linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
To: Huang Ying <ying.huang@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>, "H. Peter Anvin" <hpa@zytor.com>,
	Andi Kleen <ak@linux.intel.com>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: [PATCH 08/10] x86, mce: change barriers in mce_log
Date: Mon, 05 Oct 2009 15:41:23 +0900	[thread overview]
Message-ID: <4AC99513.2010504@jp.fujitsu.com> (raw)
In-Reply-To: <4AC990E1.7030708@jp.fujitsu.com>

Long time ago, smp_wmb() was replaced to wmb() by following commit:
  commit 7644143cd6f7e029f3a8ea64f5fb0ab33ec39f72
  Author: Mike Waychison <mikew@google.com>
  Date:   Fri Sep 30 00:01:27 2005 +0200
  [PATCH] x86_64: Fix mce_log
  > AK: turned the smp_wmbs into true wmbs to make sure they are not
  > reordered by the compiler on UP.

Change them back to original form, and put comments.

(This piece originates from Huang's patch, titled:
 "x86, MCE: Fix bugs and issues of MCE log ring buffer")

Originally-From: Huang Ying <ying.huang@intel.com>
Signed-off-by: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
---
 arch/x86/kernel/cpu/mcheck/mce.c |   10 +++++++---
 1 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
index 87b2e29..655915b 100644
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -141,8 +141,9 @@ void mce_log(struct mce *mce)
 	struct mce_log_cpu *mcelog_cpu = &__get_cpu_var(mce_log_cpus);
 	unsigned next, entry;
 
+	/* mce->finished must be set to 0 before written to buffer */
 	mce->finished = 0;
-	wmb();
+	smp_wmb();
 
 	do {
 		entry = mcelog_cpu->next;
@@ -161,9 +162,12 @@ void mce_log(struct mce *mce)
 
 	memcpy(mcelog_cpu->entry + entry, mce, sizeof(struct mce));
 
-	wmb();
+	/* ".finished" of MCE record in buffer must be set after copy */
+	smp_wmb();
 	mcelog_cpu->entry[entry].finished = 1;
-	wmb();
+
+	/* bit 0 of notify_user should be set after finished be set */
+	smp_wmb();
 	mce->finished = 1;
 	set_bit(0, &mce_need_notify);
 }
-- 
1.6.4.3



  parent reply	other threads:[~2009-10-05  6:42 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-09-18 10:20 [BUGFIX -v7] x86, MCE: Fix bugs and issues of MCE log ring buffer Huang Ying
2009-09-18 11:09 ` Ingo Molnar
2009-09-21  5:37   ` Huang Ying
2009-09-22 13:39     ` [PATCH] x86: mce: New MCE logging design Ingo Molnar
2009-10-05  6:23 ` [BUGFIX -v7] x86, MCE: Fix bugs and issues of MCE log ring buffer Hidetoshi Seto
2009-10-05  6:33   ` [PATCH 01/10] x86, mce: remove tsc handling from mce_read Hidetoshi Seto
2009-10-05  6:34   ` [PATCH 02/10] x86, mce: mce_read can check args without mutex Hidetoshi Seto
2009-10-05  6:35   ` [PATCH 03/10] x86, mce: change writer timeout in mce_read Hidetoshi Seto
2009-10-05  6:36   ` [PATCH 04/10] x86, mce: use do-while in mce_log Hidetoshi Seto
2009-10-05  6:37   ` [PATCH 05/10] x86, mce: make mce_log buffer to per-CPU, prep Hidetoshi Seto
2009-10-05  6:38   ` [PATCH 06/10] x86, mce: make mce_log buffer to per-CPU Hidetoshi Seto
2009-10-05  7:06     ` Andi Kleen
2009-10-05  7:50       ` Hidetoshi Seto
2009-10-09  1:45         ` Huang Ying
2009-10-09  5:34           ` Hidetoshi Seto
2009-10-05  6:40   ` [PATCH 07/10] x86, mce: remove for-loop in mce_log Hidetoshi Seto
2009-10-05  6:41   ` Hidetoshi Seto [this message]
2009-10-05  6:42   ` [PATCH 09/10] x86, mce: make mce_log buffer to ring buffer Hidetoshi Seto
2009-10-05  6:44   ` [PATCH 10/10] x86, mce: move mce_log_init() into mce_cap_init() Hidetoshi Seto
2009-10-05  7:07   ` [BUGFIX -v7] x86, MCE: Fix bugs and issues of MCE log ring buffer Hidetoshi Seto
2009-10-05  8:51   ` Frédéric Weisbecker
2009-10-05 15:16     ` Andi Kleen
2009-10-06  5:46     ` Hidetoshi Seto

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=4AC99513.2010504@jp.fujitsu.com \
    --to=seto.hidetoshi@jp.fujitsu.com \
    --cc=ak@linux.intel.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=ying.huang@intel.com \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).