[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140709091917.GC9918@twins.programming.kicks-ass.net>
Date: Wed, 9 Jul 2014 11:19:17 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: kan.liang@...el.com
Cc: andi@...stfloor.org, linux-kernel@...r.kernel.org,
kvm@...r.kernel.org
Subject: Re: [PATCH V4 1/2] perf ignore LBR and extra_regs.
On Tue, Jul 08, 2014 at 09:49:40AM -0700, kan.liang@...el.com wrote:
> + /*
> + * Access LBR MSR may cause #GP under certain circumstances.
> + * E.g. KVM doesn't support LBR MSR
> + * Check all LBT MSR here.
> + * Disable LBR access if any LBR MSRs can not be accessed.
> + */
> + if (x86_pmu.lbr_nr) {
> + access = check_msr(x86_pmu.lbr_tos);
> + for (i = 0; i < x86_pmu.lbr_nr; i++) {
> + access &= check_msr(x86_pmu.lbr_from + i);
> + access &= check_msr(x86_pmu.lbr_to + i);
> + }
> + if (!access)
> + x86_pmu.lbr_nr = 0;
> + }
You _could_ write that as:
if (x86_pmu.lbr_nr) {
if (!check_msr(x86_pmu.lbr_tos))
x86_pmu.lbr_nr = 0;
for (i = 0; i < x86_pmu.lbr_nr; i++) {
if (!(check_msr(x86_pmu.lbr_from + i) &&
check_msr(x86_pmu.lbr_to + i)))
x86_pmu.lbr_nr = 0;
}
}
There is no point in checking any more MSRs after the first fail after
all.
Content of type "application/pgp-signature" skipped
Powered by blists - more mailing lists