All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jaswinder Singh Rajput <jaswinder@kernel.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Peter Zijlstra <peterz@infradead.org>,
	x86 maintainers <x86@kernel.org>,
	LKML <linux-kernel@vger.kernel.org>
Subject: [PATCH 2/6 -tip] perf stat: treat same behaviour for all CYCLES and CLOCKS
Date: Wed, 01 Jul 2009 15:06:17 +0530	[thread overview]
Message-ID: <1246440977.3403.7.camel@hpdv5.satnam> (raw)
In-Reply-To: <1246440909.3403.5.camel@hpdv5.satnam>


For normalization also added SW_CPU_CLOCK and HW_BUS_CYCLES

For nsec_printout also added SW_CPU_CLOCK

Added helper functions to check counter unit as cycles and instructions

Signed-off-by: Jaswinder Singh Rajput <jaswinderrajput@gmail.com>
---
 tools/perf/builtin-stat.c |   49 +++++++++++++++++++++++++++++++-------------
 1 files changed, 34 insertions(+), 15 deletions(-)

diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index 6bf2b80..af61c29 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -144,6 +144,29 @@ static inline int nsec_counter(int counter)
 }
 
 /*
+ * Does the counter have cycles as a unit?
+ */
+static inline int cycle_counter(int counter)
+{
+	if (MATCH_EVENT(HARDWARE, HW_CPU_CYCLES, counter) ||
+	    MATCH_EVENT(HARDWARE, HW_BUS_CYCLES, counter))
+		return 1;
+
+	return 0;
+}
+
+/*
+ * Does the counter have instructions as a unit?
+ */
+static inline int instruction_counter(int counter)
+{
+	if (MATCH_EVENT(HARDWARE, HW_INSTRUCTIONS, counter))
+		return 1;
+
+	return 0;
+}
+
+/*
  * Read out the results of a single counter:
  */
 static void read_counter(int counter)
@@ -191,9 +214,9 @@ static void read_counter(int counter)
 	/*
 	 * Save the full runtime - to allow normalization during printout:
 	 */
-	if (MATCH_EVENT(SOFTWARE, SW_TASK_CLOCK, counter))
+	if (nsec_counter(counter))
 		runtime_nsecs[run_idx] = count[0];
-	if (MATCH_EVENT(HARDWARE, HW_CPU_CYCLES, counter))
+	else if (cycle_counter(counter))
 		runtime_cycles[run_idx] = count[0];
 }
 
@@ -287,11 +310,10 @@ static void nsec_printout(int counter, u64 *count, u64 *noise)
 
 	fprintf(stderr, " %14.6f  %-24s", msecs, event_name(counter));
 
-	if (MATCH_EVENT(SOFTWARE, SW_TASK_CLOCK, counter)) {
-		if (walltime_nsecs_avg)
-			fprintf(stderr, " # %10.3f CPUs ",
-				(double)count[0] / (double)walltime_nsecs_avg);
-	}
+	if (nsec_counter(counter) && walltime_nsecs_avg)
+		fprintf(stderr, " # %10.3f CPUs ",
+			(double)count[0] / (double)walltime_nsecs_avg);
+
 	print_noise(count, noise);
 }
 
@@ -299,16 +321,13 @@ static void abs_printout(int counter, u64 *count, u64 *noise)
 {
 	fprintf(stderr, " %14Ld  %-24s", count[0], event_name(counter));
 
-	if (runtime_cycles_avg &&
-	    MATCH_EVENT(HARDWARE, HW_INSTRUCTIONS, counter)) {
+	if (instruction_counter(counter) && runtime_cycles_avg)
 		fprintf(stderr, " # %10.3f IPC  ",
 			(double)count[0] / (double)runtime_cycles_avg);
-	} else {
-		if (runtime_nsecs_avg) {
-			fprintf(stderr, " # %10.3f M/sec",
-				(double)count[0]/runtime_nsecs_avg*1000.0);
-		}
-	}
+	else if (runtime_nsecs_avg)
+		fprintf(stderr, " # %10.3f M/sec",
+			(double)count[0]/runtime_nsecs_avg*1000.0);
+
 	print_noise(count, noise);
 }
 
-- 
1.6.0.6




  reply	other threads:[~2009-07-01  9:47 UTC|newest]

Thread overview: 41+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-07-01  9:33 [GIT-PULL -tip][PATCH 0/6] perf_counter patches Jaswinder Singh Rajput
2009-07-01  9:35 ` [PATCH 1/6 -tip] perf stat: define MATCH_EVENT for easy attrs checking Jaswinder Singh Rajput
2009-07-01  9:36   ` Jaswinder Singh Rajput [this message]
2009-07-01  9:37     ` [PATCH 3/6 -tip] perf_counter: Add Generalized Hardware vectored co-processor support for AMD Jaswinder Singh Rajput
2009-07-01  9:38       ` [PATCH 4/6 -tip] perf_counter: Add Generalized Hardware interrupt " Jaswinder Singh Rajput
2009-07-01  9:38         ` [PATCH 5/6 -tip] perf_counter: Add hardware vector events for nehalem Jaswinder Singh Rajput
2009-07-01  9:40           ` [PATCH 6/6 -tip] perf_counter: Add hardware interrupt events for nehalem, core2 and atom Jaswinder Singh Rajput
2009-07-01 11:24         ` [PATCH 4/6 -tip] perf_counter: Add Generalized Hardware interrupt support for AMD Ingo Molnar
2009-07-03 12:01           ` Jaswinder Singh Rajput
2009-07-04 10:22             ` Ingo Molnar
2009-07-04 14:17               ` Jaswinder Singh Rajput
2009-07-05  1:11                 ` Ingo Molnar
2009-07-05  4:29                   ` Jaswinder Singh Rajput
2009-07-05  8:04                     ` Ingo Molnar
2009-07-05  9:01                       ` Jaswinder Singh Rajput
2009-07-05  9:55                       ` Jaswinder Singh Rajput
2009-07-01 11:20       ` [PATCH 3/6 -tip] perf_counter: Add Generalized Hardware vectored co-processor " Ingo Molnar
2009-07-01 11:27         ` Ingo Molnar
2009-07-01 11:40           ` Jaswinder Singh Rajput
2009-07-01 11:49             ` Ingo Molnar
2009-07-02  9:44               ` [PATCH 1/2 -tip] perf_counter: Add generalized hardware vectored co-processor support for AMD and Intel Corei7/Nehalem Jaswinder Singh Rajput
2009-07-02  9:45                 ` [PATCH 2/2 -tip] perf_counter: Add generalized hardware interrupt support for AMD and Intel Corei7/Nehalem, Core2 and Atom Jaswinder Singh Rajput
2009-07-03 10:33                   ` Ingo Molnar
2009-07-03  7:38                 ` [PATCH 1/2 -tip] perf_counter: Add generalized hardware vectored co-processor support for AMD and Intel Corei7/Nehalem Jaswinder Singh Rajput
2009-07-03  9:30                   ` Ingo Molnar
2009-07-03 10:10                     ` Jaswinder Singh Rajput
2009-07-03 12:17                     ` [PATCH 3/3 -tip] perf list: avoid replicating functions Jaswinder Singh Rajput
2009-07-04  9:50                       ` Ingo Molnar
2009-07-03 10:29                 ` [PATCH 1/2 -tip] perf_counter: Add generalized hardware vectored co-processor support for AMD and Intel Corei7/Nehalem Ingo Molnar
2009-07-03 11:55                   ` Jaswinder Singh Rajput
2009-07-03 12:49                     ` Jaswinder Singh Rajput
2009-07-03 13:25                       ` Jaswinder Singh Rajput
2009-07-04 10:03                         ` Ingo Molnar
2009-07-04 14:05                           ` Jaswinder Singh Rajput
2009-07-04  9:49                     ` Ingo Molnar
2009-07-04 13:54                       ` Jaswinder Singh Rajput
2009-07-01 11:39     ` [PATCH 2/6 -tip] perf stat: treat same behaviour for all CYCLES and CLOCKS Ingo Molnar
2009-07-03  8:18       ` Paul Mackerras
2009-07-03  8:27         ` Ingo Molnar
2009-07-01 11:30   ` [tip:perfcounters/urgent] perf stat: Define MATCH_EVENT for easy attr checking tip-bot for Jaswinder Singh Rajput
2009-07-01 11:45 ` [GIT-PULL -tip][PATCH 0/6] perf_counter patches Ingo Molnar

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=1246440977.3403.7.camel@hpdv5.satnam \
    --to=jaswinder@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.