mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Lai Jiangshan <laijs@cn.fujitsu.com>
To: Tejun Heo <tj@kernel.org>, linux-kernel@vger.kernel.org
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Subject: [PATCH 1/7 V6] workqueue: ensure the wq_worker_sleeping() see the right flags
Date: Sun, 9 Sep 2012 01:12:50 +0800	[thread overview]
Message-ID: <1347124383-18723-2-git-send-email-laijs@cn.fujitsu.com> (raw)
In-Reply-To: <1347124383-18723-1-git-send-email-laijs@cn.fujitsu.com>

The compiler may compile this code into TWO write/modify instructions.
		worker->flags &= ~WORKER_UNBOUND;
		worker->flags |= WORKER_REBIND;

so the other CPU may see the temporary of worker->flags which has
not WORKER_UNBOUND nor WORKER_REBIND, it will wrongly do local wake up.

so we use one write/modify instruction explicitly instead.

This bug will not occur on idle workers, because they have another
WORKER_NOT_RUNNING flags.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
---
 kernel/workqueue.c |    7 +++++--
 1 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 85bd340..050b2a5 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -1739,10 +1739,13 @@ retry:
 	for_each_busy_worker(worker, i, pos, gcwq) {
 		struct work_struct *rebind_work = &worker->rebind_work;
 		struct workqueue_struct *wq;
+		unsigned long worker_flags = worker->flags;
 
 		/* morph UNBOUND to REBIND */
-		worker->flags &= ~WORKER_UNBOUND;
-		worker->flags |= WORKER_REBIND;
+		worker_flags &= ~WORKER_UNBOUND;
+		worker_flags |= WORKER_REBIND;
+		/* ensure the wq_worker_sleeping() see the right flags */
+		ACCESS_ONCE(worker->flags) = worker_flags;
 
 		if (test_and_set_bit(WORK_STRUCT_PENDING_BIT,
 				     work_data_bits(rebind_work)))
-- 
1.7.4.4


  reply	other threads:[~2012-09-08 17:11 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-09-08 17:12 [PATCH 0/7 V6] workqueue: fix hoplug things Lai Jiangshan
2012-09-08 17:12 ` Lai Jiangshan [this message]
2012-09-08 17:12 ` [PATCH 2/7 V6] workqueue: async idle rebinding Lai Jiangshan
2012-09-08 17:12 ` [PATCH 3/7 V6] workqueue: add manager pointer for worker_pool Lai Jiangshan
2012-09-08 17:12 ` [PATCH 4/7 V6] workqueue: fix idle worker depletion Lai Jiangshan
2012-09-08 17:40   ` Tejun Heo
2012-09-08 17:50     ` Lai Jiangshan
2012-09-08 17:53       ` Tejun Heo
2012-09-08 18:07         ` Lai Jiangshan
2012-09-08 18:11           ` Tejun Heo
2012-09-08 18:34             ` Lai Jiangshan
2012-09-08 19:02               ` Tejun Heo
2012-09-09  4:09                 ` Lai Jiangshan
2012-09-09  6:30                   ` Tejun Heo
2012-09-08 17:12 ` [PATCH 5/7 V6] workqueue: rename manager_mutex to assoc_mutex Lai Jiangshan
2012-09-08 17:12 ` [PATCH 6/7 V6] workqueue: new day don't need WORKER_REBIND for busy rebinding Lai Jiangshan
2012-09-08 17:12 ` [PATCH 7/7 V6] workqueue: remove WORKER_REBIND Lai Jiangshan
2012-09-08 17:27 ` [PATCH 0/7 V6] workqueue: fix hoplug things Lai Jiangshan
2012-09-08 17:34   ` Lai Jiangshan
2012-09-08 17:37   ` Tejun Heo
2012-09-08 17:46     ` Lai Jiangshan
2012-09-08 17:50       ` Tejun Heo
2012-09-08 17:51         ` Tejun Heo
2012-09-08 17:56         ` Lai Jiangshan
2012-09-08 17:59           ` Tejun Heo
2012-09-08 18:21             ` Lai Jiangshan

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=1347124383-18723-2-git-send-email-laijs@cn.fujitsu.com \
    --to=laijs@cn.fujitsu.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=tj@kernel.org \
    /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

all inboxes | Powered by JetHome®