mirror of
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
synced 2026-05-10 22:22:51 -04:00
74069a01609ef0f499fa6bf89bb1b017eba16b0d
1351601 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Howard Chu
|
74069a0160 |
perf record --off-cpu: Dump the remaining PERF_SAMPLE_ in sample_type from BPF's stack trace map
Dump the remaining PERF_SAMPLE_ data, as if it is dumping a direct sample. Put the stack trace, tid, off-cpu time and cgroup id into the raw_data section, just like a direct off-cpu sample coming from BPF's bpf_perf_event_output(). This ensures that evsel__parse_sample() correctly parses both direct samples and accumulated samples. Suggested-by: Namhyung Kim <namhyung@kernel.org> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Gautam Menghani <gautam@linux.ibm.com> Tested-by: Ian Rogers <irogers@google.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241108204137.2444151-10-howardchu95@gmail.com Link: https://lore.kernel.org/r/20250501022809.449767-9-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Howard Chu
|
8ae7a5769b |
perf script: Display off-cpu samples correctly
No PERF_SAMPLE_CALLCHAIN in sample_type, but 'perf script' needs to
display a callchain, have to specify manually.
Also, prefer displaying a callchain:
gvfs-afc-volume 2267 [001] 3829232.955656: 1001115340 offcpu-time:
77f05292603f __pselect+0xbf (/usr/lib/x86_64-linux-gnu/libc.so.6)
77f052a1801c [unknown] (/usr/lib/x86_64-linux-gnu/libusbmuxd-2.0.so.6.0.0)
77f052a18d45 [unknown] (/usr/lib/x86_64-linux-gnu/libusbmuxd-2.0.so.6.0.0)
77f05289ca94 start_thread+0x384 (/usr/lib/x86_64-linux-gnu/libc.so.6)
77f052929c3c clone3+0x2c (/usr/lib/x86_64-linux-gnu/libc.so.6)
to a raw binary BPF output:
BPF output: 0000: dd 08 00 00 db 08 00 00 <DD>...<DB>...
0008: cc ce ab 3b 00 00 00 00 <CC>Ϋ;....
0010: 06 00 00 00 00 00 00 00 ........
0018: 00 fe ff ff ff ff ff ff .<FE><FF><FF><FF><FF><FF><FF>
0020: 3f 60 92 52 f0 77 00 00 ?`.R<F0>w..
0028: 1c 80 a1 52 f0 77 00 00 ..<A1>R<F0>w..
0030: 45 8d a1 52 f0 77 00 00 E.<A1>R<F0>w..
0038: 94 ca 89 52 f0 77 00 00 .<CA>.R<F0>w..
0040: 3c 9c 92 52 f0 77 00 00 <..R<F0>w..
0048: 00 00 00 00 00 00 00 00 ........
0050: 00 00 00 00 ....
Reviewed-by: Ian Rogers <irogers@google.com>
Signed-off-by: Howard Chu <howardchu95@gmail.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Gautam Menghani <gautam@linux.ibm.com>
Tested-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@linaro.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20241108204137.2444151-9-howardchu95@gmail.com
Link: https://lore.kernel.org/r/20250501022809.449767-8-howardchu95@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Howard Chu
|
7de1a87f1e |
perf record --off-cpu: Disable perf_event's callchain collection
There is a check in evsel.c that does this: if (evsel__is_offcpu_event(evsel)) evsel->core.attr.sample_type &= OFFCPU_SAMPLE_TYPES; This along with: #define OFFCPU_SAMPLE_TYPES (PERF_SAMPLE_IDENTIFIER | PERF_SAMPLE_IP | \ PERF_SAMPLE_TID | PERF_SAMPLE_TIME | \ PERF_SAMPLE_ID | PERF_SAMPLE_CPU | \ PERF_SAMPLE_PERIOD | PERF_SAMPLE_CALLCHAIN | \ PERF_SAMPLE_CGROUP) will tell perf_event to collect callchain. We don't need the callchain from perf_event when collecting off-cpu samples, because it's prev's callchain, not next's callchain. (perf_event) (task_storage) (needed) prev next | | ---sched_switch----> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Gautam Menghani <gautam@linux.ibm.com> Tested-by: Ian Rogers <irogers@google.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241108204137.2444151-8-howardchu95@gmail.com Link: https://lore.kernel.org/r/20250501022809.449767-7-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Howard Chu
|
7f8f56475d |
perf evsel: Assemble off-cpu samples
Use the data in bpf-output samples, to assemble off-cpu samples.
In evsel__is_offcpu_event(), check if sample_type is PERF_SAMPLE_RAW to
support off-cpu sample data created by an older version of perf.
Testing compatibility on off-cpu samples collected by perf before this patch series:
See below, the sample_type still uses PERF_SAMPLE_CALLCHAIN
$ perf script --header -i ./perf.data.ptn | grep "event : name = offcpu-time"
# event : name = offcpu-time, , id = { 237917, 237918, 237919, 237920 }, type = 1 (software), size = 136, config = 0xa (PERF_COUNT_SW_BPF_OUTPUT), { sample_period, sample_freq } = 1, sample_type = IP|TID|TIME|CALLCHAIN|CPU|PERIOD|IDENTIFIER, read_format = ID|LOST, disabled = 1, freq = 1, sample_id_all = 1
The output is correct.
$ perf script -i ./perf.data.ptn | grep offcpu-time
gmain 2173 [000] 18446744069.414584: 100102015 offcpu-time:
NetworkManager 901 [000] 18446744069.414584: 5603579 offcpu-time:
Web Content 1183550 [000] 18446744069.414584: 46278 offcpu-time:
gnome-control-c 2200559 [000] 18446744069.414584: 11998247014 offcpu-time:
<SNIP>
$
And after this patch series:
$ perf script --header -i ./perf.data.off-cpu-v9 | grep "event : name = offcpu-time"
# event : name = offcpu-time, , id = { 237959, 237960, 237961, 237962 }, type = 1 (software), size = 136, config = 0xa (PERF_COUNT_SW_BPF_OUTPUT), { sample_period, sample_freq } = 1, sample_type = IP|TID|TIME|CPU|PERIOD|RAW|IDENTIFIER, read_format = ID|LOST, disabled = 1, freq = 1, sample_id_all = 1
$ ./perf script -i ./perf.data.off-cpu-v9 | grep offcpu-time
gnome-shell 1875 [001] 4789616.361225: 100097057 offcpu-time:
gnome-shell 1875 [001] 4789616.461419: 100107463 offcpu-time:
firefox 2206821 [002] 4789616.475690: 255257245 offcpu-time:
$
Committer testing:
The command to record those samples:
root@number:~# perf record --off-cpu -a sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 2.092 MB perf.data (1552 samples) ]
root@number:~#
Then, before this patch series, the sample_type for the "offcpu-time" event is:
root@number:~# perf evlist -v | grep offcpu-time
offcpu-time: type: 1 (PERF_TYPE_SOFTWARE), size: 136, config: 0xa (PERF_COUNT_SW_BPF_OUTPUT), { sample_period, sample_freq }: 1, sample_type: IP|TID|TIME|CALLCHAIN|CPU|PERIOD|IDENTIFIER, read_format: ID|LOST, disabled: 1, freq: 1, sample_id_all: 1
root@number:~#
And after it, after recording it again:
root@number:~# perf record --off-cpu -a sleep 1 ; perf evlist -v | grep offcpu-time
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 2.151 MB perf.data (2843 samples) ]
offcpu-time: type: 1 (PERF_TYPE_SOFTWARE), size: 136, config: 0xa (PERF_COUNT_SW_BPF_OUTPUT), { sample_period, sample_freq }: 1, sample_type: IP|TID|TIME|CPU|PERIOD|IDENTIFIER, read_format: ID|LOST, disabled: 1, sample_id_all: 1
root@number:~#
Suggested-by: Namhyung Kim <namhyung@kernel.org>
Reviewed-by: Ian Rogers <irogers@google.com>
Signed-off-by: Howard Chu <howardchu95@gmail.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Gautam Menghani <gautam@linux.ibm.com>
Tested-by: Ian Rogers <irogers@google.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: James Clark <james.clark@linaro.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20241108204137.2444151-7-howardchu95@gmail.com
Link: https://lore.kernel.org/r/20250501022809.449767-6-howardchu95@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Howard Chu
|
d6948f2af2 |
perf record --off-cpu: Dump off-cpu samples in BPF
Collect tid, period, callchain, and cgroup id and dump them when off-cpu time threshold is reached. We don't collect the off-cpu time twice (the delta), it's either in direct samples, or accumulated samples that are dumped at the end of perf.data. Suggested-by: Namhyung Kim <namhyung@kernel.org> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Gautam Menghani <gautam@linux.ibm.com> Tested-by: Ian Rogers <irogers@google.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241108204137.2444151-6-howardchu95@gmail.com Link: https://lore.kernel.org/r/20250501022809.449767-5-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Howard Chu
|
282c195906 |
perf record --off-cpu: Preparation of off-cpu BPF program
Set the perf_event map in BPF for dumping off-cpu samples, and set the offcpu_thresh to specify the threshold. Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Gautam Menghani <gautam@linux.ibm.com> Tested-by: Ian Rogers <irogers@google.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241108204137.2444151-5-howardchu95@gmail.com Link: https://lore.kernel.org/r/20250501022809.449767-4-howardchu95@gmail.com [ Added some missing iteration variables to off_cpu_config() and fixed up a manually edited patch hunk line boundary line ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Howard Chu
|
0f72027bb9 |
perf record --off-cpu: Parse off-cpu event
Parse the off-cpu event using parse_event(), as bpf-output. Call evlist__enable_evsel() on off-cpu event. This fixes the inability to collect direct off-cpu samples on a workload, as reported by Arnaldo Carvalho de Melo <acme@redhat.com>. The reason being, workload sets enable_on_exec instead of calling evlist__enable(), but off-cpu event does not attach to an executable and execve won't be called, so the fds from perf_event_open() are not enabled. no-inherit should be set to 1, here's the reason: We update the BPF perf_event map for direct off-cpu sample dumping (in following patches), it executes as follows: bpf_map_update_value() bpf_fd_array_map_update_elem() perf_event_fd_array_get_ptr() perf_event_read_local() In perf_event_read_local(), there is: int perf_event_read_local(struct perf_event *event, u64 *value, u64 *enabled, u64 *running) { ... /* * It must not be an event with inherit set, we cannot read * all child counters from atomic context. */ if (event->attr.inherit) { ret = -EOPNOTSUPP; goto out; } Which means no-inherit has to be true for updating the BPF perf_event map. Moreover, for bpf-output events, we primarily want a system-wide event instead of a per-task event. The reason is that in BPF's bpf_perf_event_output(), BPF uses the CPU index to retrieve the perf_event file descriptor it outputs to. Making a bpf-output event system-wide naturally satisfies this requirement by mapping CPU appropriately. Suggested-by: Namhyung Kim <namhyung@kernel.org> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Gautam Menghani <gautam@linux.ibm.com> Tested-by: Ian Rogers <irogers@google.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241108204137.2444151-4-howardchu95@gmail.com Link: https://lore.kernel.org/r/20250501022809.449767-3-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Howard Chu
|
671e943452 |
perf evsel: Expose evsel__is_offcpu_event() for future use
Expose evsel__is_offcpu_event() so it can be used in off_cpu_config(), evsel__parse_sample() and 'perf script'. Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Howard Chu <howardchu95@gmail.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Tested-by: Gautam Menghani <gautam@linux.ibm.com> Tested-by: Ian Rogers <irogers@google.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20241108204137.2444151-3-howardchu95@gmail.com Link: https://lore.kernel.org/r/20250501022809.449767-2-howardchu95@gmail.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
fa9c4977fb |
perf symbol-minimal: Fix double free in filename__read_build_id
Running the "perf script task-analyzer tests" with address sanitizer
showed a double free:
```
FAIL: "test_csv_extended_times" Error message: "Failed to find required string:'Out-Out;'."
=================================================================
==19190==ERROR: AddressSanitizer: attempting double-free on 0x50b000017b10 in thread T0:
#0 0x55da9601c78a in free (perf+0x26078a) (BuildId: e7ef50e08970f017a96fde6101c5e2491acc674a)
#1 0x55da96640c63 in filename__read_build_id tools/perf/util/symbol-minimal.c:221:2
0x50b000017b10 is located 0 bytes inside of 112-byte region [0x50b000017b10,0x50b000017b80)
freed by thread T0 here:
#0 0x55da9601ce40 in realloc (perf+0x260e40) (BuildId: e7ef50e08970f017a96fde6101c5e2491acc674a)
#1 0x55da96640ad6 in filename__read_build_id tools/perf/util/symbol-minimal.c:204:10
previously allocated by thread T0 here:
#0 0x55da9601ca23 in malloc (perf+0x260a23) (BuildId: e7ef50e08970f017a96fde6101c5e2491acc674a)
#1 0x55da966407e7 in filename__read_build_id tools/perf/util/symbol-minimal.c:181:9
SUMMARY: AddressSanitizer: double-free (perf+0x26078a) (BuildId: e7ef50e08970f017a96fde6101c5e2491acc674a) in free
==19190==ABORTING
FAIL: "invocation of perf script report task-analyzer --csv-summary csvsummary --summary-extended command failed" Error message: ""
FAIL: "test_csvsummary_extended" Error message: "Failed to find required string:'Out-Out;'."
---- end(-1) ----
132: perf script task-analyzer tests : FAILED!
```
The buf_size if always set to phdr->p_filesz, but that may be 0
causing a free and realloc to return NULL. This is treated in
filename__read_build_id like a failure and the buffer is freed again.
To avoid this problem only grow buf, meaning the buf_size will never
be 0. This also reduces the number of memory (re)allocations.
Fixes:
|
||
|
Namhyung Kim
|
f7458176a7 |
perf mem: Add 'dtlb' output field
This is a breakdown of perf_mem_data_src.mem_dtlb values. It assumes
PMU drivers would set PERF_MEM_TLB_HIT bit with an appropriate level.
And having PERF_MEM_TLB_MISS means that it failed to find one in any
levels of TLB. For now, it doesn't use PERF_MEM_TLB_{WK,OS} bits.
Also it seems Intel machines don't distinguish L1 or L2 precisely. So I
added ANY_HIT (printed as "L?-Hit") to handle the case.
$ perf mem report -F overhead,dtlb,dso --stdio
...
# --- D-TLB ----
# Overhead L?-Hit Miss Shared Object
# ........ .............. .................
#
67.03% 99.5% 0.5% [unknown]
31.23% 99.2% 0.8% [kernel.kallsyms]
1.08% 97.8% 2.2% [i915]
0.36% 100.0% 0.0% [JIT] tid 6853
0.12% 100.0% 0.0% [drm]
0.05% 100.0% 0.0% [drm_kms_helper]
0.05% 100.0% 0.0% [ext4]
0.02% 100.0% 0.0% [aesni_intel]
0.02% 100.0% 0.0% [crc32c_intel]
0.02% 100.0% 0.0% [dm_crypt]
...
Committer testing:
# perf report --header | grep cpudesc
# cpudesc : AMD Ryzen 9 9950X3D 16-Core Processor
# perf mem report -F overhead,dtlb,dso --stdio | head -20
# To display the perf.data header info, please use --header/--header-only options.
#
#
# Total Lost Samples: 0
#
# Samples: 2K of event 'cycles:P'
# Total weight : 2637
# Sort order : local_weight,mem,sym,dso,symbol_daddr,dso_daddr,snoop,tlb,locked,blocked,local_ins_lat,local_p_stage_cyc
#
# ---------- D-TLB -----------
# Overhead L1-Hit L2-Hit Miss Other Shared Object
# ........ ............................ .................................
#
77.47% 18.4% 0.1% 0.6% 80.9% [kernel.kallsyms]
5.61% 36.5% 0.7% 1.4% 61.5% libxul.so
2.77% 39.7% 0.0% 12.3% 47.9% libc.so.6
2.01% 34.0% 1.9% 1.9% 62.3% libglib-2.0.so.0.8400.1
1.93% 31.4% 2.0% 2.0% 64.7% [amdgpu]
1.63% 48.8% 0.0% 0.0% 51.2% [JIT] tid 60168
1.14% 3.3% 0.0% 0.0% 96.7% [vdso]
#
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20250430205548.789750-12-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
5e424a0178 |
perf mem: Add 'snoop' output field
This is a breakdown of perf_mem_data_src.mem_snoop values. For now, it
doesn't use mem_snoopx values like FWD and PEER.
$ perf mem report -F overhead,snoop,comm --stdio
...
# ---------- Snoop -----------
# Overhead Hit HitM Miss Other Command
# ........ ............................ ...............
#
34.24% 0.6% 0.0% 0.0% 99.4% gnome-shell
12.02% 1.0% 0.0% 0.0% 99.0% chrome
9.32% 1.0% 0.0% 0.3% 98.7% Isolated Web Co
6.85% 1.0% 0.3% 0.0% 98.6% swapper
6.30% 0.8% 0.8% 0.0% 98.5% Xorg
3.02% 2.4% 0.0% 0.0% 97.6% VizCompositorTh
2.35% 0.0% 0.0% 0.0% 100.0% firefox-esr
2.04% 0.0% 0.0% 0.0% 100.0% JS Helper
1.51% 3.2% 0.0% 0.0% 96.8% threaded-ml
1.44% 0.0% 0.0% 0.0% 100.0% AudioIP~allback
...
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20250430205548.789750-11-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
abe4dc24a8 |
perf mem: Add 'cache' and 'memory' output fields
This is a breakdown of perf_mem_data_src.mem_lvl_num. But it's also
divided into two parts because the combination is bigger than 8.
Since there are many entries for different cache levels, 'cache' field
focuses on them. I generalized buffers like LFB, MAB and MHB to L1-buf
and L2-buf.
The rest goes to 'memory' field which can be RAM, CXL, PMEM, IO, etc.
$ perf mem report -F cache,mem,dso --stdio
...
#
# -------------- Cache -------------- --- Memory ---
# L1 L2 L3 L1-buf Other RAM Other Shared Object
# ................................... .............. ....................................
#
53.9% 3.6% 16.2% 21.6% 4.8% 4.8% 95.2% [kernel.kallsyms]
64.7% 1.7% 3.5% 17.4% 12.8% 12.8% 87.2% chrome (deleted)
78.3% 2.8% 0.0% 1.0% 17.9% 17.9% 82.1% libc.so.6
39.6% 1.5% 0.0% 5.7% 53.2% 53.2% 46.8% libxul.so
26.2% 0.0% 0.0% 0.0% 73.8% 73.8% 26.2% [unknown]
85.5% 0.0% 0.0% 14.5% 0.0% 0.0% 100.0% libspa-audioconvert.so
66.3% 4.4% 0.0% 29.4% 0.0% 0.0% 100.0% libglib-2.0.so.0.8200.1 (deleted)
1.9% 0.0% 0.0% 0.0% 98.1% 98.1% 1.9% libmutter-cogl-15.so.0.0.0 (deleted)
10.6% 0.0% 0.0% 89.4% 0.0% 0.0% 100.0% libpulsecommon-16.1.so
0.0% 0.0% 0.0% 100.0% 0.0% 0.0% 100.0% libfreeblpriv3.so (deleted)
...
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20250430205548.789750-10-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
225772c17c |
perf hist: Hide unused mem stat columns
Some mem_stat types don't use all 8 columns. And there are cases only
samples in certain kinds of mem_stat types are available only. For that
case hide columns which has no samples.
The new output for the previous data would be:
$ perf mem report -F overhead,op,comm --stdio
...
# ------ Mem Op -------
# Overhead Load Store Other Command
# ........ ..................... ...............
#
44.85% 21.1% 30.7% 48.3% swapper
26.82% 98.8% 0.3% 0.9% netsli-prober
7.19% 51.7% 13.7% 34.6% perf
5.81% 89.7% 2.2% 8.1% qemu-system-ppc
4.77% 100.0% 0.0% 0.0% notifications_c
1.77% 95.9% 1.2% 3.0% MemoryReleaser
0.77% 71.6% 4.1% 24.3% DefaultEventMan
0.19% 66.7% 22.2% 11.1% gnome-shell
...
On Intel machines, the event is only for loads or stores so it'll have
only one column:
# Mem Op
# Overhead Load Command
# ........ ....... ...............
#
20.55% 100.0% swapper
17.13% 100.0% chrome
9.02% 100.0% data-loop.0
6.26% 100.0% pipewire-pulse
5.63% 100.0% threaded-ml
5.47% 100.0% GraphRunner
5.37% 100.0% AudioIP~allback
5.30% 100.0% Chrome_ChildIOT
3.17% 100.0% Isolated Web Co
...
Committer testing:
# grep "model name" -m1 /proc/cpuinfo
model name : AMD Ryzen 9 9950X3D 16-Core Processo
# perf mem report -F overhead,op,comm --stdio
# Total Lost Samples: 0
#
# Samples: 2K of event 'cycles:P'
# Total weight : 2637
# Sort order : local_weight,mem,sym,dso,symbol_daddr,dso_daddr,snoop,tlb,locked,blocked,local_ins_lat,local_p_stage_cyc
#
# ------ Mem Op -------
# Overhead Load Store Other Command
# ........ ..................... ...............
#
61.02% 14.4% 25.5% 60.1% swapper
5.61% 26.4% 13.5% 60.1% Isolated Web Co
5.50% 21.4% 29.7% 49.0% perf
4.74% 27.2% 15.2% 57.6% gnome-shell
4.63% 33.6% 11.5% 54.9% mdns_service
4.29% 28.3% 12.4% 59.3% ptyxis
2.16% 24.6% 19.3% 56.1% DOM Worker
0.99% 23.1% 34.6% 42.3% firefox
0.72% 26.3% 15.8% 57.9% IPC I/O Parent
0.61% 12.5% 12.5% 75.0% kworker/u130:20
0.61% 37.5% 18.8% 43.8% podman
0.57% 33.3% 6.7% 60.0% Timer
0.53% 14.3% 7.1% 78.6% KMS thread
0.49% 30.8% 7.7% 61.5% kworker/u130:3-
0.46% 41.7% 33.3% 25.0% IPDL Background
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20250430205548.789750-9-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
1e6569dca5 |
perf mem: Add 'op' output field
This is an actual example of the he_mem_stat based sample breakdown. It
uses 'mem_op' field of union perf_mem_data_src which means memory
operations.
It'd have basically 'load' or 'store' which can be useful if PMU doesn't
have separate events for them like IBS or SPE. In addition, there's an
entry in case load and store happen at the same time. Also adds entries
for prefetching and execution.
$ perf mem report -F +op -s comm --stdio
# To display the perf.data header info, please use --header/--header-only options.
#
#
# Total Lost Samples: 0
#
# Samples: 4K of event 'ibs_op//'
# Total weight : 9559
# Sort order : comm
#
# --------------------- Mem Op ----------------------
# Overhead Samples Load Store Ld+St Pfetch Exec Other N/A N/A Command
# ........ ....... ................................................... ...............
#
44.85% 4077 21.1% 30.7% 0.0% 0.0% 0.0% 48.3% 0.0% 0.0% swapper
26.82% 45 98.8% 0.3% 0.0% 0.0% 0.0% 0.9% 0.0% 0.0% netsli-prober
7.19% 442 51.7% 13.7% 0.0% 0.0% 0.0% 34.6% 0.0% 0.0% perf
5.81% 75 89.7% 2.2% 0.0% 0.0% 0.0% 8.1% 0.0% 0.0% qemu-system-ppc
4.77% 1 100.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% notifications_c
1.77% 10 95.9% 1.2% 0.0% 0.0% 0.0% 3.0% 0.0% 0.0% MemoryReleaser
0.77% 32 71.6% 4.1% 0.0% 0.0% 0.0% 24.3% 0.0% 0.0% DefaultEventMan
0.19% 10 66.7% 22.2% 0.0% 0.0% 0.0% 11.1% 0.0% 0.0% gnome-shell
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20250430205548.789750-8-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
b1fc83ca43 |
perf hist: Implement output fields for mem stats
This is a preparation for later changes to support mem_stat output. The new fields will need two lines for the header - the first line will show type of mem stat and the second line will show the name of each item which is returned by mem_stat_name(). Each element in the mem_stat array will be printed in percentage for the hist_entry and their sum would be 100%. Add new output field dimension only for SORT_MODE__MEM using mem_stat. To handle possible name conflict with existing sort keys, move the order of checking output field dimensions after the sort dimensions when it looks for sort keys. Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Link: https://lore.kernel.org/r/20250430205548.789750-7-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Namhyung Kim
|
9fcb43e27c |
perf hist: Basic support for mem_stat accounting
Add a logic to account he->mem_stat based on mem_stat_type in hists. Each mem_stat entry will have different meaning based on the type so the index in the array is calculated at runtime using the corresponding value in the sample.data_src. Still hists has no mem_stat_types yet so this code won't work for now. Later hists->mem_stat_types will be allocated based on what users want in the output actually. Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Link: https://lore.kernel.org/r/20250430205548.789750-6-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Namhyung Kim
|
930d4c45c6 |
perf hist: Add struct he_mem_stat
The 'struct he_mem_stat' is to save detailed information about memory instruction. It'll be used to show breakdown of various data from PERF_SAMPLE_DATA_SRC. Note that this structure is generic and the contents will be different depending on actual data it'll use later. The information about the actual data will be saved in 'struct hists' and its length is in nr_mem_stats. This commit just adds ground works and does nothing since hists->nr_mem_stats is 0 for now. Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Link: https://lore.kernel.org/r/20250430205548.789750-5-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Namhyung Kim
|
29e6392ec3 |
perf hist: Support multi-line header
This is a preparation to support multi-line headers in 'perf mem report'. Normal sort keys and output fields that don't have contents for multi- line will print the header string at the last line only. As we don't use multi-line headers normally, it should not have any changes in the output. Signed-off-by: Namhyung Kim <namhyung@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Leo Yan <leo.yan@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Ravi Bangoria <ravi.bangoria@amd.com> Link: https://lore.kernel.org/r/20250430205548.789750-4-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Namhyung Kim
|
43a6446998 |
perf record: Add --sample-mem-info option
There's no way to enable PERF_SAMPLE_DATA_SRC without PERF_SAMPLE_ADDR
which brings a lot of overhead due to the number of MMAP[2] records.
Let's add a new option to enable this information separately.
Committer testing:
# perf record -a --sample-mem-info
^C[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 1.815 MB perf.data (2637 samples) ]
#
# perf evlist -v
cycles:P: type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0 (PERF_COUNT_HW_CPU_CYCLES), { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|CPU|PERIOD|IDENTIFIER|DATA_SRC, read_format: ID|LOST, disabled: 1, freq: 1, precise_ip: 2, sample_id_all: 1
dummy:u: type: 1 (PERF_TYPE_SOFTWARE), size: 136, config: 0x9 (PERF_COUNT_SW_DUMMY), { sample_period, sample_freq }: 1, sample_type: IP|TID|TIME|CPU|IDENTIFIER|DATA_SRC, read_format: ID|LOST, exclude_kernel: 1, exclude_hv: 1, mmap: 1, comm: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1
#
# perf report -D |& grep -w PERF_RECORD_SAMPLE -A3 -m1
0 44675164447282 0x1a7590 [0x40]: PERF_RECORD_SAMPLE(IP, 0x4001): 107299/107299: 0xffffffffac4a5e11 period: 144 addr: 0
. data_src: 0x229080142
... thread: perf:107299
...... dso: /lib/modules/6.15.0-rc4+/build/vmlinux
#
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Leo Yan <leo.yan@arm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ravi Bangoria <ravi.bangoria@amd.com>
Link: https://lore.kernel.org/r/20250430205548.789750-3-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
3761e7fe98 |
perf hist: Remove output field from sort-list properly
When it removes an output format for cancelled children or latency, it
should delete itself from the sort list as well. Otherwise assertion
in fmt_free() will fire.
$ perf report -H --stdio
perf: ui/hist.c:603: fmt_free: Assertion `!(!list_empty(&fmt->sort_list))' failed.
Aborted (core dumped)
Also convert to perf_hpp__column_unregister() for the same open codes.
Committer notes:
Before this patch:
# perf test hierarchy
83: perf report --hierarchy : FAILED!
# perf test -v hierarchy
--- start ---
test child forked, pid 102242
perf report --hierarchy
Linux
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.025 MB /tmp/perf-test-report.HX0N85TlPq/perf-report-hierarchy-perf.data (6 samples) ]
perf: ui/hist.c:603: fmt_free: Assertion `!(!list_empty(&fmt->sort_list))' failed.
/home/acme/libexec/perf-core/tests/shell/perf-report-hierarchy.sh: line 34: 102250 Aborted (core dumped) perf report --hierarchy > /dev/null
--- Cleaning up ---
---- end(-1) ----
83: perf report --hierarchy : FAILED!
#
After:
# perf test hierarchy
83: perf report --hierarchy : Ok
#
Fixes:
|
||
|
Arnaldo Carvalho de Melo
|
bb5ae52e53 |
perf test perf-report-hierarchy: Add new test
Super simple test to check that at least we're not segfaulting when trying to use 'perf report --hierarchy', more subtests should be added to make sure the output is the expected one. This is being merged right before a fix for that that this test detects: # perf test hierarchy 83: perf report --hierarchy : FAILED! # perf test -v hierarchy --- start --- test child forked, pid 102242 perf report --hierarchy Linux [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 0.025 MB /tmp/perf-test-report.HX0N85TlPq/perf-report-hierarchy-perf.data (6 samples) ] perf: ui/hist.c:603: fmt_free: Assertion `!(!list_empty(&fmt->sort_list))' failed. /home/acme/libexec/perf-core/tests/shell/perf-report-hierarchy.sh: line 34: 102250 Aborted (core dumped) perf report --hierarchy > /dev/null --- Cleaning up --- ---- end(-1) ---- 83: perf report --hierarchy : FAILED! # Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Link: https://lore.kernel.org/lkml/20250430180321.736939-1-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ravi Bangoria
|
35db59fa8e |
perf test amd ibs: Add sample period unit test
IBS Fetch and IBS Op PMUs has various constraints on supported sample
periods. Add perf unit tests to test those.
Running it in parallel with other tests causes intermittent failures.
Mark it exclusive to force it to run sequentially. Sample output on a
Zen5 machine:
Without kernel fixes:
$ sudo ./perf test -vv 112
112: AMD IBS sample period:
--- start ---
test child forked, pid 8774
Using CPUID AuthenticAMD-26-2-1
IBS config tests:
-----------------
Fetch PMU tests:
0xffff : Ok (nr samples: 1078)
0x1000 : Ok (nr samples: 17030)
0xff : Ok (nr samples: 41068)
0x1 : Ok (nr samples: 40543)
0x0 : Ok
0x10000 : Ok
Op PMU tests:
0x0 : Ok
0x1 : Fail
0x8 : Fail
0x9 : Ok (nr samples: 40543)
0xf : Ok (nr samples: 40543)
0x1000 : Ok (nr samples: 18736)
0xffff : Ok (nr samples: 1168)
0x10000 : Ok
0x100000 : Fail (nr samples: 14)
0xf00000 : Fail (nr samples: 1)
0xf0ffff : Fail (nr samples: 1)
0x1f0ffff : Fail (nr samples: 1)
0x7f0ffff : Fail (nr samples: 0)
0x8f0ffff : Ok
0x17f0ffff : Ok
IBS sample period constraint tests:
-----------------------------------
Fetch PMU test:
freq 0, sample_freq 0: Ok
freq 0, sample_freq 1: Fail
freq 0, sample_freq 15: Fail
freq 0, sample_freq 16: Ok (nr samples: 1604)
freq 0, sample_freq 17: Ok (nr samples: 1604)
freq 0, sample_freq 143: Ok (nr samples: 1604)
freq 0, sample_freq 144: Ok (nr samples: 1604)
freq 0, sample_freq 145: Ok (nr samples: 1604)
freq 0, sample_freq 1234: Ok (nr samples: 1566)
freq 0, sample_freq 4103: Ok (nr samples: 1119)
freq 0, sample_freq 65520: Ok (nr samples: 2264)
freq 0, sample_freq 65535: Ok (nr samples: 2263)
freq 0, sample_freq 65552: Ok (nr samples: 1166)
freq 0, sample_freq 8388607: Ok (nr samples: 268)
freq 0, sample_freq 268435455: Ok (nr samples: 8)
freq 1, sample_freq 0: Ok
freq 1, sample_freq 1: Ok (nr samples: 4)
freq 1, sample_freq 15: Ok (nr samples: 4)
freq 1, sample_freq 16: Ok (nr samples: 4)
freq 1, sample_freq 17: Ok (nr samples: 4)
freq 1, sample_freq 143: Ok (nr samples: 5)
freq 1, sample_freq 144: Ok (nr samples: 5)
freq 1, sample_freq 145: Ok (nr samples: 5)
freq 1, sample_freq 1234: Ok (nr samples: 7)
freq 1, sample_freq 4103: Ok (nr samples: 35)
freq 1, sample_freq 65520: Ok (nr samples: 642)
freq 1, sample_freq 65535: Ok (nr samples: 636)
freq 1, sample_freq 65552: Ok (nr samples: 651)
freq 1, sample_freq 8388607: Ok
Op PMU test:
freq 0, sample_freq 0: Ok
freq 0, sample_freq 1: Fail
freq 0, sample_freq 15: Fail
freq 0, sample_freq 16: Fail
freq 0, sample_freq 17: Fail
freq 0, sample_freq 143: Fail
freq 0, sample_freq 144: Ok (nr samples: 1604)
freq 0, sample_freq 145: Ok (nr samples: 1604)
freq 0, sample_freq 1234: Ok (nr samples: 1604)
freq 0, sample_freq 4103: Ok (nr samples: 1604)
freq 0, sample_freq 65520: Ok (nr samples: 2227)
freq 0, sample_freq 65535: Ok (nr samples: 2296)
freq 0, sample_freq 65552: Ok (nr samples: 2213)
freq 0, sample_freq 8388607: Ok (nr samples: 250)
freq 0, sample_freq 268435455: Ok (nr samples: 8)
freq 1, sample_freq 0: Ok
freq 1, sample_freq 1: Fail (nr samples: 4)
freq 1, sample_freq 15: Fail (nr samples: 4)
freq 1, sample_freq 16: Fail (nr samples: 4)
freq 1, sample_freq 17: Fail (nr samples: 4)
freq 1, sample_freq 143: Fail (nr samples: 5)
freq 1, sample_freq 144: Fail (nr samples: 5)
freq 1, sample_freq 145: Fail (nr samples: 5)
freq 1, sample_freq 1234: Fail (nr samples: 8)
freq 1, sample_freq 4103: Fail (nr samples: 33)
freq 1, sample_freq 65520: Fail (nr samples: 546)
freq 1, sample_freq 65535: Fail (nr samples: 544)
freq 1, sample_freq 65552: Fail (nr samples: 555)
freq 1, sample_freq 8388607: Ok
IBS ioctl() tests:
------------------
Fetch PMU tests
ioctl(period = 0x0 ): Ok
ioctl(period = 0x1 ): Fail
ioctl(period = 0xf ): Fail
ioctl(period = 0x10 ): Ok
ioctl(period = 0x11 ): Fail
ioctl(period = 0x1f ): Fail
ioctl(period = 0x20 ): Ok
ioctl(period = 0x80 ): Ok
ioctl(period = 0x8f ): Fail
ioctl(period = 0x90 ): Ok
ioctl(period = 0x91 ): Fail
ioctl(period = 0x100 ): Ok
ioctl(period = 0xfff0 ): Ok
ioctl(period = 0xffff ): Fail
ioctl(period = 0x10000 ): Ok
ioctl(period = 0x1fff0 ): Ok
ioctl(period = 0x1fff5 ): Fail
ioctl(freq = 0x0 ): Ok
ioctl(freq = 0x1 ): Ok
ioctl(freq = 0xf ): Ok
ioctl(freq = 0x10 ): Ok
ioctl(freq = 0x11 ): Ok
ioctl(freq = 0x1f ): Ok
ioctl(freq = 0x20 ): Ok
ioctl(freq = 0x80 ): Ok
ioctl(freq = 0x8f ): Ok
ioctl(freq = 0x90 ): Ok
ioctl(freq = 0x91 ): Ok
ioctl(freq = 0x100 ): Ok
Op PMU tests
ioctl(period = 0x0 ): Ok
ioctl(period = 0x1 ): Fail
ioctl(period = 0xf ): Fail
ioctl(period = 0x10 ): Fail
ioctl(period = 0x11 ): Fail
ioctl(period = 0x1f ): Fail
ioctl(period = 0x20 ): Fail
ioctl(period = 0x80 ): Fail
ioctl(period = 0x8f ): Fail
ioctl(period = 0x90 ): Ok
ioctl(period = 0x91 ): Fail
ioctl(period = 0x100 ): Ok
ioctl(period = 0xfff0 ): Ok
ioctl(period = 0xffff ): Fail
ioctl(period = 0x10000 ): Ok
ioctl(period = 0x1fff0 ): Ok
ioctl(period = 0x1fff5 ): Fail
ioctl(freq = 0x0 ): Ok
ioctl(freq = 0x1 ): Ok
ioctl(freq = 0xf ): Ok
ioctl(freq = 0x10 ): Ok
ioctl(freq = 0x11 ): Ok
ioctl(freq = 0x1f ): Ok
ioctl(freq = 0x20 ): Ok
ioctl(freq = 0x80 ): Ok
ioctl(freq = 0x8f ): Ok
ioctl(freq = 0x90 ): Ok
ioctl(freq = 0x91 ): Ok
ioctl(freq = 0x100 ): Ok
IBS freq (negative) tests:
--------------------------
freq 1, sample_freq 200000: Fail
IBS L3MissOnly test: (takes a while)
--------------------
Fetch L3MissOnly: Fail (nr_samples: 1213)
Op L3MissOnly: Ok (nr_samples: 1193)
---- end(-1) ----
112: AMD IBS sample period : FAILED!
With kernel fixes:
$ sudo ./perf test -vv 112
112: AMD IBS sample period:
--- start ---
test child forked, pid 6939
Using CPUID AuthenticAMD-26-2-1
IBS config tests:
-----------------
Fetch PMU tests:
0xffff : Ok (nr samples: 969)
0x1000 : Ok (nr samples: 15540)
0xff : Ok (nr samples: 40555)
0x1 : Ok (nr samples: 40543)
0x0 : Ok
0x10000 : Ok
Op PMU tests:
0x0 : Ok
0x1 : Ok
0x8 : Ok
0x9 : Ok (nr samples: 40543)
0xf : Ok (nr samples: 40543)
0x1000 : Ok (nr samples: 19156)
0xffff : Ok (nr samples: 1169)
0x10000 : Ok
0x100000 : Ok (nr samples: 1151)
0xf00000 : Ok (nr samples: 76)
0xf0ffff : Ok (nr samples: 73)
0x1f0ffff : Ok (nr samples: 33)
0x7f0ffff : Ok (nr samples: 10)
0x8f0ffff : Ok
0x17f0ffff : Ok
IBS sample period constraint tests:
-----------------------------------
Fetch PMU test:
freq 0, sample_freq 0: Ok
freq 0, sample_freq 1: Ok
freq 0, sample_freq 15: Ok
freq 0, sample_freq 16: Ok (nr samples: 1203)
freq 0, sample_freq 17: Ok (nr samples: 1604)
freq 0, sample_freq 143: Ok (nr samples: 1604)
freq 0, sample_freq 144: Ok (nr samples: 1604)
freq 0, sample_freq 145: Ok (nr samples: 1604)
freq 0, sample_freq 1234: Ok (nr samples: 1604)
freq 0, sample_freq 4103: Ok (nr samples: 1343)
freq 0, sample_freq 65520: Ok (nr samples: 2254)
freq 0, sample_freq 65535: Ok (nr samples: 2136)
freq 0, sample_freq 65552: Ok (nr samples: 1158)
freq 0, sample_freq 8388607: Ok (nr samples: 257)
freq 0, sample_freq 268435455: Ok (nr samples: 8)
freq 1, sample_freq 0: Ok
freq 1, sample_freq 1: Ok (nr samples: 4)
freq 1, sample_freq 15: Ok (nr samples: 4)
freq 1, sample_freq 16: Ok (nr samples: 4)
freq 1, sample_freq 17: Ok (nr samples: 4)
freq 1, sample_freq 143: Ok (nr samples: 5)
freq 1, sample_freq 144: Ok (nr samples: 5)
freq 1, sample_freq 145: Ok (nr samples: 5)
freq 1, sample_freq 1234: Ok (nr samples: 8)
freq 1, sample_freq 4103: Ok (nr samples: 34)
freq 1, sample_freq 65520: Ok (nr samples: 458)
freq 1, sample_freq 65535: Ok (nr samples: 628)
freq 1, sample_freq 65552: Ok (nr samples: 396)
freq 1, sample_freq 8388607: Ok
Op PMU test:
freq 0, sample_freq 0: Ok
freq 0, sample_freq 1: Ok
freq 0, sample_freq 15: Ok
freq 0, sample_freq 16: Ok
freq 0, sample_freq 17: Ok
freq 0, sample_freq 143: Ok
freq 0, sample_freq 144: Ok (nr samples: 1604)
freq 0, sample_freq 145: Ok (nr samples: 1604)
freq 0, sample_freq 1234: Ok (nr samples: 1604)
freq 0, sample_freq 4103: Ok (nr samples: 1604)
freq 0, sample_freq 65520: Ok (nr samples: 2250)
freq 0, sample_freq 65535: Ok (nr samples: 2158)
freq 0, sample_freq 65552: Ok (nr samples: 2296)
freq 0, sample_freq 8388607: Ok (nr samples: 243)
freq 0, sample_freq 268435455: Ok (nr samples: 6)
freq 1, sample_freq 0: Ok
freq 1, sample_freq 1: Ok (nr samples: 4)
freq 1, sample_freq 15: Ok (nr samples: 4)
freq 1, sample_freq 16: Ok (nr samples: 4)
freq 1, sample_freq 17: Ok (nr samples: 4)
freq 1, sample_freq 143: Ok (nr samples: 4)
freq 1, sample_freq 144: Ok (nr samples: 5)
freq 1, sample_freq 145: Ok (nr samples: 4)
freq 1, sample_freq 1234: Ok (nr samples: 6)
freq 1, sample_freq 4103: Ok (nr samples: 27)
freq 1, sample_freq 65520: Ok (nr samples: 542)
freq 1, sample_freq 65535: Ok (nr samples: 550)
freq 1, sample_freq 65552: Ok (nr samples: 552)
freq 1, sample_freq 8388607: Ok
IBS ioctl() tests:
------------------
Fetch PMU tests
ioctl(period = 0x0 ): Ok
ioctl(period = 0x1 ): Ok
ioctl(period = 0xf ): Ok
ioctl(period = 0x10 ): Ok
ioctl(period = 0x11 ): Ok
ioctl(period = 0x1f ): Ok
ioctl(period = 0x20 ): Ok
ioctl(period = 0x80 ): Ok
ioctl(period = 0x8f ): Ok
ioctl(period = 0x90 ): Ok
ioctl(period = 0x91 ): Ok
ioctl(period = 0x100 ): Ok
ioctl(period = 0xfff0 ): Ok
ioctl(period = 0xffff ): Ok
ioctl(period = 0x10000 ): Ok
ioctl(period = 0x1fff0 ): Ok
ioctl(period = 0x1fff5 ): Ok
ioctl(freq = 0x0 ): Ok
ioctl(freq = 0x1 ): Ok
ioctl(freq = 0xf ): Ok
ioctl(freq = 0x10 ): Ok
ioctl(freq = 0x11 ): Ok
ioctl(freq = 0x1f ): Ok
ioctl(freq = 0x20 ): Ok
ioctl(freq = 0x80 ): Ok
ioctl(freq = 0x8f ): Ok
ioctl(freq = 0x90 ): Ok
ioctl(freq = 0x91 ): Ok
ioctl(freq = 0x100 ): Ok
Op PMU tests
ioctl(period = 0x0 ): Ok
ioctl(period = 0x1 ): Ok
ioctl(period = 0xf ): Ok
ioctl(period = 0x10 ): Ok
ioctl(period = 0x11 ): Ok
ioctl(period = 0x1f ): Ok
ioctl(period = 0x20 ): Ok
ioctl(period = 0x80 ): Ok
ioctl(period = 0x8f ): Ok
ioctl(period = 0x90 ): Ok
ioctl(period = 0x91 ): Ok
ioctl(period = 0x100 ): Ok
ioctl(period = 0xfff0 ): Ok
ioctl(period = 0xffff ): Ok
ioctl(period = 0x10000 ): Ok
ioctl(period = 0x1fff0 ): Ok
ioctl(period = 0x1fff5 ): Ok
ioctl(freq = 0x0 ): Ok
ioctl(freq = 0x1 ): Ok
ioctl(freq = 0xf ): Ok
ioctl(freq = 0x10 ): Ok
ioctl(freq = 0x11 ): Ok
ioctl(freq = 0x1f ): Ok
ioctl(freq = 0x20 ): Ok
ioctl(freq = 0x80 ): Ok
ioctl(freq = 0x8f ): Ok
ioctl(freq = 0x90 ): Ok
ioctl(freq = 0x91 ): Ok
ioctl(freq = 0x100 ): Ok
IBS freq (negative) tests:
--------------------------
freq 1, sample_freq 200000: Ok
IBS L3MissOnly test: (takes a while)
--------------------
Fetch L3MissOnly: Ok (nr_samples: 1301)
Op L3MissOnly: Ok (nr_samples: 1590)
---- end(0) ----
112: AMD IBS sample period : Ok
Committer notes:
Avoid using PAGE_SIZE as that define is also in sys/user.h
Make it a variable not to call sysconf() multiple times.
Also cast func to void * when passing it as the first arg to memcpy to
avoid this with some versions of clang:
arch/x86/tests/amd-ibs-period.c:81:3: error: no matching function for call to 'memcpy'
memcpy(func, insn1, sizeof(insn1));
^~~~~~
/usr/include/string.h:27:7: note: candidate function not viable: no known conversion from 'int (*)(void)' to 'void *' for 1st argument
void *memcpy (void *__restrict, const void *__restrict, size_t);
^
/usr/include/fortify/string.h:40:27: note: candidate function not viable: no known conversion from 'int (*)(void)' to 'void *const' for 1st argument
_FORTIFY_FN(memcpy) void *memcpy(void * _FORTIFY_POS0 __od,
^
arch/x86/tests/amd-ibs-period.c:87:3: error: no matching function for call to 'memcpy'
This one, for instance:
Alpine clang version 19.1.4
Target: x86_64-alpine-linux-musl
Thread model: posix
InstalledDir: /usr/lib/llvm19/bin
Configuration file: /etc/clang19/x86_64-alpine-linux-musl.cfg
System configuration file directory: /etc/clang19
Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Ananth Narayan <ananth.narayan@amd.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Joe Mario <jmario@redhat.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sandipan Das <sandipan.das@amd.com>
Cc: Santosh Shukla <santosh.shukla@amd.com>
Cc: Stephane Eranian <eranian@google.com>
Link: https://lore.kernel.org/r/20250429035938.1301-5-ravi.bangoria@amd.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Ravi Bangoria
|
fa1332a801 |
perf mem/c2c amd: Add ldlat support
'perf mem/c2c' uses IBS Op PMU on AMD platforms. IBS Op PMU on Zen5 uarch has added support for Load Latency filtering. Implement 'perf mem/c2c' --ldlat using IBS Op Load Latency filtering capability. Some subtle differences between AMD and other arch: o --ldlat is disabled by default on AMD o Supported values are 128 to 2048. Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com> Cc: Ananth Narayan <ananth.narayan@amd.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Joe Mario <jmario@redhat.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sandipan Das <sandipan.das@amd.com> Cc: Santosh Shukla <santosh.shukla@amd.com> Cc: Stephane Eranian <eranian@google.com> Link: https://lore.kernel.org/r/20250429035938.1301-4-ravi.bangoria@amd.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ravi Bangoria
|
fc481adc97 |
perf amd ibs: Incorporate Zen5 DTLB and PageSize information
IBS Op PMU on Zen5 reports DTLB and page size information differently
compared to prior generation.
IBS_OP_DATA3 Zen3/4 Zen5
----------------------------------------------------------------
19 IbsDcL2TlbHit1G Reserved
----------------------------------------------------------------
6 IbsDcL2tlbHit2M Reserved
----------------------------------------------------------------
5 IbsDcL1TlbHit1G PageSize:
4 IbsDcL1TlbHit2M 0 - 4K
1 - 2M
2 - 1G
3 - Reserved
Valid only if
IbsDcPhyAddrValid = 1
----------------------------------------------------------------
3 IbsDcL2TlbMiss IbsDcL2TlbMiss
Valid only if
IbsDcPhyAddrValid = 1
----------------------------------------------------------------
2 IbsDcL1tlbMiss IbsDcL1tlbMiss
Valid only if
IbsDcPhyAddrValid = 1
----------------------------------------------------------------
Kernel expose this change as "dtlb_pgsize" capability in PMU sysfs.
Change IBS register raw-dump logic according to new bit definitions.
Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com>
Cc: Ananth Narayan <ananth.narayan@amd.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Joe Mario <jmario@redhat.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Sandipan Das <sandipan.das@amd.com>
Cc: Santosh Shukla <santosh.shukla@amd.com>
Cc: Stephane Eranian <eranian@google.com>
Link: https://lore.kernel.org/r/20250429035938.1301-3-ravi.bangoria@amd.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Ravi Bangoria
|
eeefc13c71 |
perf amd ibs: Add Load Latency bits in raw dump
IBS OP PMU on Zen5 supports Load Latency filtering. Decode and dump Load Latency filtering related bits into perf script raw dump. Also add oneliner example in the perf-amd-ibs man page. Signed-off-by: Ravi Bangoria <ravi.bangoria@amd.com> Cc: Ananth Narayan <ananth.narayan@amd.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Joe Mario <jmario@redhat.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Sandipan Das <sandipan.das@amd.com> Cc: Santosh Shukla <santosh.shukla@amd.com> Cc: Stephane Eranian <eranian@google.com> Link: https://lore.kernel.org/r/20250429035938.1301-2-ravi.bangoria@amd.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Arnaldo Carvalho de Melo
|
4d728bb93b |
perf symbols: Handle 'u' and 'l' symbols in /proc/kallsyms
I started seeing this in recent Fedora 42 kernels: # uname -a Linux number 6.14.3-300.fc42.x86_64 #1 SMP PREEMPT_DYNAMIC Sun Apr 20 16:08:39 UTC 2025 x86_64 GNU/Linux # # perf test vmlinux 1: vmlinux symtab matches kallsyms : FAILED! # Where we have Rust enabled: # grep CONFIG_RUST /boot/config-6.14.3-300.fc42.x86_64 CONFIG_RUSTC_VERSION=108600 CONFIG_RUST_IS_AVAILABLE=y CONFIG_RUSTC_LLVM_VERSION=200101 CONFIG_RUSTC_HAS_COERCE_POINTEE=y CONFIG_RUST=y CONFIG_RUSTC_VERSION_TEXT="rustc 1.86.0 (05f9846f8 2025-03-31) (Fedora 1.86.0-1.fc42)" CONFIG_RUST_FW_LOADER_ABSTRACTIONS=y CONFIG_RUST_PHYLIB_ABSTRACTIONS=y # CONFIG_RUST_DEBUG_ASSERTIONS is not set CONFIG_RUST_OVERFLOW_CHECKS=y # CONFIG_RUST_BUILD_ASSERT_ALLOW is not set # Looking at the reason for the failure: # perf test -v vmlinux |& grep ^ERR ERR : 0xffffffff99efc7d0: __pfx__RNCINvNtNtNtCsf5tcb0XGUW4_4core4iter8adapters3map12map_try_foldjNtCsagR6JbSOIa9_12drm_panic_qr7VersionuINtNtNtBa_3ops12control_flow11ControlFlowB10_ENcB10_0NCINvNvNtNtNtB8_6traits8iterator8Iterator4find5checkB10_NCNvMB12_B10_13from_segments0E0E0B12_ not on kallsyms ERR : 0xffffffff99efc7e0: _RNCINvNtNtNtCsf5tcb0XGUW4_4core4iter8adapters3map12map_try_foldjNtCsagR6JbSOIa9_12drm_panic_qr7VersionuINtNtNtBa_3ops12control_flow11ControlFlowB10_ENcB10_0NCINvNvNtNtNtB8_6traits8iterator8Iterator4find5checkB10_NCNvMB12_B10_13from_segments0E0E0B12_ not on kallsyms # But: # grep -w u /proc/kallsyms ffffffff99efc7d0 u __pfx__RNCINvNtNtNtCsf5tcb0XGUW4_4core4iter8adapters3map12map_try_foldjNtCsagR6JbSOIa9_12drm_panic_qr7VersionuINtNtNtBa_3ops12control_flow11ControlFlowB10_ENcB10_0NCINvNvNtNtNtB8_6traits8iterator8Iterator4find5checkB10_NCNvMB12_B10_13from_segments0E0E0B12_ ffffffff99efc7e0 u _RNCINvNtNtNtCsf5tcb0XGUW4_4core4iter8adapters3map12map_try_foldjNtCsagR6JbSOIa9_12drm_panic_qr7VersionuINtNtNtBa_3ops12control_flow11ControlFlowB10_ENcB10_0NCINvNvNtNtNtB8_6traits8iterator8Iterator4find5checkB10_NCNvMB12_B10_13from_segments0E0E0B12_ # The test checks that "vmlinux symtab matches kallsyms", so it finds those two symbols in vmlinux: # pahole --running_kernel_vmlinux /usr/lib/debug/lib/modules/6.14.3-300.fc42.x86_64/vmlinux # # readelf -sW /usr/lib/debug/lib/modules/6.14.3-300.fc42.x86_64/vmlinux | grep -Ew '(__pfx__RNCINvNtNtNtCsf5tcb0XGUW4_4core4iter8adapters3map12map_try_foldjNtCsagR6JbSOIa9_12drm_panic_qr7VersionuINtNtNtBa_3ops12control_flow11ControlFlowB10_ENcB10_0NCINvNvNtNtNtB8_6traits8iterator8Iterator4find5checkB10_NCNvMB12_B10_13from_segments0E0E0B12_|_RNCINvNtNtNtCsf5tcb0XGUW4_4core4iter8adapters3map12map_try_foldjNtCsagR6JbSOIa9_12drm_panic_qr7VersionuINtNtNtBa_3ops12control_flow11ControlFlowB10_ENcB10_0NCINvNvNtNtNtB8_6traits8iterator8Iterator4find5checkB10_NCNvMB12_B10_13from_segments0E0E0B12_)' 81844: ffffffff81efc7e0 524 FUNC LOCAL DEFAULT 1 _RNCINvNtNtNtCsf5tcb0XGUW4_4core4iter8adapters3map12map_try_foldjNtCsagR6JbSOIa9_12drm_panic_qr7VersionuINtNtNtBa_3ops12control_flow11ControlFlowB10_ENcB10_0NCINvNvNtNtNtB8_6traits8iterator8Iterator4find5checkB10_NCNvMB12_B10_13from_segments0E0E0B12_ 144259: ffffffff81efc7d0 16 FUNC LOCAL DEFAULT 1 __pfx__RNCINvNtNtNtCsf5tcb0XGUW4_4core4iter8adapters3map12map_try_foldjNtCsagR6JbSOIa9_12drm_panic_qr7VersionuINtNtNtBa_3ops12control_flow11ControlFlowB10_ENcB10_0NCINvNvNtNtNtB8_6traits8iterator8Iterator4find5checkB10_NCNvMB12_B10_13from_segments0E0E0B12_ # It is there. From the nm documentation we can see that: "U" The symbol is undefined. "u" The symbol is a unique global symbol. This is a GNU extension to the standard set of ELF symbol bindings. For such a symbol the dynamic linker will make sure that in the entire process there is just one symbol with this name and type in use. So lets consider 'u' symbols in /proc/kallsyms when loading it to cover this case. Fedora:40 shows this as a 'l' symbol, so consider that as well. With this patch 'perf test 1' is happy again: # perf test vmlinux 1: vmlinux symtab matches kallsyms : Ok # Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Masahiro Yamada <masahiroy@kernel.org> Cc: Miguel Ojeda <ojeda@kernel.org> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Namhyung Kim <namhyung@kernel.org> Link: https://lore.kernel.org/r/aBE_n0PGl3g6h-cS@x1 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Jakub Brnak
|
92b664dcef |
perf test probe_vfs_getname: Skip if no suitable line detected
In some cases when calling function add_probe_vfs_getname, line number
can't be detected by 'perf probe -L getname_flags':
78 atomic_set(&result->refcnt, 1);
// one of the following lines should have line number
// but sometimes it does not because of optimization
result->uptr = filename;
result->aname = NULL;
81 audit_getname(result);
To prevent false failures, skip the affected tests if no suitable line
numbers can be detected.
Signed-off-by: Jakub Brnak <jbrnak@redhat.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Michael Petlan <mpetlan@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Tomas Glozar <tglozar@redhat.com>
Link: https://lore.kernel.org/r/20250324144523.597557-1-jbrnak@redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
13f35928a4 |
perf lock contention: Symbolize zone->lock using BTF
The struct zone is embedded in struct pglist_data which can be allocated
for each NUMA node early in the boot process. As it's not a slab object
nor a global lock, this was not symbolized.
Since the zone->lock is often contended, it'd be nice if we can
symbolize it. On NUMA systems, node_data array will have pointers for
struct pglist_data. By following the pointer, it can calculate the
address of each zone and its lock using BTF. On UMA, it can just use
contig_page_data and its zones.
The following example shows the zone lock contention at the end.
$ sudo ./perf lock con -abl -E 5 -- ./perf bench sched messaging
# Running 'sched/messaging' benchmark:
# 20 sender and receiver processes per group
# 10 groups == 400 processes run
Total time: 0.038 [sec]
contended total wait max wait avg wait address symbol
5167 18.17 ms 10.27 us 3.52 us ffff953340052d00 &kmem_cache_node (spinlock)
38 11.75 ms 465.49 us 309.13 us ffff95334060c480 &sock_inode_cache (spinlock)
3916 10.13 ms 10.43 us 2.59 us ffff953342aecb40 &kmem_cache_node (spinlock)
2963 10.02 ms 13.75 us 3.38 us ffff9533d2344098 &kmalloc-rnd-08-2k (spinlock)
216 5.05 ms 99.49 us 23.39 us ffff9542bf7d65d0 zone_lock (spinlock)
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Song Liu <song@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: bpf@vger.kernel.org
Cc: linux-mm@kvack.org
Link: https://lore.kernel.org/r/20250401063055.7431-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
2d099ccaad |
perf test: Add perf trace summary test
$ sudo ./perf test -vv 'trace summary' 109: perf trace summary: --- start --- test child forked, pid 3501572 testing: perf trace -s -- true testing: perf trace -S -- true testing: perf trace -s --summary-mode=thread -- true testing: perf trace -S --summary-mode=total -- true testing: perf trace -as --summary-mode=thread --no-bpf-summary -- true testing: perf trace -as --summary-mode=total --no-bpf-summary -- true testing: perf trace -as --summary-mode=thread --bpf-summary -- true testing: perf trace -as --summary-mode=total --bpf-summary -- true testing: perf trace -aS --summary-mode=total --bpf-summary -- true ---- end(0) ---- 109: perf trace summary : Ok Reviewed-by: Howard Chu <howardchu95@gmail.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Song Liu <song@kernel.org> Cc: bpf@vger.kernel.org Link: https://lore.kernel.org/r/20250326044001.3503432-2-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Namhyung Kim
|
1bec43f523 |
perf trace: Implement syscall summary in BPF
When -s/--summary option is used, it doesn't need (augmented) arguments
of syscalls. Let's skip the augmentation and load another small BPF
program to collect the statistics in the kernel instead of copying the
data to the ring-buffer to calculate the stats in userspace. This will
be much more light-weight than the existing approach and remove any lost
events.
Let's add a new option --bpf-summary to control this behavior. I cannot
make it default because there's no way to get e_machine in the BPF which
is needed for detecting different ABIs like 32-bit compat mode.
No functional changes intended except for no more LOST events. :)
$ sudo ./perf trace -as --summary-mode=total --bpf-summary sleep 1
Summary of events:
total, 6194 events
syscall calls errors total min avg max stddev
(msec) (msec) (msec) (msec) (%)
--------------- -------- ------ -------- --------- --------- --------- ------
epoll_wait 561 0 4530.843 0.000 8.076 520.941 18.75%
futex 693 45 4317.231 0.000 6.230 500.077 21.98%
poll 300 0 1040.109 0.000 3.467 120.928 17.02%
clock_nanosleep 1 0 1000.172 1000.172 1000.172 1000.172 0.00%
ppoll 360 0 872.386 0.001 2.423 253.275 41.91%
epoll_pwait 14 0 384.349 0.001 27.453 380.002 98.79%
pselect6 14 0 108.130 7.198 7.724 8.206 0.85%
nanosleep 39 0 43.378 0.069 1.112 10.084 44.23%
...
Reviewed-by: Howard Chu <howardchu95@gmail.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Howard Chu <howardchu95@gmail.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Song Liu <song@kernel.org>
Link: https://lore.kernel.org/r/20250326044001.3503432-1-namhyung@kernel.org
[ Added fixup sent from Namhyung in response to my report to make it also dependent on CONFIG_TRACE ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Junhao He
|
85447f68a1 |
MAINTAINERS: Add hisilicon PMU JSON events under its entry
The all hisilicon PMU JSON events were missing to be listed there. Reviewed-by: James Clark <james.clark@linaro.org> Signed-off-by: Junhao He <hejunhao3@huawei.com> Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Acked-by: Yicong Yang <yangyicong@hisilicon.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: John Garry <john.g.garry@oracle.com> Cc: Leo Yan <leo.yan@arm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Will Deacon <will@kernel.org> Link: https://lore.kernel.org/r/20250418070812.3771441-4-hejunhao3@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Junhao He
|
c756441c35 |
perf vendor events arm64: Drop hip08 PublicDescription if same as BriefDescription
If BriefDescription and PublicDescription are the same, only BriefDescription is needed. It will be used for both long and short format outputs. Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Reviewed-by: Yicong Yang <yangyicong@hisilicon.com> Signed-off-by: Junhao He <hejunhao3@huawei.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: John Garry <john.g.garry@oracle.com> Cc: Leo Yan <leo.yan@arm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Will Deacon <will@kernel.org> Link: https://lore.kernel.org/r/20250418070812.3771441-3-hejunhao3@huawei.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Junhao He
|
43fff3e948 |
perf vendor events arm64: Fill up Desc field for Hisi hip08 hha pmu
In the same PMU, when some JSON events have the "BriefDescription" field
populated while others do not, the cmp_sevent() function will split these
two types of events into separate groups. As a result, when using perf
list to display events, the two types of events cannot be grouped together
in the output.
before patch:
$ perf list pmu
...
uncore hha:
hisi_sccl1_hha2/sdir-hit/
hisi_sccl1_hha2/sdir-lookup/
...
uncore hha:
edir-hit
[Count of The number of HHA E-Dir hit operations. Unit: hisi_sccl1_hha2]
...
after patch:
$ perf list pmu
...
uncore hha:
edir-hit
[Count of The number of HHA E-Dir hit operations. Unit: hisi_sccl1_hha2]
sdir-hit
[Count of The number of HHA S-Dir hit operations. Unit: hisi_sccl1_hha2]
sdir-lookup
[Count of the number of HHA S-Dir lookup operations. Unit: hisi_sccl1_hha2]
...
Reviewed-by: James Clark <james.clark@linaro.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Yicong Yang <yangyicong@hisilicon.com>
Signed-off-by: Junhao He <hejunhao3@huawei.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: John Garry <john.g.garry@oracle.com>
Cc: Leo Yan <leo.yan@arm.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Mike Leach <mike.leach@linaro.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20250418070812.3771441-2-hejunhao3@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Ian Rogers
|
022d270bb6 |
perf bench evlist-open-close: Reduce scope of 2 variables
Make 2 global variables local. Reduces ELF binary size by removing relocations. For a no flags build, the perf binary size is reduced by 4,144 bytes on x86-64. Reviewed-by: Arnaldo Carvalho de Melo <acme@redhat.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Dapeng Mi <dapeng1.mi@linux.intel.com> Cc: Dominique Martinet <asmadeus@codewreck.org> Cc: Dr. David Alan Gilbert <linux@treblig.org> Cc: Hao Ge <gehao@kylinos.cn> Cc: Howard Chu <howardchu95@gmail.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: James Clark <james.clark@linaro.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Levi Yun <yeoreum.yun@arm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Tengda Wu <wutengda@huaweicloud.com> Cc: Thomas Richter <tmricht@linux.ibm.com> Cc: Veronika Molnarova <vmolnaro@redhat.com> Cc: Weilin Wang <weilin.wang@intel.com> Cc: Xu Yang <xu.yang_2@nxp.com> Link: https://lore.kernel.org/r/20250410173631.1713627-3-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
be8aefad33 |
perf tests record: Cleanup improvements
Remove the script output file. Add a trap debug message. Minor style consistency changes. Signed-off-by: Ian Rogers <irogers@google.com> Link: https://lore.kernel.org/r/20250410173631.1713627-2-irogers@google.com Cc: Mark Rutland <mark.rutland@arm.com> Cc: Levi Yun <yeoreum.yun@arm.com> Cc: Dominique Martinet <asmadeus@codewreck.org> Cc: Howard Chu <howardchu95@gmail.com> Cc: Tengda Wu <wutengda@huaweicloud.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Weilin Wang <weilin.wang@intel.com> Cc: Arnaldo Carvalho de Melo <acme@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Hao Ge <gehao@kylinos.cn> Cc: James Clark <james.clark@linaro.org> Cc: Thomas Richter <tmricht@linux.ibm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Dapeng Mi <dapeng1.mi@linux.intel.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Xu Yang <xu.yang_2@nxp.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Veronika Molnarova <vmolnaro@redhat.com> Cc: Dr. David Alan Gilbert <linux@treblig.org> Cc: bpf@vger.kernel.org Cc: linux-kernel@vger.kernel.org Cc: linux-perf-users@vger.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Thomas Richter
|
ccd4b5cdf0 |
perf tests metric-only perf stat: Fix tests 84 and 86 s390
On s390x KVM and z/VM machines the CPU Measurement Facility is not
available. Events cycles and instructions do not exist. Running above
tests on s390 KVM and z/VM guests always fail with this error:
# ./perf test 84 86
84: perf stat JSON output linter : FAILED!
86: perf stat STD output linter : FAILED!
#
Root cause is command:
# perf stat -j --metric-only -e instructions,cycles -- true
{"metric-value" : "none"}
#
Which fails due to unsupported events and returns "none".
Do not execute this test case on s390 KVM and z/VM machines.
Output after:
# ./perf test 84 86
84: perf stat JSON output linter : Ok
86: perf stat STD output linter : Ok
#
Fixes:
|
||
|
Ian Rogers
|
68cb156743 |
perf tool_pmu: Fix aggregation on duration_time
evsel__count_has_error() fails counters when the enabled or running time
are 0. The duration_time event reads 0 when the cpu_map_idx != 0 to
avoid aggregating time over CPUs. Change the enable and running time
to always have a ratio of 100% so that evsel__count_has_error won't
fail.
Before:
```
$ sudo /tmp/perf/perf stat --per-core -a -M UNCORE_FREQ sleep 1
Performance counter stats for 'system wide':
S0-D0-C0 1 2,615,819,485 UNC_CLOCK.SOCKET # 2.61 UNCORE_FREQ
S0-D0-C0 2 <not counted> duration_time
1.002111784 seconds time elapsed
```
After:
```
$ perf stat --per-core -a -M UNCORE_FREQ sleep 1
Performance counter stats for 'system wide':
S0-D0-C0 1 758,160,296 UNC_CLOCK.SOCKET # 0.76 UNCORE_FREQ
S0-D0-C0 2 1,003,438,246 duration_time
1.002486017 seconds time elapsed
```
Note: the metric reads the value a different way and isn't impacted.
Fixes:
|
||
|
Chun-Tse Shao
|
b1b26ce8bb |
perf session: Skip unsupported new event types
`perf report` currently halts with an error when encountering unsupported new event types (`event.type >= PERF_RECORD_HEADER_MAX`). This patch modifies the behavior to skip these samples and continue processing the remaining events. Additionally, stops reporting if the new event size is not 8-byte aligned. Suggested-by: Arnaldo Carvalho de Melo <acme@kernel.org> Suggested-by: Namhyung Kim <namhyung@kernel.org> Reviewed-by: Ian Rogers <irogers@google.com> Signed-off-by: Chun-Tse Shao <ctshao@google.com> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Ben Gainey <ben.gainey@arm.com> Cc: Dmitriy Vyukov <dvyukov@google.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20250414173921.2905822-1-ctshao@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Namhyung Kim
|
0ef8091f17 |
perf hist: Allow custom output fields in hierarchy mode
Now it can handle multiple output fields and sort keys in separate
levels, so it should be ok to use it in the hierarchy mode. This
allows fully customized output format.
$ perf report -F latency,comm,parallelism -H --stdio
...
# Latency Command / Parallelism
# ........... .....................
#
31.84% cc1
29.96% 5
1.24% 4
0.37% 6
0.26% 3
0.02% 2
24.68% as
22.39% 5
1.12% 2
0.98% 4
0.12% 3
0.07% 6
...
Committer testing:
Before:
$ perf report -F latency,comm,parallelism -H --stdio
Error: --hierarchy and --fields options cannot be used together
Usage: perf report [<options>]
-F, --fields <key[,keys...]>
output field(s): overhead latency overhead_sys overhead_us
overhead_guest_sys overhead_guest_us overhead_children
latency_children sample period weight1 weight2 weight3
<SNIP>
-H, --hierarchy Show entries in a hierarchy
$
After:
$ perf report -F latency,comm,parallelism -H --stdio
# Total Lost Samples: 0
#
# Samples: 1K of event 'cycles:Pu'
# Event count (approx.): 1581450138
#
# Latency Command / Parallelism
# ........... .....................
#
97.66% git
96.95% 1
0.55% 2
0.04% 5
0.03% 8
0.03% 4
0.02% 3
0.01% 9
0.01% 7
0.01% 6
0.01% 10
0.00% 12
2.34% git-remote-http
2.24% 1
0.07% 5
0.02% 2
0.00% 4
#
# (Tip: To analyze particular parallelism levels, try: perf report --latency --parallelism=32-64)
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20250331073722.4695-5-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Namhyung Kim
|
390627dda7 |
perf hist: Set levels in output_field_add()
It turns out that the output fields didn't consider the hierarchy mode and put all the fields in the same level. To support hierarchy, each non-output field should be in a separate level. Pass a pointer to level to output_field_add() and make it increase the level when it sees non-output fields. Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20250331073722.4695-4-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Namhyung Kim
|
b09124e2e1 |
perf hist: Remove formats in hierarchy when cancel latency
Likewise, it should remove latency output fields in hierarchy list. Pass evlist to perf_hpp__cancel_latency() to handle them properly. Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20250331073722.4695-3-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Namhyung Kim
|
dbd11b6bda |
perf hist: Remove formats in hierarchy when cancel children
This is to support hierarchy options with custom output fields. Currently perf_hpp__cancel_cumulate() only removes accumulated overhead and latency fields from the global perf_hpp_list. This is not used in the hierarchy mode because each evsel's hist has its own separate hpp_list. So it needs to remove the fields from the lists too. Pass evlist to the function so that it can iterate the evsels. Signed-off-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Ian Rogers <irogers@google.com> Cc: Ingo Molnar <mingo@kernel.org> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Link: https://lore.kernel.org/r/20250331073722.4695-2-namhyung@kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
92504d927d |
perf record: Retirement latency cleanup in evsel__config
'perf record' will fail with retirement latency events as the open doesn't do a perf_event_open system call. Use evsel__config() to set up such events for recording by removing the flag and enabling sample weights - the sample weights containing the retirement latency. Reviewed-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Weilin Wang <weilin.wang@intel.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Andreas Färber <afaerber@suse.de> Cc: Caleb Biggers <caleb.biggers@intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: Perry Taylor <perry.taylor@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250414174134.3095492-17-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
fc807b6bde |
perf pmu-events: Add retirement latency to JSON events inside of perf
The updated Intel vendor events add retirement latency for graniterapids: https://lore.kernel.org/lkml/20250322063403.364981-14-irogers@google.com/ This change makes those values available within an alias/event within a PMU and saves them into the evsel at event parse time. When no TPEBS data is available the default values are substituted in for TMA metrics that are using retirement latency events - currently just those on graniterapids. Reviewed-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Weilin Wang <weilin.wang@intel.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Andreas Färber <afaerber@suse.de> Cc: Caleb Biggers <caleb.biggers@intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: Perry Taylor <perry.taylor@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250414174134.3095492-16-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
f19306f065 |
perf stat: Add mean, min, max and last --tpebs-mode options
Add command line configuration option for how retirement latency
events are combined.
The default "mean" gives the average of retirement latency.
"min" or "max" give the smallest or largest retirment latency times
respectively.
"last" uses the last retirment latency sample's time.
Committer notes:
Enclose parse_tpebs_mode() under HAVE_ARCH_X86_64_SUPPORT to match the
ifdef block where it is used, fixing the build in systems like:
20 5.60 debian:experimental-x-mips : FAIL gcc version 14.2.0 (Debian 14.2.0-1)
builtin-stat.c:2330:12: error: 'parse_tpebs_mode' defined but not used [-Werror=unused-function]
2330 | static int parse_tpebs_mode(const struct option *opt, const char *str,
| ^~~~~~~~~~~~~~~~
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Weilin Wang <weilin.wang@intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexandre Torgue <alexandre.torgue@foss.st.com>
Cc: Andreas Färber <afaerber@suse.de>
Cc: Caleb Biggers <caleb.biggers@intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com>
Cc: Perry Taylor <perry.taylor@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Falcon <thomas.falcon@intel.com>
Link: https://lore.kernel.org/r/20250414174134.3095492-15-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Ian Rogers
|
3533b56d22 |
perf intel-tpebs: Use stats for retirement latency statistics
struct stats provides access to mean, min and max. It also provides uniformity with statistics code used elsewhere in perf. Reviewed-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Weilin Wang <weilin.wang@intel.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Andreas Färber <afaerber@suse.de> Cc: Caleb Biggers <caleb.biggers@intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: Perry Taylor <perry.taylor@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250414174134.3095492-14-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
1ddf95f6d8 |
perf intel-tpebs: Don't close record on read
Factor sending record control fd code into its own function. Rather than killing the record process send it a ping when reading. Timeouts were witnessed if done too frequently, so only ping for the first tpebs events. Don't kill the record command send it a stop command. As close isn't reliably called also close on evsel__exit. Add extra checks on the record being terminated to avoid warnings. Adjust the locking as needed and incorporate extra -Wthread-safety checks. Check to do six 500ms poll timeouts when sending commands, rather than the larger 3000ms, to allow the record process terminating to be better witnessed. Reviewed-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Weilin Wang <weilin.wang@intel.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Andreas Färber <afaerber@suse.de> Cc: Caleb Biggers <caleb.biggers@intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: Perry Taylor <perry.taylor@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250414174134.3095492-13-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
8174392049 |
perf intel-tpebs: Add mutex for tpebs_results
Ensure sample reader isn't racing with events being added/removed. Reviewed-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Weilin Wang <weilin.wang@intel.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Andreas Färber <afaerber@suse.de> Cc: Caleb Biggers <caleb.biggers@intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: Perry Taylor <perry.taylor@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250414174134.3095492-12-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
ea61db61d9 |
perf intel-tpebs: Add support for updating counts in evsel__tpebs_read
Rename to reflect evsel argument and for consistency with other tpebs functions. Update count from prev_raw_counts when available. Eventually this will allow inteval mode support. Reviewed-by: Kan Liang <kan.liang@linux.intel.com> Signed-off-by: Ian Rogers <irogers@google.com> Tested-by: Weilin Wang <weilin.wang@intel.com> Acked-by: Namhyung Kim <namhyung@kernel.org> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Alexandre Torgue <alexandre.torgue@foss.st.com> Cc: Andreas Färber <afaerber@suse.de> Cc: Caleb Biggers <caleb.biggers@intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com> Cc: Perry Taylor <perry.taylor@intel.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Falcon <thomas.falcon@intel.com> Link: https://lore.kernel.org/r/20250414174134.3095492-11-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
bb1c0f1b43 |
perf intel-tpebs: Refactor tpebs_results list
evsel names and metric-ids are used for matching but this can be
problematic, for example, multiple occurrences of the same retirement
latency event become a single event for the record.
Change the name of the record events so they are unique and reflect the
evsel of the retirement latency event that opens them (the retirement
latency event's evsel address is embedded within them).
This allows an evsel based close to close the event when the retirement
latency event is closed.
This is important as 'perf stat' has an evlist and the session listen to
the record events has an evlist, knowing which event should remove the
tpebs_retire_lat can't be tied to an evlist list as there is more than
1, so closing which evlist should cause the tpebs to stop?
Using the evsel and the last one out doing the tpebs_stop is cleaner.
Committer notes:
Fix the build on 32-bit systems by using unsigned long when converting
pointers to integers instead of uint64_t. Fixes:
20 4.97 debian:experimental-x-mips : FAIL gcc version 14.2.0 (Debian 14.2.0-13)
util/intel-tpebs.c: In function 'tpebs_retire_lat__find':
util/intel-tpebs.c:377:21: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast]
377 | if ((uint64_t)t->evsel == num)
| ^
cc1: all warnings being treated as errors
Reviewed-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Weilin Wang <weilin.wang@intel.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexandre Torgue <alexandre.torgue@foss.st.com>
Cc: Andreas Färber <afaerber@suse.de>
Cc: Caleb Biggers <caleb.biggers@intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Maxime Coquelin <mcoquelin.stm32@gmail.com>
Cc: Perry Taylor <perry.taylor@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Falcon <thomas.falcon@intel.com>
Link: https://lore.kernel.org/r/20250414174134.3095492-10-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|