On Fri, 2015-08-14 at 00:13 +0300, Nikolay Aleksandrov wrote:
> > On Aug 13, 2015, at 11:44 PM, Eric Dumazet wrote:
> >
> > On Thu, 2015-08-13 at 13:19 -0700, Eric Dumazet wrote:
> >
> >>
> >> A caller handler can not call del_timer_sync()
> >
> > A timer handler can not call del_timer_sync()
> On Aug 13, 2015, at 11:44 PM, Eric Dumazet wrote:
>
> On Thu, 2015-08-13 at 13:19 -0700, Eric Dumazet wrote:
>
>>
>> A caller handler can not call del_timer_sync()
>
> A timer handler can not call del_timer_sync()
>
> I am testing a minimal fix :
>
> diff --git a/net/ipv4/inet_connection_
On Thu, 2015-08-13 at 13:19 -0700, Eric Dumazet wrote:
>
> A caller handler can not call del_timer_sync()
A timer handler can not call del_timer_sync()
I am testing a minimal fix :
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 05e3145f7dc3..134957159c27 1
On Mon, 2015-08-10 at 21:18 -0700, David Miller wrote:
> From: Eric Dumazet
> Date: Mon, 10 Aug 2015 09:09:13 -0700
>
> > From: Eric Dumazet
> >
> > reqsk_queue_destroy() and reqsk_queue_unlink() should use
> > del_timer_sync() instead of del_timer() before calling reqsk_put(),
> > otherwise we
From: Eric Dumazet
Date: Mon, 10 Aug 2015 09:09:13 -0700
> From: Eric Dumazet
>
> reqsk_queue_destroy() and reqsk_queue_unlink() should use
> del_timer_sync() instead of del_timer() before calling reqsk_put(),
> otherwise we could free a req still used by another cpu.
>
> But before doing so,
From: Eric Dumazet
reqsk_queue_destroy() and reqsk_queue_unlink() should use
del_timer_sync() instead of del_timer() before calling reqsk_put(),
otherwise we could free a req still used by another cpu.
But before doing so, reqsk_queue_destroy() must release syn_wait_lock
spinlock or risk a dead
