From: Peter Zijlstra <peterz@infradead.org>
To: lkp@lists.01.org
Subject: Re: [PATCH] netdev: Fix sleeping inside wait event
Date: Wed, 29 Oct 2014 18:31:10 +0100 [thread overview]
Message-ID: <20141029173110.GE15602@worktop.programming.kicks-ass.net> (raw)
In-Reply-To: <20141029171345.GO12706@worktop.programming.kicks-ass.net>
[-- Attachment #1: Type: text/plain, Size: 2754 bytes --]
On Wed, Oct 29, 2014 at 06:13:45PM +0100, Peter Zijlstra wrote:
> On Wed, Oct 29, 2014 at 09:29:55AM -0700, Cong Wang wrote:
> > While you are on it, please fix rtnl_lock_unregistering_all() too?
>
> Ah, that's hidden someplace else, sure I can do that. Thanks for
> pointing it out.
Here goes..
---
Subject: netdev: Fix sleeping inside wait event
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed Oct 29 17:04:56 CET 2014
rtnl_lock_unregistering*() take rtnl_lock() -- a mutex -- inside a
wait loop. The wait loop relies on current->state to function, but so
does mutex_lock(), nesting them makes for the inner to destroy the
outer state.
Fix this using the new wait_woken() bits.
Cc: David Miller <davem@davemloft.net>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
net/core/dev.c | 10 +++++-----
net/core/rtnetlink.c | 10 +++++-----
2 files changed, 10 insertions(+), 10 deletions(-)
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7196,11 +7196,10 @@ static void __net_exit rtnl_lock_unregis
*/
struct net *net;
bool unregistering;
- DEFINE_WAIT(wait);
+ DEFINE_WAIT_FUNC(wait, woken_wake_function);
+ add_wait_queue(&netdev_unregistering_wq, &wait);
for (;;) {
- prepare_to_wait(&netdev_unregistering_wq, &wait,
- TASK_UNINTERRUPTIBLE);
unregistering = false;
rtnl_lock();
list_for_each_entry(net, net_list, exit_list) {
@@ -7212,9 +7211,10 @@ static void __net_exit rtnl_lock_unregis
if (!unregistering)
break;
__rtnl_unlock();
- schedule();
+
+ wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
}
- finish_wait(&netdev_unregistering_wq, &wait);
+ remove_wait_queue(&netdev_unregistering_wq, &wait);
}
static void __net_exit default_device_exit_batch(struct list_head *net_list)
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -365,11 +365,10 @@ static void rtnl_lock_unregistering_all(
{
struct net *net;
bool unregistering;
- DEFINE_WAIT(wait);
+ DEFINE_WAIT_FUNC(wait, woken_wake_function);
+ add_wait_queue(&netdev_unregistering_wq, &wait);
for (;;) {
- prepare_to_wait(&netdev_unregistering_wq, &wait,
- TASK_UNINTERRUPTIBLE);
unregistering = false;
rtnl_lock();
for_each_net(net) {
@@ -381,9 +380,10 @@ static void rtnl_lock_unregistering_all(
if (!unregistering)
break;
__rtnl_unlock();
- schedule();
+
+ wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
}
- finish_wait(&netdev_unregistering_wq, &wait);
+ remove_wait_queue(&netdev_unregistering_wq, &wait);
}
/**
WARNING: multiple messages have this Message-ID (diff)
From: Peter Zijlstra <peterz@infradead.org>
To: Cong Wang <xiyou.wangcong@gmail.com>
Cc: Fengguang Wu <fengguang.wu@intel.com>, LKP <lkp@01.org>,
LKML <linux-kernel@vger.kernel.org>,
"oleg@redhat.com" <oleg@redhat.com>,
"Eric W. Biederman" <ebiederm@xmission.com>,
David Miller <davem@davemloft.net>,
Linux Kernel Network Developers <netdev@vger.kernel.org>
Subject: Re: [PATCH] netdev: Fix sleeping inside wait event
Date: Wed, 29 Oct 2014 18:31:10 +0100 [thread overview]
Message-ID: <20141029173110.GE15602@worktop.programming.kicks-ass.net> (raw)
In-Reply-To: <20141029171345.GO12706@worktop.programming.kicks-ass.net>
On Wed, Oct 29, 2014 at 06:13:45PM +0100, Peter Zijlstra wrote:
> On Wed, Oct 29, 2014 at 09:29:55AM -0700, Cong Wang wrote:
> > While you are on it, please fix rtnl_lock_unregistering_all() too?
>
> Ah, that's hidden someplace else, sure I can do that. Thanks for
> pointing it out.
Here goes..
---
Subject: netdev: Fix sleeping inside wait event
From: Peter Zijlstra <peterz@infradead.org>
Date: Wed Oct 29 17:04:56 CET 2014
rtnl_lock_unregistering*() take rtnl_lock() -- a mutex -- inside a
wait loop. The wait loop relies on current->state to function, but so
does mutex_lock(), nesting them makes for the inner to destroy the
outer state.
Fix this using the new wait_woken() bits.
Cc: David Miller <davem@davemloft.net>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Eric Biederman <ebiederm@xmission.com>
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
net/core/dev.c | 10 +++++-----
net/core/rtnetlink.c | 10 +++++-----
2 files changed, 10 insertions(+), 10 deletions(-)
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -7196,11 +7196,10 @@ static void __net_exit rtnl_lock_unregis
*/
struct net *net;
bool unregistering;
- DEFINE_WAIT(wait);
+ DEFINE_WAIT_FUNC(wait, woken_wake_function);
+ add_wait_queue(&netdev_unregistering_wq, &wait);
for (;;) {
- prepare_to_wait(&netdev_unregistering_wq, &wait,
- TASK_UNINTERRUPTIBLE);
unregistering = false;
rtnl_lock();
list_for_each_entry(net, net_list, exit_list) {
@@ -7212,9 +7211,10 @@ static void __net_exit rtnl_lock_unregis
if (!unregistering)
break;
__rtnl_unlock();
- schedule();
+
+ wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
}
- finish_wait(&netdev_unregistering_wq, &wait);
+ remove_wait_queue(&netdev_unregistering_wq, &wait);
}
static void __net_exit default_device_exit_batch(struct list_head *net_list)
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -365,11 +365,10 @@ static void rtnl_lock_unregistering_all(
{
struct net *net;
bool unregistering;
- DEFINE_WAIT(wait);
+ DEFINE_WAIT_FUNC(wait, woken_wake_function);
+ add_wait_queue(&netdev_unregistering_wq, &wait);
for (;;) {
- prepare_to_wait(&netdev_unregistering_wq, &wait,
- TASK_UNINTERRUPTIBLE);
unregistering = false;
rtnl_lock();
for_each_net(net) {
@@ -381,9 +380,10 @@ static void rtnl_lock_unregistering_all(
if (!unregistering)
break;
__rtnl_unlock();
- schedule();
+
+ wait_woken(&wait, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT);
}
- finish_wait(&netdev_unregistering_wq, &wait);
+ remove_wait_queue(&netdev_unregistering_wq, &wait);
}
/**
next prev parent reply other threads:[~2014-10-29 17:31 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-10-28 14:25 WARNING: CPU: 0 PID: 61 at kernel/sched/core.c:7312 __might_sleep() Fengguang Wu
2014-10-28 14:25 ` Fengguang Wu
2014-10-29 16:16 ` [PATCH] netdev: Fix sleeping inside wait event Peter Zijlstra
2014-10-29 16:16 ` Peter Zijlstra
2014-10-29 16:29 ` Cong Wang
2014-10-29 16:29 ` Cong Wang
2014-10-29 17:13 ` Peter Zijlstra
2014-10-29 17:13 ` Peter Zijlstra
2014-10-29 17:31 ` Peter Zijlstra [this message]
2014-10-29 17:31 ` Peter Zijlstra
2014-10-29 17:38 ` David Miller
2014-10-29 17:38 ` David Miller
2014-11-04 16:09 ` [tip:sched/core] netdev, sched/wait: " tip-bot for Peter Zijlstra
2014-10-29 17:37 ` [PATCH] netdev: " David Miller
2014-10-29 17:37 ` David Miller
2015-02-01 3:43 ` __might_sleep() warnings on v3.19-rc6 Fengguang Wu
2015-02-01 3:43 ` Fengguang Wu
2015-02-01 23:03 ` NeilBrown
2015-02-01 23:03 ` NeilBrown
2015-02-02 4:58 ` NeilBrown
2015-02-02 4:58 ` NeilBrown
2015-02-02 5:08 ` Linus Torvalds
2015-02-02 5:08 ` Linus Torvalds
2015-02-02 6:09 ` NeilBrown
2015-02-02 6:09 ` NeilBrown
2015-02-05 16:16 ` Chris Mason
2015-02-05 16:16 ` Chris Mason
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=20141029173110.GE15602@worktop.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=lkp@lists.01.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 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.