linux-kernel - Re: VMX Preemption Timer appears to be buggy on SKX, CLX, and ICX

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: <8a34f1d4-9f43-4fa7-9566-144b5eeda4d9@intel.com>
Date: Wed, 3 Jul 2024 13:14:09 -0700
From: Reinette Chatre <reinette.chatre@...el.com>
To: Sean Christopherson <seanjc@...gle.com>
CC: <isaku.yamahata@...el.com>, <pbonzini@...hat.com>,
	<erdemaktas@...gle.com>, <vkuznets@...hat.com>, <vannapurve@...gle.com>,
	<jmattson@...gle.com>, <mlevitsk@...hat.com>, <xiaoyao.li@...el.com>,
	<chao.gao@...el.com>, <rick.p.edgecombe@...el.com>, <yuan.yao@...el.com>,
	<kvm@...r.kernel.org>, <linux-kernel@...r.kernel.org>, "Hao, Xudong"
	<xudong.hao@...el.com>
Subject: Re: VMX Preemption Timer appears to be buggy on SKX, CLX, and ICX

Hi Sean,

On 6/28/24 5:39 PM, Sean Christopherson wrote:
> Forking this off to try and avoid confusion...
> 
> On Wed, Jun 12, 2024, Reinette Chatre wrote:

...

>> +
>> +		freq = (tmict - tmcct) * tdcrs[i].divide_count * tsc_hz / (tsc1 - tsc0);
>> +		/* Check if measured frequency is within 1% of configured frequency. */
>> +		GUEST_ASSERT(freq < apic_hz * 101 / 100);
>> +		GUEST_ASSERT(freq > apic_hz * 99 / 100);
>> +	}
> 
> This test fails on our SKX, CLX, and ICX systems due to what appears to be a CPU
> bug.  It looks like something APICv related is clobbering internal VMX timer state?
> Or maybe there's a tearing or truncation issue?

It has been a few days. Just a note to let you know that we are investigating this.
On my side I have not yet been able to reproduce this issue. I tested
kvm-x86-next-2024.06.28 on an ICX and an CLX system by running 100 iterations of
apic_bus_clock_test and they all passed. Since I have lack of experience here there are
some Intel virtualization experts helping out with this investigation and I hope that
they will be some insights from the analysis and testing that you already provided.

Reinette