linux-kernel - Re: Wierdness - linux-next KVM patch breaks Dell Latitude D820, KVM not in kernel

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Wed, 09 Sep 2009 14:59:12 -0400
From:	Valdis.Kletnieks@...edu
To:	Avi Kivity <avi@...hat.com>
Cc:	Beth Kon <eak@...ibm.com>, rostedt@...dmis.org,
	Andrew Morton <akpm@...ux-foundation.org>,
	linux-kernel@...r.kernel.org
Subject: Re: Wierdness - linux-next KVM patch breaks Dell Latitude D820, KVM not in kernel

On Wed, 09 Sep 2009 20:38:51 +0300, Avi Kivity said:

> > And *every single one* of those 4,000 commits between 'good' and 'bad' was a
> > KVM commit, stretching all the way back to 2007.  Unless that's an artifact
> > of the way linux-next is built, I would have expected only "new-ish" commits
> > to be in the bisect window, and not just out of one tree...
> >
> > So I ended up doing a 'WTF?', an 'rm -r', and am cloning over today's linux-next
> > and will see if that brings me any better joy.

> Oh, it's an artifact of kvm.git, not linux-next.  I really should feed 
> something better to -next.

OK... knowing that my kernel doesn't have kvm, and every single kvm commit
appears to be in one long stream in linux-next, I can probably use 'git bisect
skip rev..rev' to just lop out that tree entirely, and re-try.

I looked closer at my failure mode on 'bad' commits - I suspect what actually
happened was:

1) My *original* 'bad' was a solid hang at boot.
2) 'git bisect' dropped me into the KVM tree.
3) Bisecting in there hit an area where it would triple-fault reboot instead
because I basically had a 31-rc8 "rest of tree" and the kvm tree was somewhere
back to 2007, causing a mismatch kersplat.

So hopefully excluding kvm.git I'll track down the *original* hang...

Wish me luck. ;)

Content of type "application/pgp-signature" skipped