list == &dev->adj_list.lower)
+		return NULL;
+
+	*iter = &lower->list;
+
+	return lower->dev;
+}
+EXPORT_SYMBOL(netdev_lower_get_next);
+
+/**
  * netdev_lower_get_first_private_rcu - Get the first ->private from the
  *				       lower neighbour list, RCU
  *				       variant
@@ -5073,6 +5124,30 @@
 }
 EXPORT_SYMBOL(netdev_lower_dev_get_private);
 
+
+int dev_get_nest_level(struct net_device *dev,
+		       bool (*type_check)(struct net_device *dev))
+{
+	struct net_device *lower = NULL;
+	struct list_head *iter;
+	int max_nest = -1;
+	int nest;
+
+	ASSERT_RTNL();
+
+	netdev_for_each_lower_dev(dev, lower, iter) {
+		nest = dev_get_nest_level(lower, type_check);
+		if (max_nest < nest)
+			max_nest = nest;
+	}
+
+	if (type_check(dev))
+		max_nest++;
+
+	return max_nest;
+}
+EXPORT_SYMBOL(dev_get_nest_level);
+
 static void dev_change_rx_flags(struct net_device *dev, int flags)
 {
 	const struct net_device_ops *ops = dev->netdev_ops;
@@ -5238,7 +5313,6 @@
 	if (ops->ndo_set_rx_mode)
 		ops->ndo_set_rx_mode(dev);
 }
-EXPORT_SYMBOL(__dev_set_rx_mode);
 
 void dev_set_rx_mode(struct net_device *dev)
 {
@@ -5543,7 +5617,7 @@
 
 /* Delayed registration/unregisteration */
 static LIST_HEAD(net_todo_list);
-static DECLARE_WAIT_QUEUE_HEAD(netdev_unregistering_wq);
+DECLARE_WAIT_QUEUE_HEAD(netdev_unregistering_wq);
 
 static void net_set_todo(struct net_device *dev)
 {
diff --git a/net/core/neighbour.c b/net/core/neighbour.c
index 8f8a96e..32d872e 100644
--- a/net/core/neighbour.c
+++ b/net/core/neighbour.c
@@ -1248,8 +1248,8 @@
 	neigh->updated = jiffies;
 	if (!(neigh->nud_state & NUD_FAILED))
 		return;
-	neigh->nud_state = NUD_PROBE;
-	atomic_set(&neigh->probes, NEIGH_VAR(neigh->parms, UCAST_PROBES));
+	neigh->nud_state = NUD_INCOMPLETE;
+	atomic_set(&neigh->probes, neigh_max_probes(neigh));
 	neigh_add_timer(neigh,
 			jiffies + NEIGH_VAR(neigh->parms, RETRANS_TIME));
 }
diff --git a/net/core/net_namespace.c b/net/core/net_namespace.c
index 81d3a9a..7c8ffd9 100644
--- a/net/core/net_namespace.c
+++ b/net/core/net_namespace.c
@@ -24,7 +24,7 @@
 
 static LIST_HEAD(pernet_list);
 static struct list_head *first_device = &pernet_list;
-static DEFINE_MUTEX(net_mutex);
+DEFINE_MUTEX(net_mutex);
 
 LIST_HEAD(net_namespace_list);
 EXPORT_SYMBOL_GPL(net_namespace_list);
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 9837beb..2d8d8fc 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -353,15 +353,46 @@
 }
 EXPORT_SYMBOL_GPL(__rtnl_link_unregister);
 
+/* Return with the rtnl_lock held when there are no network
+ * devices unregistering in any network namespace.
+ */
+static void rtnl_lock_unregistering_all(void)
+{
+	struct net *net;
+	bool unregistering;
+	DEFINE_WAIT(wait);
+
+	for (;;) {
+		prepare_to_wait(&netdev_unregistering_wq, &wait,
+				TASK_UNINTERRUPTIBLE);
+		unregistering = false;
+		rtnl_lock();
+		for_each_net(net) {
+			if (net->dev_unreg_count > 0) {
+				unregistering = true;
+				break;
+			}
+		}
+		if (!unregistering)
+			break;
+		__rtnl_unlock();
+		schedule();
+	}
+	finish_wait(&netdev_unregistering_wq, &wait);
+}
+
 /**
  * rtnl_link_unregister - Unregister rtnl_link_ops from rtnetlink.
  * @ops: struct rtnl_link_ops * to unregister
  */
 void rtnl_link_unregister(struct rtnl_link_ops *ops)
 {
-	rtnl_lock();
+	/* Close the race with cleanup_net() */
+	mutex_lock(&net_mutex);
+	rtnl_lock_unregistering_all();
 	__rtnl_link_unregister(ops);
 	rtnl_unlock();
+	mutex_unlock(&net_mutex);
 }
 EXPORT_SYMBOL_GPL(rtnl_link_unregister);
 
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 1b62343..8383b2b 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3076,7 +3076,7 @@
 	if (unlikely(p->len + len >= 65536))
 		return -E2BIG;
 
-	lp = NAPI_GRO_CB(p)->last ?: p;
+	lp = NAPI_GRO_CB(p)->last;
 	pinfo = skb_shinfo(lp);
 
 	if (headlen <= offset) {
@@ -3192,7 +3192,7 @@
 
 	__skb_pull(skb, offset);
 
-	if (!NAPI_GRO_CB(p)->last)
+	if (NAPI_GRO_CB(p)->last == p)
 		skb_shinfo(p)->frag_list = skb;
 	else
 		NAPI_GRO_CB(p)->last->next = skb;
diff --git a/net/core/utils.c b/net/core/utils.c
index 2f737bf..eed3433 100644
--- a/net/core/utils.c
+++ b/net/core/utils.c
@@ -348,8 +348,8 @@
 {
 	struct __net_random_once_work *work =
 		container_of(w, struct __net_random_once_work, work);
-	if (!static_key_enabled(work->key))
-		static_key_slow_inc(work->key);
+	BUG_ON(!static_key_enabled(work->key));
+	static_key_slow_dec(work->key);
 	kfree(work);
 }
 
@@ -367,7 +367,7 @@
 }
 
 bool __net_get_random_once(void *buf, int nbytes, bool *done,
-			   struct static_key *done_key)
+			   struct static_key *once_key)
 {
 	static DEFINE_SPINLOCK(lock);
 	unsigned long flags;
@@ -382,7 +382,7 @@
 	*done = true;
 	spin_unlock_irqrestore(&lock, flags);
 
-	__net_random_once_disable_jump(done_key);
+	__net_random_once_disable_jump(once_key);
 
 	return true;
 }
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index 0eb5d5e..5db37ce 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -406,8 +406,9 @@
 		goto out_free;
 	}
 
-	chip_index = 0;
+	chip_index = -1;
 	for_each_available_child_of_node(np, child) {
+		chip_index++;
 		cd = &pd->chip[chip_index];
 
 		cd->mii_bus = &mdio_bus->dev;
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 8c54870..6d6dd34 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -1650,6 +1650,39 @@
 	return register_pernet_subsys(&ipv4_mib_ops);
 }
 
+static __net_init int inet_init_net(struct net *net)
+{
+	/*
+	 * Set defaults for local port range
+	 */
+	seqlock_init(&net->ipv4.ip_local_ports.lock);
+	net->ipv4.ip_local_ports.range[0] =  32768;
+	net->ipv4.ip_local_ports.range[1] =  61000;
+
+	seqlock_init(&net->ipv4.ping_group_range.lock);
+	/*
+	 * Sane defaults - nobody may create ping sockets.
+	 * Boot scripts should set this to distro-specific group.
+	 */
+	net->ipv4.ping_group_range.range[0] = make_kgid(&init_user_ns, 1);
+	net->ipv4.ping_group_range.range[1] = make_kgid(&init_user_ns, 0);
+	return 0;
+}
+
+static __net_exit void inet_exit_net(struct net *net)
+{
+}
+
+static __net_initdata struct pernet_operations af_inet_ops = {
+	.init = inet_init_net,
+	.exit = inet_exit_net,
+};
+
+static int __init init_inet_pernet_ops(void)
+{
+	return register_pernet_subsys(&af_inet_ops);
+}
+
 static int ipv4_proc_init(void);
 
 /*
@@ -1794,6 +1827,9 @@
 	if (ip_mr_init())
 		pr_crit("%s: Cannot init ipv4 mroute\n", __func__);
 #endif
+
+	if (init_inet_pernet_ops())
+		pr_crit("%s: Cannot init ipv4 inet pernet ops\n", __func__);
 	/*
 	 *	Initialise per-cpu ipv4 mibs
 	 */
diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c
index 8a043f0..b10cd43a 100644
--- a/net/ipv4/fib_semantics.c
+++ b/net/ipv4/fib_semantics.c
@@ -821,13 +821,13 @@
 	fi = kzalloc(sizeof(*fi)+nhs*sizeof(struct fib_nh), GFP_KERNEL);
 	if (fi == NULL)
 		goto failure;
+	fib_info_cnt++;
 	if (cfg->fc_mx) {
 		fi->fib_metrics = kzalloc(sizeof(u32) * RTAX_MAX, GFP_KERNEL);
 		if (!fi->fib_metrics)
 			goto failure;
 	} else
 		fi->fib_metrics = (u32 *) dst_default_metrics;
-	fib_info_cnt++;
 
 	fi->fib_net = hold_net(net);
 	fi->fib_protocol = cfg->fc_protocol;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 0d1e2cb..a56b8e6 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -37,11 +37,11 @@
 	unsigned int seq;
 
 	do {
-		seq = read_seqbegin(&net->ipv4.sysctl_local_ports.lock);
+		seq = read_seqbegin(&net->ipv4.ip_local_ports.lock);
 
-		*low = net->ipv4.sysctl_local_ports.range[0];
-		*high = net->ipv4.sysctl_local_ports.range[1];
-	} while (read_seqretry(&net->ipv4.sysctl_local_ports.lock, seq));
+		*low = net->ipv4.ip_local_ports.range[0];
+		*high = net->ipv4.ip_local_ports.range[1];
+	} while (read_seqretry(&net->ipv4.ip_local_ports.lock, seq));
 }
 EXPORT_SYMBOL(inet_get_local_port_range);
 
diff --git a/net/ipv4/ip_forward.c b/net/ipv4/ip_forward.c
index be8abe7..6f111e4 100644
--- a/net/ipv4/ip_forward.c
+++ b/net/ipv4/ip_forward.c
@@ -42,12 +42,12 @@
 static bool ip_may_fragment(const struct sk_buff *skb)
 {
 	return unlikely((ip_hdr(skb)->frag_off & htons(IP_DF)) == 0) ||
-	       !skb->local_df;
+		skb->local_df;
 }
 
 static bool ip_exceeds_mtu(const struct sk_buff *skb, unsigned int mtu)
 {
-	if (skb->len <= mtu || skb->local_df)
+	if (skb->len <= mtu)
 		return false;
 
 	if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu)
@@ -56,53 +56,6 @@
 	return true;
 }
 
-static bool ip_gso_exceeds_dst_mtu(const struct sk_buff *skb)
-{
-	unsigned int mtu;
-
-	if (skb->local_df || !skb_is_gso(skb))
-		return false;
-
-	mtu = ip_dst_mtu_maybe_forward(skb_dst(skb), true);
-
-	/* if seglen > mtu, do software segmentation for IP fragmentation on
-	 * output.  DF bit cannot be set since ip_forward would have sent
-	 * icmp error.
-	 */
-	return skb_gso_network_seglen(skb) > mtu;
-}
-
-/* called if GSO skb needs to be fragmented on forward */
-static int ip_forward_finish_gso(struct sk_buff *skb)
-{
-	struct dst_entry *dst = skb_dst(skb);
-	netdev_features_t features;
-	struct sk_buff *segs;
-	int ret = 0;
-
-	features = netif_skb_dev_features(skb, dst->dev);
-	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
-	if (IS_ERR(segs)) {
-		kfree_skb(skb);
-		return -ENOMEM;
-	}
-
-	consume_skb(skb);
-
-	do {
-		struct sk_buff *nskb = segs->next;
-		int err;
-
-		segs->next = NULL;
-		err = dst_output(segs);
-
-		if (err && ret == 0)
-			ret = err;
-		segs = nskb;
-	} while (segs);
-
-	return ret;
-}
 
 static int ip_forward_finish(struct sk_buff *skb)
 {
@@ -114,9 +67,6 @@
 	if (unlikely(opt->optlen))
 		ip_forward_options(skb);
 
-	if (ip_gso_exceeds_dst_mtu(skb))
-		return ip_forward_finish_gso(skb);
-
 	return dst_output(skb);
 }
 
diff --git a/net/ipv4/ip_fragment.c b/net/ipv4/ip_fragment.c
index c10a3ce..ed32313 100644
--- a/net/ipv4/ip_fragment.c
+++ b/net/ipv4/ip_fragment.c
@@ -232,8 +232,9 @@
 		 * "Fragment Reassembly Timeout" message, per RFC792.
 		 */
 		if (qp->user == IP_DEFRAG_AF_PACKET ||
-		    (qp->user == IP_DEFRAG_CONNTRACK_IN &&
-		     skb_rtable(head)->rt_type != RTN_LOCAL))
+		    ((qp->user >= IP_DEFRAG_CONNTRACK_IN) &&
+		     (qp->user <= __IP_DEFRAG_CONNTRACK_IN_END) &&
+		     (skb_rtable(head)->rt_type != RTN_LOCAL)))
 			goto out_rcu_unlock;
 
 
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 1cbeba5..a52f501 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -211,6 +211,48 @@
 	return -EINVAL;
 }
 
+static int ip_finish_output_gso(struct sk_buff *skb)
+{
+	netdev_features_t features;
+	struct sk_buff *segs;
+	int ret = 0;
+
+	/* common case: locally created skb or seglen is <= mtu */
+	if (((IPCB(skb)->flags & IPSKB_FORWARDED) == 0) ||
+	      skb_gso_network_seglen(skb) <= ip_skb_dst_mtu(skb))
+		return ip_finish_output2(skb);
+
+	/* Slowpath -  GSO segment length is exceeding the dst MTU.
+	 *
+	 * This can happen in two cases:
+	 * 1) TCP GRO packet, DF bit not set
+	 * 2) skb arrived via virtio-net, we thus get TSO/GSO skbs directly
+	 * from host network stack.
+	 */
+	features = netif_skb_features(skb);
+	segs = skb_gso_segment(skb, features & ~NETIF_F_GSO_MASK);
+	if (IS_ERR(segs)) {
+		kfree_skb(skb);
+		return -ENOMEM;
+	}
+
+	consume_skb(skb);
+
+	do {
+		struct sk_buff *nskb = segs->next;
+		int err;
+
+		segs->next = NULL;
+		err = ip_fragment(segs, ip_finish_output2);
+
+		if (err && ret == 0)
+			ret = err;
+		segs = nskb;
+	} while (segs);
+
+	return ret;
+}
+
 static int ip_finish_output(struct sk_buff *skb)
 {
 #if defined(CONFIG_NETFILTER) && defined(CONFIG_XFRM)
@@ -220,10 +262,13 @@
 		return dst_output(skb);
 	}
 #endif
-	if (skb->len > ip_skb_dst_mtu(skb) && !skb_is_gso(skb))
+	if (skb_is_gso(skb))
+		return ip_finish_output_gso(skb);
+
+	if (skb->len > ip_skb_dst_mtu(skb))
 		return ip_fragment(skb, ip_finish_output2);
-	else
-		return ip_finish_output2(skb);
+
+	return ip_finish_output2(skb);
 }
 
 int ip_mc_output(struct sock *sk, struct sk_buff *skb)
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index b3f85973..2acc233 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -540,9 +540,10 @@
 	unsigned int max_headroom;	/* The extra header space needed */
 	__be32 dst;
 	int err;
-	bool connected = true;
+	bool connected;
 
 	inner_iph = (const struct iphdr *)skb_inner_network_header(skb);
+	connected = (tunnel->parms.iph.daddr != 0);
 
 	dst = tnl_params->daddr;
 	if (dst == 0) {
@@ -882,6 +883,7 @@
 	 */
 	if (!IS_ERR(itn->fb_tunnel_dev)) {
 		itn->fb_tunnel_dev->features |= NETIF_F_NETNS_LOCAL;
+		itn->fb_tunnel_dev->mtu = ip_tunnel_bind_dev(itn->fb_tunnel_dev);
 		ip_tunnel_add(itn, netdev_priv(itn->fb_tunnel_dev));
 	}
 	rtnl_unlock();
diff --git a/net/ipv4/ip_vti.c b/net/ipv4/ip_vti.c
index afcee51..13ef00f 100644
--- a/net/ipv4/ip_vti.c
+++ b/net/ipv4/ip_vti.c
@@ -239,6 +239,7 @@
 static int vti4_err(struct sk_buff *skb, u32 info)
 {
 	__be32 spi;
+	__u32 mark;
 	struct xfrm_state *x;
 	struct ip_tunnel *tunnel;
 	struct ip_esp_hdr *esph;
@@ -254,6 +255,8 @@
 	if (!tunnel)
 		return -1;
 
+	mark = be32_to_cpu(tunnel->parms.o_key);
+
 	switch (protocol) {
 	case IPPROTO_ESP:
 		esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
@@ -281,7 +284,7 @@
 		return 0;
 	}
 
-	x = xfrm_state_lookup(net, skb->mark, (const xfrm_address_t *)&iph->daddr,
+	x = xfrm_state_lookup(net, mark, (const xfrm_address_t *)&iph->daddr,
 			      spi, protocol, AF_INET);
 	if (!x)
 		return 0;
diff --git a/net/ipv4/netfilter/nf_defrag_ipv4.c b/net/ipv4/netfilter/nf_defrag_ipv4.c
index 12e13bd..f40f321 100644
--- a/net/ipv4/netfilter/nf_defrag_ipv4.c
+++ b/net/ipv4/netfilter/nf_defrag_ipv4.c
@@ -22,7 +22,6 @@
 #endif
 #include <net/netfilter/nf_conntrack_zones.h>
 
-/* Returns new sk_buff, or NULL */
 static int nf_ct_ipv4_gather_frags(struct sk_buff *skb, u_int32_t user)
 {
 	int err;
@@ -33,8 +32,10 @@
 	err = ip_defrag(skb, user);
 	local_bh_enable();
 
-	if (!err)
+	if (!err) {
 		ip_send_check(ip_hdr(skb));
+		skb->local_df = 1;
+	}
 
 	return err;
 }
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 8210964..044a0dd 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -236,15 +236,15 @@
 static void inet_get_ping_group_range_net(struct net *net, kgid_t *low,
 					  kgid_t *high)
 {
-	kgid_t *data = net->ipv4.sysctl_ping_group_range;
+	kgid_t *data = net->ipv4.ping_group_range.range;
 	unsigned int seq;
 
 	do {
-		seq = read_seqbegin(&net->ipv4.sysctl_local_ports.lock);
+		seq = read_seqbegin(&net->ipv4.ping_group_range.lock);
 
 		*low = data[0];
 		*high = data[1];
-	} while (read_seqretry(&net->ipv4.sysctl_local_ports.lock, seq));
+	} while (read_seqretry(&net->ipv4.ping_group_range.lock, seq));
 }
 
 
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index db1e0da8..5e676be 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1519,7 +1519,7 @@
 	struct in_device *out_dev;
 	unsigned int flags = 0;
 	bool do_cache;
-	u32 itag;
+	u32 itag = 0;
 
 	/* get a working reference to the output device */
 	out_dev = __in_dev_get_rcu(FIB_RES_DEV(*res));
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 44eba05..5cde8f2 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -45,10 +45,10 @@
 /* Update system visible IP port range */
 static void set_local_port_range(struct net *net, int range[2])
 {
-	write_seqlock(&net->ipv4.sysctl_local_ports.lock);
-	net->ipv4.sysctl_local_ports.range[0] = range[0];
-	net->ipv4.sysctl_local_ports.range[1] = range[1];
-	write_sequnlock(&net->ipv4.sysctl_local_ports.lock);
+	write_seqlock(&net->ipv4.ip_local_ports.lock);
+	net->ipv4.ip_local_ports.range[0] = range[0];
+	net->ipv4.ip_local_ports.range[1] = range[1];
+	write_sequnlock(&net->ipv4.ip_local_ports.lock);
 }
 
 /* Validate changes from /proc interface. */
@@ -57,7 +57,7 @@
 				 size_t *lenp, loff_t *ppos)
 {
 	struct net *net =
-		container_of(table->data, struct net, ipv4.sysctl_local_ports.range);
+		container_of(table->data, struct net, ipv4.ip_local_ports.range);
 	int ret;
 	int range[2];
 	struct ctl_table tmp = {
@@ -87,14 +87,14 @@
 {
 	kgid_t *data = table->data;
 	struct net *net =
-		container_of(table->data, struct net, ipv4.sysctl_ping_group_range);
+		container_of(table->data, struct net, ipv4.ping_group_range.range);
 	unsigned int seq;
 	do {
-		seq = read_seqbegin(&net->ipv4.sysctl_local_ports.lock);
+		seq = read_seqbegin(&net->ipv4.ip_local_ports.lock);
 
 		*low = data[0];
 		*high = data[1];
-	} while (read_seqretry(&net->ipv4.sysctl_local_ports.lock, seq));
+	} while (read_seqretry(&net->ipv4.ip_local_ports.lock, seq));
 }
 
 /* Update system visible IP port range */
@@ -102,11 +102,11 @@
 {
 	kgid_t *data = table->data;
 	struct net *net =
-		container_of(table->data, struct net, ipv4.sysctl_ping_group_range);
-	write_seqlock(&net->ipv4.sysctl_local_ports.lock);
+		container_of(table->data, struct net, ipv4.ping_group_range.range);
+	write_seqlock(&net->ipv4.ip_local_ports.lock);
 	data[0] = low;
 	data[1] = high;
-	write_sequnlock(&net->ipv4.sysctl_local_ports.lock);
+	write_sequnlock(&net->ipv4.ip_local_ports.lock);
 }
 
 /* Validate changes from /proc interface. */
@@ -805,7 +805,7 @@
 	},
 	{
 		.procname	= "ping_group_range",
-		.data		= &init_net.ipv4.sysctl_ping_group_range,
+		.data		= &init_net.ipv4.ping_group_range.range,
 		.maxlen		= sizeof(gid_t)*2,
 		.mode		= 0644,
 		.proc_handler	= ipv4_ping_group_range,
@@ -819,8 +819,8 @@
 	},
 	{
 		.procname	= "ip_local_port_range",
-		.maxlen		= sizeof(init_net.ipv4.sysctl_local_ports.range),
-		.data		= &init_net.ipv4.sysctl_local_ports.range,
+		.maxlen		= sizeof(init_net.ipv4.ip_local_ports.range),
+		.data		= &init_net.ipv4.ip_local_ports.range,
 		.mode		= 0644,
 		.proc_handler	= ipv4_local_port_range,
 	},
@@ -858,20 +858,6 @@
 			table[i].data += (void *)net - (void *)&init_net;
 	}
 
-	/*
-	 * Sane defaults - nobody may create ping sockets.
-	 * Boot scripts should set this to distro-specific group.
-	 */
-	net->ipv4.sysctl_ping_group_range[0] = make_kgid(&init_user_ns, 1);
-	net->ipv4.sysctl_ping_group_range[1] = make_kgid(&init_user_ns, 0);
-
-	/*
-	 * Set defaults for local port range
-	 */
-	seqlock_init(&net->ipv4.sysctl_local_ports.lock);
-	net->ipv4.sysctl_local_ports.range[0] =  32768;
-	net->ipv4.sysctl_local_ports.range[1] =  61000;
-
 	net->ipv4.ipv4_hdr = register_net_sysctl(net, "net/ipv4", table);
 	if (net->ipv4.ipv4_hdr == NULL)
 		goto err_reg;
diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c
index 40e701f..186a8ec 100644
--- a/net/ipv4/xfrm4_output.c
+++ b/net/ipv4/xfrm4_output.c
@@ -62,10 +62,7 @@
 	if (err)
 		return err;
 
-	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
-	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE | IPSKB_XFRM_TRANSFORMED;
-
-	skb->protocol = htons(ETH_P_IP);
+	IPCB(skb)->flags |= IPSKB_XFRM_TUNNEL_SIZE;
 
 	return x->outer_mode->output2(x, skb);
 }
@@ -73,27 +70,34 @@
 
 int xfrm4_output_finish(struct sk_buff *skb)
 {
-#ifdef CONFIG_NETFILTER
-	if (!skb_dst(skb)->xfrm) {
-		IPCB(skb)->flags |= IPSKB_REROUTED;
-		return dst_output(skb);
-	}
+	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
+	skb->protocol = htons(ETH_P_IP);
 
+#ifdef CONFIG_NETFILTER
 	IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED;
 #endif
 
-	skb->protocol = htons(ETH_P_IP);
 	return xfrm_output(skb);
 }
 
+static int __xfrm4_output(struct sk_buff *skb)
+{
+	struct xfrm_state *x = skb_dst(skb)->xfrm;
+
+#ifdef CONFIG_NETFILTER
+	if (!x) {
+		IPCB(skb)->flags |= IPSKB_REROUTED;
+		return dst_output(skb);
+	}
+#endif
+
+	return x->outer_mode->afinfo->output_finish(skb);
+}
+
 int xfrm4_output(struct sock *sk, struct sk_buff *skb)
 {
-	struct dst_entry *dst = skb_dst(skb);
-	struct xfrm_state *x = dst->xfrm;
-
 	return NF_HOOK_COND(NFPROTO_IPV4, NF_INET_POST_ROUTING, skb,
-			    NULL, dst->dev,
-			    x->outer_mode->afinfo->output_finish,
+			    NULL, skb_dst(skb)->dev, __xfrm4_output,
 			    !(IPCB(skb)->flags & IPSKB_REROUTED));
 }
 
diff --git a/net/ipv4/xfrm4_protocol.c b/net/ipv4/xfrm4_protocol.c
index 7f7b243..a2ce010 100644
--- a/net/ipv4/xfrm4_protocol.c
+++ b/net/ipv4/xfrm4_protocol.c
@@ -50,8 +50,12 @@
 {
 	int ret;
 	struct xfrm4_protocol *handler;
+	struct xfrm4_protocol __rcu **head = proto_handlers(protocol);
 
-	for_each_protocol_rcu(*proto_handlers(protocol), handler)
+	if (!head)
+		return 0;
+
+	for_each_protocol_rcu(*head, handler)
 		if ((ret = handler->cb_handler(skb, err)) <= 0)
 			return ret;
 
@@ -64,15 +68,20 @@
 {
 	int ret;
 	struct xfrm4_protocol *handler;
+	struct xfrm4_protocol __rcu **head = proto_handlers(nexthdr);
 
 	XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = NULL;
 	XFRM_SPI_SKB_CB(skb)->family = AF_INET;
 	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
 
-	for_each_protocol_rcu(*proto_handlers(nexthdr), handler)
+	if (!head)
+		goto out;
+
+	for_each_protocol_rcu(*head, handler)
 		if ((ret = handler->input_handler(skb, nexthdr, spi, encap_type)) != -EINVAL)
 			return ret;
 
+out:
 	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
 
 	kfree_skb(skb);
@@ -208,6 +217,9 @@
 	int ret = -EEXIST;
 	int priority = handler->priority;
 
+	if (!proto_handlers(protocol) || !netproto(protocol))
+		return -EINVAL;
+
 	mutex_lock(&xfrm4_protocol_mutex);
 
 	if (!rcu_dereference_protected(*proto_handlers(protocol),
@@ -250,6 +262,9 @@
 	struct xfrm4_protocol *t;
 	int ret = -ENOENT;
 
+	if (!proto_handlers(protocol) || !netproto(protocol))
+		return -EINVAL;
+
 	mutex_lock(&xfrm4_protocol_mutex);
 
 	for (pprev = proto_handlers(protocol);
diff --git a/net/ipv6/ip6_offload.c b/net/ipv6/ip6_offload.c
index 59f95af..b2f0915 100644
--- a/net/ipv6/ip6_offload.c
+++ b/net/ipv6/ip6_offload.c
@@ -196,7 +196,6 @@
 	unsigned int off;
 	u16 flush = 1;
 	int proto;
-	__wsum csum;
 
 	off = skb_gro_offset(skb);
 	hlen = off + sizeof(*iph);
@@ -264,13 +263,10 @@
 
 	NAPI_GRO_CB(skb)->flush |= flush;
 
-	csum = skb->csum;
-	skb_postpull_rcsum(skb, iph, skb_network_header_len(skb));
+	skb_gro_postpull_rcsum(skb, iph, nlen);
 
 	pp = ops->callbacks.gro_receive(head, skb);
 
-	skb->csum = csum;
-
 out_unlock:
 	rcu_read_unlock();
 
diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c
index 40e7581..fbf1156 100644
--- a/net/ipv6/ip6_output.c
+++ b/net/ipv6/ip6_output.c
@@ -344,12 +344,16 @@
 
 static bool ip6_pkt_too_big(const struct sk_buff *skb, unsigned int mtu)
 {
-	if (skb->len <= mtu || skb->local_df)
+	if (skb->len <= mtu)
 		return false;
 
+	/* ipv6 conntrack defrag sets max_frag_size + local_df */
 	if (IP6CB(skb)->frag_max_size && IP6CB(skb)->frag_max_size > mtu)
 		return true;
 
+	if (skb->local_df)
+		return false;
+
 	if (skb_is_gso(skb) && skb_gso_network_seglen(skb) <= mtu)
 		return false;
 
@@ -1225,7 +1229,7 @@
 		unsigned int maxnonfragsize, headersize;
 
 		headersize = sizeof(struct ipv6hdr) +
-			     (opt ? opt->tot_len : 0) +
+			     (opt ? opt->opt_flen + opt->opt_nflen : 0) +
 			     (dst_allfrag(&rt->dst) ?
 			      sizeof(struct frag_hdr) : 0) +
 			     rt->rt6i_nfheader_len;
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index b05b609..f6a66bb 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -1557,7 +1557,7 @@
 {
 	u8 proto;
 
-	if (!data)
+	if (!data || !data[IFLA_IPTUN_PROTO])
 		return 0;
 
 	proto = nla_get_u8(data[IFLA_IPTUN_PROTO]);
diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c
index b7c0f82..6cc9f93 100644
--- a/net/ipv6/ip6_vti.c
+++ b/net/ipv6/ip6_vti.c
@@ -511,6 +511,7 @@
 		    u8 type, u8 code, int offset, __be32 info)
 {
 	__be32 spi;
+	__u32 mark;
 	struct xfrm_state *x;
 	struct ip6_tnl *t;
 	struct ip_esp_hdr *esph;
@@ -524,6 +525,8 @@
 	if (!t)
 		return -1;
 
+	mark = be32_to_cpu(t->parms.o_key);
+
 	switch (protocol) {
 	case IPPROTO_ESP:
 		esph = (struct ip_esp_hdr *)(skb->data + offset);
@@ -545,7 +548,7 @@
 	    type != NDISC_REDIRECT)
 		return 0;
 
-	x = xfrm_state_lookup(net, skb->mark, (const xfrm_address_t *)&iph->daddr,
+	x = xfrm_state_lookup(net, mark, (const xfrm_address_t *)&iph->daddr,
 			      spi, protocol, AF_INET6);
 	if (!x)
 		return 0;
@@ -1097,7 +1100,6 @@
 
 	err = xfrm6_protocol_register(&vti_esp6_protocol, IPPROTO_ESP);
 	if (err < 0) {
-		unregister_pernet_device(&vti6_net_ops);
 		pr_err("%s: can't register vti6 protocol\n", __func__);
 
 		goto out;
@@ -1106,7 +1108,6 @@
 	err = xfrm6_protocol_register(&vti_ah6_protocol, IPPROTO_AH);
 	if (err < 0) {
 		xfrm6_protocol_deregister(&vti_esp6_protocol, IPPROTO_ESP);
-		unregister_pernet_device(&vti6_net_ops);
 		pr_err("%s: can't register vti6 protocol\n", __func__);
 
 		goto out;
@@ -1116,7 +1117,6 @@
 	if (err < 0) {
 		xfrm6_protocol_deregister(&vti_ah6_protocol, IPPROTO_AH);
 		xfrm6_protocol_deregister(&vti_esp6_protocol, IPPROTO_ESP);
-		unregister_pernet_device(&vti6_net_ops);
 		pr_err("%s: can't register vti6 protocol\n", __func__);
 
 		goto out;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 09a22f4..ca8d4ea 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -851,7 +851,7 @@
 static void ndisc_recv_na(struct sk_buff *skb)
 {
 	struct nd_msg *msg = (struct nd_msg *)skb_transport_header(skb);
-	const struct in6_addr *saddr = &ipv6_hdr(skb)->saddr;
+	struct in6_addr *saddr = &ipv6_hdr(skb)->saddr;
 	const struct in6_addr *daddr = &ipv6_hdr(skb)->daddr;
 	u8 *lladdr = NULL;
 	u32 ndoptlen = skb_tail_pointer(skb) - (skb_transport_header(skb) +
@@ -944,10 +944,7 @@
 			/*
 			 * Change: router to host
 			 */
-			struct rt6_info *rt;
-			rt = rt6_get_dflt_router(saddr, dev);
-			if (rt)
-				ip6_del_rt(rt);
+			rt6_clean_tohost(dev_net(dev),  saddr);
 		}
 
 out:
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index 95f3f1d..d38e6a8 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -30,13 +30,15 @@
 		.daddr = iph->daddr,
 		.saddr = iph->saddr,
 	};
+	int err;
 
 	dst = ip6_route_output(net, skb->sk, &fl6);
-	if (dst->error) {
+	err = dst->error;
+	if (err) {
 		IP6_INC_STATS(net, ip6_dst_idev(dst), IPSTATS_MIB_OUTNOROUTES);
 		LIMIT_NETDEBUG(KERN_DEBUG "ip6_route_me_harder: No more route.\n");
 		dst_release(dst);
-		return dst->error;
+		return err;
 	}
 
 	/* Drop old route. */
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 004fffb..6ebdb7b6 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -2234,6 +2234,27 @@
 	fib6_clean_all(net, fib6_remove_prefsrc, &adni);
 }
 
+#define RTF_RA_ROUTER		(RTF_ADDRCONF | RTF_DEFAULT | RTF_GATEWAY)
+#define RTF_CACHE_GATEWAY	(RTF_GATEWAY | RTF_CACHE)
+
+/* Remove routers and update dst entries when gateway turn into host. */
+static int fib6_clean_tohost(struct rt6_info *rt, void *arg)
+{
+	struct in6_addr *gateway = (struct in6_addr *)arg;
+
+	if ((((rt->rt6i_flags & RTF_RA_ROUTER) == RTF_RA_ROUTER) ||
+	     ((rt->rt6i_flags & RTF_CACHE_GATEWAY) == RTF_CACHE_GATEWAY)) &&
+	     ipv6_addr_equal(gateway, &rt->rt6i_gateway)) {
+		return -1;
+	}
+	return 0;
+}
+
+void rt6_clean_tohost(struct net *net, struct in6_addr *gateway)
+{
+	fib6_clean_all(net, fib6_clean_tohost, gateway);
+}
+
 struct arg_dev_net {
 	struct net_device *dev;
 	struct net *net;
@@ -2709,6 +2730,9 @@
 	if (tb[RTA_OIF])
 		oif = nla_get_u32(tb[RTA_OIF]);
 
+	if (tb[RTA_MARK])
+		fl6.flowi6_mark = nla_get_u32(tb[RTA_MARK]);
+
 	if (iif) {
 		struct net_device *dev;
 		int flags = 0;
diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c
index 0d78132..8517d3c 100644
--- a/net/ipv6/tcpv6_offload.c
+++ b/net/ipv6/tcpv6_offload.c
@@ -42,7 +42,7 @@
 	if (NAPI_GRO_CB(skb)->flush)
 		goto skip_csum;
 
-	wsum = skb->csum;
+	wsum = NAPI_GRO_CB(skb)->csum;
 
 	switch (skb->ip_summed) {
 	case CHECKSUM_NONE:
diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c
index 19ef329..b930d08 100644
--- a/net/ipv6/xfrm6_output.c
+++ b/net/ipv6/xfrm6_output.c
@@ -114,12 +114,6 @@
 	if (err)
 		return err;
 
-	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
-#ifdef CONFIG_NETFILTER
-	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
-#endif
-
-	skb->protocol = htons(ETH_P_IPV6);
 	skb->local_df = 1;
 
 	return x->outer_mode->output2(x, skb);
@@ -128,11 +122,13 @@
 
 int xfrm6_output_finish(struct sk_buff *skb)
 {
+	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
+	skb->protocol = htons(ETH_P_IPV6);
+
 #ifdef CONFIG_NETFILTER
 	IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED;
 #endif
 
-	skb->protocol = htons(ETH_P_IPV6);
 	return xfrm_output(skb);
 }
 
@@ -142,6 +138,13 @@
 	struct xfrm_state *x = dst->xfrm;
 	int mtu;
 
+#ifdef CONFIG_NETFILTER
+	if (!x) {
+		IP6CB(skb)->flags |= IP6SKB_REROUTED;
+		return dst_output(skb);
+	}
+#endif
+
 	if (skb->protocol == htons(ETH_P_IPV6))
 		mtu = ip6_skb_dst_mtu(skb);
 	else
@@ -165,6 +168,7 @@
 
 int xfrm6_output(struct sock *sk, struct sk_buff *skb)
 {
-	return NF_HOOK(NFPROTO_IPV6, NF_INET_POST_ROUTING, skb, NULL,
-		       skb_dst(skb)->dev, __xfrm6_output);
+	return NF_HOOK_COND(NFPROTO_IPV6, NF_INET_POST_ROUTING, skb,
+			    NULL, skb_dst(skb)->dev, __xfrm6_output,
+			    !(IP6CB(skb)->flags & IP6SKB_REROUTED));
 }
diff --git a/net/ipv6/xfrm6_protocol.c b/net/ipv6/xfrm6_protocol.c
index 6ab989c..54d13f8 100644
--- a/net/ipv6/xfrm6_protocol.c
+++ b/net/ipv6/xfrm6_protocol.c
@@ -50,6 +50,10 @@
 {
 	int ret;
 	struct xfrm6_protocol *handler;
+	struct xfrm6_protocol __rcu **head = proto_handlers(protocol);
+
+	if (!head)
+		return 0;
 
 	for_each_protocol_rcu(*proto_handlers(protocol), handler)
 		if ((ret = handler->cb_handler(skb, err)) <= 0)
@@ -184,10 +188,12 @@
 	struct xfrm6_protocol __rcu **pprev;
 	struct xfrm6_protocol *t;
 	bool add_netproto = false;
-
 	int ret = -EEXIST;
 	int priority = handler->priority;
 
+	if (!proto_handlers(protocol) || !netproto(protocol))
+		return -EINVAL;
+
 	mutex_lock(&xfrm6_protocol_mutex);
 
 	if (!rcu_dereference_protected(*proto_handlers(protocol),
@@ -230,6 +236,9 @@
 	struct xfrm6_protocol *t;
 	int ret = -ENOENT;
 
+	if (!proto_handlers(protocol) || !netproto(protocol))
+		return -EINVAL;
+
 	mutex_lock(&xfrm6_protocol_mutex);
 
 	for (pprev = proto_handlers(protocol);
diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c
index 01e77b0..8c9d730 100644
--- a/net/iucv/af_iucv.c
+++ b/net/iucv/af_iucv.c
@@ -1830,7 +1830,7 @@
 		spin_lock_irqsave(&list->lock, flags);
 
 		while (list_skb != (struct sk_buff *)list) {
-			if (msg->tag != IUCV_SKB_CB(list_skb)->tag) {
+			if (msg->tag == IUCV_SKB_CB(list_skb)->tag) {
 				this = list_skb;
 				break;
 			}
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index 222c28b..f169b6e 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -317,6 +317,7 @@
 
 	bool started, abort, hw_begun, notified;
 	bool to_be_freed;
+	bool on_channel;
 
 	unsigned long hw_start_time;
 
diff --git a/net/mac80211/mlme.c b/net/mac80211/mlme.c
index dee50ae..27600a9 100644
--- a/net/mac80211/mlme.c
+++ b/net/mac80211/mlme.c
@@ -3598,18 +3598,24 @@
 
 	sdata_lock(sdata);
 
-	if (ifmgd->auth_data) {
+	if (ifmgd->auth_data || ifmgd->assoc_data) {
+		const u8 *bssid = ifmgd->auth_data ?
+				ifmgd->auth_data->bss->bssid :
+				ifmgd->assoc_data->bss->bssid;
+
 		/*
-		 * If we are trying to authenticate while suspending, cfg80211
-		 * won't know and won't actually abort those attempts, thus we
-		 * need to do that ourselves.
+		 * If we are trying to authenticate / associate while suspending,
+		 * cfg80211 won't know and won't actually abort those attempts,
+		 * thus we need to do that ourselves.
 		 */
-		ieee80211_send_deauth_disassoc(sdata,
-					       ifmgd->auth_data->bss->bssid,
+		ieee80211_send_deauth_disassoc(sdata, bssid,
 					       IEEE80211_STYPE_DEAUTH,
 					       WLAN_REASON_DEAUTH_LEAVING,
 					       false, frame_buf);
-		ieee80211_destroy_auth_data(sdata, false);
+		if (ifmgd->assoc_data)
+			ieee80211_destroy_assoc_data(sdata, false);
+		if (ifmgd->auth_data)
+			ieee80211_destroy_auth_data(sdata, false);
 		cfg80211_tx_mlme_mgmt(sdata->dev, frame_buf,
 				      IEEE80211_DEAUTH_FRAME_LEN);
 	}
diff --git a/net/mac80211/offchannel.c b/net/mac80211/offchannel.c
index 6fb3855..7a17dec 100644
--- a/net/mac80211/offchannel.c
+++ b/net/mac80211/offchannel.c
@@ -333,7 +333,7 @@
 		container_of(work, struct ieee80211_roc_work, work.work);
 	struct ieee80211_sub_if_data *sdata = roc->sdata;
 	struct ieee80211_local *local = sdata->local;
-	bool started;
+	bool started, on_channel;
 
 	mutex_lock(&local->mtx);
 
@@ -354,14 +354,26 @@
 	if (!roc->started) {
 		struct ieee80211_roc_work *dep;
 
-		/* start this ROC */
-		ieee80211_offchannel_stop_vifs(local);
+		WARN_ON(local->use_chanctx);
 
-		/* switch channel etc */
+		/* If actually operating on the desired channel (with at least
+		 * 20 MHz channel width) don't stop all the operations but still
+		 * treat it as though the ROC operation started properly, so
+		 * other ROC operations won't interfere with this one.
+		 */
+		roc->on_channel = roc->chan == local->_oper_chandef.chan &&
+				  local->_oper_chandef.width != NL80211_CHAN_WIDTH_5 &&
+				  local->_oper_chandef.width != NL80211_CHAN_WIDTH_10;
+
+		/* start this ROC */
 		ieee80211_recalc_idle(local);
 
-		local->tmp_channel = roc->chan;
-		ieee80211_hw_config(local, 0);
+		if (!roc->on_channel) {
+			ieee80211_offchannel_stop_vifs(local);
+
+			local->tmp_channel = roc->chan;
+			ieee80211_hw_config(local, 0);
+		}
 
 		/* tell userspace or send frame */
 		ieee80211_handle_roc_started(roc);
@@ -380,9 +392,10 @@
  finish:
 		list_del(&roc->list);
 		started = roc->started;
+		on_channel = roc->on_channel;
 		ieee80211_roc_notify_destroy(roc, !roc->abort);
 
-		if (started) {
+		if (started && !on_channel) {
 			ieee80211_flush_queues(local, NULL);
 
 			local->tmp_channel = NULL;
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index 216c45b..2b608b2 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1231,7 +1231,8 @@
 		if (ether_addr_equal(bssid, rx->sdata->u.ibss.bssid) &&
 		    test_sta_flag(sta, WLAN_STA_AUTHORIZED)) {
 			sta->last_rx = jiffies;
-			if (ieee80211_is_data(hdr->frame_control)) {
+			if (ieee80211_is_data(hdr->frame_control) &&
+			    !is_multicast_ether_addr(hdr->addr1)) {
 				sta->last_rx_rate_idx = status->rate_idx;
 				sta->last_rx_rate_flag = status->flag;
 				sta->last_rx_rate_vht_flag = status->vht_flag;
diff --git a/net/mac80211/sta_info.c b/net/mac80211/sta_info.c
index 137a192..847d92f 100644
--- a/net/mac80211/sta_info.c
+++ b/net/mac80211/sta_info.c
@@ -1148,7 +1148,8 @@
 	atomic_dec(&ps->num_sta_ps);
 
 	/* This station just woke up and isn't aware of our SMPS state */
-	if (!ieee80211_smps_is_restrictive(sta->known_smps_mode,
+	if (!ieee80211_vif_is_mesh(&sdata->vif) &&
+	    !ieee80211_smps_is_restrictive(sta->known_smps_mode,
 					   sdata->smps_mode) &&
 	    sta->known_smps_mode != sdata->bss->req_smps &&
 	    sta_info_tx_streams(sta) != 1) {
diff --git a/net/mac80211/status.c b/net/mac80211/status.c
index 00ba90b0..60cb7a6 100644
--- a/net/mac80211/status.c
+++ b/net/mac80211/status.c
@@ -314,10 +314,9 @@
 	    !is_multicast_ether_addr(hdr->addr1))
 		txflags |= IEEE80211_RADIOTAP_F_TX_FAIL;
 
-	if ((info->status.rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS) ||
-	    (info->status.rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT))
+	if (info->status.rates[0].flags & IEEE80211_TX_RC_USE_CTS_PROTECT)
 		txflags |= IEEE80211_RADIOTAP_F_TX_CTS;
-	else if (info->status.rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS)
+	if (info->status.rates[0].flags & IEEE80211_TX_RC_USE_RTS_CTS)
 		txflags |= IEEE80211_RADIOTAP_F_TX_RTS;
 
 	put_unaligned_le16(txflags, pos);
diff --git a/net/mac80211/trace.h b/net/mac80211/trace.h
index a0b0aea..cec5b60 100644
--- a/net/mac80211/trace.h
+++ b/net/mac80211/trace.h
@@ -21,10 +21,10 @@
 
 #define VIF_ENTRY	__field(enum nl80211_iftype, vif_type) __field(void *, sdata)	\
 			__field(bool, p2p)						\
-			__string(vif_name, sdata->dev ? sdata->dev->name : "<nodev>")
+			__string(vif_name, sdata->name)
 #define VIF_ASSIGN	__entry->vif_type = sdata->vif.type; __entry->sdata = sdata;	\
 			__entry->p2p = sdata->vif.p2p;					\
-			__assign_str(vif_name, sdata->dev ? sdata->dev->name : sdata->name)
+			__assign_str(vif_name, sdata->name)
 #define VIF_PR_FMT	" vif:%s(%d%s)"
 #define VIF_PR_ARG	__get_str(vif_name), __entry->vif_type, __entry->p2p ? "/p2p" : ""
 
diff --git a/net/mac80211/util.c b/net/mac80211/util.c
index 275c94f..3c36583 100644
--- a/net/mac80211/util.c
+++ b/net/mac80211/util.c
@@ -1780,7 +1780,7 @@
 	mutex_unlock(&local->mtx);
 
 	if (sched_scan_stopped)
-		cfg80211_sched_scan_stopped(local->hw.wiphy);
+		cfg80211_sched_scan_stopped_rtnl(local->hw.wiphy);
 
 	/*
 	 * If this is for hw restart things are still running.
diff --git a/net/mac80211/vht.c b/net/mac80211/vht.c
index e9e36a2..9265adfd 100644
--- a/net/mac80211/vht.c
+++ b/net/mac80211/vht.c
@@ -129,9 +129,12 @@
 	if (!vht_cap_ie || !sband->vht_cap.vht_supported)
 		return;
 
-	/* A VHT STA must support 40 MHz */
-	if (!(sta->sta.ht_cap.cap & IEEE80211_HT_CAP_SUP_WIDTH_20_40))
-		return;
+	/*
+	 * A VHT STA must support 40 MHz, but if we verify that here
+	 * then we break a few things - some APs (e.g. Netgear R6300v2
+	 * and others based on the BCM4360 chipset) will unset this
+	 * capability bit when operating in 20 MHz.
+	 */
 
 	vht_cap->vht_supported = true;
 
diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c
index ccc46fa..5857963 100644
--- a/net/netfilter/nf_conntrack_netlink.c
+++ b/net/netfilter/nf_conntrack_netlink.c
@@ -1336,6 +1336,9 @@
 #ifdef CONFIG_NF_NAT_NEEDED
 	int ret;
 
+	if (!cda[CTA_NAT_DST] && !cda[CTA_NAT_SRC])
+		return 0;
+
 	ret = ctnetlink_parse_nat_setup(ct, NF_NAT_MANIP_DST,
 					cda[CTA_NAT_DST]);
 	if (ret < 0)
diff --git a/net/netfilter/nf_tables_core.c b/net/netfilter/nf_tables_core.c
index 8041053..345acfb 100644
--- a/net/netfilter/nf_tables_core.c
+++ b/net/netfilter/nf_tables_core.c
@@ -66,20 +66,6 @@
 	int			rulenum;
 };
 
-static inline void
-nft_chain_stats(const struct nft_chain *this, const struct nft_pktinfo *pkt,
-		struct nft_jumpstack *jumpstack, unsigned int stackptr)
-{
-	struct nft_stats __percpu *stats;
-	const struct nft_chain *chain = stackptr ? jumpstack[0].chain : this;
-
-	rcu_read_lock_bh();
-	stats = rcu_dereference(nft_base_chain(chain)->stats);
-	__this_cpu_inc(stats->pkts);
-	__this_cpu_add(stats->bytes, pkt->skb->len);
-	rcu_read_unlock_bh();
-}
-
 enum nft_trace {
 	NFT_TRACE_RULE,
 	NFT_TRACE_RETURN,
@@ -117,13 +103,14 @@
 unsigned int
 nft_do_chain(struct nft_pktinfo *pkt, const struct nf_hook_ops *ops)
 {
-	const struct nft_chain *chain = ops->priv;
+	const struct nft_chain *chain = ops->priv, *basechain = chain;
 	const struct nft_rule *rule;
 	const struct nft_expr *expr, *last;
 	struct nft_data data[NFT_REG_MAX + 1];
 	unsigned int stackptr = 0;
 	struct nft_jumpstack jumpstack[NFT_JUMP_STACK_SIZE];
-	int rulenum = 0;
+	struct nft_stats __percpu *stats;
+	int rulenum;
 	/*
 	 * Cache cursor to avoid problems in case that the cursor is updated
 	 * while traversing the ruleset.
@@ -131,6 +118,7 @@
 	unsigned int gencursor = ACCESS_ONCE(chain->net->nft.gencursor);
 
 do_chain:
+	rulenum = 0;
 	rule = list_entry(&chain->rules, struct nft_rule, list);
 next_rule:
 	data[NFT_REG_VERDICT].verdict = NFT_CONTINUE;
@@ -156,8 +144,10 @@
 		switch (data[NFT_REG_VERDICT].verdict) {
 		case NFT_BREAK:
 			data[NFT_REG_VERDICT].verdict = NFT_CONTINUE;
-			/* fall through */
+			continue;
 		case NFT_CONTINUE:
+			if (unlikely(pkt->skb->nf_trace))
+				nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE);
 			continue;
 		}
 		break;
@@ -183,37 +173,44 @@
 		jumpstack[stackptr].rule  = rule;
 		jumpstack[stackptr].rulenum = rulenum;
 		stackptr++;
-		/* fall through */
+		chain = data[NFT_REG_VERDICT].chain;
+		goto do_chain;
 	case NFT_GOTO:
+		if (unlikely(pkt->skb->nf_trace))
+			nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RULE);
+
 		chain = data[NFT_REG_VERDICT].chain;
 		goto do_chain;
 	case NFT_RETURN:
 		if (unlikely(pkt->skb->nf_trace))
 			nft_trace_packet(pkt, chain, rulenum, NFT_TRACE_RETURN);
-
-		/* fall through */
+		break;
 	case NFT_CONTINUE:
+		if (unlikely(pkt->skb->nf_trace && !(chain->flags & NFT_BASE_CHAIN)))
+			nft_trace_packet(pkt, chain, ++rulenum, NFT_TRACE_RETURN);
 		break;
 	default:
 		WARN_ON(1);
 	}
 
 	if (stackptr > 0) {
-		if (unlikely(pkt->skb->nf_trace))
-			nft_trace_packet(pkt, chain, ++rulenum, NFT_TRACE_RETURN);
-
 		stackptr--;
 		chain = jumpstack[stackptr].chain;
 		rule  = jumpstack[stackptr].rule;
 		rulenum = jumpstack[stackptr].rulenum;
 		goto next_rule;
 	}
-	nft_chain_stats(chain, pkt, jumpstack, stackptr);
 
 	if (unlikely(pkt->skb->nf_trace))
-		nft_trace_packet(pkt, chain, ++rulenum, NFT_TRACE_POLICY);
+		nft_trace_packet(pkt, basechain, -1, NFT_TRACE_POLICY);
 
-	return nft_base_chain(chain)->policy;
+	rcu_read_lock_bh();
+	stats = rcu_dereference(nft_base_chain(basechain)->stats);
+	__this_cpu_inc(stats->pkts);
+	__this_cpu_add(stats->bytes, pkt->skb->len);
+	rcu_read_unlock_bh();
+
+	return nft_base_chain(basechain)->policy;
 }
 EXPORT_SYMBOL_GPL(nft_do_chain);
 
diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c
index e009087..23ef77c 100644
--- a/net/netfilter/nfnetlink.c
+++ b/net/netfilter/nfnetlink.c
@@ -256,15 +256,15 @@
 #endif
 		{
 			nfnl_unlock(subsys_id);
-			kfree_skb(nskb);
-			return netlink_ack(skb, nlh, -EOPNOTSUPP);
+			netlink_ack(skb, nlh, -EOPNOTSUPP);
+			return kfree_skb(nskb);
 		}
 	}
 
 	if (!ss->commit || !ss->abort) {
 		nfnl_unlock(subsys_id);
-		kfree_skb(nskb);
-		return netlink_ack(skb, nlh, -EOPNOTSUPP);
+		netlink_ack(skb, nlh, -EOPNOTSUPP);
+		return kfree_skb(skb);
 	}
 
 	while (skb->len >= nlmsg_total_size(0)) {
diff --git a/net/rxrpc/ar-key.c b/net/rxrpc/ar-key.c
index 7633a75..0ad0807 100644
--- a/net/rxrpc/ar-key.c
+++ b/net/rxrpc/ar-key.c
@@ -99,7 +99,7 @@
 	_debug("tktlen: %x", tktlen);
 	if (tktlen > AFSTOKEN_RK_TIX_MAX)
 		return -EKEYREJECTED;
-	if (8 * 4 + tktlen != toklen)
+	if (toklen < 8 * 4 + tktlen)
 		return -EKEYREJECTED;
 
 	plen = sizeof(*token) + sizeof(*token->kad) + tktlen;
diff --git a/net/sched/cls_tcindex.c b/net/sched/cls_tcindex.c
index eed8404..f435a88 100644
--- a/net/sched/cls_tcindex.c
+++ b/net/sched/cls_tcindex.c
@@ -188,6 +188,12 @@
 	[TCA_TCINDEX_CLASSID]		= { .type = NLA_U32 },
 };
 
+static void tcindex_filter_result_init(struct tcindex_filter_result *r)
+{
+	memset(r, 0, sizeof(*r));
+	tcf_exts_init(&r->exts, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE);
+}
+
 static int
 tcindex_set_parms(struct net *net, struct tcf_proto *tp, unsigned long base,
 		  u32 handle, struct tcindex_data *p,
@@ -207,15 +213,11 @@
 		return err;
 
 	memcpy(&cp, p, sizeof(cp));
-	memset(&new_filter_result, 0, sizeof(new_filter_result));
-	tcf_exts_init(&new_filter_result.exts, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE);
+	tcindex_filter_result_init(&new_filter_result);
 
+	tcindex_filter_result_init(&cr);
 	if (old_r)
-		memcpy(&cr, r, sizeof(cr));
-	else {
-		memset(&cr, 0, sizeof(cr));
-		tcf_exts_init(&cr.exts, TCA_TCINDEX_ACT, TCA_TCINDEX_POLICE);
-	}
+		cr.res = r->res;
 
 	if (tb[TCA_TCINDEX_HASH])
 		cp.hash = nla_get_u32(tb[TCA_TCINDEX_HASH]);
@@ -267,9 +269,14 @@
 	err = -ENOMEM;
 	if (!cp.perfect && !cp.h) {
 		if (valid_perfect_hash(&cp)) {
+			int i;
+
 			cp.perfect = kcalloc(cp.hash, sizeof(*r), GFP_KERNEL);
 			if (!cp.perfect)
 				goto errout;
+			for (i = 0; i < cp.hash; i++)
+				tcf_exts_init(&cp.perfect[i].exts, TCA_TCINDEX_ACT,
+					      TCA_TCINDEX_POLICE);
 			balloc = 1;
 		} else {
 			cp.h = kcalloc(cp.hash, sizeof(f), GFP_KERNEL);
@@ -295,14 +302,17 @@
 		tcf_bind_filter(tp, &cr.res, base);
 	}
 
-	tcf_exts_change(tp, &cr.exts, &e);
+	if (old_r)
+		tcf_exts_change(tp, &r->exts, &e);
+	else
+		tcf_exts_change(tp, &cr.exts, &e);
 
 	tcf_tree_lock(tp);
 	if (old_r && old_r != r)
-		memset(old_r, 0, sizeof(*old_r));
+		tcindex_filter_result_init(old_r);
 
 	memcpy(p, &cp, sizeof(cp));
-	memcpy(r, &cr, sizeof(cr));
+	r->res = cr.res;
 
 	if (r == &new_filter_result) {
 		struct tcindex_filter **fp;
diff --git a/net/wireless/scan.c b/net/wireless/scan.c
index 7d09a71..88f108e 100644
--- a/net/wireless/scan.c
+++ b/net/wireless/scan.c
@@ -284,14 +284,22 @@
 }
 EXPORT_SYMBOL(cfg80211_sched_scan_results);
 
-void cfg80211_sched_scan_stopped(struct wiphy *wiphy)
+void cfg80211_sched_scan_stopped_rtnl(struct wiphy *wiphy)
 {
 	struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy);
 
+	ASSERT_RTNL();
+
 	trace_cfg80211_sched_scan_stopped(wiphy);
 
-	rtnl_lock();
 	__cfg80211_stop_sched_scan(rdev, true);
+}
+EXPORT_SYMBOL(cfg80211_sched_scan_stopped_rtnl);
+
+void cfg80211_sched_scan_stopped(struct wiphy *wiphy)
+{
+	rtnl_lock();
+	cfg80211_sched_scan_stopped_rtnl(wiphy);
 	rtnl_unlock();
 }
 EXPORT_SYMBOL(cfg80211_sched_scan_stopped);
diff --git a/net/wireless/sme.c b/net/wireless/sme.c
index acdcb4a8..3546a77 100644
--- a/net/wireless/sme.c
+++ b/net/wireless/sme.c
@@ -234,7 +234,6 @@
 					NULL, 0, NULL, 0,
 					WLAN_STATUS_UNSPECIFIED_FAILURE,
 					false, NULL);
-			cfg80211_sme_free(wdev);
 		}
 		wdev_unlock(wdev);
 	}
@@ -648,6 +647,7 @@
 			cfg80211_unhold_bss(bss_from_pub(bss));
 			cfg80211_put_bss(wdev->wiphy, bss);
 		}
+		cfg80211_sme_free(wdev);
 		return;
 	}
 
diff --git a/tools/Makefile b/tools/Makefile
index bcae806..9a617ad 100644
--- a/tools/Makefile
+++ b/tools/Makefile
@@ -44,6 +44,9 @@
 cgroup firewire hv guest usb virtio vm net: FORCE
 	$(call descend,$@)
 
+liblockdep: FORCE
+	$(call descend,lib/lockdep)
+
 libapikfs: FORCE
 	$(call descend,lib/api)
 
@@ -91,6 +94,9 @@
 cgroup_clean hv_clean firewire_clean lguest_clean usb_clean virtio_clean vm_clean net_clean:
 	$(call descend,$(@:_clean=),clean)
 
+liblockdep_clean:
+	$(call descend,lib/lockdep,clean)
+
 libapikfs_clean:
 	$(call descend,lib/api,clean)
 
diff --git a/tools/lib/lockdep/Makefile b/tools/lib/lockdep/Makefile
index cb09d3f..bba2f52 100644
--- a/tools/lib/lockdep/Makefile
+++ b/tools/lib/lockdep/Makefile
@@ -1,8 +1,7 @@
 # file format version
 FILE_VERSION = 1
 
-MAKEFLAGS += --no-print-directory
-LIBLOCKDEP_VERSION=$(shell make -sC ../../.. kernelversion)
+LIBLOCKDEP_VERSION=$(shell make --no-print-directory -sC ../../.. kernelversion)
 
 # Makefiles suck: This macro sets a default value of $(2) for the
 # variable named by $(1), unless the variable has been set by
@@ -231,7 +230,7 @@
 install: install_lib
 
 clean:
-	$(RM) *.o *~ $(TARGETS) *.a *.so $(VERSION_FILES) .*.d
+	$(RM) *.o *~ $(TARGETS) *.a *liblockdep*.so* $(VERSION_FILES) .*.d
 	$(RM) tags TAGS
 
 endif # skip-makefile