Re: [06/44] numa: fix slab_node(MPOL_BIND)

From: Eric Dumazet
Date: Wed Dec 08 2010 - 00:07:19 EST


Le mercredi 08 dÃcembre 2010 Ã 05:33 +0100, Eric Dumazet a Ãcrit :

> By the way, anybody knows how I can emulate a memoryless node on a dual
> node x86_64 machine (with memory present on both nodes) ?
>
>

this hack works for me :

diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c
index a35cb9d..1087333 100644
--- a/arch/x86/mm/srat_64.c
+++ b/arch/x86/mm/srat_64.c
@@ -261,6 +261,7 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma)
end = start + ma->length;
pxm = ma->proximity_domain;
node = setup_node(pxm);
+ node = 0;
if (node < 0) {
printk(KERN_ERR "SRAT: Too many proximity domains.\n");
bad_srat();

[ 0.000000] SRAT: PXM 0 -> APIC 0x00 -> Node 0
[ 0.000000] SRAT: PXM 0 -> APIC 0x01 -> Node 0
[ 0.000000] SRAT: PXM 0 -> APIC 0x02 -> Node 0
[ 0.000000] SRAT: PXM 0 -> APIC 0x03 -> Node 0
[ 0.000000] SRAT: PXM 0 -> APIC 0x04 -> Node 0
[ 0.000000] SRAT: PXM 0 -> APIC 0x05 -> Node 0
[ 0.000000] SRAT: PXM 0 -> APIC 0x06 -> Node 0
[ 0.000000] SRAT: PXM 0 -> APIC 0x07 -> Node 0
[ 0.000000] SRAT: PXM 1 -> APIC 0x10 -> Node 1
[ 0.000000] SRAT: PXM 1 -> APIC 0x11 -> Node 1
[ 0.000000] SRAT: PXM 1 -> APIC 0x12 -> Node 1
[ 0.000000] SRAT: PXM 1 -> APIC 0x13 -> Node 1
[ 0.000000] SRAT: PXM 1 -> APIC 0x14 -> Node 1
[ 0.000000] SRAT: PXM 1 -> APIC 0x15 -> Node 1
[ 0.000000] SRAT: PXM 1 -> APIC 0x16 -> Node 1
[ 0.000000] SRAT: PXM 1 -> APIC 0x17 -> Node 1
[ 0.000000] SRAT: Node 0 PXM 0 0-80000000
[ 0.000000] SRAT: Node 0 PXM 1 80000000-e0000000
[ 0.000000] SRAT: Node 0 PXM 1 100000000-120000000
[ 0.000000] SRAT: Node 0 [0,80000000) + [80000000,e0000000) -> [0,e0000000)
[ 0.000000] SRAT: Node 0 [0,e0000000) + [100000000,120000000) -> [0,120000000)
[ 0.000000] NUMA: Using 63 for the hash shift.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/