summaryrefslogtreecommitdiffstats
path: root/fs/orangefs
diff options
context:
space:
mode:
authorMartin Brandenburg <martin@omnibond.com>2017-04-25 21:38:07 +0200
committerMike Marshall <hubcap@omnibond.com>2017-04-26 20:33:01 +0200
commitb5a9d61eebdd0016ccb383b25a5c3d04977a6549 (patch)
treea25c822b778b2b9fa1d1d3c9887195385cb8591f /fs/orangefs
parentorangefs: return from orangefs_devreq_read quickly if possible (diff)
downloadlinux-b5a9d61eebdd0016ccb383b25a5c3d04977a6549.tar.xz
linux-b5a9d61eebdd0016ccb383b25a5c3d04977a6549.zip
orangefs: do not wait for timeout if umounting
When the computer is turned off, all the processes are killed and then all the filesystems are umounted. OrangeFS should not wait for the userspace daemon to come back in that case. This only works for plain umount(2). To actually take advantage of this interactively, `umount -f' is needed; otherwise umount will issue a statfs first, which will wait for the userspace daemon to come back. Signed-off-by: Martin Brandenburg <martin@omnibond.com> Signed-off-by: Mike Marshall <hubcap@omnibond.com>
Diffstat (limited to 'fs/orangefs')
-rw-r--r--fs/orangefs/waitqueue.c9
1 files changed, 8 insertions, 1 deletions
diff --git a/fs/orangefs/waitqueue.c b/fs/orangefs/waitqueue.c
index abcfa3fa9992..61e2ca7fec55 100644
--- a/fs/orangefs/waitqueue.c
+++ b/fs/orangefs/waitqueue.c
@@ -124,7 +124,14 @@ retry_servicing:
gossip_debug(GOSSIP_WAIT_DEBUG,
"%s:client core is NOT in service.\n",
__func__);
- timeout = op_timeout_secs * HZ;
+ /*
+ * Don't wait for the userspace component to return if
+ * the filesystem is being umounted anyway.
+ */
+ if (op->upcall.type == ORANGEFS_VFS_OP_FS_UMOUNT)
+ timeout = 0;
+ else
+ timeout = op_timeout_secs * HZ;
}
spin_unlock(&orangefs_request_list_lock);