Received-SPF: pass (google.com: domain of linux-kernel+bounces-199982-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:40f1:3f00::1 as permitted sender) client-ip=2604:1380:40f1:3f00::1;
Precedence: bulk
MIME-Version: 1.0
References: <20240604031124.2261-1-yang.yang@vivo.com>
In-Reply-To: <20240604031124.2261-1-yang.yang@vivo.com>
From: Ming Lei <ming.lei@redhat.com>
Date: Tue, 4 Jun 2024 11:25:24 +0800
Message-ID: <CAFj5m9KV7OJ4_KjbSkpdtfrKamoLzV6EH-mJP3=y+VvoYOzC3w@mail.gmail.com>
Subject: Re: [PATCH v2] sbitmap: fix io hung due to race on sbitmap_word::cleared
To: Yang Yang <yang.yang@vivo.com>
Cc: Jens Axboe <axboe@kernel.dk>, Andrew Morton <akpm@linux-foundation.org>, 
	Pavel Begunkov <asml.silence@gmail.com>, linux-kernel@vger.kernel.org, 
	linux-block@vger.kernel.org, Ming Lei <ming.lei@redhat.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Tue, Jun 4, 2024 at 11:12=E2=80=AFAM Yang Yang <yang.yang@vivo.com> wrot=
e:
>
> Configuration for sbq:
>   depth=3D64, wake_batch=3D6, shift=3D6, map_nr=3D1
>
> 1. There are 64 requests in progress:
>   map->word =3D 0xFFFFFFFFFFFFFFFF
> 2. After all the 64 requests complete, and no more requests come:
>   map->word =3D 0xFFFFFFFFFFFFFFFF, map->cleared =3D 0xFFFFFFFFFFFFFFFF
> 3. Now two tasks try to allocate requests:
>   T1:                                       T2:
>   __blk_mq_get_tag                          .
>   __sbitmap_queue_get                       .
>   sbitmap_get                               .
>   sbitmap_find_bit                          .
>   sbitmap_find_bit_in_word                  .
>   __sbitmap_get_word  -> nr=3D-1              __blk_mq_get_tag
>   sbitmap_deferred_clear                    __sbitmap_queue_get
>   /* map->cleared=3D0xFFFFFFFFFFFFFFFF */     sbitmap_find_bit
>     if (!READ_ONCE(map->cleared))           sbitmap_find_bit_in_word
>       return false;                         __sbitmap_get_word -> nr=3D-1
>     mask =3D xchg(&map->cleared, 0)           sbitmap_deferred_clear
>     atomic_long_andnot()                    /* map->cleared=3D0 */
>                                               if (!(map->cleared))
>                                                 return false;
>                                      /*
>                                       * map->cleared is cleared by T1
>                                       * T2 fail to acquire the tag
>                                       */
>
> 4. T2 is the sole tag waiter. When T1 puts the tag, T2 cannot be woken
> up due to the wake_batch being set at 6. If no more requests come, T1
> will wait here indefinitely.
>
> To fix this issue, simply revert commit 661d4f55a794 ("sbitmap:
> remove swap_lock"), which causes this issue.

I'd suggest to add the following words in commit log:

Check on ->cleared and update on both ->cleared and ->word need to be
done atomically, and using spinlock could be the simplest solution.

Otherwise, the patch looks fine for me.

Thanks,