xfs: take the invalidate lock while mapping a pNFS layout

xfs_fs_map_blocks() flushes the page cache under the iolock, but a
write fault takes only the invalidate lock. A fault on a shared mapping
can then add a delalloc extent before the range is mapped. That trips
an ASSERT, or nfsd warns and refuses the layout.

Take the invalidate lock with the iolock, as __xfs_file_fallocate()
does. Not for stable: pNFS already leaves local writers to shared
mappings unsynchronized.

Suggested-by: Darrick J. Wong <djwong@kernel.org>
Link: https://lore.kernel.org/r/20260519145949.GH9555@frogsfrogsfrogs
Signed-off-by: Daejun Park <daejun7.park@samsung.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Carlos Maiolino <cem@kernel.org>

0 files changed, 0 insertions, 0 deletions