[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20161016135039.GK3142@twins.programming.kicks-ass.net>
Date: Sun, 16 Oct 2016 15:50:39 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Imre Palik <imrep.amz@...il.com>
Cc: Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
"H. Peter Anvin" <hpa@...or.com>, x86@...nel.org,
Andi Kleen <ak@...ux.intel.com>,
Stephane Eranian <eranian@...gle.com>,
Kan Liang <kan.liang@...el.com>,
David Carrillo-Cisneros <davidcc@...gle.com>,
linux-kernel@...r.kernel.org,
Alexander Kozyrev <alexander.kozyrev@...el.com>,
Artyom Kuanbekov <artyom.kuanbekov@...el.com>,
Imre Palik <imrep@...zon.de>, Matt Wilson <msw@...zon.com>
Subject: Re: [RFC PATCH] perf: honouring the cpuid for number of fixed
counters in hypervisors
On Thu, Oct 13, 2016 at 01:28:09AM -0700, Imre Palik wrote:
> + if (version > 1) {
> + unsigned int ecx = cpuid_ecx(1);
> +
> + if (ecx >> 31)
What is this magic, undocumented gunk doing? Is that supposed to be
static_cpu_has(X86_FEATURE_HYPERVISOR) ?
Powered by blists - more mailing lists