linux-kernel - Re: [PATCH] x86/tsc: mark tsc reliable on CoffeeLake

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] [day] [month] [year] [list]

Date:   Fri, 12 Apr 2019 11:28:12 +0800
From:   You-Sheng Yang <vicamo.yang@...onical.com>
To:     Thomas Gleixner <tglx@...utronix.de>,
        You-Sheng Yang <vicamo@...il.com>
Cc:     Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        x86@...nel.org, Pavel Tatashin <pasha.tatashin@...cle.com>,
        Peter Zijlstra <peterz@...radead.org>,
        "Rafael J. Wysocki" <rafael.j.wysocki@...el.com>,
        Dou Liyang <douly.fnst@...fujitsu.com>,
        Daniel Vacek <neelx@...hat.com>,
        Mike Travis <mike.travis@....com>,
        Xiaoming Gao <gxm.linux.kernel@...il.com>,
        "H. Peter Anvin" <hpa@...or.com>,
        Chuanhua Lei <chuanhua.lei@...ux.intel.com>,
        Juergen Gross <jgross@...e.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] x86/tsc: mark tsc reliable on CoffeeLake

On 2019/4/8 8:03 PM, Thomas Gleixner wrote:
> On Mon, 8 Apr 2019, You-Sheng Yang wrote:
>> +	/*
>> +	 * On Intel CoffeeLake, tsc may be marked unstable unexpectedly after
>> +	 * entering PC10.
>> +	 */
>> +	if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
>> +	    (boot_cpu_data.x86_model == INTEL_FAM6_KABYLAKE_MOBILE ||
>> +	     boot_cpu_data.x86_model == INTEL_FAM6_KABYLAKE_DESKTOP) &&
>> +	    boot_cpu_data.x86_stepping >= 0x0a)
>> +		tsc_clocksource_reliable = 1;
> 
> No. We are not starting that family/model/stepping game especially not
> with random stepping cutoffs which are pulled out of thin air.  That's
> going to spiral out of control sooner than later.

What about we simply disable clocksource watchdog if this is an
invariant TSC?

> There must be a better way to do that. Rafael?
> 
> Thanks,
> 
> 	tglx

You-Sheng Yang



Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)