By default, _dmerror_load_*_table() suspends the dm device with
"--nolockfs" option. Callers have to feed two arguments to these
functions to change the behavior, with the second being 1, but the
first argument is not used at all, which doesn't make sense.
Fix it by checking if the first argument is "lockfs" and removing
"--nolockfs" option if so, or passing all options to dmsetup.
Signed-off-by: Zorro Lang <zlang@redhat.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
The test program src/test-nextquota.c relies on atoi() to convert a
string to an *unsigned* int. If the string represents an integer
which is greater than INT_MAX, it is undefined how atoi(3) works,
and it turns out that:
uint id = atoi("2147483649");
results in id == 2147483649 on x86_64, and id == 2147483647 on a
32-bit x86 platform.
So use strtoul(3) instead, which is portable and technically correct
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
When btrfs hits EDQUOTA when reserving data space, it will leak
already reserved data space.
This test case will check it by using more restrict enospc_debug
mount option to trigger kernel warning at umount time.
Signed-off-by: Qu Wenruo <quwenruo@cn.fujitsu.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
The test does the following:
Initialize a RAID5 with some data
Re-mount RAID5 degraded with _dev3_ missing and write data.
Save md5sum checkpoint1
Re-mount healthy RAID5
Let balance fix degraded blocks.
Save md5sum checkpoint2
Re-mount RAID1 degraded now with _dev1_ missing.
Save md5sum checkpoint3
Verify if all three md5sum matches
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
The test does the following:
Initialize a RAID1 with some data
Re-mount RAID1 degraded with _dev1_ and write up to
half of the FS capacity
Save md5sum checkpoint1
Re-mount healthy RAID1
Let balance re-silver.
Save md5sum checkpoint2
Re-mount RAID1 degraded with _dev2_
Save md5sum checkpoint3
Verify if all three md5sum match
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
We need btrfs to be a module so that it can unloaded and reloaded,
so that we can clean up the btrfs internal in memory device list.
This patch adds _require_btrfs_unloadable() and _reload_btrfs_ko()
to help with the same.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Below patches added helper function to get the requested
number of devices for scratch and spare device
btrfs: add functions to get and put a device for replace target
btrfs: add functions to set and reset required number of SCRATCH_DEV_POOL
This patch makes use of them.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
For the replace tests we need a device as a spare device,
here functions _spare_dev_get() and _spare_dev_put()
will get it from the SCRATCH_DEV_POOL_SAVED, which is set
when _scratch_dev_pool_get() is called, and is based on how
many has already been assigned to SCRATCH_DEV_POOL.
usage:
_scratch_dev_pool_get 3
_spare_dev_get
SPARE_DEV will have a device set which can be
used as the replace target device.
_spare_dev_put
_scratch_dev_pool_put
_spare_dev_get() will pick the next device after SCRATCH_DEV_POOL
devices, from the SCRATCH_DEV_POOL_SAVED, and assigns it to
SPARE_DEV. _spare_dev_put() will set to SPARE_DEV to null.
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
This patch provides functions
_scratch_dev_pool_get()
_scratch_dev_pool_put()
Which will help to set/reset SCRATCH_DEV_POOL with the required
number of devices. SCRATCH_DEV_POOL_SAVED will hold all the devices.
Usage:
_scratch_dev_pool_get() <ndevs>
:: do stuff
_scratch_dev_pool_put()
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
XFS has a bug where directory readahead completions can occur after
unmount. This can lead to a crash or panic because metadata read
verification attempts to access core XFS data structures (e.g., the
log) after they have been freed and certain pointers have been
reset.
Add a test that triggers directory readahead, delays the readahead
I/O and immediately unmounts the filesystem. This test is part of
the dangerous group as it will cause kernels affected by the bug to
crash.
[eguan replaced touch with echo to speedup file creation]
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Add some infrastructure in common/dmdelay to support use of the
dm-delay device-mapper module within tests. This is effectively
copied from the existing infrastructure in common/dmflakey. This
provides the ability to delay I/O. It only supports delaying read
I/O as that is all that is required at this point in time.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Test the removal of a dir which contains whiteout from lower dir,
which could crash the kernel.
The following kernel commit fixed this issue
84889d4 ovl: check dentry positiveness in ovl_cleanup_whiteouts()
Reviewed-by: Xiong Zhou <xzhou@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
overlayfs was leaking dentry on non-directories when using the
'default_permissions' mount option, which resulted in "BUG Dentry
still in use".
The following kernel commit fixed it:
ovl: fix dentry leak for default_permissions
Reviewed-by: Xiong Zhou <xzhou@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
This test has been removed from the auto group because it fails without
an expected fix. Remove it from the quick group as well, as the quick
group should be a subset of the auto group.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
btrfsprogs v4.5.3 changed the formatting of some error messages.
This patch extends the filter for btrfs prop to handle those.
Signed-off-by: Jeff Mahoney <jeffm@suse.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Add _require_runas and _runas functions instead of open-coding it in
test cases.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Commit 8469a8c1: "xfs/133-4: filter redundant projid 0 quota report
info out" added _filter_project_quota to common/filter, and this
conflicted with a _filter_project_quota function in generic/082. So
rename the function in generic/082 to avoid the conflict.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Teach _require_odirect to skip the O_DIRECT tests if the ext4 mount
option "test_dummy_encryption" is enabled, since O_DIRECT doesen't
make sense when file system level encryption is enabled, and several
tests which use O_DIRECT --- ext4/301, ext4/304, generic/250 and
generic/252 --- fail when encryption is enabled.
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
ext4/271 runs in no journal mode (-onoload), so running test with
journal related mount options makes no sense, and test fails after
kernel commit 1e381f60dad9 ("ext4: do not allow journal_opts for fs
w/o journal"), journal related mount options are not allowed in no
journal mode. So _notrun if there're journal related mount options.
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Some tests require that there's no certain mount option in
MKFS_OPTIONS, so introduce a new helper
_exclude_scratch_mount_option() to do the check on $MOUNT_OPTIONS.
Also convert generic/192 and xfs/134 to use this helper.
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
btrfs fi sync /mnt, now does not output anything for success,
so the 006.out should be updated.
This change in btrfs-progs was introduced in the commit
b005ca024990569d2de459485682158633937928
btrfs-progs: fi sync: make it silent by default
which was integrated at btrfs-progs version v4.5.2
Signed-off-by: Anand Jain <anand.jain@oracle.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Create file/dir over whiteout by another user, uid/gid of the
file/dir should be current fsuid/fsgid, not the mounter's.
This upstream kernel patch fixed the issue
d0e13f5 ovl: fix uid/gid when creating over whiteout
Reviewed-by: Xiong Zhou <xzhou@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Test getcwd() failure after unsuccessful rmdir, the getcwd() call
should return the correct path.
Upstream commit ce9113bbcbf4 ("ovl: fix getcwd() failure after
unsuccessful rmdir") fixed this issue.
Reviewed-by: Xiong Zhou <xzhou@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Zorro Lang
Theodore Ts'o
Qu Wenruo
Anand Jain
Brian Foster
Eryu Guan
Christoph Hellwig
Jeff Mahoney
Andreas Gruenbacher
Omer Zilberberg