Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: "Agovic, Sanimir" <sanimir.agovic@intel.com>
To: 'Yao Qi' <yao@codesourcery.com>
Cc: "gdb-patches@sourceware.org" <gdb-patches@sourceware.org>
Subject: RE: ping: [PATCH 0/4 V3] GDB Performance testing
Date: Thu, 24 Oct 2013 08:50:00 -0000	[thread overview]
Message-ID: <0377C58828D86C4588AEEC42FC3B85A71769EA52@IRSMSX105.ger.corp.intel.com> (raw)
In-Reply-To: <52688937.4030600@codesourcery.com>

Looks good to me, looking forward to your commit.

A minor comment in your perf framework:
diff --git a/gdb/testsuite/gdb.perf/lib/perftest/measure.py b/gdb/testsuite/gdb.perf/lib/perftest/measure.py
new file mode 100644
index 0000000..7478be5
--- /dev/null
+++ b/gdb/testsuite/gdb.perf/lib/perftest/measure.py
@@ -0,0 +1,146 @@
+# Copyright (C) 2013 Free Software Foundation, Inc.
+
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation; either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+import time
+import os
+import gc
+
+class Measure(object):
+    """A class that measure and collect the interesting data for a given testcase.
+
+    An instance of Measure has a collection of measurements, and each
+    of them is to measure a given aspect, such as time and memory.
+    """
+
+    def __init__(self, measurements):
+        """Constructor of measure.
+
+        measurements is a collection of Measurement objects.
+        """
+
+        self.measurements = measurements
+
+    def measure(self, func, id):
+        """Measure the operations done by func with a collection of measurements."""
+        for m in self.measurements:
+            m.start(id)
+
+        # Enable GC, force GC and disable GC before running test in order to reduce
+        # the interference from GC.
+        gc.enable()
+        gc.collect()
+        gc.disable()
+
+        func()
+
+        gc.enable()
I`d rather exclude gc cycles completely from the measurements or include gc collection
_after_ you run your perf test. Otherwise the results will depend on the order the perf
tests are run.

 -Sanimir

> -----Original Message-----
> From: gdb-patches-owner@sourceware.org [mailto:gdb-patches-owner@sourceware.org] On Behalf
> Of Yao Qi
> Sent: Thursday, October 24, 2013 04:43 AM
> To: gdb-patches@sourceware.org
> Subject: ping: [PATCH 0/4 V3] GDB Performance testing
> 
> On 10/16/2013 03:09 PM, Yao Qi wrote:
> > Here is the V3 of GDB performance testing.  The changes in V3 can be
> > found in each patch, and they address all review comments.
> >
> > The basic usages of performance testing are unchanged,
> >
> > $ make check-perf
> > $ make check-perf RUNTESTFLAGS="--target_board=native-gdbserver solib.exp"
> > $ make check-perf RUNTESTFLAGS="solib.exp SOLIB_COUNT=512"
> >
> > We can skip compilation step like,
> >
> > $ make check-perf RUNTESTFLAGS='solib.exp GDB_PERFORMANCE=run'
> >
> > or only compile test cases,
> >
> > $ make check-perf RUNTESTFLAGS='solib.exp GDB_PERFORMANCE=compile'
> >
> > in default, GDB_PERFORMANCE=both.
> >
> > If the perf test case takes much time, we can specify timeout by
> > GDB_PERFORMANCE_TIMEOUT=1000.
> >
> > $ make check-perf RUNTESTFLAGS="--target_board=native-gdbserver
> GDB_PERFORMANCE_TIMEOUT=4000 solib.exp"
> >
> > Looks the perf test framework is in shape, so I add the patch to
> > testsuite/README to describe perf test.
> 
> Ping.  https://sourceware.org/ml/gdb-patches/2013-10/msg00470.html
> 
> --
> Yao (齐尧)
Intel GmbH
Dornacher Strasse 1
85622 Feldkirchen/Muenchen, Deutschland
Sitz der Gesellschaft: Feldkirchen bei Muenchen
Geschaeftsfuehrer: Christian Lamprechter, Hannes Schwaderer, Douglas Lusk
Registergericht: Muenchen HRB 47456
Ust.-IdNr./VAT Registration No.: DE129385895
Citibank Frankfurt a.M. (BLZ 502 109 00) 600119052

  reply	other threads:[~2013-10-24  8:50 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-10-16  7:10 Yao Qi
2013-10-16  7:10 ` [PATCH 1/4] New make target 'check-perf' and new dir gdb.perf Yao Qi
2013-10-25  3:04   ` Doug Evans
2013-10-25 17:34     ` Pedro Alves
2013-10-16  7:10 ` [PATCH 2/4] Perf test framework Yao Qi
2013-10-16  9:05   ` Yao Qi
2013-10-24 14:20     ` Yao Qi
2013-10-25  6:33     ` Doug Evans
2013-10-25  6:32   ` Doug Evans
2013-10-25  8:57     ` Yao Qi
2013-10-16  7:10 ` [PATCH 3/4] Mention perf test in testsuite/README Yao Qi
2013-10-25  6:43   ` Doug Evans
2013-10-16  7:10 ` [PATCH 4/4] Test on solib load and unload Yao Qi
2013-10-25  7:17   ` Doug Evans
2013-10-24  2:44 ` ping: [PATCH 0/4 V3] GDB Performance testing Yao Qi
2013-10-24  8:50   ` Agovic, Sanimir [this message]
2013-10-24 14:13     ` Yao Qi
2013-10-25  2:59 ` Doug Evans

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=0377C58828D86C4588AEEC42FC3B85A71769EA52@IRSMSX105.ger.corp.intel.com \
    --to=sanimir.agovic@intel.com \
    --cc=gdb-patches@sourceware.org \
    --cc=yao@codesourcery.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