rpms/rt-tests
7
0
Fork 0
Code Issues Pull Requests Releases Activity

import rt-tests-1.10-3.el8

Browse Source
This commit is contained in:
CentOS Sources 2021-02-20 06:12:32 +00:00 committed by Andrew Lukoshko
parent e27c24c836
commit c2800d82c9
5 changed files with 231 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
SOURCES/oslat-Use-cpuset-size-as-upper-bound.patch Normal file
View File

@ -0,0 +1,38 @@
From 6d5aa2b00d41ecdb6eb1355309737647e177e5e6 Mon Sep 17 00:00:00 2001
From: Daniel Wagner <dwagner@suse.de>
Date: Wed, 10 Feb 2021 17:54:07 +0100
Subject: [PATCH 2/3] oslat: Use cpuset size as upper bound
To assign the threads to the correct CPU we need to use the cpuset
size as upper bound for the loop and not the number of threads.
Fixes: 85b0763dacd9 ("oslat: Use parse_cpumask() from rt-numa.h")
Reported-by: Peter Xu <peterx@redhat.com>
Signed-off-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: John Kacur <jkacur@redhat.com>
---
src/oslat/oslat.c | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/src/oslat/oslat.c b/src/oslat/oslat.c
index 7826c277f26d..2fe550b3ee12 100644
--- a/src/oslat/oslat.c
+++ b/src/oslat/oslat.c
@@ -747,9 +747,12 @@ int main(int argc, char *argv[])
n_cores = numa_bitmask_weight(cpu_set);
TEST(threads = calloc(1, n_cores * sizeof(threads[0])));
- for (i = 0; i < n_cores; ++i)
- if (numa_bitmask_isbitset(cpu_set, i) && move_to_core(i) == 0)
+ for (i = 0; n_cores && i < cpu_set->size; i++) {
+ if (numa_bitmask_isbitset(cpu_set, i) && move_to_core(i) == 0) {
threads[g.n_threads_total++].core_i = i;
+ n_cores--;
+ }
+ }
if (numa_bitmask_isbitset(cpu_set, 0) && g.rtprio)
printf("WARNING: Running SCHED_FIFO workload on CPU 0 may hang the thread\n");
--
2.26.2

73
SOURCES/oslat-allow-scheduling-on-all-possible-cores.patch Normal file
View File

@ -0,0 +1,73 @@
From 29884cff6352856fee9fffecb4a715efd70e08f5 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Thu, 18 Feb 2021 14:27:29 -0500
Subject: [PATCH] oslat: Fix --cpu-list won't allow to schedule on all possible
cores
parse_cpumask() is too strict for oslat, in that use_current_cpuset() will
filter out all the cores that are not allowed for current process to run. This
seems to be unnecessary at least for oslat. For example, the bash process that
runs the oslat program may have a sched affinity of 0-2, however it's still
legal to have it start a oslat thread running on the cores outside 0-2 as long
as the follow up sched_setaffinity() will succeed.
numa_parse_cpustring_all() suites exactly for this case, which should already
have considered sysconf(_SC_NPROCESSORS_ONLN) limit. Use that instead.
Since at it, also remove initialization of cpu_set variable otherwise it's
leaked in previous parse_cpumask too: numa_parse_cpustring_all() will return a
newly allocated buffer already. Quotting from manual:
numa_parse_nodestring() parses a character string list of nodes into a bit
mask. The bit mask is allocated by numa_allocate_nodemask().
numa_parse_nodestring_all() is similar to numa_parse_nodestring, but can
parse all possible nodes, not only current nodeset.
Cc: John Kacur <jkacur@redhat.com>
Cc: Daniel Wagner <dwagner@suse.de>
Cc: Clark Williams <williams@redhat.com>
Reported-by: Pradipta Kumar Sahoo <psahoo@redhat.com>
Reported-by: Mike Stowell <mstowell@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Signed-off-by: John Kacur <jkacur@redhat.com>
---
src/oslat/oslat.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/src/oslat/oslat.c b/src/oslat/oslat.c
index b2c5373388fb..465a694cdd1d 100644
--- a/src/oslat/oslat.c
+++ b/src/oslat/oslat.c
@@ -785,7 +785,6 @@ int main(int argc, char *argv[])
struct thread *threads;
int i, n_cores;
struct bitmask *cpu_set = NULL;
- int max_cpus = sysconf(_SC_NPROCESSORS_ONLN);
#ifdef FRC_MISSING
printf("This architecture is not yet supported. "
@@ -797,10 +796,6 @@ int main(int argc, char *argv[])
exit(1);
}
- cpu_set = numa_allocate_cpumask();
- if (!cpu_set)
- fatal("oslat: Could not allocate cpumask\n");
-
g.app_name = argv[0];
g.rtprio = 0;
g.bucket_size = BUCKET_SIZE;
@@ -817,7 +812,8 @@ int main(int argc, char *argv[])
if (!g.cpu_list)
g.cpu_list = strdup("all");
- if (parse_cpumask(g.cpu_list, max_cpus, &cpu_set) != 0)
+ cpu_set = numa_parse_cpustring_all(g.cpu_list);
+ if (!cpu_set)
fatal("oslat: parse_cpumask failed.\n");
n_cores = numa_bitmask_weight(cpu_set);
--
2.26.2

47
SOURCES/rt-tests-oslat-Allocate-memory-for-cpu_set.patch Normal file
View File

@ -0,0 +1,47 @@
From 5821269dde6a778b0af06c172bc2f19bbe324bda Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Fri, 12 Feb 2021 12:22:23 -0500
Subject: [PATCH 3/3] rt-tests: oslat: Allocate memory for cpu_set
- cpu_set is a pointer to a bitmask struct
Memory needs to be allocated for the struct, so call
numa_allocate_cpumask()
- use rt-tests fatal to exit on error conditions
Reviewed-by: Daniel Wagner <dwagner@suse.de>
Signed-off-by: John Kacur <jkacur@redhat.com>
---
src/oslat/oslat.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/src/oslat/oslat.c b/src/oslat/oslat.c
index 2fe550b3ee12..2a3be393a268 100644
--- a/src/oslat/oslat.c
+++ b/src/oslat/oslat.c
@@ -727,6 +727,10 @@ int main(int argc, char *argv[])
exit(1);
}
+ cpu_set = numa_allocate_cpumask();
+ if (!cpu_set)
+ fatal("oslat: Could not allocate cpumask\n");
+
g.app_name = argv[0];
g.rtprio = 0;
g.bucket_size = BUCKET_SIZE;
@@ -742,8 +746,9 @@ int main(int argc, char *argv[])
if (!g.cpu_list)
g.cpu_list = strdup("all");
- if (parse_cpumask(g.cpu_list, max_cpus, &cpu_set))
- exit(1);
+
+ if (parse_cpumask(g.cpu_list, max_cpus, &cpu_set) != 0)
+ fatal("oslat: parse_cpumask failed.\n");
n_cores = numa_bitmask_weight(cpu_set);
TEST(threads = calloc(1, n_cores * sizeof(threads[0])));
--
2.26.2

54
SOURCES/rt-tests-oslat-print-version-string.patch Normal file
View File

@ -0,0 +1,54 @@
From 2f4d564fb5557f7a420c183ddd3938647c231a8c Mon Sep 17 00:00:00 2001
From: John Kacur <jkacur@redhat.com>
Date: Wed, 10 Feb 2021 22:18:41 -0500
Subject: [PATCH 1/3] rt-tests: oslat: print version string
During the streamlining of the command line options something went awry
with the version. The author of oslat wishes to always print the version
string. This allows us to just exit in the case of -v
Fixes e411219d27b1
Reported-by: Pradipta Kumar Sahoo <psahoo@redhat.com>
Reported-by: Reported-by: Peter Xu <peterx@redhat.com>
Signed-off-by: John Kacur <jkacur@redhat.com>
---
src/oslat/oslat.c | 7 +++----
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/src/oslat/oslat.c b/src/oslat/oslat.c
index 5b7e0d5b5d5c..7826c277f26d 100644
--- a/src/oslat/oslat.c
+++ b/src/oslat/oslat.c
@@ -512,7 +512,6 @@ static void handle_alarm(int code)
static void usage(int error)
{
- printf("oslat V %1.2f\n", VERSION);
printf("Usage:\n"
"oslat <options>\n\n"
"This is an OS latency detector by running busy loops on specified cores.\n"
@@ -657,8 +656,8 @@ static void parse_options(int argc, char *argv[])
break;
case 'v':
/*
- * Because we always dump the version even before parsing options,
- * what we need to do is to quit..
+ * We always print the version before parsing options
+ * so just exit
*/
exit(0);
break;
@@ -736,7 +735,7 @@ int main(int argc, char *argv[])
g.workload_mem_size = WORKLOAD_MEM_SIZE;
/* Run the main thread on cpu0 by default */
g.cpu_main_thread = 0;
-
+ printf("oslat V %1.2f\n", VERSION);
parse_options(argc, argv);
TEST(mlockall(MCL_CURRENT | MCL_FUTURE) == 0);
--
2.26.2

20
SPECS/rt-tests.spec
View File

@ -6,7 +6,7 @@ Name: rt-tests
# Numa argument to make: NUMA=1 # Numa argument to make: NUMA=1
# #
Version: 1.10 Version: 1.10
Release: 1%{?dist} Release: 3%{?dist}
License: GPLv2 License: GPLv2
Group: Development/Tools Group: Development/Tools
URL: git://git.kernel.org/pub/scm/utils/rt-tests/rt-tests.git URL: git://git.kernel.org/pub/scm/utils/rt-tests/rt-tests.git
@ -21,6 +21,10 @@ BuildRequires: python3-devel
Requires: bash bc Requires: bash bc
#Patches #Patches
Patch1: rt-tests-oslat-print-version-string.patch
Patch2: oslat-Use-cpuset-size-as-upper-bound.patch
Patch3: rt-tests-oslat-Allocate-memory-for-cpu_set.patch
Patch4: oslat-allow-scheduling-on-all-possible-cores.patch
%description %description
rt-tests is a set of programs that test and measure various components of rt-tests is a set of programs that test and measure various components of
@ -29,6 +33,10 @@ latency. It also tests the functioning of priority-inheritance mutexes.
%prep %prep
%setup -q -n %{name}-%{version} %setup -q -n %{name}-%{version}
%patch1 -p1
%patch2 -p1
%patch3 -p1
%patch4 -p1
%build %build
%set_build_flags %set_build_flags
@ -83,6 +91,16 @@ rm -rf ${build_root}
%{_mandir}/man8/determine_maximum_mpps.8.* %{_mandir}/man8/determine_maximum_mpps.8.*
%changelog %changelog
* Fri Feb 19 2021 John Kacur <jkacur@redhat.com> - 1.10-3
- parse_cpumask() is too strict for oslat, allow all possible cores
Resolves: rhbz#1926578
* Thu Feb 18 2021 John Kacur <jkacur@redhat.com> - 1.10-2
- print the version number in oslat everytime.
- use cpuset size as upper bound in loop in oslat
- allocate memory for cpu_set in oslat
Resolves: rhbz#1926578
* Mon Jan 11 2021 John Kacur <jkacur@redhat.com> - 1.10-1 * Mon Jan 11 2021 John Kacur <jkacur@redhat.com> - 1.10-1
- Upgrade to upstream rt-tests-1.10 - Upgrade to upstream rt-tests-1.10
Resolves: rhbz#1890556 Resolves: rhbz#1890556