See the comment below. This fixes LVM because on read-only mounts,
blkdev_get is now only called with a READ argument. We had been
calling it a 2nd time with READ|WRITE, and lvm didn't like this for
it's read-only snapshots.
So, don't call it a second time in XFS (the right thing to do, anyway)
and then LVM is happy as well.
Date: Fri Aug 16 09:32:00 PDT 2002
Workarea:
stout.americas.sgi.com:/localhome/src/sandeen/2.4.x-xfs/workarea-alwaysclean
The following file(s) were checked into:
bonnie.engr.sgi.com:/isms/slinx/2.4.x-xfs
Modid: 2.4.x-xfs:slinx:125261a
linux/fs/xfs/xfs_vfsops.c - 1.371
linux/fs/xfs/xfs_mount.c - 1.298
linux/fs/xfs/pagebuf/page_buf_locking.c - 1.27
linux/fs/xfs/pagebuf/page_buf.h - 1.37
- Add do_blkdev_get/do_blkdev_put arguments to pagebuf_lock_enable
and pagebuf_lock_disable, so that we don't get/put a second time
for the data device - the vfs does this for us. We still need to
do it ourselves for the logdev and rtdev, though.
This also means that the blkdev_get for the data device is now
read-only
for read-only mounts.
|