direct-io.hg
changeset 7549:dc36edf1102f
Don't schedule work for reboot if work is already scheduled. This fixes a
problem whereby migrated domains would shut themselves down immediately,
because the watch from the removal of the control/shutdown node would fire
after the workqueue had started, but before the domain was suspended, meaning
that an additional piece of work was scheduled, which ran immediately after the
domain was resumed.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
problem whereby migrated domains would shut themselves down immediately,
because the watch from the removal of the control/shutdown node would fire
after the workqueue had started, but before the domain was suspended, meaning
that an additional piece of work was scheduled, which ran immediately after the
domain was resumed.
Signed-off-by: Ewan Mellor <ewan@xensource.com>
author | emellor@leeni.uk.xensource.com |
---|---|
date | Sun Oct 30 13:52:38 2005 +0100 (2005-10-30) |
parents | 9f44b2aa98f3 |
children | bcccadcc56e5 |
files | linux-2.6-xen-sparse/arch/xen/kernel/reboot.c |
line diff
1.1 --- a/linux-2.6-xen-sparse/arch/xen/kernel/reboot.c Sun Oct 30 13:48:34 2005 +0100 1.2 +++ b/linux-2.6-xen-sparse/arch/xen/kernel/reboot.c Sun Oct 30 13:52:38 2005 +0100 1.3 @@ -283,7 +283,7 @@ static void shutdown_handler(struct xenb 1.4 int err; 1.5 1.6 if (shutting_down != SHUTDOWN_INVALID) 1.7 - goto out; 1.8 + return; 1.9 1.10 again: 1.11 xbt = xenbus_transaction_start();