Skip to contentSkip to conversation

XFS assertion during testing

DISCUSSION Jan Kara 2 participants 3 messages Sep 27, 2016

lore
Share & tools
lore

Jan Kara

author
#0 lore raw

Hi,

I've hit the following assertion:

XFS: Assertion failed: dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count),
file: fs/xfs/xfs_dir2_readdir.c, line: 87

during my testing. Now take this with a grain of salt since this is with my DAX patches applied on top of for-next branch of Dave's tree from yesterday. But I didn't touch XFS directory code even remotely which is why I'm reporting this.

The error happened while running xfstest xfs/083 on 20G ramdisk with dax enabled.

Honza

Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR
#1 lore raw
Quoted 7 lines · Jan Kara

Hi,

I've hit the following assertion:

XFS: Assertion failed: dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count),
file: fs/xfs/xfs_dir2_readdir.c, line: 87

Quoted 7 lines

during my testing. Now take this with a grain of salt since this is with my DAX patches applied on top of for-next branch of Dave's tree from yesterday. But I didn't touch XFS directory code even remotely which is why I'm reporting this.

The error happened while running xfstest xfs/083 on 20G ramdisk with dax enabled.

xfs/083 == "corrupt random stuff and see if anything notices" test, which is why you get asserts in the directory code while testing DAX. :)

That assert probably ought to be:

if (dp->i_d.di_size < xfs_dir2_sf_hdr_size(sfp->i8count))

	return -EFSCORRUPTED;

--D

Quoted 10 lines

Honza

Signature · 6 lines
Jan Kara <[email protected]>
SUSE Labs, CR
--
To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
the body of a message to [email protected]
More majordomo info at  http://vger.kernel.org/majordomo-info.html
author
#2 lore raw
Quoted 13 lines · Darrick J. Wong
Quoted 12 lines · Jan Kara

Hi,

I've hit the following assertion:

XFS: Assertion failed: dp->i_d.di_size >= xfs_dir2_sf_hdr_size(sfp->i8count),
file: fs/xfs/xfs_dir2_readdir.c, line: 87

during my testing. Now take this with a grain of salt since this is with my DAX patches applied on top of for-next branch of Dave's tree from yesterday. But I didn't touch XFS directory code even remotely which is why I'm reporting this.

Quoted 5 lines
Quoted 2 lines

The error happened while running xfstest xfs/083 on 20G ramdisk with dax enabled.

xfs/083 == "corrupt random stuff and see if anything notices" test, which is why you get asserts in the directory code while testing DAX. :)

Yeah, I just figured out it should not BUG, as it basically took down the machine (due to bugging when holding locks...) and that's impractical even for debugging runs.

Quoted 4 lines

That assert probably ought to be:

if (dp->i_d.di_size < xfs_dir2_sf_hdr_size(sfp->i8count))

	return -EFSCORRUPTED;

Yeah, that would have been nice ;)

Honza

Signature · 2 lines
Jan Kara <[email protected]>
SUSE Labs, CR