linux-kernel - Re: [PATCH] md/bitmap: use mddev_suspend/resume instead of ->quiesce()

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: <20180928183044.GA12491@kernel.org>
Date:   Fri, 28 Sep 2018 11:30:44 -0700
From:   Shaohua Li <shli@...nel.org>
To:     Jack Wang <xjtuwjp@...il.com>
Cc:     neilb@...e.com, linux-raid@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        Jack Wang <jinpu.wang@...fitbricks.com>
Subject: Re: [PATCH] md/bitmap: use mddev_suspend/resume instead of
 ->quiesce()

On Thu, Sep 27, 2018 at 10:07:57AM +0200, Jack Wang wrote:
> From: Jack Wang <jinpu.wang@...fitbricks.com>
> 
> After 9e1cc0a54556 ("md: use mddev_suspend/resume instead of ->quiesce()")
> We still have similar left in bitmap functions.
> 
> Replace quiesce() with mddev_suspend/resume.
> 
> Also move md_bitmap_create out of mddev_suspend. and move mddev_resume
> after md_bitmap_destroy. as we did in set_bitmap_file.
> 
> Signed-off-by: Jack Wang <jinpu.wang@...fitbricks.com>
> Reviewed-by: Gioh Kim <gi-oh.kim@...fitbricks.com>
> 
> ---
> v2->v1: add reviewed-by.
> ---
>  drivers/md/md-bitmap.c | 13 +++++++------
>  1 file changed, 7 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/md/md-bitmap.c b/drivers/md/md-bitmap.c
> index 2fc8c113977f..c369f1753ea6 100644
> --- a/drivers/md/md-bitmap.c
> +++ b/drivers/md/md-bitmap.c
> @@ -2123,7 +2123,7 @@ int md_bitmap_resize(struct bitmap *bitmap, sector_t blocks,
>  	}
>  
>  	if (!init)
> -		bitmap->mddev->pers->quiesce(bitmap->mddev, 1);
> +		mddev_suspend(bitmap->mddev);


mddev_suspend is supposed to be called with reconfig_mutex hold. At least one
place this isn't true with this change, for example, raid_preresume doesn't
call md_bitmap_resize with the lock hold. Could you please double check the
lock usage?
 
>  	store.file = bitmap->storage.file;
>  	bitmap->storage.file = NULL;
> @@ -2239,7 +2239,7 @@ int md_bitmap_resize(struct bitmap *bitmap, sector_t blocks,
>  
>  	if (!init) {
>  		md_bitmap_unplug(bitmap);
> -		bitmap->mddev->pers->quiesce(bitmap->mddev, 0);
> +		mddev_resume(bitmap->mddev);
>  	}
>  	ret = 0;
>  err:
> @@ -2288,9 +2288,9 @@ location_store(struct mddev *mddev, const char *buf, size_t len)
>  			goto out;
>  		}
>  		if (mddev->pers) {
> -			mddev->pers->quiesce(mddev, 1);
> +			mddev_suspend(mddev);
>  			md_bitmap_destroy(mddev);
> -			mddev->pers->quiesce(mddev, 0);
> +			mddev_resume(mddev);
>  		}
>  		mddev->bitmap_info.offset = 0;
>  		if (mddev->bitmap_info.file) {
> @@ -2327,8 +2327,8 @@ location_store(struct mddev *mddev, const char *buf, size_t len)
>  			mddev->bitmap_info.offset = offset;
>  			if (mddev->pers) {
>  				struct bitmap *bitmap;
> -				mddev->pers->quiesce(mddev, 1);
>  				bitmap = md_bitmap_create(mddev, -1);
> +				mddev_suspend(mddev);
>  				if (IS_ERR(bitmap))
>  					rv = PTR_ERR(bitmap);
>  				else {
> @@ -2337,11 +2337,12 @@ location_store(struct mddev *mddev, const char *buf, size_t len)
>  					if (rv)
>  						mddev->bitmap_info.offset = 0;
>  				}
> -				mddev->pers->quiesce(mddev, 0);
>  				if (rv) {
>  					md_bitmap_destroy(mddev);
> +					mddev_resume(mddev);
>  					goto out;
>  				}
> +				mddev_resume(mddev);
>  			}
>  		}
>  	}
> -- 
> 2.7.4
>