From: Ingo Molnar <mingo@elte.hu>
To: David Miller <davem@davemloft.net>
Cc: a.p.zijlstra@chello.nl, paulus@samba.org, tglx@linutronix.de,
linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org,
akpm@linux-foundation.org, eranian@googlemail.com,
dada1@cosmosbay.com, robert.richter@amd.com, arjan@infradead.org,
hpa@zytor.com, rostedt@goodmis.org
Subject: Re: [patch 0/3] [Announcement] Performance Counters for Linux
Date: Fri, 5 Dec 2008 09:11:37 +0100 [thread overview]
Message-ID: <20081205081137.GB2030@elte.hu> (raw)
In-Reply-To: <20081205.000716.40104924.davem@davemloft.net>
* David Miller <davem@davemloft.net> wrote:
> From: Peter Zijlstra <a.p.zijlstra@chello.nl>
> Date: Fri, 05 Dec 2008 09:03:36 +0100
>
> > On Fri, 2008-12-05 at 18:57 +1100, Paul Mackerras wrote:
> > > Peter Zijlstra writes:
> > >
> > > > So, while most people would not consider two consecutive read() ops to
> > > > be close or near the same time, due to preemption and such, that is
> > > > taken away by the fact that the counters are task local time based - so
> > > > preemption doesn't affect thing. Right?
> > >
> > > I'm sorry, I don't follow the argument here. What do you mean by
> > > "task local time based"?
> >
> > time only flows when the task is running.
>
> These things aren't measuring time, or even just cycles, they are
> measuring things like L2 cache misses, cpu cycles, and other similar
> kinds of events.
>
> So these counters are going to measure all of the damn crap assosciated
> with doing the read() call as well as the real work the task does.
that's wrong, look at the example we posted - see it pasted below.
When monitoring another task it does _not_ count the read() done in the
monitoring task, it does _not_ include it in the event count. It is a
fundamental property of our code to be as unintrusive as possible. It
only measures the work done by that task.
( You _can_ measure your own overhead of course too, if you want to. It's
a natural special-case of our performance counter abstraction. )
Ingo
---
/*
* Performance counters monitoring test case
*/
#include <sys/types.h>
#include <sys/stat.h>
#include <sys/time.h>
#include <unistd.h>
#include <stdint.h>
#include <stdlib.h>
#include <string.h>
#include <getopt.h>
#include <fcntl.h>
#include <stdio.h>
#include <errno.h>
#define __user
#include "sys.h"
static int count = 10000;
static int eventid;
static int tid;
static char *debuginfo;
static void display_help(void)
{
printf("monitor\n");
printf("Usage:\n"
"monitor options threadid\n\n"
"-e EID --eventid=EID eventid\n"
"-c CNT --count=CNT event count on which IP is sampled\n"
"-d FILE --debug=FILE path to binary file with debug info\n");
exit(0);
}
static void process_options (int argc, char *argv[])
{
int error = 0;
for (;;) {
int option_index = 0;
/** Options for getopt */
static struct option long_options[] = {
{"count", required_argument, NULL, 'c'},
{"debug", required_argument, NULL, 'd'},
{"eventid", required_argument, NULL, 'e'},
{"help", no_argument, NULL, 'h'},
{NULL, 0, NULL, 0}
};
int c = getopt_long(argc, argv, "c:d:e:",
long_options, &option_index);
if (c == -1)
break;
switch (c) {
case 'c': count = atoi(optarg); break;
case 'd': debuginfo = strdup(optarg); break;
case 'e': eventid = atoi(optarg); break;
default: error = 1; break;
}
}
if (error || optind == argc)
display_help ();
tid = atoi(argv[optind]);
}
int main(int argc, char *argv[])
{
char str[256];
uint64_t ip;
ssize_t res;
int fd;
process_options(argc, argv);
fd = perf_counter_open(eventid, count, 1, tid, -1);
if (fd < 0) {
perror("Create counter");
exit(-1);
}
while (1) {
res = read(fd, (char *) &ip, sizeof(ip));
if (res != sizeof(ip)) {
perror("Read counter");
break;
}
if (!debuginfo) {
printf("IP: 0x%016llx\n", (unsigned long long)ip);
} else {
sprintf(str, "addr2line -e %s 0x%llx\n", debuginfo,
(unsigned long long)ip);
system(str);
}
}
close(fd);
exit(0);
}
next prev parent reply other threads:[~2008-12-05 8:12 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-12-04 23:44 [patch 0/3] [Announcement] Performance Counters for Linux Thomas Gleixner
2008-12-04 23:44 ` [patch 1/3] performance counters: core code Thomas Gleixner
2008-12-05 10:55 ` Paul Mackerras
2008-12-05 11:20 ` Ingo Molnar
2008-12-04 23:44 ` [patch 2/3] performance counters: documentation Thomas Gleixner
2008-12-05 0:33 ` Paul Mackerras
2008-12-05 0:37 ` David Miller
2008-12-05 2:50 ` Arjan van de Ven
2008-12-05 3:26 ` David Miller
2008-12-05 2:33 ` Andi Kleen
2008-12-04 23:45 ` [patch 3/3] performance counters: x86 support Thomas Gleixner
2008-12-05 0:22 ` [patch 0/3] [Announcement] Performance Counters for Linux Paul Mackerras
2008-12-05 6:31 ` Ingo Molnar
2008-12-05 7:02 ` Arjan van de Ven
2008-12-05 7:52 ` David Miller
2008-12-05 7:03 ` Ingo Molnar
2008-12-05 7:03 ` Ingo Molnar
2008-12-05 7:16 ` Peter Zijlstra
2008-12-05 7:57 ` Paul Mackerras
2008-12-05 8:03 ` Peter Zijlstra
2008-12-05 8:07 ` David Miller
2008-12-05 8:11 ` Ingo Molnar [this message]
2008-12-05 8:17 ` David Miller
2008-12-05 8:24 ` Ingo Molnar
2008-12-05 8:27 ` David Miller
2008-12-05 8:42 ` Ingo Molnar
2008-12-05 8:49 ` David Miller
2008-12-05 12:13 ` Ingo Molnar
2008-12-05 12:13 ` Ingo Molnar
2008-12-05 12:39 ` Andi Kleen
2008-12-05 20:08 ` David Miller
2008-12-10 3:48 ` Paul Mundt
2008-12-10 4:42 ` Paul Mackerras
2008-12-10 8:43 ` Mikael Pettersson
2008-12-10 10:28 ` Andi Kleen
2008-12-10 10:23 ` Paul Mundt
2008-12-10 11:03 ` Andi Kleen
2008-12-10 11:03 ` Andi Kleen
2008-12-10 10:28 ` Andi Kleen
2008-12-05 15:00 ` Arjan van de Ven
2008-12-05 9:16 ` Paul Mackerras
2008-12-05 7:57 ` David Miller
2008-12-05 8:18 ` Ingo Molnar
2008-12-05 8:20 ` David Miller
2008-12-05 7:54 ` Paul Mackerras
2008-12-05 8:08 ` Ingo Molnar
2008-12-05 8:15 ` David Miller
2008-12-05 13:25 ` Ingo Molnar
2008-12-05 9:10 ` Paul Mackerras
2008-12-05 12:07 ` Ingo Molnar
2008-12-06 0:05 ` Paul Mackerras
2008-12-06 1:23 ` Mikael Pettersson
2008-12-06 12:34 ` Peter Zijlstra
2008-12-07 5:15 ` Paul Mackerras
2008-12-08 7:18 ` stephane eranian
2008-12-08 11:11 ` Ingo Molnar
2008-12-08 11:58 ` David Miller
2008-12-09 0:21 ` stephane eranian
2008-12-09 0:21 ` stephane eranian
2008-12-05 0:22 ` H. Peter Anvin
2008-12-05 0:43 ` Paul Mackerras
2008-12-05 1:12 ` David Miller
2008-12-05 6:10 ` Ingo Molnar
2008-12-05 7:50 ` David Miller
2008-12-05 9:34 ` Paul Mackerras
2008-12-05 10:41 ` Ingo Molnar
2008-12-05 10:05 ` Ingo Molnar
2008-12-05 3:30 ` Andrew Morton
2008-12-06 2:36 ` stephane eranian
2008-12-08 2:12 ` [perfmon2] [patch 0/3] [Announcement] Performance Counters forLinux Dan Terpstra
2008-12-10 16:27 ` [patch 0/3] [Announcement] Performance Counters for Linux Rob Fowler
2008-12-10 16:27 ` [perfmon2] " Rob Fowler
2008-12-10 17:11 ` Andi Kleen
2008-12-10 17:11 ` 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=20081205081137.GB2030@elte.hu \
--to=mingo@elte.hu \
--cc=a.p.zijlstra@chello.nl \
--cc=akpm@linux-foundation.org \
--cc=arjan@infradead.org \
--cc=dada1@cosmosbay.com \
--cc=davem@davemloft.net \
--cc=eranian@googlemail.com \
--cc=hpa@zytor.com \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=paulus@samba.org \
--cc=robert.richter@amd.com \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
/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