From: John Kacur <jkacur@redhat.com>
To: frank rowand <frank.rowand@am.sony.com>
Cc: Frank Rowand <Frank_Rowand@sonyusa.com>,
linux-rt-users@vger.kernel.org,
Garrett Smith <garrett@vmware.com>,
Lenin Singaravelu <lsingara@vmware.com>,
Bhavesh Davda <bhavesh@vmware.com>,
williams@redhat.com
Subject: Re: [PATCH] cyclictest histogram overflow instance tracking
Date: Mon, 15 Oct 2012 14:13:36 -0400 (EDT) [thread overview]
Message-ID: <215411254.459752.1350324816725.JavaMail.root@redhat.com> (raw)
In-Reply-To: <507C51B5.2060505@am.sony.com>
Ok, I'm in the process of gathering up patches - and I'll pass the to Clark.
Thanks
John
----- Original Message -----
> On 10/15/12 09:38, Bhavesh Davda wrote:
> > Hello Frank,
> >
> > Friendly ping, to make sure this doesn't fall through the cracks.
>
> You need to ping Clark Williams or John Kacur (not sure which...).
>
> >
> > Thanks
> >
> > - Bhavesh
> >
> > On Oct 11, 2012, at 1:59 PM, Bhavesh Davda wrote:
> >
> >> Thanks for the review, Frank.
> >>
> >> On Oct 11, 2012, at 12:57 PM, Frank Rowand wrote:
> >>> Nice idea…
> >> Thanks
> >>
> >>> Suggest:
> >>> printf("# Histogram Overflows at cycle number:");
> >>>
> >> Done
> >>
> >>>> - printf(" %05lu", par[j]->stats->hist_overflow);
> >>>> + printf(" %05lu", par[j]->stats->hist_overflow);
> >>>
> >>> Am I blind, or is that a whitespace change? (The latest version
> >>> I have does not seem
> >>> to have any white space issues in that line.)
> >>>
> >> Turns out I had removed extraneous whitespace. My vim settings
> >> allow me to see extraneous whitespaces in red, so I get annoyed
> >> by them and can fix them up. So I've removed a few more in the
> >> revised diffs (sorry!)
> >>
> >>> Suggest:
> >>>
> >>> + printf("# thread %d:", j);
> >>>
> >> Done
> >>
> >>> Nit: You don't need stats->outliers_overflow. You can calculate
> >>> it from
> >>> stats->hist_overflow - stats->num_outliers:
> >>>
> >> Nice catch! I've removed outliers_overflow now.
> >>
> >>
> >> From: Bhavesh Davda <bhavesh@vmware.com>
> >>
> >> Add feature to cyclictest histogram mode to track cycle counts
> >> every time a
> >> sample overflows the histogram limit. This should help identify if
> >> there is a
> >> timing pattern to jitters in cyclictest runs.
> >>
> >> Example output (with -h 10):
> >> ...
> >> Histogram Overflows: 00000 00253 00000 00005 00000 00024 00003
> >> 00005
> >> Histogram Overflow at cycle number:
> >> Thread 0:
> >> Thread 1: 00023 00028 00337 00338 00339 00340 00341 00342 00343
> >> 00344 # 00243 others
> >> Thread 2:
> >> Thread 3: 10486 10487 10488 10489 10490
> >> Thread 4:
> >> Thread 5: 00002 00004 00008 00012 00178 10458 10459 10460 10461
> >> 10462 # 00014 others
> >> Thread 6: 05954 08954 29955
> >> Thread 7: 20536 20537 20538 20539 20540
> >> ...
> >>
> >> Signed-off-by: Bhavesh Davda <bhavesh@vmware.com>
> >> ---
> >> src/cyclictest/cyclictest.c | 38
> >> +++++++++++++++++++++++++++++---------
> >> 1 files changed, 29 insertions(+), 9 deletions(-)
> >>
> >> diff --git a/src/cyclictest/cyclictest.c
> >> b/src/cyclictest/cyclictest.c
> >> index 731b4bd..dc985de 100644
> >> --- a/src/cyclictest/cyclictest.c
> >> +++ b/src/cyclictest/cyclictest.c
> >> @@ -147,6 +147,7 @@ struct thread_stat {
> >> double avg;
> >> long *values;
> >> long *hist_array;
> >> + unsigned long *outliers;
> >> pthread_t thread;
> >> int threadstarted;
> >> int tid;
> >> @@ -154,6 +155,7 @@ struct thread_stat {
> >> long redmax;
> >> long cycleofmax;
> >> long hist_overflow;
> >> + long num_outliers;
> >> };
> >>
> >> static int shutdown;
> >> @@ -756,8 +758,11 @@ void *timerthread(void *param)
> >>
> >> /* Update the histogram */
> >> if (histogram) {
> >> - if (diff >= histogram)
> >> + if (diff >= histogram) {
> >> stat->hist_overflow++;
> >> + if (stat->num_outliers < histogram)
> >> + stat->outliers[stat->num_outliers++] = stat->cycles - 1;
> >> + }
> >> else
> >> stat->hist_array[diff]++;
> >> }
> >> @@ -811,7 +816,7 @@ static void display_help(int error)
> >> if (kernvar(O_RDONLY, "available_tracers", tracers,
> >> sizeof(tracers)))
> >> strcpy(tracers, "none");
> >> }
> >> -
> >> +
> >> printf("cyclictest V %1.2f\n", VERSION_STRING);
> >> printf("Usage:\n"
> >> "cyclictest <options>\n\n"
> >> @@ -1188,7 +1193,7 @@ static int check_kernel(void)
> >> kv = KV_30;
> >> strcpy(functiontracer, "function");
> >> strcpy(traceroptions, "trace_options");
> >> -
> >> +
> >> } else
> >> kv = KV_NOT_SUPPORTED;
> >>
> >> @@ -1226,7 +1231,7 @@ static void print_tids(struct thread_param
> >> *par[], int nthreads)
> >>
> >> static void print_hist(struct thread_param *par[], int nthreads)
> >> {
> >> - int i, j;
> >> + int i, j, k;
> >> unsigned long long int log_entries[nthreads+1];
> >> unsigned long maxmax, alloverflows;
> >>
> >> @@ -1270,7 +1275,7 @@ static void print_hist(struct thread_param
> >> *par[], int nthreads)
> >> printf("# Max Latencies:");
> >> maxmax = 0;
> >> for (j = 0; j < nthreads; j++) {
> >> - printf(" %05lu", par[j]->stats->max);
> >> + printf(" %05lu", par[j]->stats->max);
> >> if (par[j]->stats->max > maxmax)
> >> maxmax = par[j]->stats->max;
> >> }
> >> @@ -1280,9 +1285,20 @@ static void print_hist(struct thread_param
> >> *par[], int nthreads)
> >> printf("# Histogram Overflows:");
> >> alloverflows = 0;
> >> for (j = 0; j < nthreads; j++) {
> >> - printf(" %05lu", par[j]->stats->hist_overflow);
> >> + printf(" %05lu", par[j]->stats->hist_overflow);
> >> alloverflows += par[j]->stats->hist_overflow;
> >> }
> >> + printf("\n");
> >> + printf("# Histogram Overflow at cycle number:\n");
> >> + for (j = 0; j < nthreads; j++) {
> >> + printf("# Thread %d:", j);
> >> + for (k = 0; k < par[j]->stats->num_outliers; k++)
> >> + printf(" %05lu", par[j]->stats->outliers[k]);
> >> + if (par[j]->stats->num_outliers < par[j]->stats->hist_overflow)
> >> + printf(" # %05lu others", par[j]->stats->hist_overflow -
> >> par[j]->stats->num_outliers);
> >> + printf("\n");
> >> + }
> >> +
> >> if (histofall && nthreads > 1)
> >> printf(" %05lu", alloverflows);
> >> printf("\n");
> >> @@ -1434,10 +1450,12 @@ int main(int argc, char **argv)
> >> int bufsize = histogram * sizeof(long);
> >>
> >> stat->hist_array = threadalloc(bufsize, node);
> >> - if (stat->hist_array == NULL)
> >> + stat->outliers = threadalloc(bufsize, node);
> >> + if (stat->hist_array == NULL || stat->outliers == NULL)
> >> fatal("failed to allocate histogram of size %d on node %d\n",
> >> histogram, i);
> >> memset(stat->hist_array, 0, bufsize);
> >> + memset(stat->outliers, 0, bufsize);
> >> }
> >>
> >> if (verbose) {
> >> @@ -1553,8 +1571,10 @@ int main(int argc, char **argv)
> >>
> >> if (histogram) {
> >> print_hist(parameters, num_threads);
> >> - for (i = 0; i < num_threads; i++)
> >> + for (i = 0; i < num_threads; i++) {
> >> threadfree(statistics[i]->hist_array, histogram*sizeof(long),
> >> parameters[i]->node);
> >> + threadfree(statistics[i]->outliers, histogram*sizeof(long),
> >> parameters[i]->node);
> >> + }
> >> }
> >>
> >> if (tracelimit) {
> >> @@ -1564,7 +1584,7 @@ int main(int argc, char **argv)
> >> printf("# Break value: %lu\n", break_thread_value);
> >> }
> >> }
> >> -
> >> +
> >>
> >> for (i=0; i < num_threads; i++) {
> >> if (!statistics[i])
> >> --
> >> 1.7.1
> >>
> >
> > --
> > Bhavesh Davda
> > bhavesh@vmware.com
>
>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2012-10-15 18:13 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-11 18:42 [PATCH] cyclictest histogram overflow instance tracking Bhavesh Davda
2012-10-11 19:57 ` Frank Rowand
2012-10-11 20:59 ` Bhavesh Davda
2012-10-15 16:38 ` Bhavesh Davda
2012-10-15 18:11 ` Frank Rowand
2012-10-15 18:13 ` John Kacur [this message]
2012-10-16 0:33 ` John Kacur
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=215411254.459752.1350324816725.JavaMail.root@redhat.com \
--to=jkacur@redhat.com \
--cc=Frank_Rowand@sonyusa.com \
--cc=bhavesh@vmware.com \
--cc=frank.rowand@am.sony.com \
--cc=garrett@vmware.com \
--cc=linux-rt-users@vger.kernel.org \
--cc=lsingara@vmware.com \
--cc=williams@redhat.com \
/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