xfs
[Top] [All Lists]

Re: PATCH 09/13 V2] libxfs: remove ASSERT on ftype read from disk

To: Eric Sandeen <sandeen@xxxxxxxxxxx>
Subject: Re: PATCH 09/13 V2] libxfs: remove ASSERT on ftype read from disk
From: Brian Foster <bfoster@xxxxxxxxxx>
Date: Tue, 24 Mar 2015 08:07:43 -0400
Cc: Eric Sandeen <sandeen@xxxxxxxxxx>, xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <551073F9.6060908@xxxxxxxxxxx>
References: <1426624395-8258-1-git-send-email-sandeen@xxxxxxxxxx> <1426624395-8258-10-git-send-email-sandeen@xxxxxxxxxx> <551073F9.6060908@xxxxxxxxxxx>
User-agent: Mutt/1.5.23 (2014-03-12)
On Mon, Mar 23, 2015 at 03:13:45PM -0500, Eric Sandeen wrote:
> This one is already fixed in the kernel, with
>       fb04013 xfs: don't ASSERT on corrupt ftype
> but that kernel<->userspace merge is still pending.
> 
> In the meantime, just fix it as a one-off here, because ASSERTing
> on bad on-disk values when running xfs_repair is a very unfriendly
> thing to do.
> 
> Signed-off-by: Eric Sandeen <sandeen@xxxxxxxxxx>
> ---

Reviewed-by: Brian Foster <bfoster@xxxxxxxxxx>

> 
> Remove it, don't cplusplus-comment it out!  o_O
> 
> diff --git a/include/xfs_da_format.h b/include/xfs_da_format.h
> index 89a1a21..11f1420 100644
> --- a/include/xfs_da_format.h
> +++ b/include/xfs_da_format.h
> @@ -561,7 +561,6 @@ xfs_dir3_dirent_get_ftype(
>       if (xfs_sb_version_hasftype(&mp->m_sb)) {
>               __uint8_t       type = dep->name[dep->namelen];
>  
> -             ASSERT(type < XFS_DIR3_FT_MAX);
>               if (type < XFS_DIR3_FT_MAX)
>                       return type;
>  
> 
> _______________________________________________
> xfs mailing list
> xfs@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/xfs

<Prev in Thread] Current Thread [Next in Thread>