Thank you for providing review comment.
On Tue, Feb 8, 2011 at 10:17 AM, Alex Elder <aelder@xxxxxxx> wrote:
> On Thu, 2011-02-03 at 15:17 +0900, Ajeet Yadav wrote:
>> xfsprogs: unhandled error check in libxfs_trans_read_buf
>
> Sorry you didn't get a response earlier. ÂReporting
> problems like this--especially if they come with a
> proposed fix--is always appreciated.
>
>> libxfs_trans_read_buf() is used in both mkfs.xfs & xfs_repair.
>> During stability testing we found some time occur pagefault in mkfs.xfs,
>> code inspection shows that if libxfs_readbuf() fails then occurs a
>> page fault in xfs_buf_item_init() called in libxfs_trans_read_buf().
>>
>> mkfs.xfs: unhandled page fault (11) at 0x00000070, code 0x017
>>
>> Added NULL check and errno handling.
>
> I expect there are other similar problems lurking in
> the libxfs code.
>
> I think your change looks good, with one exception,
> noted below. ÂI will gladly adjust your patch for
> you if you consent to the change I suggest.
>
> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â-Alex
>
>> Signed-off-by: Ajeet Yadav <ajeet.yadav.77@xxxxxxxxx>
>>
>> diff -Nurp xfsprogs/libxfs/rdwr.c xfsprogs-dirty/libxfs/rdwr.c
>> --- xfsprogs/libxfs/rdwr.c   Â2011-01-28 20:22:11.000000000 +0900
>> +++ xfsprogs-dirty/libxfs/rdwr.c    Â2011-02-02 16:59:16.000000000 +0900
>> @@ -207,10 +207,11 @@ libxfs_trace_readbuf(const char *func, c
>> Â{
>>     xfs_buf_t    *bp = libxfs_readbuf(dev, blkno, len, flags);
>>
>> - Â Â Â bp->b_func = func;
>> - Â Â Â bp->b_file = file;
>> - Â Â Â bp->b_line = line;
>> -
>> + Â Â Â if (bp){
>> + Â Â Â Â Â Â Â bp->b_func = func;
>> + Â Â Â Â Â Â Â bp->b_file = file;
>> + Â Â Â Â Â Â Â bp->b_line = line;
>> + Â Â Â }
>> Â Â Â Â return bp;
>> Â}
>>
>> @@ -485,6 +486,7 @@ libxfs_readbuf(dev_t dev, xfs_daddr_t bl
>> Â Â Â Â Â Â Â Â error = libxfs_readbufr(dev, blkno, bp, len, flags);
>> Â Â Â Â Â Â Â Â if (error) {
>> Â Â Â Â Â Â Â Â Â Â Â Â libxfs_putbuf(bp);
>> + Â Â Â Â Â Â Â Â Â Â Â errno = error;
>
> I think that libxfs_readbuf() simply returns the
> value of the special global "errno" if there is
> an error. ÂAnd I believe that at this point it
> will not have been changed, so there's no need
> to assign it here.
>
> In other words, I'd like to remove this one piece
> of your patch.
>
>> Â Â Â Â Â Â Â Â Â Â Â Â return NULL;
>> Â Â Â Â Â Â Â Â }
>> Â Â Â Â }
>> diff -Nurp xfsprogs/libxfs/trans.c xfsprogs-dirty/libxfs/trans.c
>> --- xfsprogs/libxfs/trans.c   2011-01-28 20:22:11.000000000 +0900
>> +++ xfsprogs-dirty/libxfs/trans.c    2011-02-02 17:00:42.000000000 +0900
>> @@ -508,6 +508,10 @@ libxfs_trans_read_buf(
>> Â Â Â Â }
>>
>> Â Â Â Â bp = libxfs_readbuf(dev, blkno, len, flags);
>> + Â Â Â if (!bp){
>> + Â Â Â Â Â Â Â *bpp = NULL;
>> + Â Â Â Â Â Â Â return errno;
>> + Â Â Â }
>> Â#ifdef XACT_DEBUG
>> Â Â Â Â fprintf(stderr, "trans_read_buf buffer %p, transaction %p\n", bp, tp);
>> Â#endif
>>
>> _______________________________________________
>> xfs mailing list
>> xfs@xxxxxxxxxxx
>> http://oss.sgi.com/mailman/listinfo/xfs
>
>
>
>
_______________________________________________
xfs mailing list
xfs@xxxxxxxxxxx
http://oss.sgi.com/mailman/listinfo/xfs
