From: tip-bot for Peter Zijlstra <tipbot@zytor.com>
To: linux-tip-commits@vger.kernel.org
Cc: aherrmann@suse.com, mingo@kernel.org, jencce.kernel@gmail.com,
hpa@zytor.com, peterz@infradead.org, tglx@linutronix.de,
luto@kernel.org, linux-kernel@vger.kernel.org, bp@suse.de
Subject: [tip:x86/urgent] x86/topology: Fix AMD core count
Date: Tue, 29 Mar 2016 01:48:43 -0700 [thread overview]
Message-ID: <tip-ee6825c80e870fff1a370c718ec77022ade0889b@git.kernel.org> (raw)
In-Reply-To: <20160317095220.GO6344@twins.programming.kicks-ass.net>
Commit-ID: ee6825c80e870fff1a370c718ec77022ade0889b
Gitweb: http://git.kernel.org/tip/ee6825c80e870fff1a370c718ec77022ade0889b
Author: Peter Zijlstra <peterz@infradead.org>
AuthorDate: Fri, 25 Mar 2016 15:52:34 +0100
Committer: Thomas Gleixner <tglx@linutronix.de>
CommitDate: Tue, 29 Mar 2016 10:45:04 +0200
x86/topology: Fix AMD core count
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.
Boris: fixup ras/mce_amd_inj.c too, to compute the Node Base Core
properly, according to the new nomenclature.
Fixes: 1f12e32f4cd5 ("x86/topology: Create logical package id")
Reported-by: Xiong Zhou <jencce.kernel@gmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Cc: Andreas Herrmann <aherrmann@suse.com>
Cc: Andy Lutomirski <luto@kernel.org>
Link: http://lkml.kernel.org/r/20160317095220.GO6344@twins.programming.kicks-ass.net
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
arch/x86/include/asm/smp.h | 1 +
arch/x86/kernel/cpu/amd.c | 8 ++++----
arch/x86/ras/mce_amd_inj.c | 3 ++-
3 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/arch/x86/include/asm/smp.h b/arch/x86/include/asm/smp.h
index 20a3de5..66b0573 100644
--- a/arch/x86/include/asm/smp.h
+++ b/arch/x86/include/asm/smp.h
@@ -155,6 +155,7 @@ static inline int wbinvd_on_all_cpus(void)
wbinvd();
return 0;
}
+#define smp_num_siblings 1
#endif /* CONFIG_SMP */
extern unsigned disabled_cpus;
diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c
index 6e47e3a..4d0087f 100644
--- a/arch/x86/kernel/cpu/amd.c
+++ b/arch/x86/kernel/cpu/amd.c
@@ -312,9 +312,9 @@ static void amd_get_topology(struct cpuinfo_x86 *c)
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;
@@ -329,8 +329,8 @@ static void amd_get_topology(struct cpuinfo_x86 *c)
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;
diff --git a/arch/x86/ras/mce_amd_inj.c b/arch/x86/ras/mce_amd_inj.c
index 55d38cf..9e02dca 100644
--- a/arch/x86/ras/mce_amd_inj.c
+++ b/arch/x86/ras/mce_amd_inj.c
@@ -20,6 +20,7 @@
#include <linux/pci.h>
#include <asm/mce.h>
+#include <asm/smp.h>
#include <asm/amd_nb.h>
#include <asm/irq_vectors.h>
@@ -206,7 +207,7 @@ 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();
+ cores_per_node = (c->x86_max_cores * smp_num_siblings) / amd_get_nodes_per_socket();
return cores_per_node * node_id;
}
next prev parent reply other threads:[~2016-03-29 8:49 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-16 6:48 4.5.0+ panic when setup loop device Xiong Zhou
2016-03-16 15:26 ` Thomas Gleixner
2016-03-17 1:56 ` Xiong Zhou
2016-03-17 9:52 ` Peter Zijlstra
2016-03-17 9:56 ` Peter Zijlstra
2016-03-17 10:21 ` Thomas Gleixner
2016-03-17 10:26 ` Peter Zijlstra
2016-03-17 11:39 ` Thomas Gleixner
2016-03-17 11:51 ` Peter Zijlstra
2016-03-17 11:57 ` Borislav Petkov
2016-03-17 12:01 ` Thomas Gleixner
2016-03-17 16:42 ` Jens Axboe
2016-03-17 18:26 ` Jens Axboe
2016-03-17 20:20 ` Thomas Gleixner
2016-03-17 20:23 ` Jens Axboe
2016-03-17 20:30 ` Thomas Gleixner
2016-03-17 20:41 ` Jens Axboe
2016-03-18 2:31 ` Xiong Zhou
2016-03-18 4:11 ` Mike Galbraith
2016-03-18 7:51 ` Peter Zijlstra
2016-03-18 10:15 ` Peter Zijlstra
2016-03-18 12:39 ` Mike Galbraith
2016-03-18 13:32 ` Peter Zijlstra
2016-03-18 14:07 ` Mike Galbraith
2016-03-18 11:55 ` Thomas Gleixner
2016-03-18 12:39 ` Mike Galbraith
2016-03-19 9:31 ` [tip:x86/urgent] x86/topology: Use total_cpus not nr_cpu_ids for logical packages tip-bot for Thomas Gleixner
2016-03-29 8:48 ` tip-bot for Peter Zijlstra [this message]
-- strict thread matches above, loose matches on Subject: below --
2016-03-18 15:03 [PATCH 0/3] x86 topology fixes Peter Zijlstra
2016-03-18 15:03 ` [PATCH 1/3] x86/topology: Fix logical pkg mapping Peter Zijlstra
2016-03-19 9:30 ` [tip:x86/urgent] x86/topology: Fix logical package mapping tip-bot for Peter Zijlstra
2016-03-18 15:03 ` [PATCH 2/3] x86/topology: Fix AMD core count Peter Zijlstra
2016-03-18 16:41 ` Borislav Petkov
2016-03-21 3:07 ` Huang Rui
2016-03-21 3:46 ` Huang Rui
2016-03-21 8:26 ` Borislav Petkov
2016-03-21 9:18 ` Huang Rui
2016-03-21 8:56 ` Thomas Gleixner
2016-03-21 8:21 ` Peter Zijlstra
2016-03-21 9:46 ` Huang Rui
2016-03-21 13:57 ` Borislav Petkov
2016-03-22 8:10 ` Sherry Hurwitz
2016-03-22 11:22 ` Borislav Petkov
2016-03-21 8:23 ` Borislav Petkov
2016-03-21 10:05 ` Huang Rui
2016-03-21 10:23 ` Borislav Petkov
2016-03-19 9:24 ` Thomas Gleixner
2016-03-19 15:56 ` Borislav Petkov
2016-03-20 10:39 ` Peter Zijlstra
2016-03-20 11:04 ` Borislav Petkov
2016-03-20 12:32 ` Peter Zijlstra
2016-03-20 12:46 ` Peter Zijlstra
2016-03-20 13:09 ` Borislav Petkov
2016-03-20 17:08 ` Peter Zijlstra
2016-03-20 18:46 ` Borislav Petkov
2016-03-29 8:49 ` [tip:x86/urgent] perf/x86/amd: Cleanup Fam10h NB event constraints tip-bot for Peter Zijlstra
2016-03-22 7:56 ` [PATCH 2/3] x86/topology: Fix AMD core count Huang Rui
2016-03-18 15:03 ` [PATCH 3/3] x86/topology: Fix Intel HT disable Peter Zijlstra
2016-03-19 9:31 ` [tip:x86/urgent] " tip-bot for Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=tip-ee6825c80e870fff1a370c718ec77022ade0889b@git.kernel.org \
--to=tipbot@zytor.com \
--cc=aherrmann@suse.com \
--cc=bp@suse.de \
--cc=hpa@zytor.com \
--cc=jencce.kernel@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.