On Thu, Apr 04, 2024 at 11:15:25AM +0200, Mahmoud Adam wrote:
Amir Goldstein amir73il@gmail.com writes:
On Wed, Apr 3, 2024 at 9:18 PM Darrick J. Wong djwong@kernel.org wrote:
To the group: Who's the appropriate person to handle these?
Mahmoud: If the answer to the above is "???" or silence, would you be willing to take on stable testing and maintenance?
Probably there is an answer now :). But Yes, I'm okay with doing that, Xfstests is already part for our nightly 6.1 testing.
Mahmoud,
I assume that you are running xfstests on LTS kernels regularly? In that case, you should have an established baseline for failing/passing tests on 6.1.y. Did you run these backports against all tests to verify no regressions? If you did - then please include this information (also which xfs configurations were tested) in the posting of backport candidates to xfs list.
Yes, I did run the full xfstests to confirm no regression. we do regularly run the latest stable xfstests version with loopback setup. and we run 'xfs/quick' group over x86_64 & arm64 to catch any regression. I'll make sure to post to xfs list first next time :)
our setup looks similar to this:
sudo fallocate -l 5G $MOUNT_POINT/block-xfs.img sudo mkfs.xfs -f -m reflink=1 $MOUNT_POINT/block-xfs.img sudo losetup -f $MOUNT_POINT/block-xfs.img sudo mkdir -p $MOUNT_POINT/test sudo mount /dev/loop0 $MOUNT_POINT/test
sudo fallocate -l 5G $MOUNT_POINT/block-xfs-scratch.img sudo losetup -f $MOUNT_POINT/block-xfs-scratch.img
local.config: export DISABLE_UDF_TEST=1 export TEST_DEV=/dev/loop0 export TEST_DIR=$MOUNT_POINT/test export SCRATCH_MNT=$MOUNT_POINT/scratch export SCRATCH_DEV=/dev/loop1
So does this mean we should take these for stable inclusion, or are they going to need some other tests/acks for us to be able to do this?
thanks,
greg k-h