* [PATCH RESEND] x86: Fix sibling map with NumaChip
@ 2015-03-03 15:18 Daniel J Blueman
2015-03-03 16:29 ` Borislav Petkov
0 siblings, 1 reply; 2+ messages in thread
From: Daniel J Blueman @ 2015-03-03 15:18 UTC (permalink / raw)
To: Ingo Molnar, Thomas Gleixner, H. Peter Anvin
Cc: x86, linux-kernel, Steffen Persvold, Daniel J Blueman
On NumaChip systems, the physical processor ID assignment wasn't accounting
for the number of nodes in AMD multi-module processors, giving an incorrect
sibling map:
$ cd /sys/devices/system/cpu/cpu29/topology
$ grep . *
core_id:5
core_siblings:00000000,ff000000
core_siblings_list:24-31
physical_package_id:3
thread_siblings:00000000,30000000
thread_siblings_list:28-29
After fixing:
core_id:5
core_siblings:00000000,ffff0000
core_siblings_list:16-31
physical_package_id:1
thread_siblings:00000000,30000000
thread_siblings_list:28-29
Candidate for stable.
Signed-off-by: Daniel J Blueman <daniel@numascale.com>
---
arch/x86/kernel/apic/apic_numachip.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/arch/x86/kernel/apic/apic_numachip.c b/arch/x86/kernel/apic/apic_numachip.c
index 19f893f..a59768b 100644
--- a/arch/x86/kernel/apic/apic_numachip.c
+++ b/arch/x86/kernel/apic/apic_numachip.c
@@ -180,11 +180,15 @@ static int __init numachip_probe(void)
static void fixup_cpu_id(struct cpuinfo_x86 *c, int node)
{
-
- if (c->phys_proc_id != node) {
- c->phys_proc_id = node;
- per_cpu(cpu_llc_id, smp_processor_id()) = node;
- }
+ u64 val;
+ u32 nodes;
+
+ per_cpu(cpu_llc_id, smp_processor_id()) = node;
+
+ /* Account for nodes per socket in multi-core-module processors */
+ rdmsrl(MSR_FAM10H_NODE_ID, val);
+ nodes = ((val >> 3) & 7) + 1;
+ c->phys_proc_id = node / nodes;
}
static int parse_oemn(struct acpi_table_header *table)
--
2.1.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH RESEND] x86: Fix sibling map with NumaChip
2015-03-03 15:18 [PATCH RESEND] x86: Fix sibling map with NumaChip Daniel J Blueman
@ 2015-03-03 16:29 ` Borislav Petkov
0 siblings, 0 replies; 2+ messages in thread
From: Borislav Petkov @ 2015-03-03 16:29 UTC (permalink / raw)
To: Daniel J Blueman
Cc: Ingo Molnar, Thomas Gleixner, H. Peter Anvin, x86, linux-kernel,
Steffen Persvold
On Tue, Mar 03, 2015 at 11:18:13PM +0800, Daniel J Blueman wrote:
> @@ -180,11 +180,15 @@ static int __init numachip_probe(void)
>
> static void fixup_cpu_id(struct cpuinfo_x86 *c, int node)
> {
> -
> - if (c->phys_proc_id != node) {
> - c->phys_proc_id = node;
> - per_cpu(cpu_llc_id, smp_processor_id()) = node;
> - }
> + u64 val;
> + u32 nodes;
> +
> + per_cpu(cpu_llc_id, smp_processor_id()) = node;
> +
> + /* Account for nodes per socket in multi-core-module processors */
> + rdmsrl(MSR_FAM10H_NODE_ID, val);
You need to check a CPUID bit before accessing that MSR, see
amd_get_topology(). get_apic_id() in apic_numachip.c should be corrected
too.
Thanks.
--
Regards/Gruss,
Boris.
ECO tip #101: Trim your mails when you reply.
--
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2015-03-03 16:30 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-03-03 15:18 [PATCH RESEND] x86: Fix sibling map with NumaChip Daniel J Blueman
2015-03-03 16:29 ` Borislav Petkov
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox