pnfs/blocklayout: update last_write_offset atomically with extents

Browse source

Block/SCSI layout write completion may add committable extents to the
extent tree before updating the layout's last-written byte under the inode
lock.  If a sync happens before this value is updated, then
prepare_layoutcommit may find and encode these extents which would produce
a LAYOUTCOMMIT request whose encoded extents are larger than the request's
loca_length.

Fix this by using a last-written byte value that is updated atomically with
the extent tree so that commitable extents always match.

Signed-off-by: Benjamin Coddington <bcodding@redhat.com>
Signed-off-by: Trond Myklebust <trond.myklebust@primarydata.com>

...

This commit is contained in:

Benjamin Coddington 2016-08-22 14:11:16 -04:00 • committed by Trond Myklebust

parent b88fa69eaa

commit 41963c10c4

3 changed files with 10 additions and 5 deletions

Show all changes Ignore whitespace when comparing lines Ignore changes in amount of whitespace Ignore changes in whitespace at EOL

Download patch file Download diff file Expand all files Collapse all files

2

fs/nfs/blocklayout/blocklayout.c

View file

3

fs/nfs/blocklayout/blocklayout.h

View file

10

fs/nfs/blocklayout/extent_tree.c

View file