x86: get mp_bus_to_node early
Currently, on an amd k8 system with multi ht chains, the numa_node of
pci devices under /sys/devices/pci0000:80/* is always 0, even if that
chain is on node 1 or 2 or 3.
Workaround: pcibus_to_node(bus) is used when we want to get the node that
pci_device is on.
In struct device, we already have numa_node member, and we could use
dev_to_node()/set_dev_node() to get and set numa_node in the device.
set_dev_node is called in pci_device_add() with pcibus_to_node(bus),
and pcibus_to_node uses bus->sysdata for nodeid.
The problem is when pci_add_device is called, bus->sysdata is not assigned
correct nodeid yet. The result is that numa_node will always be 0.
pcibios_scan_root and pci_scan_root could take sysdata. So we need to get
mp_bus_to_node mapping before these two are called, and thus
get_mp_bus_to_node could get correct node for sysdata in root bus.
In scanning of the root bus, all child busses will take parent bus sysdata.
So all pci_device->dev.numa_node will be assigned correctly and automatically.
Later we could use dev_to_node(&pci_dev->dev) to get numa_node, and we
could also could make other bus specific device get the correct numa_node
too.
This is an updated version of pci_sysdata and Jeff's pci_domain patch.
[ mingo@elte.hu: build fix ]
Signed-off-by: Yinghai Lu <yinghai.lu@sun.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
diff --git a/arch/x86/pci/k8-bus_64.c b/arch/x86/pci/k8-bus_64.c
index 9cc813e..3903efbc 100644
--- a/arch/x86/pci/k8-bus_64.c
+++ b/arch/x86/pci/k8-bus_64.c
@@ -1,7 +1,9 @@
#include <linux/init.h>
#include <linux/pci.h>
+#include <asm/pci-direct.h>
#include <asm/mpspec.h>
#include <linux/cpumask.h>
+#include <linux/topology.h>
/*
* This discovers the pcibus <-> node mapping on AMD K8.
@@ -20,64 +22,102 @@
#define SUBORDINATE_LDT_BUS_NUMBER(dword) ((dword >> 16) & 0xFF)
#define PCI_DEVICE_ID_K8HTCONFIG 0x1100
+#ifdef CONFIG_NUMA
+
+#define BUS_NR 256
+
+static int mp_bus_to_node[BUS_NR];
+
+void set_mp_bus_to_node(int busnum, int node)
+{
+ if (busnum >= 0 && busnum < BUS_NR)
+ mp_bus_to_node[busnum] = node;
+}
+
+int get_mp_bus_to_node(int busnum)
+{
+ int node = -1;
+
+ if (busnum < 0 || busnum > (BUS_NR - 1))
+ return node;
+
+ node = mp_bus_to_node[busnum];
+
+ /*
+ * let numa_node_id to decide it later in dma_alloc_pages
+ * if there is no ram on that node
+ */
+ if (node != -1 && !node_online(node))
+ node = -1;
+
+ return node;
+}
+
+#endif
+
/**
- * fill_mp_bus_to_cpumask()
+ * early_fill_mp_bus_to_node()
+ * called before pcibios_scan_root and pci_scan_bus
* fills the mp_bus_to_cpumask array based according to the LDT Bus Number
* Registers found in the K8 northbridge
*/
__init static int
-fill_mp_bus_to_cpumask(void)
+early_fill_mp_bus_to_node(void)
{
- struct pci_dev *nb_dev = NULL;
+#ifdef CONFIG_NUMA
int i, j;
+ unsigned slot;
u32 ldtbus, nid;
+ u32 id;
static int lbnr[3] = {
LDT_BUS_NUMBER_REGISTER_0,
LDT_BUS_NUMBER_REGISTER_1,
LDT_BUS_NUMBER_REGISTER_2
};
- while ((nb_dev = pci_get_device(PCI_VENDOR_ID_AMD,
- PCI_DEVICE_ID_K8HTCONFIG, nb_dev))) {
- pci_read_config_dword(nb_dev, NODE_ID_REGISTER, &nid);
+ for (i = 0; i < BUS_NR; i++)
+ mp_bus_to_node[i] = -1;
+
+ if (!early_pci_allowed())
+ return -1;
+
+ for (slot = 0x18; slot < 0x20; slot++) {
+ id = read_pci_config(0, slot, 0, PCI_VENDOR_ID);
+ if (id != (PCI_VENDOR_ID_AMD | (PCI_DEVICE_ID_K8HTCONFIG<<16)))
+ break;
+ nid = read_pci_config(0, slot, 0, NODE_ID_REGISTER);
for (i = 0; i < NR_LDT_BUS_NUMBER_REGISTERS; i++) {
- pci_read_config_dword(nb_dev, lbnr[i], &ldtbus);
+ ldtbus = read_pci_config(0, slot, 0, lbnr[i]);
/*
* if there are no busses hanging off of the current
* ldt link then both the secondary and subordinate
* bus number fields are set to 0.
- *
+ *
* RED-PEN
* This is slightly broken because it assumes
- * HT node IDs == Linux node ids, which is not always
+ * HT node IDs == Linux node ids, which is not always
* true. However it is probably mostly true.
*/
if (!(SECONDARY_LDT_BUS_NUMBER(ldtbus) == 0
&& SUBORDINATE_LDT_BUS_NUMBER(ldtbus) == 0)) {
for (j = SECONDARY_LDT_BUS_NUMBER(ldtbus);
j <= SUBORDINATE_LDT_BUS_NUMBER(ldtbus);
- j++) {
- struct pci_bus *bus;
- struct pci_sysdata *sd;
-
- long node = NODE_ID(nid);
- /* Algorithm a bit dumb, but
- it shouldn't matter here */
- bus = pci_find_bus(0, j);
- if (!bus)
- continue;
- if (!node_online(node))
- node = 0;
-
- sd = bus->sysdata;
- sd->node = node;
- }
+ j++) {
+ int node = NODE_ID(nid);
+ mp_bus_to_node[j] = (unsigned char)node;
+ }
}
}
}
+ for (i = 0; i < BUS_NR; i++) {
+ int node = mp_bus_to_node[i];
+ if (node >= 0)
+ printk(KERN_DEBUG "bus: %02x to node: %02x\n", i, node);
+ }
+#endif
return 0;
}
-fs_initcall(fill_mp_bus_to_cpumask);
+postcore_initcall(early_fill_mp_bus_to_node);