linux-kernel - Re: [REGRESSION] 6.7.1: md: raid5 hang and unresponsive system; successfully bisected

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: <20240310051321.9812-1-dan@danm.net>
Date: Sat,  9 Mar 2024 22:13:21 -0700
From: Dan Moulding <dan@...m.net>
To: junxiao.bi@...cle.com
Cc: dan@...m.net,
	gregkh@...uxfoundation.org,
	linux-kernel@...r.kernel.org,
	linux-raid@...r.kernel.org,
	regressions@...ts.linux.dev,
	song@...nel.org,
	stable@...r.kernel.org
Subject: Re: [REGRESSION] 6.7.1: md: raid5 hang and unresponsive system; successfully bisected

> Dan, can you try the following patch?
> 
> diff --git a/block/blk-core.c b/block/blk-core.c
> index de771093b526..474462abfbdc 100644
> --- a/block/blk-core.c
> +++ b/block/blk-core.c
> @@ -1183,6 +1183,7 @@ void __blk_flush_plug(struct blk_plug *plug, bool 
> from_schedule)
>          if (unlikely(!rq_list_empty(plug->cached_rq)))
>                  blk_mq_free_plug_rqs(plug);
>   }
> +EXPORT_SYMBOL(__blk_flush_plug);
> 
>   /**
>    * blk_finish_plug - mark the end of a batch of submitted I/O
> diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
> index 8497880135ee..26e09cdf46a3 100644
> --- a/drivers/md/raid5.c
> +++ b/drivers/md/raid5.c
> @@ -6773,6 +6773,11 @@ static void raid5d(struct md_thread *thread)
> spin_unlock_irq(&conf->device_lock);
>                          md_check_recovery(mddev);
>                          spin_lock_irq(&conf->device_lock);
> +               } else {
> + spin_unlock_irq(&conf->device_lock);
> +                       blk_flush_plug(&plug, false);
> +                       cond_resched();
> +                       spin_lock_irq(&conf->device_lock);
>                  }
>          }
>          pr_debug("%d stripes handled\n", handled);

This patch seems to work! I can no longer reproduce the problem after
applying this.

Thanks,

-- Dan