linux-kernel - Re: [LKP] [x86/pci] 7ffb31888c: PANIC:early

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: <alpine.DEB.2.21.1809201138230.3243@nanos.tec.linutronix.de>
Date:   Thu, 20 Sep 2018 11:39:17 +0200 (CEST)
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Pu Wen <puwen@...on.cn>
cc:     Rong Chen <rong.a.chen@...el.com>, bp@...en8.de, mingo@...hat.com,
        hpa@...or.com, x86@...nel.org, thomas.lendacky@....com,
        helgaas@...nel.org, linux-kernel@...r.kernel.org,
        linux-arch@...r.kernel.org, linux-pci@...r.kernel.org, lkp@...org
Subject: Re: [LKP] [x86/pci] 7ffb31888c: PANIC:early_exception

On Thu, 20 Sep 2018, Pu Wen wrote:
> On 2018/9/20 8:26, Rong Chen wrote:
> > On 09/19/2018 09:53 PM, Pu Wen wrote:
> > > I cannot reproduce this panic on Hygon Dhyana platform. I tired lkp-tests
> > > both in Ubuntu 16.04 (with gcc-5) and Ubuntu 18.04 (with gcc-7).
> > > What kind of host do you use to run the qemu? A Intel one or AMD one?
> > It's a Intel one,
> > 
> > model: Nehalem-EX
> > memory: 224G
> > nr_node: 4
> > nr_cpu: 64
> > nr_socket: 4
> > brand: Intel(R) Xeon(R) CPU X7560 @ 2.27GHz
> 
> I reproduced the PANIC by running lkp-tests with a Intel machine.
> 
> The reason is that the function early_is_amd_nb is called even if
> running on Intel machine. At this case the misc_ids is NULL so the
> PANIC occur.
> 
> I'll rework a patch to solve this problem, either restore the default
> value of misc_ids with the original AMD one or not call early_is_amd_nb
> on Intel platform.

Return when !AMD and !HYGON is the right thing to do. No point in checking
device ids which cannot happen in the first place.

Thanks,

	tglx