mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: LKML <linux-kernel@vger.kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>,
	x86@kernel.org, Borislav Petkov <bp@alien8.de>,
	Stephane Eranian <eranian@google.com>,
	Harish Chegondi <harish.chegondi@intel.com>,
	Kan Liang <kan.liang@intel.com>,
	Andi Kleen <andi.kleen@intel.com>,
	Jacob Pan <jacob.jun.pan@linux.intel.com>
Subject: [patch V2 21/28] x86/perf/intel/rapl: Calculate timing once
Date: Mon, 22 Feb 2016 11:06:58 -0000	[thread overview]
Message-ID: <20160222110441.967549490@linutronix.de> (raw)
In-Reply-To: <20160222110337.198341739@linutronix.de>

[-- Attachment #1: x86-perf-intel-rapl--Calculate-timing-once.patch --]
[-- Type: text/plain, Size: 2976 bytes --]

No point in doing the same calculation over and over. Do it once in
rapl_check_hw_unit().

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/cpu/perf_event_intel_rapl.c |   39 ++++++++++++----------------
 1 file changed, 18 insertions(+), 21 deletions(-)

Index: b/arch/x86/kernel/cpu/perf_event_intel_rapl.c
===================================================================
--- a/arch/x86/kernel/cpu/perf_event_intel_rapl.c
+++ b/arch/x86/kernel/cpu/perf_event_intel_rapl.c
@@ -129,6 +129,7 @@ static int rapl_hw_unit[NR_RAPL_DOMAINS]
 static struct pmu rapl_pmu_class;
 static cpumask_t rapl_cpu_mask;
 static int rapl_cntr_mask;
+static u64 rapl_timer_ms;
 
 static DEFINE_PER_CPU(struct rapl_pmu *, rapl_pmu);
 static DEFINE_PER_CPU(struct rapl_pmu *, rapl_pmu_to_free);
@@ -558,7 +559,6 @@ static int rapl_cpu_prepare(int cpu)
 {
 	struct rapl_pmu *pmu = per_cpu(rapl_pmu, cpu);
 	int phys_id = topology_physical_package_id(cpu);
-	u64 ms;
 
 	if (pmu)
 		return 0;
@@ -575,19 +575,7 @@ static int rapl_cpu_prepare(int cpu)
 
 	pmu->pmu = &rapl_pmu_class;
 
-	/*
-	 * use reference of 200W for scaling the timeout
-	 * to avoid missing counter overflows.
-	 * 200W = 200 Joules/sec
-	 * divide interval by 2 to avoid lockstep (2 * 100)
-	 * if hw unit is 32, then we use 2 ms 1/200/2
-	 */
-	if (rapl_hw_unit[0] < 32)
-		ms = (1000 / (2 * 100)) * (1ULL << (32 - rapl_hw_unit[0] - 1));
-	else
-		ms = 2;
-
-	pmu->timer_interval = ms_to_ktime(ms);
+	pmu->timer_interval = ms_to_ktime(rapl_timer_ms);
 
 	rapl_hrtimer_init(pmu);
 
@@ -676,6 +664,19 @@ static int rapl_check_hw_unit(void (*qui
 	/* Apply cpu model quirk */
 	if (quirk)
 		quirk();
+
+	/*
+	 * Calculate the timer rate:
+	 * Use reference of 200W for scaling the timeout to avoid counter
+	 * overflows. 200W = 200 Joules/sec
+	 * Divide interval by 2 to avoid lockstep (2 * 100)
+	 * if hw unit is 32, then we use 2 ms 1/200/2
+	 */
+	rapl_timer_ms = 2;
+	if (rapl_hw_unit[0] < 32) {
+		rapl_timer_ms = (1000 / (2 * 100));
+		rapl_timer_ms *= (1ULL << (32 - rapl_hw_unit[0] - 1));
+	}
 	return 0;
 }
 
@@ -695,9 +696,7 @@ static const struct x86_cpu_id rapl_cpu_
 static int __init rapl_pmu_init(void)
 {
 	void (*quirk)(void) = NULL;
-	struct rapl_pmu *pmu;
-	int cpu, ret;
-	int i;
+	int cpu, ret, i;
 
 	/*
 	 * check for Intel processor family 6
@@ -758,15 +757,14 @@ static int __init rapl_pmu_init(void)
 	}
 
 	__perf_cpu_notifier(rapl_cpu_notifier);
-
-	pmu = __this_cpu_read(rapl_pmu);
+	cpu_notifier_register_done();
 
 	pr_info("RAPL PMU detected,"
 		" API unit is 2^-32 Joules,"
 		" %d fixed counters"
 		" %llu ms ovfl timer\n",
 		hweight32(rapl_cntr_mask),
-		ktime_to_ms(pmu->timer_interval));
+		rapl_timer_ms);
 	for (i = 0; i < NR_RAPL_DOMAINS; i++) {
 		if (rapl_cntr_mask & (1 << i)) {
 			pr_info("hw unit of domain %s 2^-%d Joules\n",
@@ -774,7 +772,6 @@ static int __init rapl_pmu_init(void)
 		}
 	}
 
-	cpu_notifier_register_done();
 	return 0;
 
 out:

  parent reply	other threads:[~2016-02-22 11:08 UTC|newest]

Thread overview: 47+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-02-22 11:06 [patch V2 00/28] x86/perf/intel/uncore|rapl: Fix error handling and sanitize pmu management Thomas Gleixner
2016-02-22 11:06 ` [patch V2 01/28] x86/perf/intel/uncore: Remove pointless mask check Thomas Gleixner
2016-02-22 11:06 ` [patch V2 03/28] x86/perf/intel/uncore: Fix error handling Thomas Gleixner
2016-02-22 11:06 ` [patch V2 02/28] x86/perf/intel/uncore: Simplify error rollback Thomas Gleixner
2016-02-22 11:06 ` [patch V2 04/28] x86/perf/intel/uncore: Add sanity checks for pci dev package id Thomas Gleixner
2016-02-22 11:06 ` [patch V2 05/28] x86/perf/intel_uncore: Cleanup hardware on exit Thomas Gleixner
2016-02-22 11:06 ` [patch V2 07/28] x86/perf/intel_uncore: Make code readable Thomas Gleixner
2016-02-22 11:06 ` [patch V2 06/28] x86/perf/intel/uncore: Drop pointless hotplug priority Thomas Gleixner
2016-02-22 11:06 ` [patch V2 08/28] x86/perf/uncore: Make uncore_pcibus_to_physid static Thomas Gleixner
2016-02-22 11:06 ` [patch V2 09/28] perf: Allow storage of pmu private data in event Thomas Gleixner
2016-02-22 11:06 ` [patch V2 11/28] x86/topology: Create logical package id Thomas Gleixner
2016-02-22 18:54   ` Andi Kleen
2016-02-22 19:05     ` Thomas Gleixner
2016-02-22 11:06 ` [patch V2 10/28] x86/perf/intel_uncore: Store box in event->pmu_private Thomas Gleixner
2016-02-22 11:45   ` Peter Zijlstra
2016-02-22 11:58     ` Thomas Gleixner
2016-02-22 11:52   ` Peter Zijlstra
2016-02-22 12:00     ` Thomas Gleixner
2016-02-22 11:06 ` [patch V2 12/28] x86/perf/uncore: Track packages not per cpu data Thomas Gleixner
2016-02-22 11:06 ` [patch V2 13/28] x86/perf/intel_uncore: Clear all hardware state on exit Thomas Gleixner
2016-02-22 11:58   ` Peter Zijlstra
2016-02-22 11:59     ` Thomas Gleixner
2016-02-22 11:06 ` [patch V2 15/28] cpumask: Export cpumask_any_but Thomas Gleixner
2016-02-22 11:06 ` [patch V2 14/28] x86/perf/intel_uncore: Make PCI and MSR uncore independent Thomas Gleixner
2016-02-22 11:06 ` [patch V2 16/28] x86/perf/intel_uncore: Make it modular Thomas Gleixner
2016-02-22 11:06 ` [patch V2 17/28] x86/perf/cqm: Get rid of the silly for_each_cpu lookups Thomas Gleixner
2016-02-22 11:06 ` [patch V2 18/28] x86/perf/intel_rapl: Make Knights Landings support functional Thomas Gleixner
2016-02-22 20:38   ` Borislav Petkov
2016-02-22 11:06 ` [patch V2 19/28] x86/perf/intel/rapl: Add proper error handling Thomas Gleixner
2016-02-22 11:06 ` [patch V2 20/28] x86/perf/intel/rapl: Sanitize the quirk handling Thomas Gleixner
2016-02-22 11:06 ` Thomas Gleixner [this message]
2016-02-22 11:06 ` [patch V2 22/28] x86/perf/intel/rapl: Cleanup the printk output Thomas Gleixner
2016-02-22 18:44   ` Andi Kleen
2016-02-22 18:58     ` Thomas Gleixner
2016-02-22 11:06 ` [patch V2 23/28] x86/perf/intel/rapl: Refactor code some more Thomas Gleixner
2016-02-22 11:07 ` [patch V2 24/28] x86/perf/intel/rapl: Make pmu lock raw Thomas Gleixner
2016-02-22 11:07 ` [patch V2 25/28] x86/perf/intel/rapl: Utilize event->pmu_private Thomas Gleixner
2016-02-22 11:07 ` [patch V2 26/28] x86/perf/intel/rapl: Convert it to a per package facility Thomas Gleixner
2016-02-22 12:08   ` Peter Zijlstra
2016-02-22 15:52     ` Thomas Gleixner
2016-02-22 16:04       ` Peter Zijlstra
2016-02-22 11:07 ` [patch V2 27/28] perf: Export perf_event_sysfs_show Thomas Gleixner
2016-02-22 12:09   ` Peter Zijlstra
2016-02-22 11:07 ` [patch V2 28/28] x86/perf/intel/rapl: Make it modular Thomas Gleixner
2016-02-22 18:40   ` Andi Kleen
2016-02-22 19:11     ` Thomas Gleixner
2016-02-22 18:55 ` [patch V2 00/28] x86/perf/intel/uncore|rapl: Fix error handling and sanitize pmu management Andi Kleen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20160222110441.967549490@linutronix.de \
    --to=tglx@linutronix.de \
    --cc=andi.kleen@intel.com \
    --cc=bp@alien8.de \
    --cc=eranian@google.com \
    --cc=harish.chegondi@intel.com \
    --cc=jacob.jun.pan@linux.intel.com \
    --cc=kan.liang@intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=peterz@infradead.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®