tools/perf/builtin-record.c | 26 ++++ tools/perf/builtin-script.c | 5 +- tools/perf/tests/builtin-test.c | 1 + tools/perf/tests/shell/record_offcpu.sh | 27 ++++ tools/perf/tests/tests.h | 1 + tools/perf/tests/workloads/Build | 1 + tools/perf/tests/workloads/offcpu.c | 16 +++ tools/perf/util/bpf_off_cpu.c | 176 +++++++++++++++--------- tools/perf/util/bpf_skel/off_cpu.bpf.c | 135 ++++++++++++++++++ tools/perf/util/evsel.c | 49 ++++--- tools/perf/util/evsel.h | 13 ++ tools/perf/util/header.c | 12 ++ tools/perf/util/off_cpu.h | 10 +- tools/perf/util/record.h | 1 + tools/perf/util/session.c | 16 ++- 15 files changed, 401 insertions(+), 88 deletions(-) create mode 100644 tools/perf/tests/workloads/offcpu.c
Changes in v4:
- Minimize the size of data output by perf_event_output()
- Keep only one off-cpu event
- Change off-cpu threshold's unit to microseconds
- Set a default off-cpu threshold
- Print the correct error message for the field 'embed' in perf data header
Changes in v3:
- Add off-cpu-thresh argument
- Process direct off-cpu samples in post
Changes in v2:
- Remove unnecessary comments.
- Rename function off_cpu_change_type to off_cpu_prepare_parse
v1:
As mentioned in: https://bugzilla.kernel.org/show_bug.cgi?id=207323
Currently, off-cpu samples are dumped when perf record is exiting. This
results in off-cpu samples being after the regular samples. This patch
series makes possible dumping off-cpu samples on-the-fly, directly into
perf ring buffer. And it dispatches those samples to the correct format
for perf.data consumers.
Before:
```
migration/0 21 [000] 27981.041319: 2944637851 cycles:P: ffffffff90d2e8aa record_times+0xa ([kernel.kallsyms])
perf 770116 [001] 27981.041375: 1 cycles:P: ffffffff90ee4960 event_function+0xf0 ([kernel.kallsyms])
perf 770116 [001] 27981.041377: 1 cycles:P: ffffffff90c184b1 intel_bts_enable_local+0x31 ([kernel.kallsyms])
perf 770116 [001] 27981.041379: 51611 cycles:P: ffffffff91a160b0 native_sched_clock+0x30 ([kernel.kallsyms])
migration/1 26 [001] 27981.041400: 4227682775 cycles:P: ffffffff90d06a74 wakeup_preempt+0x44 ([kernel.kallsyms])
migration/2 32 [002] 27981.041477: 4159401534 cycles:P: ffffffff90d11993 update_load_avg+0x63 ([kernel.kallsyms])
sshd 708098 [000] 18446744069.414584: 286392 offcpu-time:
79a864f1c8bb ppoll+0x4b (/usr/lib/libc.so.6)
585690935cca [unknown] (/usr/bin/sshd)
```
After:
```
perf 774767 [003] 28178.033444: 497 cycles:P: ffffffff91a160c3 native_sched_clock+0x43 ([kernel.kallsyms])
perf 774767 [003] 28178.033445: 399440 cycles:P: ffffffff91c01f8d nmi_restore+0x25 ([kernel.kallsyms])
swapper 0 [001] 28178.036639: 376650973 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms])
swapper 0 [003] 28178.182921: 348779378 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms])
blueman-tray 1355 [000] 28178.627906: 100184571 offcpu-time:
7528eef1c39d __poll+0x4d (/usr/lib/libc.so.6)
7528edf7d8fd [unknown] (/usr/lib/libglib-2.0.so.0.8000.2)
7528edf1af95 g_main_context_iteration+0x35 (/usr/lib/libglib-2.0.so.0.8000.2)
7528eda4ab86 g_application_run+0x1f6 (/usr/lib/libgio-2.0.so.0.8000.2)
7528ee6aa596 [unknown] (/usr/lib/libffi.so.8.1.4)
7fff24e862d8 [unknown] ([unknown])
blueman-tray 1355 [000] 28178.728137: 100187539 offcpu-time:
7528eef1c39d __poll+0x4d (/usr/lib/libc.so.6)
7528edf7d8fd [unknown] (/usr/lib/libglib-2.0.so.0.8000.2)
7528edf1af95 g_main_context_iteration+0x35 (/usr/lib/libglib-2.0.so.0.8000.2)
7528eda4ab86 g_application_run+0x1f6 (/usr/lib/libgio-2.0.so.0.8000.2)
7528ee6aa596 [unknown] (/usr/lib/libffi.so.8.1.4)
7fff24e862d8 [unknown] ([unknown])
swapper 0 [000] 28178.463253: 195945410 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms])
dbus-broker 412 [002] 28178.464855: 376737008 cycles:P: ffffffff91c000a0 entry_SYSCALL_64+0x20 ([kernel.kallsyms])
```
Howard Chu (9):
perf evsel: Set BPF output to system-wide
perf record --off-cpu: Add --off-cpu-thresh
perf record --off-cpu: Parse offcpu-time event
perf record off-cpu: Dump direct off-cpu samples in BPF
perf record --off-cpu: Dump total off-cpu time at the end.
perf evsel: Delete unnecessary = 0
perf record --off-cpu: Parse BPF output embedded data
perf header: Add field 'embed'
perf test: Add direct off-cpu dumping test
tools/perf/builtin-record.c | 26 ++++
tools/perf/builtin-script.c | 5 +-
tools/perf/tests/builtin-test.c | 1 +
tools/perf/tests/shell/record_offcpu.sh | 27 ++++
tools/perf/tests/tests.h | 1 +
tools/perf/tests/workloads/Build | 1 +
tools/perf/tests/workloads/offcpu.c | 16 +++
tools/perf/util/bpf_off_cpu.c | 176 +++++++++++++++---------
tools/perf/util/bpf_skel/off_cpu.bpf.c | 135 ++++++++++++++++++
tools/perf/util/evsel.c | 49 ++++---
tools/perf/util/evsel.h | 13 ++
tools/perf/util/header.c | 12 ++
tools/perf/util/off_cpu.h | 10 +-
tools/perf/util/record.h | 1 +
tools/perf/util/session.c | 16 ++-
15 files changed, 401 insertions(+), 88 deletions(-)
create mode 100644 tools/perf/tests/workloads/offcpu.c
--
2.45.2
On Wed, Aug 7, 2024 at 8:38 AM Howard Chu <howardchu95@gmail.com> wrote: > > Changes in v4: > - Minimize the size of data output by perf_event_output() > - Keep only one off-cpu event > - Change off-cpu threshold's unit to microseconds > - Set a default off-cpu threshold > - Print the correct error message for the field 'embed' in perf data header Hi Howard, I think this all looks good. There is an outstanding comment for patch 2/9 where a definition in a later patch needs moving to patch 2/9. There is also a comment in 3/9 where you say you will turn a "=" back to an "&=". These seem like small things to fix, could you rebase and post the series? Are there other outstanding issues I'm missing? I'd quite like to see this land. Thanks, Ian > Changes in v3: > - Add off-cpu-thresh argument > - Process direct off-cpu samples in post > > Changes in v2: > - Remove unnecessary comments. > - Rename function off_cpu_change_type to off_cpu_prepare_parse > > v1: > > As mentioned in: https://bugzilla.kernel.org/show_bug.cgi?id=207323 > > Currently, off-cpu samples are dumped when perf record is exiting. This > results in off-cpu samples being after the regular samples. This patch > series makes possible dumping off-cpu samples on-the-fly, directly into > perf ring buffer. And it dispatches those samples to the correct format > for perf.data consumers. > > Before: > ``` > migration/0 21 [000] 27981.041319: 2944637851 cycles:P: ffffffff90d2e8aa record_times+0xa ([kernel.kallsyms]) > perf 770116 [001] 27981.041375: 1 cycles:P: ffffffff90ee4960 event_function+0xf0 ([kernel.kallsyms]) > perf 770116 [001] 27981.041377: 1 cycles:P: ffffffff90c184b1 intel_bts_enable_local+0x31 ([kernel.kallsyms]) > perf 770116 [001] 27981.041379: 51611 cycles:P: ffffffff91a160b0 native_sched_clock+0x30 ([kernel.kallsyms]) > migration/1 26 [001] 27981.041400: 4227682775 cycles:P: ffffffff90d06a74 wakeup_preempt+0x44 ([kernel.kallsyms]) > migration/2 32 [002] 27981.041477: 4159401534 cycles:P: ffffffff90d11993 update_load_avg+0x63 ([kernel.kallsyms]) > > sshd 708098 [000] 18446744069.414584: 286392 offcpu-time: > 79a864f1c8bb ppoll+0x4b (/usr/lib/libc.so.6) > 585690935cca [unknown] (/usr/bin/sshd) > ``` > > After: > ``` > perf 774767 [003] 28178.033444: 497 cycles:P: ffffffff91a160c3 native_sched_clock+0x43 ([kernel.kallsyms]) > perf 774767 [003] 28178.033445: 399440 cycles:P: ffffffff91c01f8d nmi_restore+0x25 ([kernel.kallsyms]) > swapper 0 [001] 28178.036639: 376650973 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms]) > swapper 0 [003] 28178.182921: 348779378 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms]) > blueman-tray 1355 [000] 28178.627906: 100184571 offcpu-time: > 7528eef1c39d __poll+0x4d (/usr/lib/libc.so.6) > 7528edf7d8fd [unknown] (/usr/lib/libglib-2.0.so.0.8000.2) > 7528edf1af95 g_main_context_iteration+0x35 (/usr/lib/libglib-2.0.so.0.8000.2) > 7528eda4ab86 g_application_run+0x1f6 (/usr/lib/libgio-2.0.so.0.8000.2) > 7528ee6aa596 [unknown] (/usr/lib/libffi.so.8.1.4) > 7fff24e862d8 [unknown] ([unknown]) > > > blueman-tray 1355 [000] 28178.728137: 100187539 offcpu-time: > 7528eef1c39d __poll+0x4d (/usr/lib/libc.so.6) > 7528edf7d8fd [unknown] (/usr/lib/libglib-2.0.so.0.8000.2) > 7528edf1af95 g_main_context_iteration+0x35 (/usr/lib/libglib-2.0.so.0.8000.2) > 7528eda4ab86 g_application_run+0x1f6 (/usr/lib/libgio-2.0.so.0.8000.2) > 7528ee6aa596 [unknown] (/usr/lib/libffi.so.8.1.4) > 7fff24e862d8 [unknown] ([unknown]) > > > swapper 0 [000] 28178.463253: 195945410 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms]) > dbus-broker 412 [002] 28178.464855: 376737008 cycles:P: ffffffff91c000a0 entry_SYSCALL_64+0x20 ([kernel.kallsyms]) > ``` > > > Howard Chu (9): > perf evsel: Set BPF output to system-wide > perf record --off-cpu: Add --off-cpu-thresh > perf record --off-cpu: Parse offcpu-time event > perf record off-cpu: Dump direct off-cpu samples in BPF > perf record --off-cpu: Dump total off-cpu time at the end. > perf evsel: Delete unnecessary = 0 > perf record --off-cpu: Parse BPF output embedded data > perf header: Add field 'embed' > perf test: Add direct off-cpu dumping test > > tools/perf/builtin-record.c | 26 ++++ > tools/perf/builtin-script.c | 5 +- > tools/perf/tests/builtin-test.c | 1 + > tools/perf/tests/shell/record_offcpu.sh | 27 ++++ > tools/perf/tests/tests.h | 1 + > tools/perf/tests/workloads/Build | 1 + > tools/perf/tests/workloads/offcpu.c | 16 +++ > tools/perf/util/bpf_off_cpu.c | 176 +++++++++++++++--------- > tools/perf/util/bpf_skel/off_cpu.bpf.c | 135 ++++++++++++++++++ > tools/perf/util/evsel.c | 49 ++++--- > tools/perf/util/evsel.h | 13 ++ > tools/perf/util/header.c | 12 ++ > tools/perf/util/off_cpu.h | 10 +- > tools/perf/util/record.h | 1 + > tools/perf/util/session.c | 16 ++- > 15 files changed, 401 insertions(+), 88 deletions(-) > create mode 100644 tools/perf/tests/workloads/offcpu.c > > -- > 2.45.2 >
Hello Ian, On Tue, Sep 24, 2024 at 8:04 PM Ian Rogers <irogers@google.com> wrote: > > On Wed, Aug 7, 2024 at 8:38 AM Howard Chu <howardchu95@gmail.com> wrote: > > > > Changes in v4: > > - Minimize the size of data output by perf_event_output() > > - Keep only one off-cpu event > > - Change off-cpu threshold's unit to microseconds > > - Set a default off-cpu threshold > > - Print the correct error message for the field 'embed' in perf data header > > Hi Howard, > > I think this all looks good. There is an outstanding comment for patch > 2/9 where a definition in a later patch needs moving to patch 2/9. > There is also a comment in 3/9 where you say you will turn a "=" back > to an "&=". These seem like small things to fix, could you rebase and > post the series? Are there other outstanding issues I'm missing? I'd Thank you, I fixed them. Namhyung said the sample_embed thing should be handled with care, so I'll make this sample_embed a special case for off-cpu, and hard code something, so patches will look more concise. Will do this, rebase, and send the series. Sorry for taking so long. (〒︿〒) Thanks, Howard > quite like to see this land. > > Thanks, > Ian > > > Changes in v3: > > - Add off-cpu-thresh argument > > - Process direct off-cpu samples in post > > > > Changes in v2: > > - Remove unnecessary comments. > > - Rename function off_cpu_change_type to off_cpu_prepare_parse > > > > v1: > > > > As mentioned in: https://bugzilla.kernel.org/show_bug.cgi?id=207323 > > > > Currently, off-cpu samples are dumped when perf record is exiting. This > > results in off-cpu samples being after the regular samples. This patch > > series makes possible dumping off-cpu samples on-the-fly, directly into > > perf ring buffer. And it dispatches those samples to the correct format > > for perf.data consumers. > > > > Before: > > ``` > > migration/0 21 [000] 27981.041319: 2944637851 cycles:P: ffffffff90d2e8aa record_times+0xa ([kernel.kallsyms]) > > perf 770116 [001] 27981.041375: 1 cycles:P: ffffffff90ee4960 event_function+0xf0 ([kernel.kallsyms]) > > perf 770116 [001] 27981.041377: 1 cycles:P: ffffffff90c184b1 intel_bts_enable_local+0x31 ([kernel.kallsyms]) > > perf 770116 [001] 27981.041379: 51611 cycles:P: ffffffff91a160b0 native_sched_clock+0x30 ([kernel.kallsyms]) > > migration/1 26 [001] 27981.041400: 4227682775 cycles:P: ffffffff90d06a74 wakeup_preempt+0x44 ([kernel.kallsyms]) > > migration/2 32 [002] 27981.041477: 4159401534 cycles:P: ffffffff90d11993 update_load_avg+0x63 ([kernel.kallsyms]) > > > > sshd 708098 [000] 18446744069.414584: 286392 offcpu-time: > > 79a864f1c8bb ppoll+0x4b (/usr/lib/libc.so.6) > > 585690935cca [unknown] (/usr/bin/sshd) > > ``` > > > > After: > > ``` > > perf 774767 [003] 28178.033444: 497 cycles:P: ffffffff91a160c3 native_sched_clock+0x43 ([kernel.kallsyms]) > > perf 774767 [003] 28178.033445: 399440 cycles:P: ffffffff91c01f8d nmi_restore+0x25 ([kernel.kallsyms]) > > swapper 0 [001] 28178.036639: 376650973 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms]) > > swapper 0 [003] 28178.182921: 348779378 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms]) > > blueman-tray 1355 [000] 28178.627906: 100184571 offcpu-time: > > 7528eef1c39d __poll+0x4d (/usr/lib/libc.so.6) > > 7528edf7d8fd [unknown] (/usr/lib/libglib-2.0.so.0.8000.2) > > 7528edf1af95 g_main_context_iteration+0x35 (/usr/lib/libglib-2.0.so.0.8000.2) > > 7528eda4ab86 g_application_run+0x1f6 (/usr/lib/libgio-2.0.so.0.8000.2) > > 7528ee6aa596 [unknown] (/usr/lib/libffi.so.8.1.4) > > 7fff24e862d8 [unknown] ([unknown]) > > > > > > blueman-tray 1355 [000] 28178.728137: 100187539 offcpu-time: > > 7528eef1c39d __poll+0x4d (/usr/lib/libc.so.6) > > 7528edf7d8fd [unknown] (/usr/lib/libglib-2.0.so.0.8000.2) > > 7528edf1af95 g_main_context_iteration+0x35 (/usr/lib/libglib-2.0.so.0.8000.2) > > 7528eda4ab86 g_application_run+0x1f6 (/usr/lib/libgio-2.0.so.0.8000.2) > > 7528ee6aa596 [unknown] (/usr/lib/libffi.so.8.1.4) > > 7fff24e862d8 [unknown] ([unknown]) > > > > > > swapper 0 [000] 28178.463253: 195945410 cycles:P: ffffffff91a1ae99 intel_idle+0x59 ([kernel.kallsyms]) > > dbus-broker 412 [002] 28178.464855: 376737008 cycles:P: ffffffff91c000a0 entry_SYSCALL_64+0x20 ([kernel.kallsyms]) > > ``` > > > > > > Howard Chu (9): > > perf evsel: Set BPF output to system-wide > > perf record --off-cpu: Add --off-cpu-thresh > > perf record --off-cpu: Parse offcpu-time event > > perf record off-cpu: Dump direct off-cpu samples in BPF > > perf record --off-cpu: Dump total off-cpu time at the end. > > perf evsel: Delete unnecessary = 0 > > perf record --off-cpu: Parse BPF output embedded data > > perf header: Add field 'embed' > > perf test: Add direct off-cpu dumping test > > > > tools/perf/builtin-record.c | 26 ++++ > > tools/perf/builtin-script.c | 5 +- > > tools/perf/tests/builtin-test.c | 1 + > > tools/perf/tests/shell/record_offcpu.sh | 27 ++++ > > tools/perf/tests/tests.h | 1 + > > tools/perf/tests/workloads/Build | 1 + > > tools/perf/tests/workloads/offcpu.c | 16 +++ > > tools/perf/util/bpf_off_cpu.c | 176 +++++++++++++++--------- > > tools/perf/util/bpf_skel/off_cpu.bpf.c | 135 ++++++++++++++++++ > > tools/perf/util/evsel.c | 49 ++++--- > > tools/perf/util/evsel.h | 13 ++ > > tools/perf/util/header.c | 12 ++ > > tools/perf/util/off_cpu.h | 10 +- > > tools/perf/util/record.h | 1 + > > tools/perf/util/session.c | 16 ++- > > 15 files changed, 401 insertions(+), 88 deletions(-) > > create mode 100644 tools/perf/tests/workloads/offcpu.c > > > > -- > > 2.45.2 > >
© 2016 - 2025 Red Hat, Inc.