linux.git/fs/xfs, branch v5.5

xfs: Make the symbol 'xfs_rtalloc_log_count' static

2019-12-20T16:07:31+00:00

Fix the following sparse warning:

fs/xfs/libxfs/xfs_trans_resv.c:206:1: warning: symbol 'xfs_rtalloc_log_count' was not declared. Should it be static?

Fixes: b1de6fc7520f ("xfs: fix log reservation overflows when allocating large rt extents")
Signed-off-by: Chen Wandun 
Reviewed-by: Darrick J. Wong 
Signed-off-by: Darrick J. Wong

xfs: don't commit sunit/swidth updates to disk if that would cause repair failures

2019-12-19T15:53:48+00:00

Alex Lyakas reported[1] that mounting an xfs filesystem with new sunit
and swidth values could cause xfs_repair to fail loudly.  The problem
here is that repair calculates the where mkfs should have allocated the
root inode, based on the superblock geometry.  The allocation decisions
depend on sunit, which means that we really can't go updating sunit if
it would lead to a subsequent repair failure on an otherwise correct
filesystem.

Port from xfs_repair some code that computes the location of the root
inode and teach mount to skip the ondisk update if it would cause
problems for repair.  Along the way we'll update the documentation,
provide a function for computing the minimum AGFL size instead of
open-coding it, and cut down some indenting in the mount code.

Note that we allow the mount to proceed (and new allocations will
reflect this new geometry) because we've never screened this kind of
thing before.  We'll have to wait for a new future incompat feature to
enforce correct behavior, alas.

Note that the geometry reporting always uses the superblock values, not
the incore ones, so that is what xfs_info and xfs_growfs will report.

[1] https://lore.kernel.org/linux-xfs/20191125130744.GA44777@bfoster/T/#m00f9594b511e076e2fcdd489d78bc30216d72a7d

Reported-by: Alex Lyakas 
Signed-off-by: Darrick J. Wong 
Reviewed-by: Brian Foster

xfs: split the sunit parameter update into two parts

2019-12-19T15:53:48+00:00

If the administrator provided a sunit= mount option, we need to validate
the raw parameter, convert the mount option units (512b blocks) into the
internal unit (fs blocks), and then validate that the (now cooked)
parameter doesn't screw anything up on disk.  The incore inode geometry
computation can depend on the new sunit option, but a subsequent patch
will make validating the cooked value depends on the computed inode
geometry, so break the sunit update into two steps.

Signed-off-by: Darrick J. Wong 
Reviewed-by: Brian Foster

xfs: refactor agfl length computation function

2019-12-19T15:53:48+00:00

Refactor xfs_alloc_min_freelist to accept a NULL @pag argument, in which
case it returns the largest possible minimum length.  This will be used
in an upcoming patch to compute the length of the AGFL at mkfs time.

Signed-off-by: Darrick J. Wong 
Reviewed-by: Brian Foster

libxfs: resync with the userspace libxfs

2019-12-19T15:53:47+00:00

Prepare to resync the userspace libxfs with the kernel libxfs.  There
were a few things I missed -- a couple of static inline directory
functions that have to be exported for xfs_repair; a couple of directory
naming functions that make porting much easier if they're /not/ static
inline; and a u16 usage that should have been uint16_t.

None of these things are bugs in their own right; this just makes
porting xfsprogs easier.

Signed-off-by: Darrick J. Wong 
Reviewed-by: Eric Sandeen

xfs: use bitops interface for buf log item AIL flag check

2019-12-19T15:53:47+00:00

The xfs_log_item flags were converted to atomic bitops as of commit
22525c17ed ("xfs: log item flags are racy"). The assert check for
AIL presence in xfs_buf_item_relse() still uses the old value based
check. This likely went unnoticed as XFS_LI_IN_AIL evaluates to 0
and causes the assert to unconditionally pass. Fix up the check.

Signed-off-by: Brian Foster 
Fixes: 22525c17ed ("xfs: log item flags are racy")
Reviewed-by: Eric Sandeen 
Reviewed-by: Darrick J. Wong 
Signed-off-by: Darrick J. Wong

xfs: fix log reservation overflows when allocating large rt extents

2019-12-17T19:19:28+00:00

Omar Sandoval reported that a 4G fallocate on the realtime device causes
filesystem shutdowns due to a log reservation overflow that happens when
we log the rtbitmap updates.  Factor rtbitmap/rtsummary updates into the
the tr_write and tr_itruncate log reservation calculation.

"The following reproducer results in a transaction log overrun warning
for me:

    mkfs.xfs -f -r rtdev=/dev/vdc -d rtinherit=1 -m reflink=0 /dev/vdb
    mount -o rtdev=/dev/vdc /dev/vdb /mnt
    fallocate -l 4G /mnt/foo

Reported-by: Omar Sandoval 
Tested-by: Omar Sandoval 
Signed-off-by: Darrick J. Wong 
Reviewed-by: Brian Foster

xfs: stabilize insert range start boundary to avoid COW writeback race

2019-12-11T21:18:42+00:00

generic/522 (fsx) occasionally fails with a file corruption due to
an insert range operation. The primary characteristic of the
corruption is a misplaced insert range operation that differs from
the requested target offset. The reason for this behavior is a race
between the extent shift sequence of an insert range and a COW
writeback completion that causes a front merge with the first extent
in the shift.

The shift preparation function flushes and unmaps from the target
offset of the operation to the end of the file to ensure no
modifications can be made and page cache is invalidated before file
data is shifted. An insert range operation then splits the extent at
the target offset, if necessary, and begins to shift the start
offset of each extent starting from the end of the file to the start
offset. The shift sequence operates at extent level and so depends
on the preparation sequence to guarantee no changes can be made to
the target range during the shift. If the block immediately prior to
the target offset was dirty and shared, however, it can undergo
writeback and move from the COW fork to the data fork at any point
during the shift. If the block is contiguous with the block at the
start offset of the insert range, it can front merge and alter the
start offset of the extent. Once the shift sequence reaches the
target offset, it shifts based on the latest start offset and
silently changes the target offset of the operation and corrupts the
file.

To address this problem, update the shift preparation code to
stabilize the start boundary along with the full range of the
insert. Also update the existing corruption check to fail if any
extent is shifted with a start offset behind the target offset of
the insert range. This prevents insert from racing with COW
writeback completion and fails loudly in the event of an unexpected
extent shift.

Signed-off-by: Brian Foster 
Reviewed-by: Darrick J. Wong 
Signed-off-by: Darrick J. Wong

Merge tag 'xfs-5.5-merge-17' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux

2019-12-08T01:05:33+00:00

Pull xfs fixes from Darrick Wong:
 "Fix a couple of resource management errors and a hang:

   - fix a crash in the log setup code when log mounting fails

   - fix a hang when allocating space on the realtime device

   - fix a block leak when freeing space on the realtime device"

* tag 'xfs-5.5-merge-17' of git://git.kernel.org/pub/scm/fs/xfs/xfs-linux:
  xfs: fix mount failure crash on invalid iclog memory access
  xfs: don't check for AG deadlock for realtime files in bunmapi
  xfs: fix realtime file data space leak

xfs: fix mount failure crash on invalid iclog memory access

2019-12-03T22:53:07+00:00

syzbot (via KASAN) reports a use-after-free in the error path of
xlog_alloc_log(). Specifically, the iclog freeing loop doesn't
handle the case of a fully initialized ->l_iclog linked list.
Instead, it assumes that the list is partially constructed and NULL
terminated.

This bug manifested because there was no possible error scenario
after iclog list setup when the original code was added.  Subsequent
code and associated error conditions were added some time later,
while the original error handling code was never updated. Fix up the
error loop to terminate either on a NULL iclog or reaching the end
of the list.

Reported-by: syzbot+c732f8644185de340492@syzkaller.appspotmail.com
Signed-off-by: Brian Foster 
Reviewed-by: Darrick J. Wong 
Signed-off-by: Darrick J. Wong