linux-kernel - Re: [PATCH RESEND v2 5/5] sbitmap: correct wake_batch recalculation to avoid potential IO hung

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]

Message-ID: <20221222134146.khucy5afnxwl75px@quack3>
Date:   Thu, 22 Dec 2022 14:41:46 +0100
From:   Jan Kara <jack@...e.cz>
To:     Kemeng Shi <shikemeng@...weicloud.com>
Cc:     axboe@...nel.dk, linux-block@...r.kernel.org,
        linux-kernel@...r.kernel.org, jack@...e.cz, kbusch@...nel.org,
        Laibin Qiu <qiulaibin@...wei.com>
Subject: Re: [PATCH RESEND v2 5/5] sbitmap: correct wake_batch recalculation
 to avoid potential IO hung

On Thu 22-12-22 22:33:53, Kemeng Shi wrote:
> Commit 180dccb0dba4f ("blk-mq: fix tag_get wait task can't be awakened")
> mentioned that in case of shared tags, there could be just one real
> active hctx(queue) because of lazy detection of tag idle. Then driver tag
> allocation may wait forever on this real active hctx(queue) if wake_batch
> is > hctx_max_depth where hctx_max_depth is available tags depth for the
> actve hctx(queue). However, the condition wake_batch > hctx_max_depth is
> not strong enough to avoid IO hung as the sbitmap_queue_wake_up will only
> wake up one wait queue for each wake_batch even though there is only one
> waiter in the woken wait queue. After this, there is only one tag to free
> and wake_batch may not be reached anymore. Commit 180dccb0dba4f ("blk-mq:
> fix tag_get wait task can't be awakened") methioned that driver tag
> allocation may wait forever. Actually, the inactive hctx(queue) will be
> truely idle after at most 30 seconds and will call blk_mq_tag_wakeup_all
> to wake one waiter per wait queue to break the hung. But IO hung for 30
> seconds is also not acceptable. Set batch size to small enough that depth
> of the shared hctx(queue) is enough to wake up all of the queues like
> sbq_calc_wake_batch do to fix this potential IO hung.
> 
> Although hctx_max_depth will be clamped to at least 4 while wake_batch
> recalculation does not do the clamp, the wake_batch will be always
> recalculated to 1 when hctx_max_depth <= 4.
> 
> Fixes: 180dccb0dba4 ("blk-mq: fix tag_get wait task can't be awakened")
> Signed-off-by: Kemeng Shi <shikemeng@...weicloud.com>

So the condition in sbitmap_queue_recalculate_wake_batch() also seemed
strange to me and the changelogs of commits 180dccb0dba4 and 10825410b95
("blk-mq: Fix wrong wakeup batch configuration which will cause hang")
didn't add much confidence about the magic batch setting to 4. Let me add
to CC original author of this code if he has any thoughts on why using
wake batch of 4 is safe for cards with say 32 tags in case active_users is
currently 32. Because I don't see why that is correct either.

								Honza

> ---
>  lib/sbitmap.c | 5 +----
>  1 file changed, 1 insertion(+), 4 deletions(-)
> 
> diff --git a/lib/sbitmap.c b/lib/sbitmap.c
> index b6d3bb1c3675..804fe99783e4 100644
> --- a/lib/sbitmap.c
> +++ b/lib/sbitmap.c
> @@ -458,13 +458,10 @@ void sbitmap_queue_recalculate_wake_batch(struct sbitmap_queue *sbq,
>  					    unsigned int users)
>  {
>  	unsigned int wake_batch;
> -	unsigned int min_batch;
>  	unsigned int depth = (sbq->sb.depth + users - 1) / users;
>  
> -	min_batch = sbq->sb.depth >= (4 * SBQ_WAIT_QUEUES) ? 4 : 1;
> -
>  	wake_batch = clamp_val(depth / SBQ_WAIT_QUEUES,
> -			min_batch, SBQ_WAKE_BATCH);
> +			1, SBQ_WAKE_BATCH);
>  
>  	WRITE_ONCE(sbq->wake_batch, wake_batch);
>  }
> -- 
> 2.30.0
> 
-- 
Jan Kara <jack@...e.com>
SUSE Labs, CR