public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Pedro Alves <pedro@palves.net>
To: gdb-patches@sourceware.org
Cc: Qingchuan Shi <qingchuan.shi@amd.com>
Subject: [PATCH 2/2] Update gdb performance testsuite to be compatible with Python 3.8
Date: Fri, 18 Jun 2021 22:12:16 +0100	[thread overview]
Message-ID: <20210618211216.544504-3-pedro@palves.net> (raw)
In-Reply-To: <20210618211216.544504-1-pedro@palves.net>

Running "make check-perf" on a system with Python 3.8 (e.g., Ubuntu
20.04) runs into this Python problem:

  Traceback (most recent call last):
    File "<string>", line 1, in <module>
    File "/home/pedro/rocm/gdb/src/gdb/testsuite/gdb.perf/lib/perftest/perftest.py", line 65, in run
      self.execute_test()
    File "<string>", line 35, in execute_test
    File "/home/pedro/rocm/gdb/src/gdb/testsuite/gdb.perf/lib/perftest/measure.py", line 45, in measure
      m.start(id)
    File "/home/pedro/rocm/gdb/src/gdb/testsuite/gdb.perf/lib/perftest/measure.py", line 102, in start
      self.start_time = time.clock()
  AttributeError: module 'time' has no attribute 'clock'
  Error while executing Python code.
  (gdb) FAIL: gdb.perf/single-step.exp: python SingleStep(1000).run()

... many times over.

The problem is that the testsuite is using time.clock(), deprecated in
Python 3.3 and finaly removed in Python 3.8.  The guidelines say to
use time.perf_counter() or time.process_time() instead depending on
requirements.  Looking at the current description of those functions,
at:

   https://docs.python.org/3.10/library/time.html

we have:

   time.perf_counter() → float

       Return the value (in fractional seconds) of a performance
       counter, i.e. a clock with the highest available resolution to
       measure a short duration. It does include time elapsed during
       sleep and is system-wide. (...)

   time.process_time() → float

       Return the value (in fractional seconds) of the sum of the
       system and user CPU time of the current process. It does not
       include time elapsed during sleep. It is process-wide by
       definition. (...)

I'm thinking that it's just best to record both instead of picking
one.  So this patch replaces the MeasurementCpuTime measurement class
with two new classes -- MeasurementPerfCounter and
MeasurementProcessTime.  Correspondingly, this changes the reports in
testsuite/perftest.log -- we have two new "perf_counter" and
"process_time" measurements and the "cpu_time" measurement is gone.  I
don't suppose breaking backward compatibility here is a big problem.
I suspect no one is really tracking long term performance using the
perf testsuite today.  And if they are, it shouldn't be hard to adjust.

For backward compatility, with Python < 3.3, both perf_counter and
process_time use the old time.clock.

gdb/testsuite/ChangeLog:
yyyy-mm-dd  Qingchuan Shi  <qingchuan.shi@amd.com>
	    Pedro Alves  <pedro@palves.net>

	* gdb.perf/lib/perftest/perftest.py: Import sys.
	(time.perf_counter, time.process_time): Map to time.clock on
	Python < 3.3.
	(MeasurementCpuTime): Delete, replaced by...
	(MeasurementPerfCounter, MeasurementProcessTime): .. these two new
	classes.
	* gdb.perf/lib/perftest/perftest.py: Import MeasurementPerfCounter
	and MeasurementProcessTime instead of MeasurementCpuTime.
	(TestCaseWithBasicMeasurements): Use MeasurementPerfCounter and
	MeasurementProcessTime instead of MeasurementCpuTime.

Co-authored-by: Qingchuan Shi <qingchuan.shi@amd.com>

Change-Id: Ia850c05d5ce57d2dada70ba5b0061f566444aa2b
---
 .../gdb.perf/lib/perftest/measure.py          | 50 +++++++++++++------
 .../gdb.perf/lib/perftest/perftest.py         |  6 ++-
 2 files changed, 40 insertions(+), 16 deletions(-)

diff --git a/gdb/testsuite/gdb.perf/lib/perftest/measure.py b/gdb/testsuite/gdb.perf/lib/perftest/measure.py
index 2a20c5eafb2..9810458fdb6 100644
--- a/gdb/testsuite/gdb.perf/lib/perftest/measure.py
+++ b/gdb/testsuite/gdb.perf/lib/perftest/measure.py
@@ -16,6 +16,13 @@
 import time
 import os
 import gc
+import sys
+
+# time.perf_counter() and time.process_time() were added in Python
+# 3.3, time.clock() was removed in Python 3.8.
+if sys.version_info < (3, 3, 0):
+    time.perf_counter = time.clock
+    time.process_time = time.clock
 
 
 class Measure(object):
@@ -85,28 +92,43 @@ class Measurement(object):
         self.result.report(reporter, name + " " + self.name)
 
 
-class MeasurementCpuTime(Measurement):
-    """Measurement on CPU time."""
+class MeasurementPerfCounter(Measurement):
+    """Measurement on performance counter."""
+
+    # Measures time in fractional seconds, using a performance
+    # counter, i.e. a clock with the highest available resolution to
+    # measure a short duration.  It includes time elapsed during sleep
+    # and is system-wide.
+
+    def __init__(self, result):
+        super(MeasurementPerfCounter, self).__init__("perf_counter", result)
+        self.start_time = 0
+
+    def start(self, id):
+        self.start_time = time.perf_counter()
+
+    def stop(self, id):
+        perf_counter = time.perf_counter() - self.start_time
+        self.result.record(id, perf_counter)
+
+
+class MeasurementProcessTime(Measurement):
+    """Measurement on process time."""
 
-    # On UNIX, time.clock() measures the amount of CPU time that has
-    # been used by the current process.  On Windows it will measure
-    # wall-clock seconds elapsed since the first call to the function.
-    # Something other than time.clock() should be used to measure CPU
-    # time on Windows.
+    # Measures the sum of the system and user CPU time of the current
+    # process.  Does not include time elapsed during sleep.  It is
+    # process-wide by definition.
 
     def __init__(self, result):
-        super(MeasurementCpuTime, self).__init__("cpu_time", result)
+        super(MeasurementProcessTime, self).__init__("process_time", result)
         self.start_time = 0
 
     def start(self, id):
-        self.start_time = time.clock()
+        self.start_time = time.process_time()
 
     def stop(self, id):
-        if os.name == "nt":
-            cpu_time = 0
-        else:
-            cpu_time = time.clock() - self.start_time
-        self.result.record(id, cpu_time)
+        process_time = time.process_time() - self.start_time
+        self.result.record(id, process_time)
 
 
 class MeasurementWallTime(Measurement):
diff --git a/gdb/testsuite/gdb.perf/lib/perftest/perftest.py b/gdb/testsuite/gdb.perf/lib/perftest/perftest.py
index 07266413fb0..202ba3f71bc 100644
--- a/gdb/testsuite/gdb.perf/lib/perftest/perftest.py
+++ b/gdb/testsuite/gdb.perf/lib/perftest/perftest.py
@@ -18,7 +18,8 @@ from __future__ import absolute_import
 import perftest.testresult as testresult
 import perftest.reporter as reporter
 from perftest.measure import Measure
-from perftest.measure import MeasurementCpuTime
+from perftest.measure import MeasurementPerfCounter
+from perftest.measure import MeasurementProcessTime
 from perftest.measure import MeasurementWallTime
 from perftest.measure import MeasurementVmSize
 
@@ -72,7 +73,8 @@ class TestCaseWithBasicMeasurements(TestCase):
     def __init__(self, name):
         result_factory = testresult.SingleStatisticResultFactory()
         measurements = [
-            MeasurementCpuTime(result_factory.create_result()),
+            MeasurementPerfCounter(result_factory.create_result()),
+            MeasurementProcessTime(result_factory.create_result()),
             MeasurementWallTime(result_factory.create_result()),
             MeasurementVmSize(result_factory.create_result()),
         ]
-- 
2.26.2


  parent reply	other threads:[~2021-06-18 21:12 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-18 21:12 [PATCH 0/2] Fix a few "make check-perf" problems Pedro Alves
2021-06-18 21:12 ` [PATCH 1/2] gdb.perf/: FAIL on Python errors, avoid "ERROR: internal buffer is full" Pedro Alves
2021-06-18 21:12 ` Pedro Alves [this message]
2021-07-06 11:13 ` [PATCH 0/2] Fix a few "make check-perf" problems Pedro Alves

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=20210618211216.544504-3-pedro@palves.net \
    --to=pedro@palves.net \
    --cc=gdb-patches@sourceware.org \
    --cc=qingchuan.shi@amd.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 read-only IMAP folder(s) and NNTP newsgroup(s).