linux-kernel - Re: [PATCH] epoll: simplify ep_poll_safewake() for CONFIG_DEBUG_LOCK

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 24 Sep 2019 19:52:18 +0200
From:   Roman Penyaev <rpenyaev@...e.de>
To:     Jason Baron <jbaron@...mai.com>
Cc:     akpm@...ux-foundation.org, linux-kernel@...r.kernel.org,
        Davidlohr Bueso <dave@...olabs.net>,
        Al Viro <viro@...iv.linux.org.uk>,
        Eric Wong <normalperson@...t.net>
Subject: Re: [PATCH] epoll: simplify ep_poll_safewake() for
 CONFIG_DEBUG_LOCK_ALLOC

On 2019-09-24 19:34, Jason Baron wrote:
> On 9/23/19 3:23 PM, Roman Penyaev wrote:
>> On 2019-09-23 17:43, Jason Baron wrote:
>>> On 9/4/19 4:22 PM, Jason Baron wrote:
>>>> Currently, ep_poll_safewake() in the CONFIG_DEBUG_LOCK_ALLOC case 
>>>> uses
>>>> ep_call_nested() in order to pass the correct subclass argument to
>>>> spin_lock_irqsave_nested(). However, ep_call_nested() adds 
>>>> unnecessary
>>>> checks for epoll depth and loops that are already verified when 
>>>> doing
>>>> EPOLL_CTL_ADD. This mirrors a conversion that was done for
>>>> !CONFIG_DEBUG_LOCK_ALLOC in: commit 37b5e5212a44 ("epoll: remove
>>>> ep_call_nested() from ep_eventpoll_poll()")
>>>> 
>>>> Signed-off-by: Jason Baron <jbaron@...mai.com>
>>>> Cc: Davidlohr Bueso <dave@...olabs.net>
>>>> Cc: Roman Penyaev <rpenyaev@...e.de>
>>>> Cc: Al Viro <viro@...iv.linux.org.uk>
>>>> Cc: Eric Wong <normalperson@...t.net>
>>>> Cc: Andrew Morton <akpm@...ux-foundation.org>
>>>> ---
>>>>  fs/eventpoll.c | 36 +++++++++++++-----------------------
>>>>  1 file changed, 13 insertions(+), 23 deletions(-)
>>>> 
>>>> diff --git a/fs/eventpoll.c b/fs/eventpoll.c
>>>> index d7f1f50..a9b2737 100644
>>>> --- a/fs/eventpoll.c
>>>> +++ b/fs/eventpoll.c
>>>> @@ -551,28 +551,23 @@ static int ep_call_nested(struct nested_calls
>>>> *ncalls,
>>>>   */
>>>>  #ifdef CONFIG_DEBUG_LOCK_ALLOC
>>>> 
>>>> -static struct nested_calls poll_safewake_ncalls;
>>>> -
>>>> -static int ep_poll_wakeup_proc(void *priv, void *cookie, int
>>>> call_nests)
>>>> -{
>>>> -    unsigned long flags;
>>>> -    wait_queue_head_t *wqueue = (wait_queue_head_t *)cookie;
>>>> -
>>>> -    spin_lock_irqsave_nested(&wqueue->lock, flags, call_nests + 1);
>>>> -    wake_up_locked_poll(wqueue, EPOLLIN);
>>>> -    spin_unlock_irqrestore(&wqueue->lock, flags);
>>>> -
>>>> -    return 0;
>>>> -}
>>>> +static DEFINE_PER_CPU(int, wakeup_nest);
>>>> 
>>>>  static void ep_poll_safewake(wait_queue_head_t *wq)
>>>>  {
>>>> -    int this_cpu = get_cpu();
>>>> -
>>>> -    ep_call_nested(&poll_safewake_ncalls,
>>>> -               ep_poll_wakeup_proc, NULL, wq, (void *) (long)
>>>> this_cpu);
>>>> +    unsigned long flags;
>>>> +    int subclass;
>>>> 
>>>> -    put_cpu();
>>>> +    local_irq_save(flags);
>>>> +    preempt_disable();
>>>> +    subclass = __this_cpu_read(wakeup_nest);
>>>> +    spin_lock_nested(&wq->lock, subclass + 1);
>>>> +    __this_cpu_inc(wakeup_nest);
>>>> +    wake_up_locked_poll(wq, POLLIN);
>>>> +    __this_cpu_dec(wakeup_nest);
>>>> +    spin_unlock(&wq->lock);
>>>> +    local_irq_restore(flags);
>>>> +    preempt_enable();
>>>>  }
>> 
>> What if reduce number of lines with something as the following:
>> 
>>    int this_cpu = get_cpu();
>>    subclass = __this_cpu_inc_return(wakeup_nest);
>>    spin_lock_irqsave_nested(&wq->lock, flags, subclass);
>>    wake_up_locked_poll(wq, POLLIN);
>>    spin_unlock_irqrestore(&wq->lock, flags);
>>    __this_cpu_dec(wakeup_nest);
>>    put_cpu();
>> 
>> Other than that looks good to me.
>> 
>> Reviewed-by: Roman Penyaev <rpenyaev@...e.de>
>> 
>> --
>> Roman
> 
> 
> Hi,
> 
> I put the local_irq_save(flags), call there first so that there 
> wouldn't
> be any nesting. For example, in your sequence, there could be an irq
> after the  __this_cpu_inc_return(), that could end up back here.

That is correct, but seems this is the original behavior of 
ep_call_nested(),
where irq can happen just after spin_unlock_irqrestore():

	spin_unlock_irqrestore(&ncalls->lock, flags);

         >>>> irq here <<<<<

	/* Call the nested function */
	error = (*nproc)(priv, cookie, call_nests);

so eventually you end up with spin_lock_irqsave_nested() call where
call_nests is not monotonically increased (not sequential) but has
a gap (depends on nesting).

So if shorter, I thought that your "local_irq_save + increment" sequence
is excessive.

--
Roman