linux/tools/perf
Athira Rajeev f58faed7fb perf bench: Fix numa bench to fix usage of affinity for machines with #CPUs > 1K
The 'perf bench numa' testcase fails on systems with more than 1K CPUs.

Testcase: perf bench numa mem -p 1 -t 3 -P 512 -s 100 -zZ0qcm --thp  1

Snippet of code:

  <<>>
  perf: bench/numa.c:302: bind_to_node: Assertion `!(ret)' failed.
  Aborted (core dumped)
  <<>>

bind_to_node() uses "sched_getaffinity" to save the original cpumask and
this call is returning EINVAL ((invalid argument).

This happens because the default mask size in glibc is 1024.  To
overcome this 1024 CPUs mask size limitation of cpu_set_t, change the
mask size using the CPU_*_S macros ie, use CPU_ALLOC to allocate
cpumask, CPU_ALLOC_SIZE for size.

Apart from fixing this for "orig_mask", apply same logic to "mask" as
well which is used to setaffinity so that mask size is large enough to
represent number of possible CPU's in the system.

sched_getaffinity is used in one more place in perf numa bench. It is in
"bind_to_cpu" function. Apply the same logic there also. Though
currently no failure is reported from there, it is ideal to change
getaffinity to work with such system configurations having CPU's more
than default mask size supported by glibc.

Also fix "sched_setaffinity" to use mask size which is large enough to
represent number of possible CPU's in the system.

Fixed all places where "bind_cpumask" which is part of "struct
thread_data" is used such that bind_cpumask works in all configuration.

Reported-by: Disha Goel <disgoel@linux.vnet.ibm.com>
Signed-off-by: Athira Jajeev <atrajeev@linux.vnet.ibm.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kajol Jain <kjain@linux.ibm.com>
Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Nageswara R Sastry <rnsastry@linux.ibm.com>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: linuxppc-dev@lists.ozlabs.org
Link: https://lore.kernel.org/r/20220412164059.42654-3-atrajeev@linux.vnet.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
2022-04-14 09:15:10 -03:00
..
arch perf: arm-spe: Fix perf report --mem-mode 2022-04-09 12:34:29 -03:00
bench perf bench: Fix numa bench to fix usage of affinity for machines with #CPUs > 1K 2022-04-14 09:15:10 -03:00
dlfilters perf dlfilter: Drop unused variable 2021-12-16 12:18:11 -03:00
Documentation perf docs: Add perf-iostat link to manpages 2022-04-09 14:20:59 -03:00
examples/bpf perf tools: Fix various typos in comments 2021-03-23 17:13:43 -03:00
include perf build: Move perf_dlfilters.h in the source tree 2021-08-11 09:35:24 -03:00
jvmti perf tools: Fix various typos in comments 2021-03-23 17:13:43 -03:00
pmu-events perf vendor events intel: Update events for TremontX 2022-03-18 11:43:51 -03:00
python perf python: Convert tracepoint.py example to python3 2022-04-01 16:19:35 -03:00
scripts perf scripts python: export-to-postgresql.py: Export all sample flags 2022-02-15 17:15:07 -03:00
tests perf unwind: Don't show unwind error messages when augmenting frame pointer stack 2022-04-09 12:34:29 -03:00
trace perf beauty: Update copy of linux/socket.h with the kernel sources 2022-04-01 16:19:34 -03:00
ui perf annotate: Avoid TUI crash when navigating in the annotation of recursive functions 2022-01-10 15:47:30 -03:00
util perf bench: Fix numa testcase to check if CPU used to bind task is online 2022-04-14 09:13:41 -03:00
.gitignore Add 'tools/perf/libbpf/' to ignored files 2021-11-08 11:33:35 -08:00
Build perf daemon: Add daemon command 2021-02-09 15:42:57 -03:00
builtin-annotate.c perf annotate: Remove redundant 'ret' variable 2022-02-16 10:38:55 -03:00
builtin-bench.c perf bench: Use unbuffered output when pipe/tee'ing to a file 2021-12-16 12:18:11 -03:00
builtin-buildid-cache.c perf record: Disable debuginfod by default 2022-01-15 17:41:25 -03:00
builtin-buildid-list.c perf tools: Remove repipe argument from perf_session__new() 2021-08-02 10:06:51 -03:00
builtin-c2c.c perf tools: Enhance the matching of sub-commands abbreviations 2022-03-26 10:55:57 -03:00
builtin-config.c
builtin-daemon.c perf daemon: Remove duplicate sys/file.h include 2021-10-08 15:14:50 -03:00
builtin-data.c perf data: Don't mention --to-ctf if it's not supported 2022-02-22 21:23:08 -03:00
builtin-diff.c perf tools: Remove repipe argument from perf_session__new() 2021-08-02 10:06:51 -03:00
builtin-evlist.c perf tools: Remove repipe argument from perf_session__new() 2021-08-02 10:06:51 -03:00
builtin-ftrace.c perf evlist: Rename cpus to user_requested_cpus 2022-04-01 16:19:35 -03:00
builtin-help.c
builtin-inject.c perf namespaces: Add functions to access nsinfo 2022-02-11 14:31:22 -03:00
builtin-kallsyms.c
builtin-kmem.c perf tools: Enhance the matching of sub-commands abbreviations 2022-03-26 10:55:57 -03:00
builtin-kvm.c perf tools: Enhance the matching of sub-commands abbreviations 2022-03-26 10:55:57 -03:00
builtin-list.c perf list: Display hybrid PMU events with cpu type 2021-10-25 13:47:42 -03:00
builtin-lock.c perf tools: Enhance the matching of sub-commands abbreviations 2022-03-26 10:55:57 -03:00
builtin-mem.c perf tools: Enhance the matching of sub-commands abbreviations 2022-03-26 10:55:57 -03:00
builtin-probe.c perf namespaces: Add functions to access nsinfo 2022-02-11 14:31:22 -03:00
builtin-record.c perf record: Fix per-thread option 2022-04-14 09:05:11 -03:00
builtin-report.c perf tui: Only support --tui with slang 2022-02-15 17:15:29 -03:00
builtin-sched.c perf tools: Enhance the matching of sub-commands abbreviations 2022-03-26 10:55:57 -03:00
builtin-script.c perf tools: Enhance the matching of sub-commands abbreviations 2022-03-26 10:55:57 -03:00
builtin-stat.c perf evlist: Rename cpus to user_requested_cpus 2022-04-01 16:19:35 -03:00
builtin-timechart.c perf tools: Enhance the matching of sub-commands abbreviations 2022-03-26 10:55:57 -03:00
builtin-top.c perf evlist: Rename cpus to user_requested_cpus 2022-04-01 16:19:35 -03:00
builtin-trace.c Merge remote-tracking branch 'torvalds/master' into perf/core 2022-02-17 18:40:54 -03:00
builtin-version.c perf version: Add a feature for libpfm4 2020-11-04 09:42:40 -03:00
builtin.h perf daemon: Add daemon command 2021-02-09 15:42:57 -03:00
check-headers.sh tools arm64: Import cputype.h 2022-03-26 10:53:45 -03:00
command-list.txt perf stat: Enable iostat mode for x86 platforms 2021-04-20 08:40:20 -03:00
CREDITS
design.txt perf design.txt: Synchronize the definition of enum perf_hw_id with code 2021-11-13 18:11:50 -03:00
Makefile perf tools: Add a build-test variant to use in builds from a tarball 2021-04-20 08:43:58 -03:00
Makefile.config perf build: Don't use -ffat-lto-objects in the python feature test when building with clang-13 2022-04-09 12:34:29 -03:00
Makefile.perf perf tools: Stop depending on .git files for building PERF-VERSION-FILE 2022-04-01 16:19:35 -03:00
MANIFEST perf MANIFEST: Add bpftool files to allow building with BUILD_BPF_SKEL=1 2021-11-07 15:39:28 -03:00
perf-archive.sh perf archive: Fix filtering of empty build-ids 2021-03-06 16:54:31 -03:00
perf-completion.sh
perf-iostat.sh perf stat: Enable iostat mode for x86 platforms 2021-04-20 08:40:20 -03:00
perf-read-vdso.c
perf-sys.h
perf-with-kcore.sh
perf.c perf tools: Add external commands to list-cmds 2022-04-09 14:21:00 -03:00
perf.h