On Fri, Mar 18, 2016 at 11:03:47PM +0800, Peter Zijlstra wrote:
> It turns out AMD gets x86_max_cores wrong when there are compute
> units.
>
> The issue is that Linux assumes:
>
> nr_logical_cpus = nr_cores * nr_siblings
>
> But AMD reports its CU unit as 2 cores, but then sets num_smp_siblings
> to 2 as well.
>
> Cc: Ingo Molnar <mi...@kernel.org>
> Cc: Borislav Petkov <b...@alien8.de>
> Cc: Thomas Gleixner <t...@linutronix.de>
> Cc: Andreas Herrmann <aherrm...@suse.com>
> Reported-by: Xiong Zhou <jencce.ker...@gmail.com>
> Fixes: 1f12e32f4cd5 ("x86/topology: Create logical package id")
> Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org>
I tested this patch both on fam15h and fam16h platforms, x86_max_cores
can report correct compute unit numbers. And
/sys/devices/system/cpu/cpuX/topology/thread_siblings_list
/sys/devices/system/cpu/cpuX/topology/core_siblings_list
are also right. Thanks.
Tested-by: Huang Rui <ray.hu...@amd.com>
> Link:
> http://lkml.kernel.org/r/20160317095220.go6...@twins.programming.kicks-ass.net
> ---
> arch/x86/kernel/cpu/amd.c | 8 ++++----
> arch/x86/kernel/smpboot.c | 11 ++++++-----
> 2 files changed, 10 insertions(+), 9 deletions(-)
>
> --- a/arch/x86/kernel/cpu/amd.c
> +++ b/arch/x86/kernel/cpu/amd.c
> @@ -313,9 +313,9 @@ static void amd_get_topology(struct cpui
> node_id = ecx & 7;
>
> /* get compute unit information */
> - smp_num_siblings = ((ebx >> 8) & 3) + 1;
> + cores_per_cu = smp_num_siblings = ((ebx >> 8) & 3) + 1;
> + c->x86_max_cores /= smp_num_siblings;
> c->compute_unit_id = ebx & 0xff;
> - cores_per_cu += ((ebx >> 8) & 3);
> } else if (cpu_has(c, X86_FEATURE_NODEID_MSR)) {
> u64 value;
>
> @@ -331,8 +331,8 @@ static void amd_get_topology(struct cpui
> u32 cus_per_node;
>
> set_cpu_cap(c, X86_FEATURE_AMD_DCM);
> - cores_per_node = c->x86_max_cores / nodes_per_socket;
> - cus_per_node = cores_per_node / cores_per_cu;
> + cus_per_node = c->x86_max_cores / nodes_per_socket;
> + cores_per_node = cus_per_node * cores_per_cu;
>
> /* store NodeID, use llc_shared_map to store sibling info */
> per_cpu(cpu_llc_id, cpu) = node_id;
>
>
