XFS can allocate significant amounts of space to files via speculative
preallocation. Such preallocation may not be reclaimed automatically on
file close() if a file is repeatedly opened and extended. For smaller
filesystems with relatively large and slow growing files, this
preallocation can linger for some time, including contributing to out of
space conditions.
Create a situation where an fs is near out of space while several files
still have lingering, significant preallocations. Verify that new
writers reclaim the preallocated space rather than return ENOSPC. Repeat
a similar test for quota limits and EDQUOT.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
cifs.ko supports server-side copy offloads via CIFS_IOC_COPYCHUNK_FILE.
In handling the ioctl, the request is split into a series of
SMB2 FSCTL_SRV_COPYCHUNK wire requests, which may be handled by the SMB
server as a local read/write, or COW clone as is the case for Samba with
vfs_btrfs.
Signed-off-by: David Disseldorp <ddiss@suse.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
The following kernel commit introduced a race condition that causes
getcwd(2) to return "/" instead of correct path
232d2d6 dcache: Translating dentry into pathname without taking rename_lock
Jan Stancek hit it once when building ltp and Mikulas Patocka could
hit it by running lvm2 test suite. Please refer to this thread
https://www.mail-archive.com/ltp-list@lists.sourceforge.net/msg17896.html
These commits fixed the bug
ede4ceb prepend_path() needs to reinitialize dentry/vfsmount/mnt on restarts
f650080 __dentry_path() fixes
Cc: Artem Savkov <asavkov@redhat.com>
Cc: Jan Stancek <jstancek@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Enter max ACL count into common/attr for JFS, with the purpose of
getting JFS to pass xfstests generic/026.
Signed-off-by: Michael L. Semon <mlsemon35@gmail.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Regression test for btrfs ioctl clone operation + fsync + log
recovery. The issue was that doing an fsync after cloning into
a file didn't gave any persistence guarantees as it should.
What happened was that the in memory metadata (extent maps)
weren't updated, which made the fsync code not able to detect
that file data has been changed and must be persisted to the
log.
This issue is fixed by the following linux kernel btrfs patch:
Btrfs: make fsync work after cloning into a file
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
sizeof(pointer) will give you the size of a pointer, not the space
allocated to it. I noticed this when gcc complained:
resvtest.c:76:33: warning: argument to 'sizeof' in 'memset' call is the same expression as the destination; did you mean to provide an explicit length? [-Wsizeof-pointer-memaccess]
memset(writebuffer, 'A', sizeof(writebuffer));
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Reviewed-by: Alex Elder <elder@linaro.org>
Signed-off-by: Dave Chinner <david@fromorbit.com>
It has been reported that test xfs/013 probably uses more space than
necessary, exhausting space if run against a several GB sized ramdisk.
xfs/013 primarily creates, links and removes inodes. Most of the space
consumption occurs via the background fsstress workload.
Remove the fsstress -w option that suppresses non-write operations. This
slightly reduces the storage footprint while still providing a
background workload for the test.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Regression test for the btrfs ioctl clone operation when the source range
contains hole(s) and the FS has the NO_HOLES feature enabled (file holes
don't need file extent items in the btree to represent them).
This issue is fixed by the following linux kernel btrfs patch:
Btrfs: fix clone to deal with holes when NO_HOLES feature is enabled
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Regression test for a btrfs incremental send issue where the difference
between the snapshots used by the incremental send consists of one of
these cases:
1) First snapshot has a directory with name X and in the second snapshot
that directory doesn't exist anymore but a subvolume/snapshot with
the same name (X) exists;
2) First snapshot has a subvolume/snapshot with name X and in the second
snapshot that subvolume/snapshot doesn't exist anymore (might have been
replaced by a directory with the same name or not).
This issue is fixed by the following linux kernel btrfs patches:
Btrfs: send, don't error in the presence of subvols/snapshots
Btrfs: set dead flag on the right root when destroying snapshot
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Verify that btrfs send is able to replicate xattrs larger than
PATH_MAX. This is possible if the b+tree leaf size is larger
than 4Kb (mkfs.btrfs's default is max(16Kb, PAGE_SIZE) as of
btrfs-progs v3.12, and max(4Kb, PAGE_SIZE in older versions).
This issue is fixed by the following linux kernel btrfs patch:
Btrfs: send, use the right limits for xattr names and values
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
This is a test to verify that the btrfs ioctl clone operation is
able to clone extents of a file to different positions of the file,
that is, the source and target files are the same. Existing tests
only cover the case where the source and target files are different.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: David Disseldorp <ddiss@suse.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
So that the same check (btrfs cloner program presence) can be reused
by other tests.
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: David Disseldorp <ddiss@suse.de>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Failure message of mount has been changed since util-linux v2.21, to
something like:
mount: mount /dev/sda5 on /mnt/scratch failed: Structure needs cleaning
Filter the output to match the golden image for newer mount binary so
that both old and new version of mount work correctly.
Signed-off-by: Eryu Guan <eguan@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Run 8 processes writing 1k files to seperate files in seperate dirs to
hit ENOSPC on small fs with little free space. Loop for 100 iterations.
Regression test for
34cf865 ext4: fix deadlock when writing in ENOSPC conditions
Signed-off-by: Eryu Guan <eguan@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Regression test for btrfs send where long paths (exceeding 230 characters)
made send produce paths with random characters from a memory buffer returned
by kmalloc, as send forgot to populate the new buffer with the path string.
This issue is fixed by the following linux kernel btrfs patch:
Btrfs: send, fix corrupted path strings for long paths
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Several tests happen to make use of loop device support without the
requisite pre-test checks. This results in spurious failures for systems
that might not have loop device support. Add _require_loop checks to
shared/298, xfs/206 and xfs/259.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Regression for btrfs send when an inode only has extended references
associated to it (no regular references present). This used to cause
incorrect access to a b+tree leaf, where an extended reference item
was accessed as if it were a regular reference item, causing unexpected
and unpredictable behaviour such as producing a random/weird path string
or a crash.
This issue is fixed by the following linux kernel btrfs patch:
Btrfs: send, fix incorrect ref access when using extrefs
Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Kill any lingering fsstress processes and wait properly should we abort
the test. This prevents the workload from inadvertently affecting
subsequent tests.
Signed-off-by: Brian Foster <bfoster@redhat.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
v2: just include xfs/libxfs.h for the conftest prerequisites
commit cc085d770adb in xfsprogs removed some header files that are
included by loggen.c. Add an autoconf test that checks to see whether
xfs_log_format.h is present. If it is, include that instead of the
deprecated headers and add an alias for XFS_TRANS_MAGIC since that
value no longer exists.
With this patch, xfstests builds on both f20 and rawhide (f21) boxes.
Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Eliminate do_unlock as it's just cut-and-paste of do_lock (down to the
bad stderr fprintf). Fix the debug messages to print the cmd and type,
and add the ability to do a WRTEST and RDTEST.
Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Ensure that the fd is marked invalid after close. Also, the "closed"
and "reopen" variables are always set to 0. Remove them.
Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
In a later patch, I'll be adding a test for F_GETLK that will require
that we open the file with a different f_mode. Change do_open not to
assume that the it needs to open the file O_RDWR.
While we're at it, fix a bug in do_open. Just because we pass in '0' for
the flags, doesn't mean that this is the initial open. Move the exit(1)
for that case to the caller.
Signed-off-by: Jeff Layton <jlayton@poochiereds.net>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
With the new config stuff we lost the selinux options being set for systems with
selinux turned on. We want the selinux context set all the time, wether we
provide a MOUNT_OPTIONS value or not, so take this logic out of _mount_opts()
and just put it in the body of common/config
Signed-off-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
Brian Foster
David Disseldorp
Eryu Guan
Michael L. Semon
Filipe David Borba Manana
Jeff Moyer
Jeff Layton
Josef Bacik