Re: bcache and hibernation

[Kent Overstreet <kmo@daterainc.com>]

On Sun, Nov 30, 2014 at 07:25:03PM +0100, Mathijs Kwik wrote:
> Kent Overstreet <kmo@daterainc.com> writes:
> 
> > On Thu, Nov 13, 2014 at 02:52:02PM +0100, Mathijs Kwik wrote:
> >> The kernel documentation on suspend is VERY clear you should NOT touch
> >> anything on disk between suspend and resume. So activating luks and LVM
> >> is probably risky already, but it apppears both luks and LVM do not make
> >> any on-disk changes when activated and any in-memory state (within the
> >> resumed image) is still valid. The benefit of activating luks and LVM
> >> before resume seems to be that it allows resuming from encrypted/lvm
> >> volumes. 
> >
> > Yeah, this is handled for in kernel stuff with the freezing mechanism, which
> > bcache uses.
> >
> >> 
> >
> > So, userspace shouldn't have to do anything to tell bcache about
> > hibernation.
> 
> I understand bcache knows when the system hibernates so it can do some
> bookkeeping/flushing. What I don't get, is how this will protect the
> system in the short phase before resume, when my initrd activates bcache
> and lvm, _before_ it checks for a resume image.

Wait what? Ohh... you must be talking about hibernate, not suspend...

> I guess bcache will just start running as usual. Maybe flushing some
> dirty buckets, cache some new stuff (when lvm searches for volumes and
> udev reads labels & more).
> 
> Then finally the resume mechanism loads the old hibernated system, while
> the cache has changed in the mean time! Won't this cause issues?

I have no idea how hibernate works, but yeah, quite possibly...

> > The dev branch is getting a true read only mode (still in progress), but this
> > isn't relevant to hibernation.
> >
> > bcache kernel threads (allocator thread, gc thread) should be correct w.r.t.
> > hibernation, but - maybe the workqueue usage isn't.
> >
> > I'm probably not going to be able to get to this in the next couple days, but
> > this is a pretty serious issue. Can you ping me again every couple days until I
> > get a fix out for this, and myabe file a bug somewhere? (i think
> > bugzilla.kernel.org has been used for bcache bugs before...)
> 
> I will file a report once I'm sure what the exact cause for my data-loss
> was. If I understand you correctly, all should be safe and well, no
> matter what (initramfs) userspace does in between and the only thing
> that might not be safe is the workqueue?

I was thinking suspend to ram, not hibernate. gonna have to look into how
hibernate works now.