[Ocfs2-devel] [PATCH 1/1] ocfs2: remove redundant and incorrect mlog_error

Tiger Yang tiger.yang at oracle.com
Wed Jun 1 21:43:40 PDT 2011


On 06/01/2011 09:43 AM, Joel Becker wrote:
> On Sat, May 28, 2011 at 12:34:52AM +0800, Tiger Yang wrote:
>> We have already mlog all error and positive status is not error.
> 	Positive status is turned into -EIO.  There are actually a
> couple of places in this function that do not mlog_errno(status) and
> rely on this print.  I think you should add them to your patch.
> 	For example:
>
> 371         status = ocfs2_qinfo_lock(oinfo, 0);
> 372         if (status<  0)
> 373                 goto out_unlock;
>
> Joel
>
>
Hi, Joel,
There is something devious about this function.
1 If status == sizeof(struct ocfs2_global_disk_dqinfo)) then positive 
status will be return.
2 After goto out_unlock, they goto out_err again.
I make a new one to fix this.  please review the attached patch.

Thanks,
Tiger


-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 0001-ocfs2-remove-redundant-and-incorrect-mlog_error.patch
Url: http://oss.oracle.com/pipermail/ocfs2-devel/attachments/20110602/88479be9/attachment.pl 


More information about the Ocfs2-devel mailing list