linux-kernel - Re: [Ocfs2-devel] [PATCH] fix dead lock caused by ocfs2_defrag

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] [day] [month] [year] [list]

Message-ID: <HK0PR06MB2561EFE93735AE62A0F5A559D50A0@HK0PR06MB2561.apcprd06.prod.outlook.com>
Date:   Tue, 28 Aug 2018 02:00:49 +0000
From:   Changwei Ge <gechangwei@...e.cn>
To:     Larry Chen <lchen@...e.com>, "mfasheh@...e.com" <mfasheh@...e.com>,
        "jlbec@...lplan.org" <jlbec@...lplan.org>
CC:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "ocfs2-devel@....oracle.com" <ocfs2-devel@....oracle.com>
Subject: Re: [Ocfs2-devel] [PATCH] fix dead lock caused by ocfs2_defrag_extent

Hi Larry,

I'd like to propose another way to fix this issue, which might be easier.
Actually, ocfs2_reserve_clusters() inside will flush truncate log if 
there is no enough clusters left.
So how about just remove __ocfs2_flush_truncate_log() in 
ocfs2_defrag_extent()?

Thanks,
Changwei


On 2018/8/27 16:01, Larry Chen wrote:
> ocfs2_defrag_extent may fall into dead lock.
>
> ocfs2_ioctl_move_extents
>    ocfs2_ioctl_move_extents	
>      ocfs2_move_extents
>        ocfs2_defrag_extent
>          ocfs2_lock_allocators_move_extents
>
>            ocfs2_reserve_clusters
>              inode_lock GLOBAL_BITMAP_SYSTEM_INODE
>
> 	  __ocfs2_flush_truncate_log
>              inode_lock GLOBAL_BITMAP_SYSTEM_INODE
>
> As back trace shows above, ocfs2_reserve_clusters will call inode_lock
> against the global bitmap if local allocator has not sufficient cluters.
> Once global bitmap could meet the demand, ocfs2_reserve_cluster will
> return success with global bitmap locked.
>
> After ocfs2_reserve_cluster, if truncate log is full,
> __ocfs2_flush_truncate_log will definitely fall into dead lock
> because it needs to inode_lock global bitmap, which has
> already been locked.
>
> To fix this bug, we could remove from ocfs2_lock_allocators_move_extents
> the code which intends to lock global allocator, and put the removed
> code after __ocfs2_flush_truncate_log
>
> The ocfs2_lock_allocators_move_extents has been refered by 2 places, one
> is here, the other does not need the data allocator context, which means
> this patch does not affect the caller so far.
>
>
>
> Signed-off-by: Larry Chen <lchen@...e.com>
> ---
>   fs/ocfs2/move_extents.c | 13 ++++++-------
>   1 file changed, 6 insertions(+), 7 deletions(-)
>
> diff --git a/fs/ocfs2/move_extents.c b/fs/ocfs2/move_extents.c
> index 7eb3b0a6347e..064dedf40d74 100644
> --- a/fs/ocfs2/move_extents.c
> +++ b/fs/ocfs2/move_extents.c
> @@ -192,13 +192,6 @@ static int ocfs2_lock_allocators_move_extents(struct inode *inode,
>   		goto out;
>   	}
>   
> -	if (data_ac) {
> -		ret = ocfs2_reserve_clusters(osb, clusters_to_move, data_ac);
> -		if (ret) {
> -			mlog_errno(ret);
> -			goto out;
> -		}
> -	}
>   
>   	*credits += ocfs2_calc_extend_credits(osb->sb, et->et_root_el);
>   
> @@ -283,6 +276,12 @@ static int ocfs2_defrag_extent(struct ocfs2_move_extents_context *context,
>   		}
>   	}
>   
> +	ret = ocfs2_reserve_clusters(osb, *len, &context->data_ac);
> +	if (ret) {
> +		mlog_errno(ret);
> +		goto out_unlock_mutex;
> +	}
> +
>   	handle = ocfs2_start_trans(osb, credits);
>   	if (IS_ERR(handle)) {
>   		ret = PTR_ERR(handle);