Re: [Bug report][fstests generic/047] Internal error !(flags & XFS_DABUF_MAP_HOLE_OK) at line 2572 of file fs/xfs/libxfs/xfs_da_btree.c. Caller xfs_dabuf_map.constprop.0+0x26c/0x368 [xfs]

[edward6@linux.ibm.com]

From: Eduard Shishkin <edward6@linux.ibm.com>

[...]

> Can you test the patch below and see if it fixes the issue? Keep
> the first verifier patch I sent, then apply the patch below. You can
> drop the debug traceprintk patch - the patch below should fix it.


Thanks for fixing it!
Tested-by: Eduard Shishkin <edward6@linux.ibm.com>

[...]

> This manifests obviously on big endian platforms (e.g. s390) because
> the log dinode is in host order and the overlap is the LSBs of the
> extent count field. It is not noticed on little endian machines
> because the overlap is at the MSB end of the extent count field and
> we need to get more than 2^^48 extents in the inode before it
> manifests. i.e. the heat death of the universe will occur before we
> see the problem in little endian machines.


This sounds too bold. I can easily imagine mountable images similar to
filesystem meta-data dumps, but not generated by scanning real
partitions, instead created by a special tool e.g. for debugging
purposes. My point is that on little-endian architectures the
manifestation of such a "sleeping" bug is a much more realistic event
than it may seem at first glance.