diff options
| author | Paul Zuchowski <[email protected]> | 2019-08-15 10:27:13 -0400 |
|---|---|---|
| committer | Brian Behlendorf <[email protected]> | 2019-08-15 08:27:13 -0600 |
| commit | e2b31b58e8777faa51561342d72a5a30127fa4b6 (patch) | |
| tree | d97edab581099d6bf3c0a49d126e038484ff1108 | |
| parent | dc04a8c757d7df91efbca05491174112540f6e7a (diff) | |
Make txg_wait_synced conditional in zfsvfs_teardown
The call to txg_wait_synced in zfsvfs_teardown should
be made conditional on the objset having dirty data.
This can prevent unnecessary txg_wait_synced during
some unmount operations.
Reviewed-by: Matt Ahrens <[email protected]>
Reviewed-by: Brian Behlendorf <[email protected]>
Signed-off-by: Paul Zuchowski <[email protected]>
Closes #9115
| -rw-r--r-- | module/zfs/zfs_vfsops.c | 11 |
1 files changed, 10 insertions, 1 deletions
diff --git a/module/zfs/zfs_vfsops.c b/module/zfs/zfs_vfsops.c index 8d728adea..af82c7bc4 100644 --- a/module/zfs/zfs_vfsops.c +++ b/module/zfs/zfs_vfsops.c @@ -1778,8 +1778,17 @@ zfsvfs_teardown(zfsvfs_t *zfsvfs, boolean_t unmounting) * Evict cached data. We must write out any dirty data before * disowning the dataset. */ - if (!zfs_is_readonly(zfsvfs)) + objset_t *os = zfsvfs->z_os; + boolean_t os_dirty = B_FALSE; + for (int t = 0; t < TXG_SIZE; t++) { + if (dmu_objset_is_dirty(os, t)) { + os_dirty = B_TRUE; + break; + } + } + if (!zfs_is_readonly(zfsvfs) && os_dirty) { txg_wait_synced(dmu_objset_pool(zfsvfs->z_os), 0); + } dmu_objset_evict_dbufs(zfsvfs->z_os); return (0); |