Re: 2.6.19-rc3: known unfixed regressions (v3)

["Michael S. Tsirkin" <mst@mellanox.co.il>]

Quoting r. Jun'ichi Nomura <j-nomura@ce.jp.nec.com>:
> Subject: Re: 2.6.19-rc3: known unfixed regressions (v3)
> 
> Hi Michael,
> 
> Michael S. Tsirkin wrote:
> >> The code is related to bd_claim_by_disk which is called when
> >> device-mapper or md tries to mark the underlying devices
> >> for exclusive use and creates symlinks from/to the devices
> >> in sysfs. The patch added error handlings which weren't in
> >> the original code.
> >>
> >> I have no idea how it affects ACPI event handling.
> > 
> > It's a mystery. Probably exposes a bug somewhere?
> > 
> >> Are you using dm and/or md on your machine?
> > 
> > The .config is attached to bugzilla.
> 
> OK, I found you disabled CONFIG_MD, which means neither
> dm.ko nor md.ko was built.
> Do you have any out-of-tree kernel modules which call either
> bd_claim_by_kobject or bd_claim_by_disk?

No, I don't have any out-of-tree modules.

> If you aren't using either of them, I'm afraid reverting
> the patch doesn't really solve your problem because the patched
> code is called only from them.

I agree this could be just papering over some issue.
The test results (of both git-bisect and reverting the patch) seem to be pretty
consistent so far though. Keep me posted if you rework the patch.

> >> Have you seen any unusual kernel messages or symptoms regarding
> >> dm/md before the ACPI problem occurs?
> > 
> > I haven't.
> 
> Thanks,

-- 
MST