Fix "allocating allocated segment" panic

Gunnar Beutner did all the hard work on this one by correctly
identifying that this issue is a race between dmu_sync() and
dbuf_dirty().

Now in all cases the caller is responsible for preventing this
race by making sure the zfs_range_lock() is held when dirtying
a buffer which may be referenced in a log record.  The mmap
case which relies on zfs_putpage() was not taking the range
lock.  This code was accidentally dropped when the function
was rewritten for the Linux VFS.

This patch adds the required range locking to zfs_putpage().

It also adds the missing ZFS_ENTER()/ZFS_EXIT() macros which
aren't strictly required due to the VFS holding a reference.
However, this makes the code more consistent with the upsteam
code and there's no harm in being extra careful here.

Original-patch-by: Gunnar Beutner <gunnar@beutner.name>
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
Closes #541
This commit is contained in:
Brian Behlendorf 2012-10-31 10:06:34 -07:00
parent e26ade5101
commit 4c837f0d93
1 changed files with 10 additions and 1 deletions

View File

@ -3790,7 +3790,8 @@ zfs_putpage(struct inode *ip, struct page *pp, struct writeback_control *wbc)
zfs_sb_t *zsb = ITOZSB(ip);
loff_t offset;
loff_t pgoff;
unsigned int pglen;
unsigned int pglen;
rl_t *rl;
dmu_tx_t *tx;
caddr_t va;
int err = 0;
@ -3799,6 +3800,8 @@ zfs_putpage(struct inode *ip, struct page *pp, struct writeback_control *wbc)
int cnt = 0;
int sync;
ZFS_ENTER(zsb);
ZFS_VERIFY_ZP(zp);
ASSERT(PageLocked(pp));
@ -3810,6 +3813,7 @@ zfs_putpage(struct inode *ip, struct page *pp, struct writeback_control *wbc)
/* Page is beyond end of file */
if (pgoff >= offset) {
unlock_page(pp);
ZFS_EXIT(zsb);
return (0);
}
@ -3832,6 +3836,7 @@ zfs_putpage(struct inode *ip, struct page *pp, struct writeback_control *wbc)
set_page_writeback(pp);
unlock_page(pp);
rl = zfs_range_lock(zp, pgoff, pglen, RL_WRITER);
tx = dmu_tx_create(zsb->z_os);
sync = ((zsb->z_os->os_sync == ZFS_SYNC_ALWAYS) ||
@ -3858,6 +3863,8 @@ zfs_putpage(struct inode *ip, struct page *pp, struct writeback_control *wbc)
if (sync)
zfs_putpage_commit_cb(pp, ECANCELED);
zfs_range_unlock(rl);
ZFS_EXIT(zsb);
return (err);
}
@ -3873,6 +3880,7 @@ zfs_putpage(struct inode *ip, struct page *pp, struct writeback_control *wbc)
zfs_log_write(zsb->z_log, tx, TX_WRITE, zp, pgoff, pglen, 0);
dmu_tx_commit(tx);
zfs_range_unlock(rl);
ASSERT3S(err, ==, 0);
if (sync) {
@ -3880,6 +3888,7 @@ zfs_putpage(struct inode *ip, struct page *pp, struct writeback_control *wbc)
zfs_putpage_commit_cb(pp, err);
}
ZFS_EXIT(zsb);
return (err);
}