linux-rt-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: GeunSik Lim <leemgs1@gmail.com>
To: Clark Williams <williams@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>
Cc: rt-users <linux-rt-users@vger.kernel.org>
Subject: Re: cyclictest - duplicate explanation of cyclictest option
Date: Mon, 6 Apr 2009 17:29:04 +0900	[thread overview]
Message-ID: <49b7c2350904060129r640a3db6r5487e72f2cf6330f@mail.gmail.com> (raw)
In-Reply-To: <49b7c2350904051956n214efbb5pc8cd9990820a534f@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 7088 bytes --]

Dear Thomas & Clark,

This is e-mail about patch to support wakeup & wakeup_rt tracing additionally
at the argument of cyclictest.
Currently, cyclictest v 0.34 support  three tracing like PREEMPTOFF ,
IRQSOFF , PREEMPTIRQSOFF.
I want you to append WAKEUP & WAKEUP_RT at the current cyclictest version.

and, cyclictest ver 0.34 have a bug about context switching  and event option.
for example,
we can not use  "-C" option without --context option.
we can not use  "-E" option without --event option.

Belows is result after patching in sequence.
 1) adjust  "remove-duplicate-h-option.patch" file.
 2) and then,  adjust
"add-wakeup-wakeuprt-tracing-and-misc-bugfix.patch"  file.

Hope this helps.  If I mistake, correct me.
Thanks.

After Patching ....

#### [ wakeup test]###############################################
[root@fedora9 rt-tests-temp0]# cat /debug/tracing/available_tracers
syscall blk kmemtrace power branch function_graph mmiotrace wakeup_rt
wakeup preemptirqsoff preemptoff irqsoff function sched_switch
initcall nop

[root@fedora9 rt-tests-temp0]# ./cyclictest  -t 2 -b 200 -W -D 5
could not set latency_hist/wakeup_latency/reset to 1
0.16 0.27 0.21 2/308 2521

T: 0 ( 2520) P: 0 I:1000 C:   4288 Min:     16 Act:   27 Avg:   28 Max:     152
T: 1 ( 2521) P: 0 I:1500 C:   2859 Min:     19 Act:   29 Avg:   27 Max:      72

[root@fedora9 rt-tests-temp0]# cat /debug/tracing/trace
# tracer: wakeup
#
# wakeup latency trace v1.1.5 on 2.6.29-rt2-ftrace
# --------------------------------------------------------------------
# latency: 20 us, #3/3, CPU#2 | (M:preempt VP:0, KP:0, SP:0 HP:0 #P:4)
#    -----------------
#    | task: gnome-terminal-3884 (uid:504 nice:0 policy:0 rt_prio:0)
#    -----------------
#
#                  _------=> CPU#
#                 / _-----=> irqs-off
#                | / _----=> need-resched
#                || / _---=> hardirq/softirq
#                ||| / _--=> preempt-depth
#                |||| /
#                |||||     delay
#  cmd     pid   ||||| time  |   caller
#     \   /      |||||   \   |   /
gnome-te-3884    2d.h.    1us+: irq_enter <-smp_apic_timer_interrupt
gnome-te-3884    2dN.1   20us : irq_exit <-smp_apic_timer_interrupt
gnome-te-3884    2dN.1   21us : trace_preempt_on <-smp_apic_timer_interrupt

#####[ Context Switching Test]#####################################

[root@fedora9 rt-tests-temp0]# ./cyclictest  -t 10 -b 100000 -C -m -D 5
could not set latency_hist/wakeup_latency/reset to 1
0.17 0.17 0.17 1/326 2706

T: 0 ( 2697) P: 0 I:1000 C:   4000 Min:     12 Act:   21 Avg:   20 Max:     107
T: 1 ( 2698) P: 0 I:1500 C:   2667 Min:     13 Act:   27 Avg:   22 Max:     138
T: 2 ( 2699) P: 0 I:2000 C:   2000 Min:     12 Act:   13 Avg:   13 Max:      40
T: 3 ( 2700) P: 0 I:2500 C:   1600 Min:     12 Act:   16 Avg:   13 Max:      41
T: 4 ( 2701) P: 0 I:3000 C:   1334 Min:     12 Act:   15 Avg:   13 Max:      39
T: 5 ( 2702) P: 0 I:3500 C:   1143 Min:     12 Act:   28 Avg:   16 Max:      61
T: 6 ( 2703) P: 0 I:4000 C:   1000 Min:     12 Act:   13 Avg:   13 Max:     102
T: 7 ( 2704) P: 0 I:4500 C:    889 Min:     12 Act:   13 Avg:   15 Max:     184
T: 8 ( 2705) P: 0 I:5000 C:    800 Min:     12 Act:   13 Avg:   13 Max:      32
T: 9 ( 2706) P: 0 I:5500 C:    728 Min:     12 Act:   15 Avg:   15 Max:      54
[root@fedora9 rt-tests-temp0]# cat /debug/tracing/trace |head
# tracer: sched_switch
#
#           TASK-PID    CPU#    TIMESTAMP  FUNCTION
#              | |       |          |         |
##### CPU 0 buffer started ####
          <idle>-0     [000]  1107.917632:      0:140:R   + [000]
10: 49:D sirq-sched/0
          <idle>-0     [000]  1107.917635:      0:140:R   + [000]
5: 49:D sirq-timer/0
          <idle>-0     [000]  1107.917642:      0:140:R ==> [000]
10: 49:R sirq-sched/0
    sirq-sched/0-10    [000]  1107.917655:     10: 49:D ==> [000]
5: 49:R sirq-timer/0
    sirq-timer/0-5     [000]  1107.917660:      5: 49:D ==> [000]
0:140:R <idle>
[root@fedora9 rt-tests-temp0]#


#################################################################################





2009/4/6 GeunSik Lim <leemgs1@gmail.com>:
> Dear Clark,
>
> This is trivial contents about option of cyclictest.
>
> Cyclictest utiltiy (ver 0.34) display the information with duplicate
> "-h" options,
> when a user run "cyclictest --help" command.
> User will confuse  exact understanding because of duplicate
> description of -h option.
> How about you remove "-h H_MAX  *****" line?
> Thanks reading.
>
> Signed-off-by: GeunSik Lim <leemgs1@gmail.com>
>
> diff --git a/src/cyclictest/cyclictest.c b/src/cyclictest/cyclictest.c
> index f878217..cedc36f 100644
> --- a/src/cyclictest/cyclictest.c
> +++ b/src/cyclictest/cyclictest.c
> @@ -688,7 +688,8 @@ static void display_help(void)
>               "-d DIST  --distance=DIST   distance of thread
> intervals in us default=500\n"
>               "-E       --event           event tracing (used with -b)\n"
>               "-f       --ftrace          function trace (when -b is active)\n"
> -              "-h H_MAX                   latency histogram size in
> us default 0 (off)\n"
> +              "-h       --histogram=US    dump a latency histogram to
> stdout after the run\n"
> +              "                           US is the max time to be be
> tracked in microseconds\n"
>               "-i INTV  --interval=INTV   base interval of thread in
> us default=1000\n"
>               "-I       --irqsoff         Irqsoff tracing (used with -b)\n"
>               "-l LOOPS --loops=LOOPS     number of loops:
> default=0(endless)\n"
> @@ -710,8 +711,6 @@ static void display_help(void)
>               "-D       --duration=t      specify a length for the test run\n"
>               "                           default is in seconds, but
> 'm', 'h', or 'd' maybe added\n"
>               "                           to modify value to minutes,
> hours or days\n"
> -              "-h       --histogram=US    dump a latency histogram to
> stdout after the run\n"
> -              "                           US is the max time to be be
> tracked in microseconds\n"
>                );
>        exit(0);
>  }
>
>
> --
> Regards,
> GeunSik Lim
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>



-- 
Regards,
GeunSik Lim
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[-- Attachment #2: add-wakeup-wakeuprt-tracing-and-misc-bugfix.patch --]
[-- Type: application/octet-stream, Size: 6055 bytes --]

Author: GeunSik Lim <leemgs1@gmail.com>
Date:   Mon Apr 6 17:15:09 2009 +0900

    Support wakeup tracing and misc bugfix.
    
    This is patch to support wakeup & wakeup_rt tracing at the argument of cyclictest additionally.
    Currently, cyclictest(ver 0.34) support three tracing like PREEMPTOFF, IRQSOFF, PREEMPTIRQSOFF.
    I want you to append WAKEUP & WAKEUP_RT tracing at the current cyclictest version.
    
    and, cyclictest ver 0.34 have a misc bug about context switching and event option.
    for example,
    we can not use  "-C" option without --context option.
    we can not use  "-E" option without --event option.
    
    Signed-off-by: GeunSik Lim <leemgs1@gmail.com>
    Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
    Signed-off-by: Clark Williams <williams@redhat.com>

diff --git a/src/cyclictest/cyclictest.c b/src/cyclictest/cyclictest.c
index cedc36f..746b44f 100644
--- a/src/cyclictest/cyclictest.c
+++ b/src/cyclictest/cyclictest.c
@@ -93,6 +93,8 @@ enum {
        IRQSOFF,
        PREEMPTOFF,
        IRQPREEMPTOFF,
+       WAKEUP,
+       WAKEUPRT,
 };
 
 #define HIST_MAX               1000000
@@ -415,6 +417,12 @@ static void setup_tracer(void)
                case CTXTSWITCH:
                        ret = settracer("sched_switch");
                        break;
+               case WAKEUP:
+                       ret = settracer("wakeup");
+                       break;
+               case WAKEUPRT:
+                       ret = settracer("wakeup_rt");
+                       break;
                default:
                        printf("cyclictest: unknown tracer!\n");
                        ret = 0;
@@ -686,6 +694,9 @@ static void display_help(void)
               "                           1 = CLOCK_REALTIME\n"
               "-C       --context         context switch tracing (used with -b)\n"
               "-d DIST  --distance=DIST   distance of thread intervals in us default=500\n"
+              "-D       --duration=t      specify a length for the test run\n"
+              "                           default is in seconds, but 'm', 'h', or 'd' maybe added\n"
+              "                           to modify value to minutes, hours or days\n"
               "-E       --event           event tracing (used with -b)\n"
               "-f       --ftrace          function trace (when -b is active)\n"
               "-h       --histogram=US    dump a latency histogram to stdout after the run\n"
@@ -708,9 +719,8 @@ static void display_help(void)
               "                           without -t default = 1\n"
               "-v       --verbose         output values on stdout for statistics\n"
               "                           format: n:c:v n=tasknum c=count v=value in us\n"
-              "-D       --duration=t      specify a length for the test run\n"
-              "                           default is in seconds, but 'm', 'h', or 'd' maybe added\n"
-              "                           to modify value to minutes, hours or days\n"
+              "-w       --wakeup          rt task wakeup tracing (used with -b)\n"
+              "-W       --wakeuprt        non_rt task wakeup_rt tracing (used with -b)\n"
                );
        exit(0);
 }
@@ -755,6 +765,7 @@ static void process_options (int argc, char *argv[])
                        {"clock", required_argument, NULL, 'c'},
                        {"context", no_argument, NULL, 'C'},
                        {"distance", required_argument, NULL, 'd'},
+                       {"duration",required_argument, NULL, 'D'},
                        {"event", no_argument, NULL, 'E'},
                        {"ftrace", no_argument, NULL, 'f'},
                        {"histogram", required_argument, NULL, 'h'},
@@ -772,11 +783,12 @@ static void process_options (int argc, char *argv[])
                        {"system", no_argument, NULL, 's'},
                        {"threads", optional_argument, NULL, 't'},
                        {"verbose", no_argument, NULL, 'v'},
-                       {"duration",required_argument, NULL, 'D'},
+                       {"wakeup", no_argument, NULL, 'w'},
+                       {"wakeuprt", no_argument, NULL, 'W'},
                        {"help", no_argument, NULL, '?'},
                        {NULL, 0, NULL, 0}
                };
-               int c = getopt_long (argc, argv, "a::b:Bc:d:fh:i:Il:nNo:p:Pmqrst::vD:",
+               int c = getopt_long (argc, argv, "a::b:BcC:dD:Efh:i:Il:mnNo:p:Pqrst::v:wW:",
                        long_options, &option_index);
                if (c == -1)
                        break;
@@ -797,12 +809,14 @@ static void process_options (int argc, char *argv[])
                case 'c': clocksel = atoi(optarg); break;
                case 'C': tracetype = CTXTSWITCH; break;
                case 'd': distance = atoi(optarg); break;
+               case 'D': duration = parse_time_string(optarg); break;
                case 'E': tracetype = EVENTS; break;
                case 'f': ftrace = 1; break;
                case 'h': histogram = atoi(optarg); break;
                case 'i': interval = atoi(optarg); break;
                case 'I': tracetype = IRQSOFF; break;
                case 'l': max_cycles = atoi(optarg); break;
+               case 'm': lockall = 1; break;
                case 'n': use_nanosleep = MODE_CLOCK_NANOSLEEP; break;
                case 'N': use_nsecs = 1; break;
                case 'o': oscope_reduction = atoi(optarg); break;
@@ -820,9 +834,8 @@ static void process_options (int argc, char *argv[])
                                num_threads = max_cpus;
                        break;
                case 'v': verbose = 1; break;
-               case 'm': lockall = 1; break;
-               case 'D': duration = parse_time_string(optarg);
-                       break;
+               case 'w': tracetype = WAKEUP; break;
+               case 'W': tracetype = WAKEUPRT; break;
                case '?': error = 1; break;
                }
        }


  reply	other threads:[~2009-04-06  8:29 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-04-06  2:56 cyclictest - duplicate explanation of cyclictest option GeunSik Lim
2009-04-06  8:29 ` GeunSik Lim [this message]
2009-04-06 23:15   ` GeunSik Lim

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=49b7c2350904060129r640a3db6r5487e72f2cf6330f@mail.gmail.com \
    --to=leemgs1@gmail.com \
    --cc=linux-rt-users@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --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;
as well as URLs for NNTP newsgroup(s).