linux-kernel - Re: printk.time causes rare kernel boot hangs

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]

Message-ID: <20230614150717.GG7636@redhat.com>
Date:   Wed, 14 Jun 2023 16:07:17 +0100
From:   "Richard W.M. Jones" <rjones@...hat.com>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Aaron Thompson <dev@...ont.org>, linux-kernel@...r.kernel.org
Subject: Re: printk.time causes rare kernel boot hangs

On Wed, Jun 14, 2023 at 04:53:48PM +0200, Peter Zijlstra wrote:
> On Wed, Jun 14, 2023 at 03:09:45PM +0200, Peter Zijlstra wrote:
> > On Wed, Jun 14, 2023 at 02:03:48PM +0100, Richard W.M. Jones wrote:
> > > On Wed, Jun 14, 2023 at 02:53:20PM +0200, Peter Zijlstra wrote:
> > > > Ooooh, what qemu version do you have? There were some really dodgy
> > > > reports all around self modifying code, all reported on 7.2, that seems
> > > > to have gone away with 8.
> > > 
> > > > Now, all of them were using TCG, and I think you're using KVM.
> > > 
> > > I'm using qemu-system-x86-8.0.0-4.fc39.x86_64 with KVM.  The host
> > > kernel is 6.4.0-0.rc5.41.fc39.x86_64.
> > > 
> > > > I've at least 36000 cycles and still nothing :-(, let me go try your
> > > > .config.
> > > 
> > > I can definitely reproduce this with your config (but AMD host), so
> > > that's odd.  Let me try with Intel hardware and your config.  You
> > > really should see this with only a few thousands iterations.
> > 
> > Yeah, I'm running on some ancient ivb-ep, let me try on the alderlake. I
> > don't really have AMD machines at hand :-/
> 
> Using v6.4-rc6-37-gb6dad5178cea, and 36000+ cycles on the ADL (affine to
> big cores) later and nothing :-(

So I have just this minute reproduced it on Intel, after around ~2000
iterations.

On this machine I am using:

Host kernel:  6.3.7-200.fc38.x86_64
Guest kernel: git commit fb054096aea0576f0c0a61c598e5e9676443ee86
Guest config: defconfig + kvm_guest.config + CONFIG_DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT=y
QEMU:         qemu-system-x86-8.0.0-4.fc39.x86_64 (using kvm_intel)
Hardware:     Intel(R) Core(TM) i7-10850H CPU @ 2.70GHz

> Clearly I'm doing something wrong.

I'm suspecting version of qemu may have something to do with it, but
honestly I don't know.  Which qemu are you using?

I'll now try with upstream qemu from git in case there's something
about the Fedora qemu.

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
Fedora Windows cross-compiler. Compile Windows programs, test, and
build Windows installers. Over 100 libraries supported.
http://fedoraproject.org/wiki/MinGW