linux-kernel - Re: GPF in intel_pmu_lbr_reset() with qemu -cpu host

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:	Sat, 22 Mar 2014 08:54:10 +0100
From:	Paolo Bonzini <pbonzini@...hat.com>
To:	"H. Peter Anvin" <hpa@...or.com>,
	Venkatesh Srinivas <venkateshs@...gle.com>,
	Peter Wu <peter@...ensteyn.nl>
CC:	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...nel.org>,
	Andi Kleen <ak@...ux.intel.com>,
	Linux Kernel Developers List <linux-kernel@...r.kernel.org>,
	kvm@...r.kernel.org, Gleb Natapov <gleb@...nel.org>
Subject: Re: GPF in intel_pmu_lbr_reset() with qemu -cpu host

Il 21/03/2014 20:09, H. Peter Anvin ha scritto:
> Calling this a bug in the PMU code is ridiculous.  If KVM tells the
> system it os a specific vendor-family-model-stepping but diverges in
> behavior then it, by definition, is broken.

Yeah, this is true.  On AMD there is processor support for virtualizing 
LBR, but Intel doesn't have it.  I'm not sure if generic load/save MSR 
support could be used to do it.

Unfortunately, LBR does not have any CPUID bit to show its presence, 
unlike a lot of other perf-related features.  So, even though calling it 
a bug in perf code is an exaggeration, using rdmsr_safe makes sense.

Paolo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/