From: Marco Elver <elver@google.com>
To: Qian Cai <cai@lca.pw>
Cc: Will Deacon <will@kernel.org>,
mingo@redhat.com, peterz@infradead.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH] locking/osq_lock: fix a data race in osq_wait_next
Date: Wed, 22 Jan 2020 23:38:51 +0100 [thread overview]
Message-ID: <20200122223851.GA45602@google.com> (raw)
In-Reply-To: <A5114711-B8DE-48DA-AFD0-62128AC08270@lca.pw>
On Wed, 22 Jan 2020, Qian Cai wrote:
>
>
> > On Jan 22, 2020, at 11:59 AM, Will Deacon <will@kernel.org> wrote:
> >
> > I don't understand this; 'next' is a local variable.
> >
> > Not keen on the onslaught of random "add a READ_ONCE() to shut the
> > sanitiser up" patches we're going to get from kcsan :(
>
> My fault. I suspect it is node->next. I’ll do a bit more testing to confirm.
If possible, decode and get the line numbers. I have observed a data
race in osq_lock before, however, this is the only one I have recently
seen in osq_lock:
read to 0xffff88812c12d3d4 of 4 bytes by task 23304 on cpu 0:
osq_lock+0x170/0x2f0 kernel/locking/osq_lock.c:143
while (!READ_ONCE(node->locked)) {
/*
* If we need to reschedule bail... so we can block.
* Use vcpu_is_preempted() to avoid waiting for a preempted
* lock holder:
*/
--> if (need_resched() || vcpu_is_preempted(node_cpu(node->prev)))
goto unqueue;
cpu_relax();
}
where
static inline int node_cpu(struct optimistic_spin_node *node)
{
--> return node->cpu - 1;
}
write to 0xffff88812c12d3d4 of 4 bytes by task 23334 on cpu 1:
osq_lock+0x89/0x2f0 kernel/locking/osq_lock.c:99
bool osq_lock(struct optimistic_spin_queue *lock)
{
struct optimistic_spin_node *node = this_cpu_ptr(&osq_node);
struct optimistic_spin_node *prev, *next;
int curr = encode_cpu(smp_processor_id());
int old;
node->locked = 0;
node->next = NULL;
--> node->cpu = curr;
Thanks,
-- Marco
next prev parent reply other threads:[~2020-01-22 22:39 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-22 16:38 [PATCH] locking/osq_lock: fix a data race in osq_wait_next Qian Cai
2020-01-22 16:59 ` Will Deacon
2020-01-22 17:08 ` Qian Cai
2020-01-22 22:38 ` Marco Elver [this message]
2020-01-22 23:54 ` Qian Cai
2020-01-23 9:39 ` Peter Zijlstra
2020-01-28 3:11 ` Qian Cai
2020-01-28 11:46 ` Marco Elver
2020-01-28 12:53 ` Qian Cai
2020-01-28 16:52 ` Peter Zijlstra
2020-01-28 16:56 ` Peter Zijlstra
2020-01-29 0:22 ` Paul E. McKenney
2020-01-29 15:29 ` Marco Elver
2020-01-29 18:40 ` Peter Zijlstra
2020-01-30 13:39 ` Marco Elver
2020-01-30 13:48 ` Peter Zijlstra
2020-01-31 3:32 ` Qian Cai
2020-01-29 18:49 ` Peter Zijlstra
2020-01-29 19:26 ` Paul E. McKenney
2020-01-23 9:36 ` Peter Zijlstra
2020-01-28 3:12 ` Qian Cai
2020-01-28 8:18 ` Marco Elver
2020-01-28 10:10 ` Qian Cai
2020-01-28 10:29 ` Marco Elver
2020-01-22 17:09 ` 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=20200122223851.GA45602@google.com \
--to=elver@google.com \
--cc=cai@lca.pw \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=will@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;
as well as URLs for NNTP newsgroup(s).