Re: [PATCH 2/3] x86/topology: Fix AMD core count

From: Borislav Petkov
Date: Sat Mar 19 2016 - 11:57:24 EST


On Sat, Mar 19, 2016 at 10:24:59AM +0100, Thomas Gleixner wrote:
> On Fri, 18 Mar 2016, 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 <mingo@xxxxxxxxxx>
> > Cc: Borislav Petkov <bp@xxxxxxxxx>
> > Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> > Cc: Andreas Herrmann <aherrmann@xxxxxxxx>
> > Reported-by: Xiong Zhou <jencce.kernel@xxxxxxxxx>
> > Fixes: 1f12e32f4cd5 ("x86/topology: Create logical package id")
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx>
> > Link: http://lkml.kernel.org/r/20160317095220.GO6344@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> > ---
> > 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;
>
> Unfortunately that will break stuff in event/amd/core.c, ras/mce_amd_inj.c
> which rely on the AMD interpretation of c->x86_max_cores.

What is the events/amd/core.c check even mean? We alloc an amd_nb per
core but not per thread?

The mce_amd_inj.c thing we could fix probably like this:

---
diff --git a/arch/x86/ras/mce_amd_inj.c b/arch/x86/ras/mce_amd_inj.c
index 55d38cfa46c2..b5a917d943c8 100644
--- a/arch/x86/ras/mce_amd_inj.c
+++ b/arch/x86/ras/mce_amd_inj.c
@@ -203,12 +203,7 @@ static void trigger_thr_int(void *info)

static u32 get_nbc_for_node(int node_id)
{
- struct cpuinfo_x86 *c = &boot_cpu_data;
- u32 cores_per_node;
-
- cores_per_node = c->x86_max_cores / amd_get_nodes_per_socket();
-
- return cores_per_node * node_id;
+ return boot_cpu_data.x86_max_cores * node_id;
}

static void toggle_nb_mca_mst_cpu(u16 nid)
---

--
Regards/Gruss,
Boris.

ECO tip #101: Trim your mails when you reply.
--