rpms/rteval
7
0
Fork 0
Code Issues Pull Requests Releases Activity

import rteval-3.3-4.el9_0

Browse Source
This commit is contained in:
CentOS Sources 2022-05-17 04:36:43 -04:00 committed by Stepan Oksanichenko
commit f954b045d0
9 changed files with 983 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
.gitignore vendored Normal file
View File

@ -0,0 +1 @@
SOURCES/rteval-3.3.tar.xz

1
.rteval.metadata Normal file
View File

@ -0,0 +1 @@
557793c55592bfb3dc0e858274895af4347a2fe9 SOURCES/rteval-3.3.tar.xz

51
SOURCES/rteval-Don-t-restrict-measurement-threads-to-inherit.patch Normal file
View File

@ -0,0 +1,51 @@
From 75667f4b92e3120bced7ae5464d97941edffd802 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Mon, 7 Feb 2022 16:09:13 -0500
Subject: [PATCH] rteval: Don't restrict measurement threads to inherited
cpumask
This patch reverses be811d28a471cfcaf7960289e00960e7f5b18f03
This patch syncs rteval to 795433f50f20ef7598db5cfe991b5386e4080d41 in
rt-tests
If the user does not specify a cpumask, rteval will still run on the
cpus according to the cpumask inherited from the runtime environment.
However, if the user specifies a cpumask, then it will overwrite the
runtime environment where allowed.
This will allow the user to run measurement threads on isolated cpus.
Signed-off-by: John Kacur <jkacur@redhat.com>
---
rteval/modules/measurement/cyclictest.py | 12 +-----------
1 file changed, 1 insertion(+), 11 deletions(-)
diff --git a/rteval/modules/measurement/cyclictest.py b/rteval/modules/measurement/cyclictest.py
index 5330d1466302..4ca310bf7490 100644
--- a/rteval/modules/measurement/cyclictest.py
+++ b/rteval/modules/measurement/cyclictest.py
@@ -217,19 +217,9 @@ class Cyclictest(rtevalModulePrototype):
else:
self.__cpus = online_cpus()
- # Sort the list of cpus to align with the order reported by
- # cyclictest
+ # Sort the list of cpus to align with the order reported by cyclictest
self.__cpus.sort(key=int)
- # Get the cpuset from the environment
- cpuset = os.sched_getaffinity(0)
-
- # Convert the elements to strings
- cpuset = [str(c) for c in cpuset]
-
- # Only include cpus that are in the cpuset
- self.__cpus = [c for c in self.__cpus if c in cpuset]
-
self.__numcores = len(self.__cpus)
info = cpuinfo()
--
2.34.1

53
SOURCES/rteval-Fix-Popen-for-python3.6-where-text-True-is-no.patch Normal file
View File

@ -0,0 +1,53 @@
From cb8b42e3328f9a06b5a18dda64969f16196b9190 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Tue, 18 Jan 2022 15:03:36 -0500
Subject: [PATCH 1/2] rteval: Fix Popen for python3.6 where text=True is not
available
Fix using text=True in Popen for python-3.6 by using
encoding='utf-8' instead
Signed-off-by: John Kacur <jkacur@redhat.com>
---
rteval/sysinfo/services.py | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/rteval/sysinfo/services.py b/rteval/sysinfo/services.py
index e5de22efc342..c85980e19165 100644
--- a/rteval/sysinfo/services.py
+++ b/rteval/sysinfo/services.py
@@ -62,11 +62,11 @@ class SystemServices:
if not [1 for p in reject if fnmatch.fnmatch(servicename, p)] \
and os.access(service, os.X_OK):
cmd = '%s -qs "\(^\|\W\)status)" %s' % (getcmdpath('grep'), service)
- c = subprocess.Popen(cmd, shell=True, text=True)
+ c = subprocess.Popen(cmd, shell=True, encoding='utf-8')
c.wait()
if c.returncode == 0:
cmd = ['env', '-i', 'LANG="%s"' % os.environ['LANG'], 'PATH="%s"' % os.environ['PATH'], 'TERM="%s"' % os.environ['TERM'], service, 'status']
- c = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
+ c = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, encoding='utf-8')
c.wait()
if c.returncode == 0 and (c.stdout.read() or c.stderr.read()):
ret_services[servicename] = 'running'
@@ -81,7 +81,7 @@ class SystemServices:
ret_services = {}
cmd = '%s list-unit-files -t service --no-legend' % getcmdpath('systemctl')
self.__log(Log.DEBUG, "cmd: %s" % cmd)
- c = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
+ c = subprocess.Popen(cmd, shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE, encoding='utf-8')
for p in c.stdout:
# p are lines like b'servicename.service status'
v = p.strip().split()
@@ -91,7 +91,7 @@ class SystemServices:
def services_get(self):
cmd = [getcmdpath('ps'), '-ocomm=', '1']
- c = subprocess.Popen(cmd, stdout=subprocess.PIPE, text=True)
+ c = subprocess.Popen(cmd, stdout=subprocess.PIPE, encoding='utf-8')
self.__init = c.stdout.read().strip()
if self.__init == 'systemd':
self.__log(Log.DEBUG, "Using systemd to get services status")
--
2.35.1

31
SOURCES/rteval-Fix-allmodconfig-build-on-machines-that-don-t.patch Normal file
View File

@ -0,0 +1,31 @@
From 5931ace285305edbf8c86508f89869c5f3be0bf0 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Wed, 9 Mar 2022 14:03:10 -0500
Subject: [PATCH 2/2] rteval: Fix allmodconfig build on machines that don't
allow sha1 signing
rteval compiles the kernel using allmodconfig as a load.
This can be a problem if sha1 signing of modules is disabled.
Fix this by signing with sha512
Suggested-by: Juri Lelli <juri.lelli@redhat.com>
Signed-off-by: John Kacur <jkacur@redhat.com>
---
rteval/modules/loads/kcompile.py | 1 +
1 file changed, 1 insertion(+)
diff --git a/rteval/modules/loads/kcompile.py b/rteval/modules/loads/kcompile.py
index 8f97d38eca56..367f8dc1ca86 100644
--- a/rteval/modules/loads/kcompile.py
+++ b/rteval/modules/loads/kcompile.py
@@ -66,6 +66,7 @@ class KBuildJob:
self.runcmd = f"make O={self.objdir} -C {self.kdir} -j{self.jobs}"
self.cleancmd = f"make O={self.objdir} -C {self.kdir} clean allmodconfig"
+ self.cleancmd += f"&& pushd {self.objdir} && {self.kdir}/scripts/config -d CONFIG_MODULE_SIG_SHA1 -e CONFIG_MODULE_SIG_SHA512 && popd && make O={self.objdir} -C {self.kdir} olddefconfig"
if self.binder:
self.runcmd = self.binder + " " + self.runcmd
self.cleancmd = self.binder + " " + self.cleancmd
--
2.35.1

32
SOURCES/rteval-Fix-test-misses-threshold-assignment.patch Normal file
View File

@ -0,0 +1,32 @@
From aca562064a8d9f5927aa39bc108eac3bf4d0a56b Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Fri, 14 Jan 2022 14:00:54 -0500
Subject: [PATCH] rteval: Fix test misses threshold assignment
Fix case where self.__cfg.threshold has a value but
instead of 'threshold' in self.__cfg,
'breaktrace' in self.__cfg.
by just checking whether self.__cfg.threshold has a value
if self._cfg.breaktrace does not have a value
Signed-off-by: John Kacur <jkacur@redhat.com>
---
rteval/modules/measurement/cyclictest.py | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/rteval/modules/measurement/cyclictest.py b/rteval/modules/measurement/cyclictest.py
index cc74b467913d..af8adeeeb402 100644
--- a/rteval/modules/measurement/cyclictest.py
+++ b/rteval/modules/measurement/cyclictest.py
@@ -299,7 +299,7 @@ class Cyclictest(rtevalModulePrototype):
if 'breaktrace' in self.__cfg and self.__cfg.breaktrace:
self.__cmd.append("-b%d" % int(self.__cfg.breaktrace))
self.__cmd.append("--tracemark")
- elif 'threshold' in self.__cfg and self.__cfg.threshold:
+ elif self.__cfg.threshold:
self.__cmd.append("-b%d" % int(self.__cfg.threshold))
# Buffer for cyclictest data written to stdout
--
2.31.1

34
SOURCES/rteval-If-the-user-doesn-t-specify-a-cpumask-use-the.patch Normal file
View File

@ -0,0 +1,34 @@
From 14b0cff41f13602579fd708748eb8a2ad93db85b Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Mon, 14 Feb 2022 22:09:21 -0500
Subject: [PATCH] rteval: If the user doesn't specify a cpumask, use the
inherited one
If the user doesn't specify a cpumask, then we need to use the inherited
one to match what cyclictest currently does
Signed-off-by: John Kacur <jkacur@redhat.com>
---
rteval/modules/measurement/cyclictest.py | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/rteval/modules/measurement/cyclictest.py b/rteval/modules/measurement/cyclictest.py
index 4ca310bf7490..8fdd5341794a 100644
--- a/rteval/modules/measurement/cyclictest.py
+++ b/rteval/modules/measurement/cyclictest.py
@@ -216,6 +216,12 @@ class Cyclictest(rtevalModulePrototype):
self.__sparse = True
else:
self.__cpus = online_cpus()
+ # Get the cpuset from the environment
+ cpuset = os.sched_getaffinity(0)
+ # Convert the elements to strings
+ cpuset = [str(c) for c in cpuset]
+ # Only include cpus that are in the cpuset
+ self.__cpus = [c for c in self.__cpus if c in cpuset]
# Sort the list of cpus to align with the order reported by cyclictest
self.__cpus.sort(key=int)
--
2.34.1

60
SOURCES/rteval-Increase-default-buckets-from-2000-to-3500.patch Normal file
View File

@ -0,0 +1,60 @@
From 9dff629f186313beebb96594d236dd9268bef1b1 Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Wed, 26 Jan 2022 10:06:33 -0500
Subject: [PATCH] rteval: Increase default buckets from 2000 to 3500
Increase the default buckets from 2000 to 3500
With commit 0292c8963611 we skip statistics reporting if we run out of
buckets. Increase the default number of buckets to make this less likely
to occur.
Signed-off-by: John Kacur <jkacur@redhat.com>
---
doc/rteval.8 | 2 +-
rteval/modules/measurement/cyclictest.py | 2 +-
rteval/rteval.conf | 2 +-
3 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/doc/rteval.8 b/doc/rteval.8
index fa24509ce7e5..25dcfcc298e7 100644
--- a/doc/rteval.8
+++ b/doc/rteval.8
@@ -122,7 +122,7 @@ Measurement thread interval in microseconds (default: 100)
Interval increment in microseconds (default: 0)
.TP
.B \-\-cyclictest-buckets=NBUCKETS
-Number of 1 microsecond histogram buckets (default: 2000)
+Number of 1 microsecond histogram buckets (default: 3500)
.TP
.B \-\-hackbench-jobspercore=N
Number of jobs per online-core for hackbench load
diff --git a/rteval/modules/measurement/cyclictest.py b/rteval/modules/measurement/cyclictest.py
index af8adeeeb402..5330d1466302 100644
--- a/rteval/modules/measurement/cyclictest.py
+++ b/rteval/modules/measurement/cyclictest.py
@@ -450,7 +450,7 @@ def ModuleParameters():
"default": 100,
"metavar": "INTV_US"},
"buckets": {"descr": "Histogram width",
- "default": 2000,
+ "default": 3500,
"metavar": "NUM"},
"priority": {"descr": "Run cyclictest with the given priority",
"default": 95,
diff --git a/rteval/rteval.conf b/rteval/rteval.conf
index 7ce6ef0b5acc..aa0a53bcfc63 100644
--- a/rteval/rteval.conf
+++ b/rteval/rteval.conf
@@ -6,7 +6,7 @@ duration: 60.0
report_interval: 600
[cyclictest]
-buckets: 2000
+buckets: 3500
interval: 100
distance: 0
priority: 95
--
2.34.1

720
SPECS/rteval.spec Normal file
View File

@ -0,0 +1,720 @@
Name: rteval
Version: 3.3
Release: 4%{?dist}
Summary: Utility to evaluate system suitability for RT Linux
Group: Development/Tools
License: GPLv2
URL: https://git.kernel.org/pub/scm/utils/rteval/rteval.git
Source0: https://www.kernel.org/pub/linux/utils/%{name}/%{name}-%{version}.tar.xz
BuildRoot: %{_tmppath}/%{name}-%{version}-%{release}-root-%(%{__id_u} -n)
BuildRequires: python3-devel
Requires: python3-ethtool
Requires: python3-lxml
Requires: python3-dmidecode >= 3.10
Requires: realtime-tests >= 2.1-1
Requires: rteval-loads >= 1.4-14
Requires: sysstat
Requires: xz bzip2 tar gzip m4 make gawk
Requires: kernel-headers
Requires: sos
Requires: numactl
Requires: gcc binutils gcc-c++ flex bison bc make
Requires: elfutils elfutils-libelf-devel
Requires: openssl openssl-devel
Requires: stress-ng
Requires: perl-interpreter, perl-devel, perl-generators
Requires: libmpc, libmpc-devel
BuildArch: noarch
#Patches
Patch1: rteval-Fix-test-misses-threshold-assignment.patch
Patch2: rteval-Increase-default-buckets-from-2000-to-3500.patch
Patch3: rteval-Don-t-restrict-measurement-threads-to-inherit.patch
Patch4: rteval-If-the-user-doesn-t-specify-a-cpumask-use-the.patch
Patch5: rteval-Fix-Popen-for-python3.6-where-text-True-is-no.patch
Patch6: rteval-Fix-allmodconfig-build-on-machines-that-don-t.patch
%description
The rteval script is a utility for measuring various aspects of
realtime behavior on a system under load. The script unpacks the
kernel source, and then goes into a loop, running hackbench and
compiling a kernel tree. During that loop the cyclictest program
is run to measure event response time. After the run time completes,
a statistical analysis of the event response times is done and printed
to the screen.
%prep
%setup -q
%patch1 -p1
%patch2 -p1
%patch3 -p1
%patch4 -p1
%patch5 -p1
%patch6 -p1
%build
%{__python3} setup.py build
%install
%{__python3} setup.py install --root=$RPM_BUILD_ROOT
%clean
rm -rf $RPM_BUILD_ROOT
%files
%defattr(-,root,root,-)
%{python3_sitelib}/*.egg-info
%doc README doc/rteval.txt
%license COPYING
%dir %{_datadir}/%{name}
%{python3_sitelib}/rteval
%{_mandir}/man8/rteval.8.gz
%config(noreplace) %{_sysconfdir}/rteval.conf
%{_datadir}/%{name}/rteval_*.xsl
%{_bindir}/rteval
%{python3_sitelib}/rteval/__pycache__/*
%changelog
* Thu Mar 10 2022 John Kacur <jkacur@redhat.com> - 3.3-4
- Fix Popen for python3.6
- Fix allmodconfig when SHA1 signing of modules is not available
Resolves: rhbz#2062457
* Tue Feb 15 2022 John Kacur <jkacur@redhat.com> - 3.3-3
- Use inherited cpumask if user doesn't specify a cpumask
Resolves: rhbz#1983783
* Mon Feb 07 2022 John Kacur <jkacur@redhat.com> - 3.3-2
- Don't restrict threads to cpumask in environment if user specifies cpumask
Resolves: rhbz#1983783
* Wed Jan 26 2022 John Kacur <jkacur@redhat.com> - 3.3-1
- Rebase to upstream rteval-3.3
- Fix case where the threshold assignment is not properly parsed
- Increase the default number of buckets from 2000 to 3500
Resolves: rhbz#2012294
* Fri Jan 14 2022 John Kacur <jkacur@redhat.com> - 3.2-10
- Fix test missing threshold assignment
Resolves: rhbz#1995195
* Wed Jan 12 2022 John Kacur <jkacur@redhat.com> - 3.2-9
- Do not pass obsolete notrace option to cyclictest
- Parse maximum latency even if outside configured buckets
- Sort the list of cpus
- Skip statistics generation if max latency outside of configured buckets
- Add --cyclictest-threshold=USEC feature
- Add libmpc and libmpc-devel to the Requires
Resolves: rhbz#1995195
* Thu Nov 04 2021 John Kacur <jkacur@redhat.com> - 3.2-8
- allow hackbench to run with warning on low mem
- clean-ups to hackbench.py
- make donotrun work correctly in load modules
- Add the idea of an exclusive load module and make stress-ng one
Resolves: rhbz#2007022
* Tue Aug 10 2021 John Kacur <jkacur@redhat.com> - 3.2-7
- Add perl, gcc-c++ and a few other utilities for kernel compilation
- Reorganize the Requires a little for clarity
Resolves: rhbz#1987037
* Tue Aug 10 2021 Mohan Boddu <mboddu@redhat.com> - 3.2-6
- Rebuilt for IMA sigs, glibc 2.34, aarch64 flags
Related: rhbz#1991688
* Fri Jul 16 2021 John Kacur <jkacur@redhat.com> - 3.2-5
- Make changes to use linux-5.13.2 kernel
- specfile changes to undo some inadvertent changes from e083e0224c00d984e9e9e7
Resolves: rhbz#1982810
* Thu Jul 15 2021 John Kacur <jkacur@redhat.com> - 3.2-4
- Fix Requires to require realtime-tests not rt-tests
Resolves: rhbz#1981875
* Thu Jul 15 2021 John Kacur <jkacur@redhat.com> - 3.2-3
- Add stress-ng as a Requires in the specfile
Resolves: rhbz#1981875
* Thu Jul 15 2021 John Kacur <jkacur@redhat.com> - 3.2-2
- Restrict measurement threads to cpus in the cpumask
Resolves: rhbz#1982707
* Fri Jul 09 2021 John Kacur <jkacur@redhat.com> - 3.2-1
- Rebase to latest upstream rteval-3.2
- Remove patches that are included in the rebase
Resolves: rhbz#1890561
* Wed Jan 13 2021 John Kacur <jkacur@redhat.com> - 3.1-4
- Remove python-schedutils requirement
- Fix spec file to make Obsoletes rteval-common cover all necessary releases
Resolves: rhbz#1869340
* Wed Jan 13 2021 John Kacur <jkacur@redhat.com> - 3.1-3
- Remove a stray line in the changelog
Resolves: rhbz#1890555
* Wed Jan 13 2021 John Kacur <jkacur@redhat.com> - 3.1-2
- Fix incorrect import in rteval/sysinfo/__init__.py
Resolves: rhbz#1890555
* Mon Jan 11 2021 John Kacur <jkacur@redhat.com> - 3.1-1
- Upgrade to rteval-3.1
- Don't create a separate rteval-common package anymore
Resolves: rhbz#1890555
* Thu Aug 27 2020 John Kacur <jkacur@redhat.com> - 3.0-13
- Parse cpuinfo correctly when the model name has a colon in it.
Resolves: rhbz#1873120
* Thu Jul 23 2020 John Kacur <jkacur@redhat.com> - 3.0-12
- Use linux-5.7 in kcompile
Resolves: rhbz#1859762
* Thu Jun 25 2020 John Kacur <jkacur@redhat.com> - 3.0-11
- Make sure "make" is available for the kcompile module
Resolves: rhbz#1850924
* Wed Jun 24 2020 John Kacur <jkacur@redhat.com> - 3.0-10
- Ensure that a recent rteval-loads with stress-ng is required
Resolves: rhbz#1847233
* Wed Jun 24 2020 John Kacur <jkacur@redhat.com> - 3.0-9
- Ensure the stressng command line is displayed in the final report
Resolves: rhbz#1850151
* Mon Jun 15 2020 John Kacur <jkacur@redhat.com> - 3.0-8
- Add the stress-ng load module
Resolves: rhbz#1816360
* Mon May 04 2020 John Kacur <jkacur@redhat.com> - 3.0-7
- Make sure openssl is available so that rteval can compile the kernel
Resolves: rhbz#1831272
* Tue Dec 10 2019 John Kacur <jkacur@redhat.com> - 3.0-6
- Iterate over nodes and not sysTop
- Explictly add a few more software requires for compiling the kernel
Resolves: rhbz#1755603
* Tue Dec 03 2019 John Kacur <jkacur@redhat.com> - 3.0-5
- Explicitly add some software requires for compiling the kernel
Resolves: rhbz#1766879
* Mon Dec 02 2019 John Kacur <jkacur@redhat.com> - 3.0-4
- In hackbench.py node in args to Popen must be a string
Resolves: rhbz#1777048
* Tue Nov 19 2019 John Kacur <jkacur@redhat.com> - 3.0-3
- Don't assume cpu0 cannot be offlined, test for it
- Drop patches that are no longer in the spec file
Resolves: rhbz#1773792
* Mon Nov 18 2019 John Kacur <jkacur@redhat.com> - 3.0-2
- Check whether a cpu is online before adding to a list
- Change hackbench to use the systopology interface for online cpus
Resolves: rhbz#1715081
* Fri Nov 15 2019 John Kacur <jkacur@redhat.com> - 3.0-1
- Sync rt-tests and rteval-loads versions in the specfile
- Upgrade to upstream rteval-3.0
Resolves: rhbz#1748955
* Fri Nov 08 2019 John Kacur <jkacur@redhat.com> - 2.14-27
- Update kcompile sources to linux-5.1
Resolves: rhbz#1770215
* Fri Nov 08 2019 John Kacur <jkacur@redhat.com> - 2.14-26
- Fix number of hackbench jobs wrt number of CPUs
- Don't run on nodes with no CPUs available
Resolves: rhbz#1770211
* Tue Apr 02 2019 Clark Williams <williams@redhat.com> - 2.14.25
- fix incorrect test logic in gating tests
Resolves: rhbz#1682426
* Tue Apr 02 2019 Clark Williams <williams@redhat.com> - 2.14.24
- add rteval-loads dependency to gating
- added second test (short_run) to gating
Resolves: rhbz#1682426
* Mon Apr 01 2019 Clark Williams <williams@redhat.com> - 2.14.23
- add missing gating.yaml
Resolves: rhbz#1682426
* Mon Apr 01 2019 Clark Williams <williams@redhat.com> - 2.14.22
- checkin OSCI gating framework
Resolves: rhbz#1682426
* Mon Dec 17 2018 John Kacur <jkacur@redhat.com> - 2.14-21
- Fix typo in debug output
Resolves: rhbz#1659974
* Tue Oct 16 2018 John Kacur <jkacur@redhat.com> - 2.14-20
- Disable options for the remote xmlrpc server, not currently supported
Resolves: rhbz#1628322
* Sat Oct 13 2018 John Kacur <jkacur@redhat.com> - 2.14-19
- Fix Requires for python3
Resolves: rhbz#1638135
* Fri Oct 12 2018 John Kacur <jkacur@redhat.com> - 2.14-18
- Fix time format in report
Resolves: rhbz#1630733
* Fri Sep 28 2018 John Kacur <jkacur@redhat.com> - 2.14-17
- Change python3 to platform-python
Resolves: rhbz#1633619
* Fri Aug 10 2018 John Kacur <jkacur@redhat.com> - 2.14-16
- remove unnecssary encode that is causing problems
Resolves: rhbz#1614384
* Tue Aug 07 2018 John Kacur <jkacur@redhat.com> - 2.14-15
- tar is required in kcompile.py. Make it a Require in the specfile
Resolves: rhbz#1612992
* Fri Aug 03 2018 John Kacur <jkacur@redhat.com> - 2.14-14
- fix python3 division of integers
Resolves: rhbz#1611813
* Fri Aug 03 2018 John Kacur <jkacur@redhat.com> - 2.14-13
-fix rtevalclient import
Resolves: rhbz#1608464
* Sat Jun 23 2018 John Kacur <jkacur@redhat.com> - 2.14-12
- More python3 changes
- Changes for the new version of rt-tests that automates --numa
Resolves: rhbz#1594287
* Tue Jun 12 2018 John Kacur jkacur@redhat.com> - 2.14-11
- More specfile changes for python3 build
Resolves: rhbz#1518699
* Wed May 30 2018 John Kacur <jkacur@redhat.com> - 2.14-10
- Chnages for a python3 build
Resolves: rhbz#1518699
* Fri Oct 27 2017 John Kacur <jkacur@redhat.com> - 2.14-9
- Remove redundant files for clarity.
Resolves: rhbz1504162
* Fri Oct 27 2017 John Kacur <jkacur@redhat.com> - 2.14-8
- Don't fail if we don't know the init system
Resolves: rhbz1504168
* Fri Oct 20 2017 John Kacur <jkacur@redhat.com> - 2.14-7
- Remove underscore from sysread function in systopology.py
Resolves: rhbz1504164
* Fri Oct 20 2017 John Kacur <jkacur@redhat.com> - 2.14-6
- Improve error handling if cyclictest fails to run
Resolves: rhbz1504159
* Fri Oct 20 2017 John Kacur <jkacur@redhat.com> - 2.14-5
- Remove trace-cmd from Requires, since it is not needed to run rteval
Resolves: rhbz1504173
* Mon Oct 16 2017 John Kacur <jkacur@redhat.com> - 2.14-4
- Don't sleep if hackbench fails to launch due to out-of-memory
- Instead, exit gracefully
Resolves: rhbz1380144
* Wed Oct 11 2017 John Kacur <jkacur@redhat.com> - 2.14-3
- Add sos as a requires since this package is needed to run sosreport
Resolves: rhbz1500722
* Wed Oct 11 2017 John Kacur <jkacur@redhat.com> - 2.14-2
- Add the contents of the kernel boot command line to the summary report
Resolves: rhbz1452788
* Thu Mar 16 2017 Clark Williams <williams@redhat.com> - 2.14-1
- removed leftover import of systopology from sysinfo
* Wed Mar 15 2017 Clark Williams <williams@redhat.com> - 2.13-2
- Updated specfile to correct version and bz [1382155]
* Tue Sep 20 2016 Clark Williams <williams@rehdat.com> - 2.12-1
- handle empty environment variables SUDO_USER and USER [1312057]
* Tue Aug 30 2016 Clark Williams <williams@rehdat.com> - 2.11-1
- make sure we return non-zero for early exit from tests
* Wed Aug 3 2016 Clark Williams <williams@rehdat.com> - 2.10-1
- bumped version for RHEL 7.3 release
* Mon May 9 2016 Clark Williams <williams@redhat.com> - 2.9.1
- default cpulist for modules if only one specified [1333831]
* Tue Apr 26 2016 Clark Williams <williams@redhat.com> - 2.8.1
- add the --version option to print the rteval version
- made the --cyclictest-breaktrace option work properly [1209986]
* Fri Apr 1 2016 Clark Williams <williams@redhat.com> - 2.7.1
- treat SIGINT and SIGTERM as valid end-of-run events [1278757]
- added cpulist options to man page
* Thu Feb 11 2016 Clark Williams <williams@redhat.com> - 2.6.1
- update to make --loads-cpulist and --measurement-cpulist work [1306437]
* Thu Dec 10 2015 Clark Williams <williams@refhat.com> - 2.5-1
- stop using old numactl --cpubind argument
* Wed Dec 9 2015 Clark Williams <williams@refhat.com> - 2.4.2
- added Require of package numactl
* Tue Nov 17 2015 Clark Williams <williams@refhat.com> - 2.4.1
- rework hackbench load to not generate cross-node traffic [1282826]
* Wed Aug 12 2015 Clark Williams <williams@redhat.com> - 2.3-1
- comment out HWLatDetect module from default config [1245699]
* Wed Jun 10 2015 Clark Williams <williams@redhat.com> - 2.2-1
- add --loads-cpulist and --measurement-cpulist to allow cpu placement [1230401]
* Thu Apr 23 2015 Luis Claudio R. Goncalves <lgoncalv@redhat.com> - 2.1-8
- load default configs when no config file is specified (Jiri kastner) [1212452]
* Wed Jan 14 2015 Clark Williams <williams@redhat.com> - 2.1-7
- added requires of bzip2 to specfile [1151567]
* Thu Jan 8 2015 Clark Williams <williams@redhat.com> - 2.1-6
- cleaned up product documentation [1173315]
* Mon Nov 10 2014 Luis Claudio R. Goncalves <lgoncalv@redhat.com> - 2.1-5
- rebuild for RHEL-7.1 (1151567)
* Thu Mar 27 2014 Clark Williams <williams@redhat.com> - 2.1-4
- cherry-picked old commit to deal with installdir problem
* Wed Mar 26 2014 Clark Williams <williams@redhat.com> - 2.1-3
- added sysstat requires to specfile
* Tue Mar 12 2013 David Sommerseth <davids@redhat.com> - 2.1-2
- Migrated from libxslt-python to python-lxml
* Fri Jan 18 2013 David Sommerseth <davids@redhat.com> - 2.1-1
- Made some log lines clearer
- cyclictest: Added --cyclictest-breaktrace feature
- cyclictest: Removed --cyclictest-distance option
- cyclictest: Use a tempfile buffer for cyclictest's stdout data
- cyclictest: Report if breaktrace was triggered
- cyclictest: Make the unit test work again
- cyclictest: Only log and show statistic data when samples are collected
- Copyright updates
* Thu Jan 17 2013 David Sommerseth <davids@redhat.com> - 2.0.1-1
- Fix up type casting in the core module code
- hwlatdetect: Add some more useful debug info
- Reworked the run logic for modules - allow them to flag they won't run
- Fixed a few log messages in load modules
- Add a 30 seconds sleep before unleashing the measurement threads
* Thu Jan 10 2013 David Sommerseth <davids@redhat.com> - 2.0-3
- Separate out RTEVAL_VERSION into rteval.version, to avoid
massive BuildRequirements
* Fri Dec 21 2012 David Sommerseth <davids@redhat.com> - 2.0-2
- Split out common files into rteval-common
* Fri Dec 21 2012 David Sommerseth <davids@redhat.com> - 2.0-1
- Updated to rteval v2.0 and reworked spec file to use setup.py directly
* Tue Oct 23 2012 Clark Williams <williams@redhat.com> - 1.36-1
- deal with system not having dmidecode python module
- make sure to cast priority parameter to int
- from Raphaël Beamonte <raphael.beamonte@gmail.com>:
- Rewrite of the get_kthreads method to make it cross-distribution
- Adds getcmdpath method to use which to locate the used commands
- Rewrite of the get_services method to make it cross-distribution
* Mon Apr 2 2012 Clark Williams <williams@redhat.com> - 1.35-1
- fix thinko where SIGINT and SIGTERM handlers were commented out
* Thu Jan 12 2012 Clark Williams <williams@redhat.com> - 1.34-1
- fix missing config merge in rteval.py to pass parameters
down to cyclictest
- modify hackbench to use helper function to start process
* Sat May 14 2011 Clark Williams <williams@redhat.com> - 1.33-1
- modify hackbench cutoff to be 0.75GB/core
* Mon Aug 23 2010 Clark Williams <williams@redhat.com> - 1.32-1
- update docs
- refactor some RTEval methods to utility functions
- modify hackbench.py not to run under low memory conditions
- clean up XML generation to deal with new hackbench code
- clean up XSL code to deal with new XML 'run' attribute
- from David Sommerseth <davids@redhat.com>:
- improve CPU socket counting logic
- delay log directory creation until actually needed
- from Gowrishankar <gowrishankar.m@in.ibm.com>:
- check if the core id really exists (multithreading fix)
* Mon Jul 26 2010 Clark Williams <williams@redhat.com> - 1.31-1
- from David Sommerseth <davids@redhat.com>:
- Updated hackbench implementation to avoid overusing resources
- Don't show NUMA node information if it's missing in the summary.xml
- Show CPU cores properly
* Wed Jul 21 2010 Clark Williams <williams@redhat.com> - 1.30-1
- added code to hackbench to try to detect and ease memory pressure
* Fri Jul 16 2010 Clark Williams <williams@redhat.com> - 1.29-1
- fixed incorrect type value in kcompile.py
* Fri Jul 16 2010 Clark Williams <williams@redhat.com> - 1.28-1
- added logic to loads to adjust number of jobs based on ratio
of memory per core
* Wed Jul 14 2010 Clark Williams <williams@redhat.com> - 1.27-1
- modified hackbench to go back to using threads rather than
processes for units of work
- added memory size, number of numa nodes and run duration to the
parameter dictionary passed to all loads and cyclictest
* Tue Jul 13 2010 Clark Williams <williams@redhat.com> - 1.26-1
- modified hackbench parameters to reduce memory consumption
* Mon Jul 12 2010 Clark Williams <williams@redhat.com> - 1.25-1
- fixed cyclictest bug that caused everything to be uniprocessor
- updated source copyrights to 2010
* Fri Jul 9 2010 Clark Williams <williams@redhat.com> - 1.24-1
- modified hackbench arguments and added new parameters for
hackbench in rteval.conf
* Thu Jul 8 2010 Clark Williams <williams@redhat.com> - 1.23-1
- version bump to deal with out-of-sync cvs issue
* Thu Jul 8 2010 Clark Williams <williams@redhat.com> - 1.22-1
- merged David Sommerseth <davids@redhat.com> changes to use
hackbench from rt-tests packages rather than carry local copy
- converted all loads and cyclictest to pass __init__ parameters
in a dictionary rather than as discrete parameters
- added logging for load output
* Tue Apr 13 2010 Clark Williams <williams@redhat.com> - 1.21-1
- from Luis Claudio Goncalves <lgoncalv@redhat.com>:
- remove unecessary wait() call in cyclictest.py
- close /dev/null after using it
- call subprocess.wait() when needed
- remove delayloop code in hackbench.py
- from David Sommerseth <davids@redhat.com>:
- add SIGINT handler
- handle non-root user case
- process DMI warnings before command line arguments
- added --annotate feature to rteval
- updates to xmlrpc code
* Tue Apr 6 2010 Clark Williams <williams@redhat.com> - 1.20-1
- code fixes from Luis Claudio Goncalves <lgoncalv@redhat.com>
- from David Sommerseth <davids@redhat.com>:
- xmlrpc server updates
- cputopology.py for recording topology in xml
- added NUMA node recording for run data
- rpmlint fixes
- added start of rteval whitepaper in docs dir
* Tue Mar 16 2010 Clark Williams <williams@redhat.com> - 1.19-1
- add ability for --summarize to read tarfiles
- from David Sommerseth <davids@redhat.com>
- gather info about loaded kernel modules for XML file
- added child tracking to hackbench to prevent zombies
* Tue Feb 16 2010 Clark Williams <williams@redhat.com> - 1.18-1
- fix usage of python 2.6 features on RHEL5 (python 2.4)
* Tue Feb 16 2010 Clark Williams <williams@redhat.com> - 1.17-1
- added logic to filter non-printables from service status output
so that we have legal XML output
- added logic to hackbench.py to cleanup properly at the end
of the test
* Thu Feb 11 2010 Clark Williams <williams@redhat.com> - 1.16-1
- fix errors in show_remaining_time() introduced because
time values are floats rather than ints
* Thu Feb 11 2010 Clark Williams <williams@redhat.com> - 1.15-1
- added logic to use --numa and --smp options of new cyclictest
- added countdown report for time remaining in a run
* Tue Feb 9 2010 Clark Williams <williams@redhat.com> - 1.14-1
- David Sommerseth <davids@redhat.com>:
merged XMLReport() changes for hwcert suite
* Tue Dec 22 2009 Clark Williams <williams@redhat.com> - 1.13-1
- added cyclictest default initializers
- added sanity checks to statistics reduction code
- updated release checklist to include origin push
- updated Makefile clean and help targets
- davids updates (mainly for v7 integration):
- Add explicit sys.path directory to the python sitelib+
'/rteval'
- Send program arguments via RtEval() constructor
- Added more DMI data into the summary.xml report
- Fixed issue with not including all devices in the
OnBoardDeviceInfo tag
* Thu Dec 3 2009 David Sommerseth <davids@redhat.com> - 1.12-2
- fixed Makefile and specfile to include and install the
rteval/rteval_histogram_raw.py source file for gaining
raw access to histogram data
- Removed xmlrpc package during merge against master_ipv4 branch
* Wed Nov 25 2009 Clark Williams <williams@redhat.com> - 1.12-1
- fix incorrect reporting of measurement thread priorities
* Mon Nov 16 2009 Clark Williams <williams@redhat.com> - 1.11-5
- ensure that no double-slashes ("//") appear in the symlink
path for /usr/bin/rteval (problem with rpmdiff)
* Tue Nov 10 2009 Clark Williams <williams@redhat.com> - 1.11-4
- changed symlink back to install and tracked by %%files
* Mon Nov 9 2009 Clark Williams <williams@redhat.com> - 1.11-3
- changed symlink generation from %%post to %%posttrans
* Mon Nov 9 2009 Clark Williams <williams@redhat.com> - 1.11-2
- fixed incorrect dependency for libxslt
* Fri Nov 6 2009 Clark Williams <williams@redhat.com> - 1.11-1
- added base OS info to XML file and XSL report
- created new package rteval-loads for the load source code
* Wed Nov 4 2009 Clark Williams <williams@redhat.com> - 1.10-1
- added config file section for cyclictest and two settable
parameters, buckets and interval
* Thu Oct 29 2009 Clark Williams <williams@redhat.com> - 1.9-1
- merged davids updates:
-H option (raw histogram data)
cleaned up xsl files
fixed cpu sorting
* Mon Oct 26 2009 David Sommerseth <davids@redhat.com> - 1.8-3
- Fixed rpmlint complaints
* Mon Oct 26 2009 David Sommerseth <davids@redhat.com> - 1.8-2
- Added xmlrpc package, containing the XML-RPC mod_python modules
* Tue Oct 20 2009 Clark Williams <williams@redhat.com> - 1.8-1
- split kcompile and hackbench into sub-packages
- reworked Makefile (and specfile) install/uninstall logic
- fixed sysreport incorrect plugin option
- catch failure when running on root-squashed NFS
* Tue Oct 13 2009 Clark Williams <williams@redhat.com> - 1.7-1
- added kthread status to xml file
- merged davids changes for option processing and additions
to xml summary
* Tue Oct 13 2009 Clark Williams <williams@redhat.com> - 1.6-1
- changed stat calculation to loop less
- added methods to grab service and kthread status
* Mon Oct 12 2009 Clark Williams <williams@redhat.com> - 1.5-1
- changed cyclictest to use less memory when doing statisics
calculations
- updated debug output to use module name prefixes
- changed option processing to only process config file once
* Fri Oct 9 2009 Clark Williams <williams@redhat.com> - 1.4-1
- changed cyclictest to use histogram rather than sample array
- calcuated statistics directly from histogram
- changed sample interval to 100us
- added -a (affinity) argument to force cpu affinity for
measurement threads
* Thu Sep 24 2009 David Sommerseth <davids@redhat.com> - 1.3-3
- Cleaned up the spec file and made rpmlint happy
* Wed Sep 23 2009 David Sommerseth <davids@redhat.com> - 1.3-2
- Removed version number from /usr/share/rteval path
* Tue Sep 22 2009 Clark Williams <williams@redhat.com> - 1.3-1
- changes from davids:
* changed report code to sort by processor id
* added report submission retry logic
* added emailer class
* Fri Sep 18 2009 Clark Williams <williams@redhat.com> - 1.2-1
- added config file handling for modifying load behavior and
setting defaults
- added units in report per IBM request
* Wed Aug 26 2009 Clark Williams <williams@redhat.com> - 1.1-2
- missed a version change in rteval/rteval.py
* Wed Aug 26 2009 Clark Williams <williams@redhat.com> - 1.1-1
- modified cyclictest.py to start cyclictest threads with a
'distance' of zero, meaning they all have the same measurement
interval
* Tue Aug 25 2009 Clark Williams <williams@redhat.com> - 1.0-1
- merged davids XMLRPC fixes
- fixed --workdir option
- verion bump to 1.0
* Thu Aug 13 2009 Clark Williams <williams@redhat.com> - 0.9-2
- fixed problem with incorrect version in rteval.py
* Tue Aug 4 2009 Clark Williams <williams@redhat.com> - 0.9-1
- merged dsommers XMLRPC and database changes
- Specify minimum python-dmidecode version, which got native XML support
- Added rteval_dmi.xsl
- Fixed permission issues in /usr/share/rteval-x.xx
* Wed Jul 22 2009 Clark Williams <williams@redhat.com> - 0.8-1
- added code to capture clocksource info
- added code to copy dmesg info to report directory
- added code to display clocksource info in report
- added --summarize option to display summary of existing report
- added helpfile target to Makefile
* Thu Mar 26 2009 Clark Williams <williams@torg> - 0.7-1
- added require for python-schedutils to specfile
- added default for cyclictest output file
- added help parameter to option parser data
- renamed xml output file to summary.xml
- added routine to create tarfile of result files
* Wed Mar 18 2009 Clark Williams <williams@torg> - 0.6-6
- added code to handle binary data coming from DMI tables
* Wed Mar 18 2009 Clark Williams <williams@torg> - 0.6-5
- fixed logic for locating XSL template (williams)
- fixed another stupid typo in specfile (williams)
* Wed Mar 18 2009 Clark Williams <williams@torg> - 0.6-4
- fixed specfile to install rteval_text.xsl in /usr/share directory
* Wed Mar 18 2009 Clark Williams <williams@torg> - 0.6-3
- added Requires for libxslt-python (williams)
- fixed race condition in xmlout constructor/destructor (williams)
* Wed Mar 18 2009 Clark Williams <williams@torg> - 0.6-2
- added Requires for libxslt (williams)
- fixed stupid typo in rteval/rteval.py (williams)
* Wed Mar 18 2009 Clark Williams <williams@torg> - 0.6-1
- added xml output logic (williams, dsommers)
- added xlst template for report generator (dsommers)
- added dmi/smbios output to report (williams)
- added __del__ method to hackbench to cleanup after run (williams)
- modified to always keep run data (williams)
* Fri Feb 20 2009 Clark Williams <williams@torg> - 0.5-1
- fixed tab/space mix problem
- added report path line to report
* Fri Feb 20 2009 Clark Williams <williams@torg> - 0.4-1
- reworked report output
- handle keyboard interrupt better
- removed duration mismatch between rteval and cyclictest
* Mon Feb 2 2009 Clark Williams <williams@torg> - 0.3-1
- initial checkin