diff options
| author | Linus Torvalds | 2022-06-04 13:33:12 -0700 |
|---|---|---|
| committer | Linus Torvalds | 2022-06-04 13:33:12 -0700 |
| commit | 45b2e5ad6837dfe4de6b9028c575bd57c132774c (patch) | |
| tree | cfad761ceebf454fa238df05d93ee431993add35 /tools/perf/builtin-lock.c | |
| parent | 032dcf09e2bf7c822be25b4abef7a6c913870d98 (diff) | |
| parent | 1bcca2b1bd67f3c0e5c3a88ed16c6389f01a5b31 (diff) | |
Merge tag 'perf-tools-for-v5.19-2022-06-04' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux
Pull more perf tools updates from Arnaldo Carvalho de Melo:
- Synthesize task events for pre-existing threads when using 'perf lock
--threads', as we need to show task names.
- Fix unwinding with ld.lld (>= version 10.0) linked objects, where
.eh_frame_hdr and .text are in different PT_LOAD program headers,
which makes perf record --call-graph dwarf fail with such obkects.
- Check if 'perf record' hangs in the ARM SPE (Statistical Profiling
Extensions) 'perf test' entry when recording a workload with forks.
- Trace physical address for Arm SPE events, needed for 'perf c2c' to
locate the memory node for samples.
- Fix sorting in percent_rmt_hitm_cmp() in 'perf c2c'.
- Further support for Intel hybrid systems in the evlist and 'perf
record' code.
- Update IBM s/390 vendor event JSON tables.
- Add metrics (JSON) for Intel Sapphirerapids.
- Update metrics for Intel Alderlake.
- Correct typo of sysf 'event_source' directory in the documentation.
* tag 'perf-tools-for-v5.19-2022-06-04' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux:
perf vendor events intel: Update metrics for Alderlake
perf vendor events intel: Add metrics for Sapphirerapids
perf c2c: Fix sorting in percent_rmt_hitm_cmp()
perf mem: Trace physical address for Arm SPE events
perf list: Update event description for IBM zEC12/zBC12 to latest level
perf list: Update event description for IBM z196/z114 to latest level
perf list: Update event description for IBM z15 to latest level
perf list: Update event description for IBM z14 to latest level
perf list: Update event description for IBM z13 to latest level
perf list: Update event description for IBM z10 to latest level
perf list: Add IBM z16 event description for s390
perf record: Support sample-read topdown metric group for hybrid platforms
perf lock: Change to synthesize task events
perf unwind: Fix segbase for ld.lld linked objects
perf test arm-spe: Check if perf-record hangs when recording workload with forks
perf docs: Correct typo of event_sources
perf evlist: Extend arch_evsel__must_be_in_group to support hybrid systems
Diffstat (limited to 'tools/perf/builtin-lock.c')
| -rw-r--r-- | tools/perf/builtin-lock.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tools/perf/builtin-lock.c b/tools/perf/builtin-lock.c index b1200b7340a6..23a33ac15e68 100644 --- a/tools/perf/builtin-lock.c +++ b/tools/perf/builtin-lock.c @@ -1083,7 +1083,7 @@ out_delete: static int __cmd_record(int argc, const char **argv) { const char *record_args[] = { - "record", "-R", "-m", "1024", "-c", "1", "--synth", "no", + "record", "-R", "-m", "1024", "-c", "1", "--synth", "task", }; unsigned int rec_argc, i, j, ret; const char **rec_argv; |