Messages in this thread | | | Date | Mon, 16 Dec 2013 21:24:40 +0800 | Subject | Re: [PATCH] libata, freezer: avoid block device removal while system is frozen | From | Ming Lei <> |
| |
On Mon, Dec 16, 2013 at 8:45 PM, Tejun Heo <tj@kernel.org> wrote: > Hello, > > On Mon, Dec 16, 2013 at 08:12:07PM +0800, Ming Lei wrote: >> Looks we should guarantee that I/O queue is emptied before system sleep, >> otherwise it may be a bug, since battery may be drained up to cause >> data loss or block devices may be disconnected during sleep. > > It's a bit beside the point as that's what journaling is for. Power > loss during suspend isn't all that different from power loss while > suspended. Also, the filesystems are already synced before entering > suspend, so it shouldn't be an issue. However, that doesn't guarantee > there aren't no pending bdi work items. We *hope* to control the
You mean there are still some write I/O scheduled after processes are frozen? by unfreezable kernel threads?
> sources with freezer but there's no guarantee about the coverage and > no way to verify that either.
Thanks, -- Ming Lei
| |