[PATCH 8/14] xfsprogs: fix issues with e0607266f23

[Eric Sandeen]

On 8/8/13 5:07 PM, Eric Sandeen wrote:
> On 8/8/13 4:53 PM, Ben Myers wrote:
>> A couple of issues found in review.
>>
>> Signed-off-by: Ben Myers <bpm at sgi.com>
>>
>> ---
>>  libxfs/xfs_alloc.c |    9 ++-------
>>  repair/dinode.c    |    9 +++------
>>  2 files changed, 5 insertions(+), 13 deletions(-)
> 
> ...
> 
> 
>> Index: b/repair/dinode.c
>> ===================================================================
>> --- a/repair/dinode.c	2013-08-06 14:43:09.910817602 -0500
>> +++ b/repair/dinode.c	2013-08-06 14:44:49.660857353 -0500
>> @@ -182,12 +182,9 @@ clear_dinode_core(struct xfs_mount *mp, 
>>  		platform_uuid_copy(&dinoc->di_uuid, &mp->m_sb.sb_uuid);
>>  	}
>>  
>> -	for (i = 0; i < 16; i++) {
>> -		if (dinoc->di_pad[i] != 0) {
>> -			__dirty_no_modify_ret(dirty);
>> -			memset(dinoc->di_pad, 0, 16);
>> -			break;
>> -		}
>> +	if (dinoc->di_pad2 != 0) {
>> +		__dirty_no_modify_ret(dirty);
>> +		dinoc->di_pad2 = 0;
> 
> this probably needs to be fixed pronto, it's a memory corruptor right?
> 
> w/ a proper commit log, 
> 
> Reviewed-by: Eric Sandeen <sandeen at redhat.com>

Actually; everywhere else that sets di_pad to 0 does it through
a memset (sizeof . . ) - that might be best, on the off chance that
di_pad ever changes, rather than setting it to = 0 ?

-Eric

> Thanks,
> -Eric
> 
>>  	}
>>  
>>  	if (be64_to_cpu(dinoc->di_flags2) != 0)  {
>>
> 
> _______________________________________________
> xfs mailing list
> xfs at oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs
>