[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <1426871484-21285-2-git-send-email-andi@firstfloor.org>
Date: Fri, 20 Mar 2015 10:11:24 -0700
From: Andi Kleen <andi@...stfloor.org>
To: peterz@...radead.org
Cc: linux-kernel@...r.kernel.org, eranian@...gle.com,
Andi Kleen <ak@...ux.intel.com>
Subject: [PATCH 2/2] perf, x86: Avoid rewriting DEBUGCTL with the same value for LBRs
From: Andi Kleen <ak@...ux.intel.com>
perf with LBRs on has a tendency to rewrite the DEBUGCTL MSR with
the same value. Add a little optimization to skip the unnecessary
write.
Signed-off-by: Andi Kleen <ak@...ux.intel.com>
---
arch/x86/kernel/cpu/perf_event_intel_lbr.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/x86/kernel/cpu/perf_event_intel_lbr.c b/arch/x86/kernel/cpu/perf_event_intel_lbr.c
index 3d53725..94e5b50 100644
--- a/arch/x86/kernel/cpu/perf_event_intel_lbr.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_lbr.c
@@ -135,7 +135,7 @@ static void intel_pmu_lbr_filter(struct cpu_hw_events *cpuc);
static void __intel_pmu_lbr_enable(bool pmi)
{
struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
- u64 debugctl, lbr_select = 0;
+ u64 debugctl, lbr_select = 0, orig_debugctl;
/*
* No need to reprogram LBR_SELECT in a PMI, as it
@@ -147,6 +147,7 @@ static void __intel_pmu_lbr_enable(bool pmi)
}
rdmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+ orig_debugctl = debugctl;
debugctl |= DEBUGCTLMSR_LBR;
/*
* LBR callstack does not work well with FREEZE_LBRS_ON_PMI.
@@ -155,7 +156,8 @@ static void __intel_pmu_lbr_enable(bool pmi)
*/
if (!(lbr_select & LBR_CALL_STACK))
debugctl |= DEBUGCTLMSR_FREEZE_LBRS_ON_PMI;
- wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
+ if (orig_debugctl != debugctl)
+ wrmsrl(MSR_IA32_DEBUGCTLMSR, debugctl);
}
static void __intel_pmu_lbr_disable(void)
--
1.9.3
--
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/
Powered by blists - more mailing lists