20 Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
Linus Torvalds
|
535a265d7f |
perf tools changes for v6.6:
perf tools maintainership:
- Add git information for perf-tools and perf-tools-next trees/branches to the
MAINTAINERS file. That is where development now takes place and myself and
Namhyung Kim have write access, more people to come as we emulate other
maintainer groups.
perf record:
- Record kernel data maps when 'perf record --data' is used, so that global variables can
be resolved and used in tools that do data profiling.
perf trace:
- Remove the old, experimental support for BPF events in which a .c file was passed as
an event: "perf trace -e hello.c" to then get compiled and loaded.
The only known usage for that, that shipped with the kernel as an example for such events,
augmented the raw_syscalls tracepoints and was converted to a libbpf skeleton, reusing all
the user space components and the BPF code connected to the syscalls.
In the end just the way to glue the BPF part and the user space type beautifiers changed,
now being performed by libbpf skeletons.
The next step is to use BTF to do pretty printing of all syscall types, as discussed with
Alan Maguire and others.
Now, on a perf built with BUILD_BPF_SKEL=1 we get most if not all path/filenames/strings,
some of the networking data structures, perf_event_attr, etc, i.e. systemwide tracing of
nanosleep calls and perf_event_open syscalls while 'perf stat' runs 'sleep' for 5 seconds:
# perf trace -a -e *nanosleep,perf* perf stat -e cycles,instructions sleep 5
0.000 ( 9.034 ms): perf/327641 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0 (PERF_COUNT_HW_CPU_CYCLES), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 327642 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 3
9.039 ( 0.006 ms): perf/327641 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0x1 (PERF_COUNT_HW_INSTRUCTIONS), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 327642 (perf-exec), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 4
? ( ): gpm/991 ... [continued]: clock_nanosleep()) = 0
10.133 ( ): sleep/327642 clock_nanosleep(rqtp: { .tv_sec: 5, .tv_nsec: 0 }, rmtp: 0x7ffd36f83ed0) ...
? ( ): pool-gsd-smart/3051 ... [continued]: clock_nanosleep()) = 0
30.276 ( ): gpm/991 clock_nanosleep(rqtp: { .tv_sec: 2, .tv_nsec: 0 }, rmtp: 0x7ffcc6f73710) ...
223.215 (1000.430 ms): pool-gsd-smart/3051 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7f6e7fffec90) = 0
30.276 (2000.394 ms): gpm/991 ... [continued]: clock_nanosleep()) = 0
1230.814 ( ): pool-gsd-smart/3051 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7f6e7fffec90) ...
1230.814 (1000.404 ms): pool-gsd-smart/3051 ... [continued]: clock_nanosleep()) = 0
2030.886 ( ): gpm/991 clock_nanosleep(rqtp: { .tv_sec: 2, .tv_nsec: 0 }, rmtp: 0x7ffcc6f73710) ...
2237.709 (1000.153 ms): pool-gsd-smart/3051 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7f6e7fffec90) = 0
? ( ): crond/1172 ... [continued]: clock_nanosleep()) = 0
3242.699 ( ): pool-gsd-smart/3051 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7f6e7fffec90) ...
2030.886 (2000.385 ms): gpm/991 ... [continued]: clock_nanosleep()) = 0
3728.078 ( ): crond/1172 clock_nanosleep(rqtp: { .tv_sec: 60, .tv_nsec: 0 }, rmtp: 0x7ffe0971dcf0) ...
3242.699 (1000.158 ms): pool-gsd-smart/3051 ... [continued]: clock_nanosleep()) = 0
4031.409 ( ): gpm/991 clock_nanosleep(rqtp: { .tv_sec: 2, .tv_nsec: 0 }, rmtp: 0x7ffcc6f73710) ...
10.133 (5000.375 ms): sleep/327642 ... [continued]: clock_nanosleep()) = 0
Performance counter stats for 'sleep 5':
2,617,347 cycles
1,855,997 instructions # 0.71 insn per cycle
5.002282128 seconds time elapsed
0.000855000 seconds user
0.000852000 seconds sys
#
perf annotate:
- Building with binutils' libopcode now is opt-in (BUILD_NONDISTRO=1) for
licensing reasons, and we missed a build test on tools/perf/tests makefile.
Since we now default to NDEBUG=1, we ended up segfaulting when building with
BUILD_NONDISTRO=1 because a needed initialization routine was being "error
checked" via an assert.
Fix it by explicitly checking the result and aborting instead if it fails.
We better back propagate the error, but at least 'perf annotate' on samples
collected for a BPF program is back working when perf is built with
BUILD_NONDISTRO=1.
perf report/top:
- Add back TUI hierarchy mode header, that is seen when using 'perf report/top --hierarchy'.
- Fix the number of entries for 'e' key in the TUI that was preventing navigation of
lines when expanding an entry.
perf report/script:
- Support cross platform register handling, allowing a perf.data file collected
on one architecture to have registers sampled correctly displayed when
analysis tools such as 'perf report' and 'perf script' are used on a different
architecture.
- Fix handling of event attributes in pipe mode, i.e. when one uses:
perf record -o - | perf report -i -
When no perf.data files are used.
- Handle files generated via pipe mode with a version of perf and then read
also via pipe mode with a different version of perf, where the event attr
record may have changed, use the record size field to properly support this
version mismatch.
perf probe:
- Accessing global variables from uprobes isn't supported, make the error
message state that instead of stating that some minimal kernel version is
needed to have that feature. This seems just a tool limitation, the kernel
probably has all that is needed.
perf tests:
- Fix a reference count related leak in the dlfilter v0 API where the result
of a thread__find_symbol_fb() is not matched with an addr_location__exit()
to drop the reference counts of the resolved components (machine, thread, map,
symbol, etc). Add a dlfilter test to make sure that doesn't regresses.
- Lots of fixes for the 'perf test' written in shell script related to problems
found with the shellcheck utility.
- Fixes for 'perf test' shell scripts testing features enabled when perf is
built with BUILD_BPF_SKEL=1, such as 'perf stat' bpf counters.
- Add perf record sample filtering test, things like the following example, that gets
implemented as a BPF filter attached to the event:
# perf record -e task-clock -c 10000 --filter 'ip < 0xffffffff00000000'
- Improve the way the task_analyzer test checks if libtraceevent is linked,
using 'perf version --build-options' instead of the more expensinve
'perf record -e "sched:sched_switch"'.
- Add support for riscv in the mmap-basic test. (This went as well via the RiscV tree, same contents).
libperf:
- Implement riscv mmap support (This went as well via the RiscV tree, same contents).
perf script:
- New tool that converts perf.data files to the firefox profiler format so that one can use
the visualizer at https://profiler.firefox.com/. Done by Anup Sharma as part of this year's
Google Summer of Code.
One can generate the output and upload it to the web interface but Anup also automated
everything:
perf script gecko -F 99 -a sleep 60
- Support syscall name parsing on arm64.
- Print "cgroup" field on the same line as "comm".
perf bench:
- Add new 'uprobe' benchmark to measure the overhead of uprobes with/without
BPF programs attached to it.
- breakpoints are not available on power9, skip that test.
perf stat:
- Add #num_cpus_online literal to be used in 'perf stat' metrics, and add this extra
'perf test' check that exemplifies its purpose:
TEST_ASSERT_VAL("#num_cpus_online",
expr__parse(&num_cpus_online, ctx, "#num_cpus_online") == 0);
TEST_ASSERT_VAL("#num_cpus", expr__parse(&num_cpus, ctx, "#num_cpus") == 0);
TEST_ASSERT_VAL("#num_cpus >= #num_cpus_online", num_cpus >= num_cpus_online);
Miscellaneous:
- Improve tool startup time by lazily reading PMU, JSON, sysfs data.
- Improve error reporting in the parsing of events, passing YYLTYPE to error routines,
so that the output can show were the parsing error was found.
- Add 'perf test' entries to check the parsing of events improvements.
- Fix various leak for things detected by -fsanitize=address, mostly things that would
be freed at tool exit, including:
- Free evsel->filter on the destructor.
- Allow tools to register a thread->priv destructor and use it in 'perf trace'.
- Free evsel->priv in 'perf trace'.
- Free string returned by synthesize_perf_probe_point() when the caller fails
to do all it needs.
- Adjust various compiler options to not consider errors some warnings when
building with broken headers found in things like python, flex, bison, as we
otherwise build with -Werror. Some for gcc, some for clang, some for some
specific version of those, some for some specific version of flex or bison, or
some specific combination of these components, bah.
- Allow customization of clang options for BPF target, this helps building on
gentoo where there are other oddities where BPF targets gets passed some compiler
options intended for the native build, so building with WERROR=0 helps while
these oddities are fixed.
- Dont pass ERR_PTR() values to perf_session__delete() in 'perf top' and 'perf lock',
fixing some segfaults when handling some odd failures.
- Add LTO build option.
- Fix format of unordered lists in the perf docs (tools/perf/Documentation).
- Overhaul the bison files, using constructs such as YYNOMEM.
- Remove unused tokens from the bison .y files.
- Add more comments to various structs.
- A few LoongArch enablement patches.
Vendor events (JSON):
- Add JSON metrics for Yitian 710 DDR (aarch64). Things like:
EventName, BriefDescription
visible_window_limit_reached_rd, "At least one entry in read queue reaches the visible window limit.",
visible_window_limit_reached_wr, "At least one entry in write queue reaches the visible window limit.",
op_is_dqsosc_mpc , "A DQS Oscillator MPC command to DRAM.",
op_is_dqsosc_mrr , "A DQS Oscillator MRR command to DRAM.",
op_is_tcr_mrr , "A Temperature Compensated Refresh(TCR) MRR command to DRAM.",
- Add AmpereOne metrics (aarch64).
- Update N2 and V2 metrics (aarch64) and events using Arm telemetry repo.
- Update scale units and descriptions of common topdown metrics on aarch64. Things like:
- "MetricExpr": "stall_slot_frontend / (#slots * cpu_cycles)",
- "BriefDescription": "Frontend bound L1 topdown metric",
+ "MetricExpr": "100 * (stall_slot_frontend / (#slots * cpu_cycles))",
+ "BriefDescription": "This metric is the percentage of total slots that were stalled due to resource constraints in the frontend of the processor.",
- Update events for intel: meteorlake to 1.04, sapphirerapids to 1.15, Icelake+ metric constraints.
- Update files for the power10 platform.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYKAB0WIQR2GiIUctdOfX2qHhGyPKLppCJ+JwUCZPfJZgAKCRCyPKLppCJ+
J1/eAP9lgtavD0V75wy1p5zyotkceOmPTkk1DYFVx2Euhxa/lAD/YW/JvuVSo0Gr
HqJP52XaV0tF8gG+YxL+Lay/Ke0P5AQ=
=d12c
-----END PGP SIGNATURE-----
Merge tag 'perf-tools-for-v6.6-1-2023-09-05' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools
Pull perf tools updates from Arnaldo Carvalho de Melo:
"perf tools maintainership:
- Add git information for perf-tools and perf-tools-next trees and
branches to the MAINTAINERS file. That is where development now
takes place and myself and Namhyung Kim have write access, more
people to come as we emulate other maintainer groups.
perf record:
- Record kernel data maps when 'perf record --data' is used, so that
global variables can be resolved and used in tools that do data
profiling.
perf trace:
- Remove the old, experimental support for BPF events in which a .c
file was passed as an event: "perf trace -e hello.c" to then get
compiled and loaded.
The only known usage for that, that shipped with the kernel as an
example for such events, augmented the raw_syscalls tracepoints and
was converted to a libbpf skeleton, reusing all the user space
components and the BPF code connected to the syscalls.
In the end just the way to glue the BPF part and the user space
type beautifiers changed, now being performed by libbpf skeletons.
The next step is to use BTF to do pretty printing of all syscall
types, as discussed with Alan Maguire and others.
Now, on a perf built with BUILD_BPF_SKEL=1 we get most if not all
path/filenames/strings, some of the networking data structures,
perf_event_attr, etc, i.e. systemwide tracing of nanosleep calls
and perf_event_open syscalls while 'perf stat' runs 'sleep' for 5
seconds:
# perf trace -a -e *nanosleep,perf* perf stat -e cycles,instructions sleep 5
0.000 ( 9.034 ms): perf/327641 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0 (PERF_COUNT_HW_CPU_CYCLES), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 327642 (perf), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 3
9.039 ( 0.006 ms): perf/327641 perf_event_open(attr_uptr: { type: 0 (PERF_TYPE_HARDWARE), size: 136, config: 0x1 (PERF_COUNT_HW_INSTRUCTIONS), sample_type: IDENTIFIER, read_format: TOTAL_TIME_ENABLED|TOTAL_TIME_RUNNING, disabled: 1, inherit: 1, enable_on_exec: 1, exclude_guest: 1 }, pid: 327642 (perf-exec), cpu: -1, group_fd: -1, flags: FD_CLOEXEC) = 4
? ( ): gpm/991 ... [continued]: clock_nanosleep()) = 0
10.133 ( ): sleep/327642 clock_nanosleep(rqtp: { .tv_sec: 5, .tv_nsec: 0 }, rmtp: 0x7ffd36f83ed0) ...
? ( ): pool-gsd-smart/3051 ... [continued]: clock_nanosleep()) = 0
30.276 ( ): gpm/991 clock_nanosleep(rqtp: { .tv_sec: 2, .tv_nsec: 0 }, rmtp: 0x7ffcc6f73710) ...
223.215 (1000.430 ms): pool-gsd-smart/3051 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7f6e7fffec90) = 0
30.276 (2000.394 ms): gpm/991 ... [continued]: clock_nanosleep()) = 0
1230.814 ( ): pool-gsd-smart/3051 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7f6e7fffec90) ...
1230.814 (1000.404 ms): pool-gsd-smart/3051 ... [continued]: clock_nanosleep()) = 0
2030.886 ( ): gpm/991 clock_nanosleep(rqtp: { .tv_sec: 2, .tv_nsec: 0 }, rmtp: 0x7ffcc6f73710) ...
2237.709 (1000.153 ms): pool-gsd-smart/3051 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7f6e7fffec90) = 0
? ( ): crond/1172 ... [continued]: clock_nanosleep()) = 0
3242.699 ( ): pool-gsd-smart/3051 clock_nanosleep(rqtp: { .tv_sec: 1, .tv_nsec: 0 }, rmtp: 0x7f6e7fffec90) ...
2030.886 (2000.385 ms): gpm/991 ... [continued]: clock_nanosleep()) = 0
3728.078 ( ): crond/1172 clock_nanosleep(rqtp: { .tv_sec: 60, .tv_nsec: 0 }, rmtp: 0x7ffe0971dcf0) ...
3242.699 (1000.158 ms): pool-gsd-smart/3051 ... [continued]: clock_nanosleep()) = 0
4031.409 ( ): gpm/991 clock_nanosleep(rqtp: { .tv_sec: 2, .tv_nsec: 0 }, rmtp: 0x7ffcc6f73710) ...
10.133 (5000.375 ms): sleep/327642 ... [continued]: clock_nanosleep()) = 0
Performance counter stats for 'sleep 5':
2,617,347 cycles
1,855,997 instructions # 0.71 insn per cycle
5.002282128 seconds time elapsed
0.000855000 seconds user
0.000852000 seconds sys
perf annotate:
- Building with binutils' libopcode now is opt-in (BUILD_NONDISTRO=1)
for licensing reasons, and we missed a build test on
tools/perf/tests makefile.
Since we now default to NDEBUG=1, we ended up segfaulting when
building with BUILD_NONDISTRO=1 because a needed initialization
routine was being "error checked" via an assert.
Fix it by explicitly checking the result and aborting instead if it
fails.
We better back propagate the error, but at least 'perf annotate' on
samples collected for a BPF program is back working when perf is
built with BUILD_NONDISTRO=1.
perf report/top:
- Add back TUI hierarchy mode header, that is seen when using 'perf
report/top --hierarchy'.
- Fix the number of entries for 'e' key in the TUI that was
preventing navigation of lines when expanding an entry.
perf report/script:
- Support cross platform register handling, allowing a perf.data file
collected on one architecture to have registers sampled correctly
displayed when analysis tools such as 'perf report' and 'perf
script' are used on a different architecture.
- Fix handling of event attributes in pipe mode, i.e. when one uses:
perf record -o - | perf report -i -
When no perf.data files are used.
- Handle files generated via pipe mode with a version of perf and
then read also via pipe mode with a different version of perf,
where the event attr record may have changed, use the record size
field to properly support this version mismatch.
perf probe:
- Accessing global variables from uprobes isn't supported, make the
error message state that instead of stating that some minimal
kernel version is needed to have that feature. This seems just a
tool limitation, the kernel probably has all that is needed.
perf tests:
- Fix a reference count related leak in the dlfilter v0 API where the
result of a thread__find_symbol_fb() is not matched with an
addr_location__exit() to drop the reference counts of the resolved
components (machine, thread, map, symbol, etc). Add a dlfilter test
to make sure that doesn't regresses.
- Lots of fixes for the 'perf test' written in shell script related
to problems found with the shellcheck utility.
- Fixes for 'perf test' shell scripts testing features enabled when
perf is built with BUILD_BPF_SKEL=1, such as 'perf stat' bpf
counters.
- Add perf record sample filtering test, things like the following
example, that gets implemented as a BPF filter attached to the
event:
# perf record -e task-clock -c 10000 --filter 'ip < 0xffffffff00000000'
- Improve the way the task_analyzer test checks if libtraceevent is
linked, using 'perf version --build-options' instead of the more
expensinve 'perf record -e "sched:sched_switch"'.
- Add support for riscv in the mmap-basic test. (This went as well
via the RiscV tree, same contents).
libperf:
- Implement riscv mmap support (This went as well via the RiscV tree,
same contents).
perf script:
- New tool that converts perf.data files to the firefox profiler
format so that one can use the visualizer at
https://profiler.firefox.com/. Done by Anup Sharma as part of this
year's Google Summer of Code.
One can generate the output and upload it to the web interface but
Anup also automated everything:
perf script gecko -F 99 -a sleep 60
- Support syscall name parsing on arm64.
- Print "cgroup" field on the same line as "comm".
perf bench:
- Add new 'uprobe' benchmark to measure the overhead of uprobes
with/without BPF programs attached to it.
- breakpoints are not available on power9, skip that test.
perf stat:
- Add #num_cpus_online literal to be used in 'perf stat' metrics, and
add this extra 'perf test' check that exemplifies its purpose:
TEST_ASSERT_VAL("#num_cpus_online",
expr__parse(&num_cpus_online, ctx, "#num_cpus_online") == 0);
TEST_ASSERT_VAL("#num_cpus", expr__parse(&num_cpus, ctx, "#num_cpus") == 0);
TEST_ASSERT_VAL("#num_cpus >= #num_cpus_online", num_cpus >= num_cpus_online);
Miscellaneous:
- Improve tool startup time by lazily reading PMU, JSON, sysfs data.
- Improve error reporting in the parsing of events, passing YYLTYPE
to error routines, so that the output can show were the parsing
error was found.
- Add 'perf test' entries to check the parsing of events
improvements.
- Fix various leak for things detected by -fsanitize=address, mostly
things that would be freed at tool exit, including:
- Free evsel->filter on the destructor.
- Allow tools to register a thread->priv destructor and use it in
'perf trace'.
- Free evsel->priv in 'perf trace'.
- Free string returned by synthesize_perf_probe_point() when the
caller fails to do all it needs.
- Adjust various compiler options to not consider errors some
warnings when building with broken headers found in things like
python, flex, bison, as we otherwise build with -Werror. Some for
gcc, some for clang, some for some specific version of those, some
for some specific version of flex or bison, or some specific
combination of these components, bah.
- Allow customization of clang options for BPF target, this helps
building on gentoo where there are other oddities where BPF targets
gets passed some compiler options intended for the native build, so
building with WERROR=0 helps while these oddities are fixed.
- Dont pass ERR_PTR() values to perf_session__delete() in 'perf top'
and 'perf lock', fixing some segfaults when handling some odd
failures.
- Add LTO build option.
- Fix format of unordered lists in the perf docs
(tools/perf/Documentation)
- Overhaul the bison files, using constructs such as YYNOMEM.
- Remove unused tokens from the bison .y files.
- Add more comments to various structs.
- A few LoongArch enablement patches.
Vendor events (JSON):
- Add JSON metrics for Yitian 710 DDR (aarch64). Things like:
EventName, BriefDescription
visible_window_limit_reached_rd, "At least one entry in read queue reaches the visible window limit.",
visible_window_limit_reached_wr, "At least one entry in write queue reaches the visible window limit.",
op_is_dqsosc_mpc , "A DQS Oscillator MPC command to DRAM.",
op_is_dqsosc_mrr , "A DQS Oscillator MRR command to DRAM.",
op_is_tcr_mrr , "A Temperature Compensated Refresh(TCR) MRR command to DRAM.",
- Add AmpereOne metrics (aarch64).
- Update N2 and V2 metrics (aarch64) and events using Arm telemetry
repo.
- Update scale units and descriptions of common topdown metrics on
aarch64. Things like:
- "MetricExpr": "stall_slot_frontend / (#slots * cpu_cycles)",
- "BriefDescription": "Frontend bound L1 topdown metric",
+ "MetricExpr": "100 * (stall_slot_frontend / (#slots * cpu_cycles))",
+ "BriefDescription": "This metric is the percentage of total slots that were stalled due to resource constraints in the frontend of the processor.",
- Update events for intel: meteorlake to 1.04, sapphirerapids to
1.15, Icelake+ metric constraints.
- Update files for the power10 platform"
* tag 'perf-tools-for-v6.6-1-2023-09-05' of git://git.kernel.org/pub/scm/linux/kernel/git/perf/perf-tools: (217 commits)
perf parse-events: Fix driver config term
perf parse-events: Fixes relating to no_value terms
perf parse-events: Fix propagation of term's no_value when cloning
perf parse-events: Name the two term enums
perf list: Don't print Unit for "default_core"
perf vendor events intel: Fix modifier in tma_info_system_mem_parallel_reads for skylake
perf dlfilter: Avoid leak in v0 API test use of resolve_address()
perf metric: Add #num_cpus_online literal
perf pmu: Remove str from perf_pmu_alias
perf parse-events: Make common term list to strbuf helper
perf parse-events: Minor help message improvements
perf pmu: Avoid uninitialized use of alias->str
perf jevents: Use "default_core" for events with no Unit
perf test stat_bpf_counters_cgrp: Enhance perf stat cgroup BPF counter test
perf test shell stat_bpf_counters: Fix test on Intel
perf test shell record_bpf_filter: Skip 6.2 kernel
libperf: Get rid of attr.id field
perf tools: Convert to perf_record_header_attr_id()
libperf: Add perf_record_header_attr_id()
perf tools: Handle old data in PERF_RECORD_ATTR
...
|
||
Arnaldo Carvalho de Melo
|
2df2707164 |
perf bench uprobe: Add benchmark to test uprobe overhead
This just adds the initial "workload", a call to libc's usleep(1000us)
function:
$ perf stat --null perf bench uprobe all
# Running uprobe/baseline benchmark...
# Executed 1000 usleep(1000) calls
Total time: 1053533 usecs
1053.533 usecs/op
Performance counter stats for 'perf bench uprobe all':
1.061042896 seconds time elapsed
0.001079000 seconds user
0.006499000 seconds sys
$
More entries will be added using a BPF skel to add various uprobes to
the usleep() function.
Acked-by: Ian Rogers <irogers@google.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andre Fredette <anfredet@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: Dave Tucker <datucker@redhat.com>
Cc: Derek Barbosa <debarbos@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/lkml/20230719204910.539044-2-acme@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
Andrei Vagin
|
7d5cb68af6 |
perf/benchmark: add a new benchmark for seccom_unotify
The benchmark is similar to the pipe benchmark. It creates two processes,
one is calling syscalls, and another process is handling them via seccomp
user notifications. It measures the time required to run a specified number
of interations.
$ ./perf bench sched seccomp-notify --sync-mode --loop 1000000
# Running 'sched/seccomp-notify' benchmark:
# Executed 1000000 system calls
Total time: 2.769 [sec]
2.769629 usecs/op
361059 ops/sec
$ ./perf bench sched seccomp-notify
# Running 'sched/seccomp-notify' benchmark:
# Executed 1000000 system calls
Total time: 8.571 [sec]
8.571119 usecs/op
116670 ops/sec
Signed-off-by: Andrei Vagin <avagin@google.com>
Acked-by: "Peter Zijlstra (Intel)" <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230308073201.3102738-7-avagin@google.com
Link: https://lore.kernel.org/r/20230630051953.454638-1-avagin@gmail.com
[kees: Added PRIu64 format string]
Signed-off-by: Kees Cook <keescook@chromium.org>
|
||
Namhyung Kim
|
f6a7bbbfe6 |
perf bench: Add pmu-scan benchmark
The pmu-scan benchmark will repeatedly scan the sysfs to get the
available PMU information.
$ ./perf bench internals pmu-scan
# Running 'internals/pmu-scan' benchmark:
Computing performance of sysfs PMU event scan for 100 times
Average PMU scanning took: 6850.990 usec (+- 48.445 usec)
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Adrian Hunter <adrian.hunter@intel.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@linaro.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20230331202949.810326-2-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
Dmitry Vyukov
|
68a6772f11 |
perf bench: Add breakpoint benchmarks
Add 2 benchmarks:
1. Performance of thread creation/exiting in presence of breakpoints.
2. Performance of breakpoint modification in presence of threads.
The benchmarks capture use cases that we are interested in:
using inheritable breakpoints in large highly-threaded applications.
The benchmarks show significant slowdown imposed by breakpoints
(even when they don't fire).
Testing on Intel 8173M with 112 HW threads show:
perf bench --repeat=56 breakpoint thread --breakpoints=0 --parallelism=56 --threads=20
78.675000 usecs/op
perf bench --repeat=56 breakpoint thread --breakpoints=4 --parallelism=56 --threads=20
12967.135714 usecs/op
That's 165x slowdown due to presence of the breakpoints.
perf bench --repeat=20000 breakpoint enable --passive=0 --active=0
1.433250 usecs/op
perf bench --repeat=20000 breakpoint enable --passive=224 --active=0
585.318400 usecs/op
perf bench --repeat=20000 breakpoint enable --passive=0 --active=111
635.953000 usecs/op
That's 408x and 444x slowdown due to presence of threads.
Profiles show some overhead in toggle_bp_slot,
but also very high contention:
90.83% breakpoint-thre [kernel.kallsyms] [k] osq_lock
4.69% breakpoint-thre [kernel.kallsyms] [k] mutex_spin_on_owner
2.06% breakpoint-thre [kernel.kallsyms] [k] __reserve_bp_slot
2.04% breakpoint-thre [kernel.kallsyms] [k] toggle_bp_slot
79.01% breakpoint-enab [kernel.kallsyms] [k] smp_call_function_single
9.94% breakpoint-enab [kernel.kallsyms] [k] llist_add_batch
5.70% breakpoint-enab [kernel.kallsyms] [k] _raw_spin_lock_irq
1.84% breakpoint-enab [kernel.kallsyms] [k] event_function_call
1.12% breakpoint-enab [kernel.kallsyms] [k] send_call_function_single_ipi
0.37% breakpoint-enab [kernel.kallsyms] [k] generic_exec_single
0.24% breakpoint-enab [kernel.kallsyms] [k] __perf_event_disable
0.20% breakpoint-enab [kernel.kallsyms] [k] _perf_event_enable
0.18% breakpoint-enab [kernel.kallsyms] [k] toggle_bp_slot
Committer notes:
Fixup struct init for older compilers:
3 32.90 alpine:3.5 : FAIL clang version 3.8.1 (tags/RELEASE_381/final)
bench/breakpoint.c:49:34: error: missing field 'size' initializer [-Werror,-Wmissing-field-initializers]
struct perf_event_attr attr = {0};
^
1 error generated.
7 37.31 alpine:3.9 : FAIL gcc version 8.3.0 (Alpine 8.3.0)
bench/breakpoint.c:49:34: error: missing field 'size' initializer [-Werror,-Wmissing-field-initializers]
struct perf_event_attr attr = {0};
^
1 error generated.
Signed-off-by: Dmitriy Vyukov <dvyukov@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Marco Elver <elver@google.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220505155745.1690906-1-dvyukov@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
Riccardo Mancini
|
4241eabf59 |
perf bench: Add benchmark for evlist open/close operations
This new benchmark finds the total time that is taken to open, mmap,
enable, disable, munmap, close an evlist (time taken for new,
create_maps, config, delete is not counted in).
The evlist can be configured as in perf-record using the
-a,-C,-e,-u,--per-thread,-t,-p options.
The events can be duplicated in the evlist to quickly test performance
with many events using the -n options.
Furthermore, also the number of iterations used to calculate the
statistics is customizable.
Examples:
- Open one dummy event system-wide:
$ sudo ./perf bench internals evlist-open-close
Number of cpus: 4
Number of threads: 1
Number of events: 1 (4 fds)
Number of iterations: 100
Average open-close took: 613.870 usec (+- 32.852 usec)
- Open the group '{cs,cycles}' on CPU 0
$ sudo ./perf bench internals evlist-open-close -e '{cs,cycles}' -C 0
Number of cpus: 1
Number of threads: 1
Number of events: 2 (2 fds)
Number of iterations: 100
Average open-close took: 8503.220 usec (+- 252.652 usec)
- Open 10 'cycles' events for user 0, calculate average over 100 runs
$ sudo ./perf bench internals evlist-open-close -e cycles -n 10 -u 0 -i 100
Number of cpus: 4
Number of threads: 328
Number of events: 10 (13120 fds)
Number of iterations: 100
Average open-close took: 180043.140 usec (+- 2295.889 usec)
Committer notes:
Replaced a deprecated bzero() call with designated initialized zeroing.
Added some missing evlist allocation checks, one noted by Riccardo on
the mailing list.
Minor cosmetic changes (sent in private).
Signed-off-by: Riccardo Mancini <rickyman7@gmail.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lore.kernel.org/lkml/20210809201101.277594-1-rickyman7@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Linus Torvalds
|
9d9af1007b |
perf tools changes for v5.10: 1st batch
- cgroup improvements for 'perf stat', allowing for compact specification of events
and cgroups in the command line.
- Support per thread topdown metrics in 'perf stat'.
- Support sample-read topdown metric group in 'perf record'
- Show start of latency in addition to its start in 'perf sched latency'.
- Add min, max to 'perf script' futex-contention output, in addition to avg.
- Allow usage of 'perf_event_attr->exclusive' attribute via the new ':e' event
modifier.
- Add 'snapshot' command to 'perf record --control', using it with Intel PT.
- Support FIFO file names as alternative options to 'perf record --control'.
- Introduce branch history "streams", to compare 'perf record' runs with
'perf diff' based on branch records and report hot streams.
- Support PE executable symbol tables using libbfd, to profile, for instance, wine binaries.
- Add filter support for option 'perf ftrace -F/--funcs'.
- Allow configuring the 'disassembler_style' 'perf annotate' knob via 'perf config'
- Update CascadelakeX and SkylakeX JSON vendor events files.
- Add support for parsing perchip/percore JSON vendor events.
- Add power9 hv_24x7 core level metric events.
- Add L2 prefetch, ITLB instruction fetch hits JSON events for AMD zen1.
- Enable Family 19h users by matching Zen2 AMD vendor events.
- Use debuginfod in 'perf probe' when required debug files not found locally.
- Display negative tid in non-sample events in 'perf script'.
- Make GTK2 support opt-in
- Add build test with GTK+
- Add missing -lzstd to the fast path feature detection
- Add scripts to auto generate 'mmap', 'mremap' string<->id tables for use in 'perf trace'.
- Show python test script in verbose mode.
- Fix uncore metric expressions
- Msan uninitialized use fixes.
- Use condition variables in 'perf bench numa'
- Autodetect python3 binary in systems without python2.
- Support md5 build ids in addition to sha1.
- Add build id 'perf test' regression test.
- Fix printable strings in python3 scripts.
- Fix off by ones in 'perf trace' in arches using libaudit.
- Fix JSON event code for events referencing std arch events.
- Introduce 'perf test' shell script for Arm CoreSight testing.
- Add rdtsc() for Arm64 for used in the PERF_RECORD_TIME_CONV metadata
event and in 'perf test tsc'.
- 'perf c2c' improvements: Add "RMT Load Hit" metric, "Total Stores", fixes
and documentation update.
- Fix usage of reloc_sym in 'perf probe' when using both kallsyms and debuginfo files.
- Do not print 'Metric Groups:' unnecessarily in 'perf list'
- Refcounting fixes in the event parsing code.
- Add expand cgroup event 'perf test' entry.
- Fix out of bounds CPU map access when handling armv8_pmu events in 'perf stat'.
- Add build-id injection 'perf bench' benchmark.
- Enter namespace when reading build-id in 'perf inject'.
- Do not load map/dso when injecting build-id speeding up the 'perf inject' process.
- Add --buildid-all option to avoid processing all samples, just the mmap metadata events.
- Add feature test to check if libbfd has buildid support
- Add 'perf test' entry for PE binary format support.
- Fix typos in power8 PMU vendor events JSON files.
- Hide libtraceevent non API functions.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Test results:
The first ones are container based builds of tools/perf with and without libelf
support. Where clang is available, it is also used to build perf with/without
libelf, and building with LIBCLANGLLVM=1 (built-in clang) with gcc and clang
when clang and its devel libraries are installed.
The objtool and samples/bpf/ builds are disabled now that I'm switching from
using the sources in a local volume to fetching them from a http server to
build it inside the container, to make it easier to build in a container cluster.
Those will come back later.
Several are cross builds, the ones with -x-ARCH and the android one, and those
may not have all the features built, due to lack of multi-arch devel packages,
available and being used so far on just a few, like
debian:experimental-x-{arm64,mipsel}.
The 'perf test' one will perform a variety of tests exercising
tools/perf/util/, tools/lib/{bpf,traceevent,etc}, as well as run perf commands
with a variety of command line event specifications to then intercept the
sys_perf_event syscall to check that the perf_event_attr fields are set up as
expected, among a variety of other unit tests.
Then there is the 'make -C tools/perf build-test' ones, that build tools/perf/
with a variety of feature sets, exercising the build with an incomplete set of
features as well as with a complete one. It is planned to have it run on each
of the containers mentioned above, using some container orchestration
infrastructure. Get in contact if interested in helping having this in place.
$ grep "model name" -m1 /proc/cpuinfo
model name: AMD Ryzen 9 3900X 12-Core Processor
$ export PERF_TARBALL=http://192.168.122.1/perf/perf-5.9.0-rc7.tar.xz
$ dm
Thu 15 Oct 2020 01:10:56 PM -03
1 67.40 alpine:3.4 : Ok gcc (Alpine 5.3.0) 5.3.0, clang version 3.8.0 (tags/RELEASE_380/final)
2 69.01 alpine:3.5 : Ok gcc (Alpine 6.2.1) 6.2.1 20160822, clang version 3.8.1 (tags/RELEASE_381/final)
3 70.79 alpine:3.6 : Ok gcc (Alpine 6.3.0) 6.3.0, clang version 4.0.0 (tags/RELEASE_400/final)
4 79.89 alpine:3.7 : Ok gcc (Alpine 6.4.0) 6.4.0, Alpine clang version 5.0.0 (tags/RELEASE_500/final) (based on LLVM 5.0.0)
5 80.88 alpine:3.8 : Ok gcc (Alpine 6.4.0) 6.4.0, Alpine clang version 5.0.1 (tags/RELEASE_501/final) (based on LLVM 5.0.1)
6 83.88 alpine:3.9 : Ok gcc (Alpine 8.3.0) 8.3.0, Alpine clang version 5.0.1 (tags/RELEASE_502/final) (based on LLVM 5.0.1)
7 107.87 alpine:3.10 : Ok gcc (Alpine 8.3.0) 8.3.0, Alpine clang version 8.0.0 (tags/RELEASE_800/final) (based on LLVM 8.0.0)
8 115.43 alpine:3.11 : Ok gcc (Alpine 9.3.0) 9.3.0, Alpine clang version 9.0.0 (https://git.alpinelinux.org/aports f7f0d2c2b8bcd6a5843401a9a702029556492689) (based on LLVM 9.0.0)
9 106.80 alpine:3.12 : Ok gcc (Alpine 9.3.0) 9.3.0, Alpine clang version 10.0.0 (https://gitlab.alpinelinux.org/alpine/aports.git 7445adce501f8473efdb93b17b5eaf2f1445ed4c)
10 114.06 alpine:edge : Ok gcc (Alpine 10.2.0) 10.2.0, Alpine clang version 10.0.1
11 70.42 alt:p8 : Ok x86_64-alt-linux-gcc (GCC) 5.3.1 20151207 (ALT p8 5.3.1-alt3.M80P.1), clang version 3.8.0 (tags/RELEASE_380/final)
12 98.70 alt:p9 : Ok x86_64-alt-linux-gcc (GCC) 8.4.1 20200305 (ALT p9 8.4.1-alt0.p9.1), clang version 10.0.0
13 80.37 alt:sisyphus : Ok x86_64-alt-linux-gcc (GCC) 9.3.1 20200518 (ALT Sisyphus 9.3.1-alt1), clang version 10.0.1
14 64.12 amazonlinux:1 : Ok gcc (GCC) 7.2.1 20170915 (Red Hat 7.2.1-2), clang version 3.6.2 (tags/RELEASE_362/final)
15 97.64 amazonlinux:2 : Ok gcc (GCC) 7.3.1 20180712 (Red Hat 7.3.1-9), clang version 7.0.1 (Amazon Linux 2 7.0.1-1.amzn2.0.2)
16 22.70 android-ndk:r12b-arm : Ok arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease)
17 22.72 android-ndk:r15c-arm : Ok arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease)
18 26.70 centos:6 : Ok gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-23)
19 31.86 centos:7 : Ok gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39)
20 113.19 centos:8 : Ok gcc (GCC) 8.3.1 20191121 (Red Hat 8.3.1-5), clang version 9.0.1 (Red Hat 9.0.1-2.module_el8.2.0+309+0c7b6b03)
21 57.23 clearlinux:latest : Ok gcc (Clear Linux OS for Intel Architecture) 10.2.1 20200908 releases/gcc-10.2.0-203-g127d693955, clang version 10.0.1
22 64.98 debian:8 : Ok gcc (Debian 4.9.2-10+deb8u2) 4.9.2, Debian clang version 3.5.0-10 (tags/RELEASE_350/final) (based on LLVM 3.5.0)
23 76.08 debian:9 : Ok gcc (Debian 6.3.0-18+deb9u1) 6.3.0 20170516, clang version 3.8.1-24 (tags/RELEASE_381/final)
24 74.49 debian:10 : Ok gcc (Debian 8.3.0-6) 8.3.0, clang version 7.0.1-8+deb10u2 (tags/RELEASE_701/final)
25 78.50 debian:experimental : Ok gcc (Debian 10.2.0-15) 10.2.0, Debian clang version 11.0.0-2
26 33.30 debian:experimental-x-arm64 : Ok aarch64-linux-gnu-gcc (Debian 10.2.0-3) 10.2.0
27 30.96 debian:experimental-x-mips64 : Ok mips64-linux-gnuabi64-gcc (Debian 9.3.0-8) 9.3.0
28 32.63 debian:experimental-x-mipsel : Ok mipsel-linux-gnu-gcc (Debian 9.3.0-8) 9.3.0
29 30.12 fedora:20 : Ok gcc (GCC) 4.8.3 20140911 (Red Hat 4.8.3-7)
30 30.99 fedora:22 : Ok gcc (GCC) 5.3.1 20160406 (Red Hat 5.3.1-6), clang version 3.5.0 (tags/RELEASE_350/final)
31 68.60 fedora:23 : Ok gcc (GCC) 5.3.1 20160406 (Red Hat 5.3.1-6), clang version 3.7.0 (tags/RELEASE_370/final)
32 78.92 fedora:24 : Ok gcc (GCC) 6.3.1 20161221 (Red Hat 6.3.1-1), clang version 3.8.1 (tags/RELEASE_381/final)
33 26.15 fedora:24-x-ARC-uClibc : Ok arc-linux-gcc (ARCompact ISA Linux uClibc toolchain 2017.09-rc2) 7.1.1 20170710
34 80.13 fedora:25 : Ok gcc (GCC) 6.4.1 20170727 (Red Hat 6.4.1-1), clang version 3.9.1 (tags/RELEASE_391/final)
35 90.68 fedora:26 : Ok gcc (GCC) 7.3.1 20180130 (Red Hat 7.3.1-2), clang version 4.0.1 (tags/RELEASE_401/final)
36 90.45 fedora:27 : Ok gcc (GCC) 7.3.1 20180712 (Red Hat 7.3.1-6), clang version 5.0.2 (tags/RELEASE_502/final)
37 100.88 fedora:28 : Ok gcc (GCC) 8.3.1 20190223 (Red Hat 8.3.1-2), clang version 6.0.1 (tags/RELEASE_601/final)
38 105.99 fedora:29 : Ok gcc (GCC) 8.3.1 20190223 (Red Hat 8.3.1-2), clang version 7.0.1 (Fedora 7.0.1-6.fc29)
39 111.05 fedora:30 : Ok gcc (GCC) 9.3.1 20200408 (Red Hat 9.3.1-2), clang version 8.0.0 (Fedora 8.0.0-3.fc30)
40 29.96 fedora:30-x-ARC-glibc : Ok arc-linux-gcc (ARC HS GNU/Linux glibc toolchain 2019.03-rc1) 8.3.1 20190225
41 27.02 fedora:30-x-ARC-uClibc : Ok arc-linux-gcc (ARCv2 ISA Linux uClibc toolchain 2019.03-rc1) 8.3.1 20190225
42 110.47 fedora:31 : Ok gcc (GCC) 9.3.1 20200408 (Red Hat 9.3.1-2), clang version 9.0.1 (Fedora 9.0.1-2.fc31)
43 88.78 fedora:32 : Ok gcc (GCC) 10.2.1 20200723 (Red Hat 10.2.1-1), clang version 10.0.0 (Fedora 10.0.0-2.fc32)
44 15.92 fedora:rawhide : FAIL gcc (GCC) 10.2.1 20200916 (Red Hat 10.2.1-4), clang version 11.0.0 (Fedora 11.0.0-0.4.rc3.fc34)
45 33.58 gentoo-stage3-amd64:latest : Ok gcc (Gentoo 9.3.0-r1 p3) 9.3.0
46 65.32 mageia:5 : Ok gcc (GCC) 4.9.2, clang version 3.5.2 (tags/RELEASE_352/final)
47 81.35 mageia:6 : Ok gcc (Mageia 5.5.0-1.mga6) 5.5.0, clang version 3.9.1 (tags/RELEASE_391/final)
48 103.94 mageia:7 : Ok gcc (Mageia 8.4.0-1.mga7) 8.4.0, clang version 8.0.0 (Mageia 8.0.0-1.mga7)
49 91.62 manjaro:latest : Ok gcc (GCC) 10.2.0, clang version 10.0.1
50 219.87 openmandriva:cooker : Ok gcc (GCC) 10.2.0 20200723 (OpenMandriva), OpenMandriva 11.0.0-0.20200909.1 clang version 11.0.0 (/builddir/build/BUILD/llvm-project-release-11.x/clang 5cb8ffbab42358a7cdb0a67acfadb84df0779579)
51 111.76 opensuse:15.0 : Ok gcc (SUSE Linux) 7.4.1 20190905 [gcc-7-branch revision 275407], clang version 5.0.1 (tags/RELEASE_501/final 312548)
52 118.03 opensuse:15.1 : Ok gcc (SUSE Linux) 7.5.0, clang version 7.0.1 (tags/RELEASE_701/final 349238)
53 107.91 opensuse:15.2 : Ok gcc (SUSE Linux) 7.5.0, clang version 9.0.1
54 102.34 opensuse:tumbleweed : Ok gcc (SUSE Linux) 10.2.1 20200825 [revision c0746a1beb1ba073c7981eb09f55b3d993b32e5c], clang version 10.0.1
55 25.33 oraclelinux:6 : Ok gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-23.0.1)
56 30.45 oraclelinux:7 : Ok gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-44.0.3)
57 104.65 oraclelinux:8 : Ok gcc (GCC) 8.3.1 20191121 (Red Hat 8.3.1-5.0.3), clang version 9.0.1 (Red Hat 9.0.1-2.0.1.module+el8.2.0+5599+9ed9ef6d)
58 26.04 ubuntu:12.04 : Ok gcc (Ubuntu/Linaro 4.6.3-1ubuntu5) 4.6.3, Ubuntu clang version 3.0-6ubuntu3 (tags/RELEASE_30/final) (based on LLVM 3.0)
59 29.49 ubuntu:14.04 : Ok gcc (Ubuntu 4.8.4-2ubuntu1~14.04.4) 4.8.4
60 72.95 ubuntu:16.04 : Ok gcc (Ubuntu 5.4.0-6ubuntu1~16.04.12) 5.4.0 20160609, clang version 3.8.0-2ubuntu4 (tags/RELEASE_380/final)
61 26.03 ubuntu:16.04-x-arm : Ok arm-linux-gnueabihf-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
62 25.15 ubuntu:16.04-x-arm64 : Ok aarch64-linux-gnu-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
63 24.88 ubuntu:16.04-x-powerpc : Ok powerpc-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
64 25.72 ubuntu:16.04-x-powerpc64 : Ok powerpc64-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
65 25.39 ubuntu:16.04-x-powerpc64el : Ok powerpc64le-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
66 25.34 ubuntu:16.04-x-s390 : Ok s390x-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609
67 84.84 ubuntu:18.04 : Ok gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0, clang version 6.0.0-1ubuntu2 (tags/RELEASE_600/final)
68 27.15 ubuntu:18.04-x-arm : Ok arm-linux-gnueabihf-gcc (Ubuntu/Linaro 7.5.0-3ubuntu1~18.04) 7.5.0
69 26.68 ubuntu:18.04-x-arm64 : Ok aarch64-linux-gnu-gcc (Ubuntu/Linaro 7.5.0-3ubuntu1~18.04) 7.5.0
70 22.38 ubuntu:18.04-x-m68k : Ok m68k-linux-gnu-gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
71 26.35 ubuntu:18.04-x-powerpc : Ok powerpc-linux-gnu-gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
72 28.58 ubuntu:18.04-x-powerpc64 : Ok powerpc64-linux-gnu-gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
73 28.18 ubuntu:18.04-x-powerpc64el : Ok powerpc64le-linux-gnu-gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
74 178.55 ubuntu:18.04-x-riscv64 : Ok riscv64-linux-gnu-gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
75 24.58 ubuntu:18.04-x-s390 : Ok s390x-linux-gnu-gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
76 26.89 ubuntu:18.04-x-sh4 : Ok sh4-linux-gnu-gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
77 24.81 ubuntu:18.04-x-sparc64 : Ok sparc64-linux-gnu-gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
78 68.90 ubuntu:19.10 : Ok gcc (Ubuntu 9.2.1-9ubuntu2) 9.2.1 20191008, clang version 8.0.1-3build1 (tags/RELEASE_801/final)
79 69.31 ubuntu:20.04 : Ok gcc (Ubuntu 9.3.0-10ubuntu2) 9.3.0, clang version 10.0.0-4ubuntu1
80 30.00 ubuntu:20.04-x-powerpc64el : Ok powerpc64le-linux-gnu-gcc (Ubuntu 10-20200411-0ubuntu1) 10.0.1 20200411 (experimental) [master revision bb87d5cc77d:75961caccb7:f883c46b4877f637e0fa5025b4d6b5c9040ec566]
81 70.34 ubuntu:20.10 : Ok gcc (Ubuntu 10.2.0-5ubuntu2) 10.2.0, Ubuntu clang version 10.0.1-1
$
# uname -a
Linux five 5.9.0+ #1 SMP Thu Oct 15 09:06:41 -03 2020 x86_64 x86_64 x86_64 GNU/Linux
# git log --oneline -1
744aec4df2c5 perf c2c: Update documentation for metrics reorganization
# perf version --build-options
perf version 5.9.rc7.g744aec4df2c5
dwarf: [ on ] # HAVE_DWARF_SUPPORT
dwarf_getlocations: [ on ] # HAVE_DWARF_GETLOCATIONS_SUPPORT
glibc: [ on ] # HAVE_GLIBC_SUPPORT
syscall_table: [ on ] # HAVE_SYSCALL_TABLE_SUPPORT
libbfd: [ on ] # HAVE_LIBBFD_SUPPORT
libelf: [ on ] # HAVE_LIBELF_SUPPORT
libnuma: [ on ] # HAVE_LIBNUMA_SUPPORT
numa_num_possible_cpus: [ on ] # HAVE_LIBNUMA_SUPPORT
libperl: [ on ] # HAVE_LIBPERL_SUPPORT
libpython: [ on ] # HAVE_LIBPYTHON_SUPPORT
libslang: [ on ] # HAVE_SLANG_SUPPORT
libcrypto: [ on ] # HAVE_LIBCRYPTO_SUPPORT
libunwind: [ on ] # HAVE_LIBUNWIND_SUPPORT
libdw-dwarf-unwind: [ on ] # HAVE_DWARF_SUPPORT
zlib: [ on ] # HAVE_ZLIB_SUPPORT
lzma: [ on ] # HAVE_LZMA_SUPPORT
get_cpuid: [ on ] # HAVE_AUXTRACE_SUPPORT
bpf: [ on ] # HAVE_LIBBPF_SUPPORT
aio: [ on ] # HAVE_AIO_SUPPORT
zstd: [ on ] # HAVE_ZSTD_SUPPORT
# perf test
1: vmlinux symtab matches kallsyms : Ok
2: Detect openat syscall event : Ok
3: Detect openat syscall event on all cpus : Ok
4: Read samples using the mmap interface : Ok
5: Test data source output : Ok
6: Parse event definition strings : Ok
7: Simple expression parser : Ok
8: PERF_RECORD_* events & perf_sample fields : Ok
9: Parse perf pmu format : Ok
10: PMU events :
10.1: PMU event table sanity : Ok
10.2: PMU event map aliases : Ok
10.3: Parsing of PMU event table metrics : Ok
10.4: Parsing of PMU event table metrics with fake PMUs : Ok
11: DSO data read : Ok
12: DSO data cache : Ok
13: DSO data reopen : Ok
14: Roundtrip evsel->name : Ok
15: Parse sched tracepoints fields : Ok
16: syscalls:sys_enter_openat event fields : Ok
17: Setup struct perf_event_attr : Ok
18: Match and link multiple hists : Ok
19: 'import perf' in python : Ok
20: Breakpoint overflow signal handler : Ok
21: Breakpoint overflow sampling : Ok
22: Breakpoint accounting : Ok
23: Watchpoint :
23.1: Read Only Watchpoint : Skip
23.2: Write Only Watchpoint : Ok
23.3: Read / Write Watchpoint : Ok
23.4: Modify Watchpoint : Ok
24: Number of exit events of a simple workload : Ok
25: Software clock events period values : Ok
26: Object code reading : Ok
27: Sample parsing : Ok
28: Use a dummy software event to keep tracking : Ok
29: Parse with no sample_id_all bit set : Ok
30: Filter hist entries : Ok
31: Lookup mmap thread : Ok
32: Share thread maps : Ok
33: Sort output of hist entries : Ok
34: Cumulate child hist entries : Ok
35: Track with sched_switch : Ok
36: Filter fds with revents mask in a fdarray : Ok
37: Add fd to a fdarray, making it autogrow : Ok
38: kmod_path__parse : Ok
39: Thread map : Ok
40: LLVM search and compile :
40.1: Basic BPF llvm compile : Ok
40.2: kbuild searching : Ok
40.3: Compile source for BPF prologue generation : Ok
40.4: Compile source for BPF relocation : Ok
41: Session topology : Ok
42: BPF filter :
42.1: Basic BPF filtering : Ok
42.2: BPF pinning : Ok
42.3: BPF prologue generation : Ok
42.4: BPF relocation checker : Ok
43: Synthesize thread map : Ok
44: Remove thread map : Ok
45: Synthesize cpu map : Ok
46: Synthesize stat config : Ok
47: Synthesize stat : Ok
48: Synthesize stat round : Ok
49: Synthesize attr update : Ok
50: Event times : Ok
51: Read backward ring buffer : Ok
52: Print cpu map : Ok
53: Merge cpu map : Ok
54: Probe SDT events : Ok
55: is_printable_array : Ok
56: Print bitmap : Ok
57: perf hooks : Ok
58: builtin clang support : Skip (not compiled in)
59: unit_number__scnprintf : Ok
60: mem2node : Ok
61: time utils : Ok
62: Test jit_write_elf : Ok
63: Test libpfm4 support : Skip (not compiled in)
64: Test api io : Ok
65: maps__merge_in : Ok
66: Demangle Java : Ok
67: Parse and process metrics : Ok
68: PE file support : Ok
69: Event expansion for cgroups : Ok
70: x86 rdpmc : Ok
71: Convert perf time to TSC : Ok
72: DWARF unwind : Ok
73: x86 instruction decoder - new instructions : Ok
74: Intel PT packet decoder : Ok
75: x86 bp modify : Ok
76: probe libc's inet_pton & backtrace it with ping : Ok
77: Check Arm CoreSight trace data recording and synthesized samples: Skip
78: Use vfs_getname probe to get syscall args filenames : Ok
79: Check open filename arg using perf trace + vfs_getname : Ok
80: Zstd perf.data compression/decompression : Ok
81: Add vfs_getname probe to get syscall args filenames : Ok
82: build id cache operations : Ok
#
$ git log --oneline -1
744aec4df2c5b4d1 (HEAD -> perf/core, quaco/perf/core) perf c2c: Update documentation for metrics reorganization
$ make -C tools/perf build-test
make: Entering directory '/home/acme/git/perf/tools/perf'
- tarpkg: ./tests/perf-targz-src-pkg .
make_install_bin_O: make install-bin
make_static_O: make LDFLAGS=-static NO_PERF_READ_VDSO32=1 NO_PERF_READ_VDSOX32=1 NO_JVMTI=1
make_no_libdw_dwarf_unwind_O: make NO_LIBDW_DWARF_UNWIND=1
make_no_newt_O: make NO_NEWT=1
make_no_libbionic_O: make NO_LIBBIONIC=1
make_no_sdt_O: make NO_SDT=1
make_debug_O: make DEBUG=1
make_perf_o_O: make perf.o
make_no_libbpf_O: make NO_LIBBPF=1
make_no_libbpf_DEBUG_O: make NO_LIBBPF=1 DEBUG=1
make_clean_all_O: make clean all
make_tags_O: make tags
make_with_babeltrace_O: make LIBBABELTRACE=1
make_with_clangllvm_O: make LIBCLANGLLVM=1
make_no_scripts_O: make NO_LIBPYTHON=1 NO_LIBPERL=1
make_no_libelf_O: make NO_LIBELF=1
make_no_libcrypto_O: make NO_LIBCRYPTO=1
make_with_libpfm4_O: make LIBPFM4=1
make_no_libunwind_O: make NO_LIBUNWIND=1
make_util_map_o_O: make util/map.o
make_no_slang_O: make NO_SLANG=1
make_with_gtk2_O: make GTK2=1
make_no_ui_O: make NO_NEWT=1 NO_SLANG=1 NO_GTK2=1
make_util_pmu_bison_o_O: make util/pmu-bison.o
make_no_backtrace_O: make NO_BACKTRACE=1
make_no_demangle_O: make NO_DEMANGLE=1
make_help_O: make help
make_pure_O: make
make_no_gtk2_O: make NO_GTK2=1
make_install_prefix_O: make install prefix=/tmp/krava
make_no_libnuma_O: make NO_LIBNUMA=1
make_no_libpython_O: make NO_LIBPYTHON=1
make_install_prefix_slash_O: make install prefix=/tmp/krava/
make_no_libaudit_O: make NO_LIBAUDIT=1
make_no_auxtrace_O: make NO_AUXTRACE=1
make_minimal_O: make NO_LIBPERL=1 NO_LIBPYTHON=1 NO_NEWT=1 NO_GTK2=1 NO_DEMANGLE=1 NO_LIBELF=1 NO_LIBUNWIND=1 NO_BACKTRACE=1 NO_LIBNUMA=1 NO_LIBAUDIT=1 NO_LIBBIONIC=1 NO_LIBDW_DWARF_UNWIND=1 NO_AUXTRACE=1 NO_LIBBPF=1 NO_LIBCRYPTO=1 NO_SDT=1 NO_JVMTI=1 NO_LIBZSTD=1 NO_LIBCAP=1 NO_SYSCALL_TABLE=1
make_install_O: make install
make_doc_O: make doc
make_no_libperl_O: make NO_LIBPERL=1
make_no_syscall_tbl_O: make NO_SYSCALL_TABLE=1
OK
make: Leaving directory '/home/acme/git/perf/tools/perf'
$
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQR2GiIUctdOfX2qHhGyPKLppCJ+JwUCX4iuzgAKCRCyPKLppCJ+
J1khAP4iMQMFCMpNsBaL6KLtj3aTOhrooYuhbNL3kajqYVyW/QD8Dws35k6m2+tB
tcOMJykFjPkQ4I13zsxKyugeJuUzSQw=
=KdSj
-----END PGP SIGNATURE-----
Merge tag 'perf-tools-for-v5.10-2020-10-15' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux
Pull perf tools updates from Arnaldo Carvalho de Melo:
- cgroup improvements for 'perf stat', allowing for compact
specification of events and cgroups in the command line.
- Support per thread topdown metrics in 'perf stat'.
- Support sample-read topdown metric group in 'perf record'
- Show start of latency in addition to its start in 'perf sched
latency'.
- Add min, max to 'perf script' futex-contention output, in addition to
avg.
- Allow usage of 'perf_event_attr->exclusive' attribute via the new
':e' event modifier.
- Add 'snapshot' command to 'perf record --control', using it with
Intel PT.
- Support FIFO file names as alternative options to 'perf record
--control'.
- Introduce branch history "streams", to compare 'perf record' runs
with 'perf diff' based on branch records and report hot streams.
- Support PE executable symbol tables using libbfd, to profile, for
instance, wine binaries.
- Add filter support for option 'perf ftrace -F/--funcs'.
- Allow configuring the 'disassembler_style' 'perf annotate' knob via
'perf config'
- Update CascadelakeX and SkylakeX JSON vendor events files.
- Add support for parsing perchip/percore JSON vendor events.
- Add power9 hv_24x7 core level metric events.
- Add L2 prefetch, ITLB instruction fetch hits JSON events for AMD
zen1.
- Enable Family 19h users by matching Zen2 AMD vendor events.
- Use debuginfod in 'perf probe' when required debug files not found
locally.
- Display negative tid in non-sample events in 'perf script'.
- Make GTK2 support opt-in
- Add build test with GTK+
- Add missing -lzstd to the fast path feature detection
- Add scripts to auto generate 'mmap', 'mremap' string<->id tables for
use in 'perf trace'.
- Show python test script in verbose mode.
- Fix uncore metric expressions
- Msan uninitialized use fixes.
- Use condition variables in 'perf bench numa'
- Autodetect python3 binary in systems without python2.
- Support md5 build ids in addition to sha1.
- Add build id 'perf test' regression test.
- Fix printable strings in python3 scripts.
- Fix off by ones in 'perf trace' in arches using libaudit.
- Fix JSON event code for events referencing std arch events.
- Introduce 'perf test' shell script for Arm CoreSight testing.
- Add rdtsc() for Arm64 for used in the PERF_RECORD_TIME_CONV metadata
event and in 'perf test tsc'.
- 'perf c2c' improvements: Add "RMT Load Hit" metric, "Total Stores",
fixes and documentation update.
- Fix usage of reloc_sym in 'perf probe' when using both kallsyms and
debuginfo files.
- Do not print 'Metric Groups:' unnecessarily in 'perf list'
- Refcounting fixes in the event parsing code.
- Add expand cgroup event 'perf test' entry.
- Fix out of bounds CPU map access when handling armv8_pmu events in
'perf stat'.
- Add build-id injection 'perf bench' benchmark.
- Enter namespace when reading build-id in 'perf inject'.
- Do not load map/dso when injecting build-id speeding up the 'perf
inject' process.
- Add --buildid-all option to avoid processing all samples, just the
mmap metadata events.
- Add feature test to check if libbfd has buildid support
- Add 'perf test' entry for PE binary format support.
- Fix typos in power8 PMU vendor events JSON files.
- Hide libtraceevent non API functions.
* tag 'perf-tools-for-v5.10-2020-10-15' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (113 commits)
perf c2c: Update documentation for metrics reorganization
perf c2c: Add metrics "RMT Load Hit"
perf c2c: Correct LLC load hit metrics
perf c2c: Change header for LLC local hit
perf c2c: Use more explicit headers for HITM
perf c2c: Change header from "LLC Load Hitm" to "Load Hitm"
perf c2c: Organize metrics based on memory hierarchy
perf c2c: Display "Total Stores" as a standalone metrics
perf c2c: Display the total numbers continuously
perf bench: Use condition variables in numa.
perf jevents: Fix event code for events referencing std arch events
perf diff: Support hot streams comparison
perf streams: Report hot streams
perf streams: Calculate the sum of total streams hits
perf streams: Link stream pair
perf streams: Compare two streams
perf streams: Get the evsel_streams by evsel_idx
perf streams: Introduce branch history "streams"
perf intel-pt: Improve PT documentation slightly
perf tools: Add support for exclusive groups/events
...
|
||
|
Namhyung Kim
|
0bf02a0d80 |
perf bench: Add build-id injection benchmark
Sometimes I can see that 'perf record' piped with 'perf inject' take a
long time processing build-ids.
So introduce a inject-build-id benchmark to the internals benchmark
suite to measure its overhead regularly.
It runs the 'perf inject' command internally and feeds the given number
of synthesized events (MMAP2 + SAMPLE basically).
Usage: perf bench internals inject-build-id <options>
-i, --iterations <n> Number of iterations used to compute average (default: 100)
-m, --nr-mmaps <n> Number of mmap events for each iteration (default: 100)
-n, --nr-samples <n> Number of sample events per mmap event (default: 100)
-v, --verbose be more verbose (show iteration count, DSO name, etc)
By default, it measures average processing time of 100 MMAP2 events
and 10000 SAMPLE events. Below is a result on my laptop.
$ perf bench internals inject-build-id
# Running 'internals/inject-build-id' benchmark:
Average build-id injection took: 25.789 msec (+- 0.202 msec)
Average time per event: 2.528 usec (+- 0.020 usec)
Average memory usage: 8411 KB (+- 7 KB)
Committer testing:
$ perf bench
Usage:
perf bench [<common options>] <collection> <benchmark> [<options>]
# List of all available benchmark collections:
sched: Scheduler and IPC benchmarks
syscall: System call benchmarks
mem: Memory access benchmarks
numa: NUMA scheduling and MM benchmarks
futex: Futex stressing benchmarks
epoll: Epoll stressing benchmarks
internals: Perf-internals benchmarks
all: All benchmarks
$ perf bench internals
# List of available benchmarks for collection 'internals':
synthesize: Benchmark perf event synthesis
kallsyms-parse: Benchmark kallsyms parsing
inject-build-id: Benchmark build-id injection
$ perf bench internals inject-build-id
# Running 'internals/inject-build-id' benchmark:
Average build-id injection took: 14.202 msec (+- 0.059 msec)
Average time per event: 1.392 usec (+- 0.006 usec)
Average memory usage: 12650 KB (+- 10 KB)
Average build-id-all injection took: 12.831 msec (+- 0.071 msec)
Average time per event: 1.258 usec (+- 0.007 usec)
Average memory usage: 11895 KB (+- 10 KB)
$
$ perf stat -r5 perf bench internals inject-build-id
# Running 'internals/inject-build-id' benchmark:
Average build-id injection took: 14.380 msec (+- 0.056 msec)
Average time per event: 1.410 usec (+- 0.006 usec)
Average memory usage: 12608 KB (+- 11 KB)
Average build-id-all injection took: 11.889 msec (+- 0.064 msec)
Average time per event: 1.166 usec (+- 0.006 usec)
Average memory usage: 11838 KB (+- 10 KB)
# Running 'internals/inject-build-id' benchmark:
Average build-id injection took: 14.246 msec (+- 0.065 msec)
Average time per event: 1.397 usec (+- 0.006 usec)
Average memory usage: 12744 KB (+- 10 KB)
Average build-id-all injection took: 12.019 msec (+- 0.066 msec)
Average time per event: 1.178 usec (+- 0.006 usec)
Average memory usage: 11963 KB (+- 10 KB)
# Running 'internals/inject-build-id' benchmark:
Average build-id injection took: 14.321 msec (+- 0.067 msec)
Average time per event: 1.404 usec (+- 0.007 usec)
Average memory usage: 12690 KB (+- 10 KB)
Average build-id-all injection took: 11.909 msec (+- 0.041 msec)
Average time per event: 1.168 usec (+- 0.004 usec)
Average memory usage: 11938 KB (+- 10 KB)
# Running 'internals/inject-build-id' benchmark:
Average build-id injection took: 14.287 msec (+- 0.059 msec)
Average time per event: 1.401 usec (+- 0.006 usec)
Average memory usage: 12864 KB (+- 10 KB)
Average build-id-all injection took: 11.862 msec (+- 0.058 msec)
Average time per event: 1.163 usec (+- 0.006 usec)
Average memory usage: 12103 KB (+- 10 KB)
# Running 'internals/inject-build-id' benchmark:
Average build-id injection took: 14.402 msec (+- 0.053 msec)
Average time per event: 1.412 usec (+- 0.005 usec)
Average memory usage: 12876 KB (+- 10 KB)
Average build-id-all injection took: 11.826 msec (+- 0.061 msec)
Average time per event: 1.159 usec (+- 0.006 usec)
Average memory usage: 12111 KB (+- 10 KB)
Performance counter stats for 'perf bench internals inject-build-id' (5 runs):
4,267.48 msec task-clock:u # 1.502 CPUs utilized ( +- 0.14% )
0 context-switches:u # 0.000 K/sec
0 cpu-migrations:u # 0.000 K/sec
102,092 page-faults:u # 0.024 M/sec ( +- 0.08% )
3,894,589,578 cycles:u # 0.913 GHz ( +- 0.19% ) (83.49%)
140,078,421 stalled-cycles-frontend:u # 3.60% frontend cycles idle ( +- 0.77% ) (83.34%)
948,581,189 stalled-cycles-backend:u # 24.36% backend cycles idle ( +- 0.46% ) (83.25%)
5,835,587,719 instructions:u # 1.50 insn per cycle
# 0.16 stalled cycles per insn ( +- 0.21% ) (83.24%)
1,267,423,636 branches:u # 296.996 M/sec ( +- 0.22% ) (83.12%)
17,484,290 branch-misses:u # 1.38% of all branches ( +- 0.12% ) (83.55%)
2.84176 +- 0.00222 seconds time elapsed ( +- 0.08% )
$
Acked-by: Jiri Olsa <jolsa@redhat.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20201012070214.2074921-2-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
Dan Williams
|
ec6347bb43 |
x86, powerpc: Rename memcpy_mcsafe() to copy_mc_to_{user, kernel}()
In reaction to a proposal to introduce a memcpy_mcsafe_fast() implementation Linus points out that memcpy_mcsafe() is poorly named relative to communicating the scope of the interface. Specifically what addresses are valid to pass as source, destination, and what faults / exceptions are handled. Of particular concern is that even though x86 might be able to handle the semantics of copy_mc_to_user() with its common copy_user_generic() implementation other archs likely need / want an explicit path for this case: On Fri, May 1, 2020 at 11:28 AM Linus Torvalds <torvalds@linux-foundation.org> wrote: > > On Thu, Apr 30, 2020 at 6:21 PM Dan Williams <dan.j.williams@intel.com> wrote: > > > > However now I see that copy_user_generic() works for the wrong reason. > > It works because the exception on the source address due to poison > > looks no different than a write fault on the user address to the > > caller, it's still just a short copy. So it makes copy_to_user() work > > for the wrong reason relative to the name. > > Right. > > And it won't work that way on other architectures. On x86, we have a > generic function that can take faults on either side, and we use it > for both cases (and for the "in_user" case too), but that's an > artifact of the architecture oddity. > > In fact, it's probably wrong even on x86 - because it can hide bugs - > but writing those things is painful enough that everybody prefers > having just one function. Replace a single top-level memcpy_mcsafe() with either copy_mc_to_user(), or copy_mc_to_kernel(). Introduce an x86 copy_mc_fragile() name as the rename for the low-level x86 implementation formerly named memcpy_mcsafe(). It is used as the slow / careful backend that is supplanted by a fast copy_mc_generic() in a follow-on patch. One side-effect of this reorganization is that separating copy_mc_64.S to its own file means that perf no longer needs to track dependencies for its memcpy_64.S benchmarks. [ bp: Massage a bit. ] Signed-off-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Borislav Petkov <bp@suse.de> Reviewed-by: Tony Luck <tony.luck@intel.com> Acked-by: Michael Ellerman <mpe@ellerman.id.au> Cc: <stable@vger.kernel.org> Link: http://lore.kernel.org/r/CAHk-=wjSqtXAqfUJxFtWNwmguFASTgB0dz1dT3V-78Quiezqbg@mail.gmail.com Link: https://lkml.kernel.org/r/160195561680.2163339.11574962055305783722.stgit@dwillia2-desk3.amr.corp.intel.com |
||
Ian Rogers
|
7c43b0c1d4 |
perf bench: Add benchmark of find_next_bit
for_each_set_bit, or similar functions like for_each_cpu, may be hot
within the kernel. If many bits were set then one could imagine on Intel
a "bt" instruction with every bit may be faster than the function call
and word length find_next_bit logic. Add a benchmark to measure this.
This benchmark on AMD rome and Intel skylakex shows "bt" is not a good
option except for very small bitmaps.
Committer testing:
# perf bench
Usage:
perf bench [<common options>] <collection> <benchmark> [<options>]
# List of all available benchmark collections:
sched: Scheduler and IPC benchmarks
syscall: System call benchmarks
mem: Memory access benchmarks
numa: NUMA scheduling and MM benchmarks
futex: Futex stressing benchmarks
epoll: Epoll stressing benchmarks
internals: Perf-internals benchmarks
all: All benchmarks
# perf bench mem
# List of available benchmarks for collection 'mem':
memcpy: Benchmark for memcpy() functions
memset: Benchmark for memset() functions
find_bit: Benchmark for find_bit() functions
all: Run all memory access benchmarks
# perf bench mem find_bit
# Running 'mem/find_bit' benchmark:
100000 operations 1 bits set of 1 bits
Average for_each_set_bit took: 730.200 usec (+- 6.468 usec)
Average test_bit loop took: 366.200 usec (+- 4.652 usec)
100000 operations 1 bits set of 2 bits
Average for_each_set_bit took: 781.000 usec (+- 24.247 usec)
Average test_bit loop took: 550.200 usec (+- 4.152 usec)
100000 operations 2 bits set of 2 bits
Average for_each_set_bit took: 1113.400 usec (+- 112.340 usec)
Average test_bit loop took: 1098.500 usec (+- 182.834 usec)
100000 operations 1 bits set of 4 bits
Average for_each_set_bit took: 843.800 usec (+- 8.772 usec)
Average test_bit loop took: 948.800 usec (+- 10.278 usec)
100000 operations 2 bits set of 4 bits
Average for_each_set_bit took: 1185.800 usec (+- 114.345 usec)
Average test_bit loop took: 1473.200 usec (+- 175.498 usec)
100000 operations 4 bits set of 4 bits
Average for_each_set_bit took: 1769.667 usec (+- 233.177 usec)
Average test_bit loop took: 1864.933 usec (+- 187.470 usec)
100000 operations 1 bits set of 8 bits
Average for_each_set_bit took: 898.000 usec (+- 21.755 usec)
Average test_bit loop took: 1768.400 usec (+- 23.672 usec)
100000 operations 2 bits set of 8 bits
Average for_each_set_bit took: 1244.900 usec (+- 116.396 usec)
Average test_bit loop took: 2201.800 usec (+- 145.398 usec)
100000 operations 4 bits set of 8 bits
Average for_each_set_bit took: 1822.533 usec (+- 231.554 usec)
Average test_bit loop took: 2569.467 usec (+- 168.453 usec)
100000 operations 8 bits set of 8 bits
Average for_each_set_bit took: 2845.100 usec (+- 441.365 usec)
Average test_bit loop took: 3023.300 usec (+- 219.575 usec)
100000 operations 1 bits set of 16 bits
Average for_each_set_bit took: 923.400 usec (+- 17.560 usec)
Average test_bit loop took: 3240.000 usec (+- 16.492 usec)
100000 operations 2 bits set of 16 bits
Average for_each_set_bit took: 1264.300 usec (+- 114.034 usec)
Average test_bit loop took: 3714.400 usec (+- 158.898 usec)
100000 operations 4 bits set of 16 bits
Average for_each_set_bit took: 1817.867 usec (+- 222.199 usec)
Average test_bit loop took: 4015.333 usec (+- 154.162 usec)
100000 operations 8 bits set of 16 bits
Average for_each_set_bit took: 2826.350 usec (+- 433.457 usec)
Average test_bit loop took: 4460.350 usec (+- 210.762 usec)
100000 operations 16 bits set of 16 bits
Average for_each_set_bit took: 4615.600 usec (+- 809.350 usec)
Average test_bit loop took: 5129.960 usec (+- 320.821 usec)
100000 operations 1 bits set of 32 bits
Average for_each_set_bit took: 904.400 usec (+- 14.250 usec)
Average test_bit loop took: 6194.000 usec (+- 29.254 usec)
100000 operations 2 bits set of 32 bits
Average for_each_set_bit took: 1252.700 usec (+- 116.432 usec)
Average test_bit loop took: 6652.400 usec (+- 154.352 usec)
100000 operations 4 bits set of 32 bits
Average for_each_set_bit took: 1824.200 usec (+- 229.133 usec)
Average test_bit loop took: 6961.733 usec (+- 154.682 usec)
100000 operations 8 bits set of 32 bits
Average for_each_set_bit took: 2823.950 usec (+- 432.296 usec)
Average test_bit loop took: 7351.900 usec (+- 193.626 usec)
100000 operations 16 bits set of 32 bits
Average for_each_set_bit took: 4552.560 usec (+- 785.141 usec)
Average test_bit loop took: 7998.360 usec (+- 305.629 usec)
100000 operations 32 bits set of 32 bits
Average for_each_set_bit took: 7557.067 usec (+- 1407.702 usec)
Average test_bit loop took: 9072.400 usec (+- 513.209 usec)
100000 operations 1 bits set of 64 bits
Average for_each_set_bit took: 896.800 usec (+- 14.389 usec)
Average test_bit loop took: 11927.200 usec (+- 68.862 usec)
100000 operations 2 bits set of 64 bits
Average for_each_set_bit took: 1230.400 usec (+- 111.731 usec)
Average test_bit loop took: 12478.600 usec (+- 189.382 usec)
100000 operations 4 bits set of 64 bits
Average for_each_set_bit took: 1844.733 usec (+- 244.826 usec)
Average test_bit loop took: 12911.467 usec (+- 206.246 usec)
100000 operations 8 bits set of 64 bits
Average for_each_set_bit took: 2779.300 usec (+- 413.612 usec)
Average test_bit loop took: 13372.650 usec (+- 239.623 usec)
100000 operations 16 bits set of 64 bits
Average for_each_set_bit took: 4423.920 usec (+- 748.240 usec)
Average test_bit loop took: 13995.800 usec (+- 318.427 usec)
100000 operations 32 bits set of 64 bits
Average for_each_set_bit took: 7580.600 usec (+- 1462.407 usec)
Average test_bit loop took: 15063.067 usec (+- 516.477 usec)
100000 operations 64 bits set of 64 bits
Average for_each_set_bit took: 13391.514 usec (+- 2765.371 usec)
Average test_bit loop took: 16974.914 usec (+- 916.936 usec)
100000 operations 1 bits set of 128 bits
Average for_each_set_bit took: 1153.800 usec (+- 124.245 usec)
Average test_bit loop took: 26959.000 usec (+- 714.047 usec)
100000 operations 2 bits set of 128 bits
Average for_each_set_bit took: 1445.200 usec (+- 113.587 usec)
Average test_bit loop took: 25798.800 usec (+- 512.908 usec)
100000 operations 4 bits set of 128 bits
Average for_each_set_bit took: 1990.933 usec (+- 219.362 usec)
Average test_bit loop took: 25589.400 usec (+- 348.288 usec)
100000 operations 8 bits set of 128 bits
Average for_each_set_bit took: 2963.000 usec (+- 419.487 usec)
Average test_bit loop took: 25690.050 usec (+- 262.025 usec)
100000 operations 16 bits set of 128 bits
Average for_each_set_bit took: 4585.200 usec (+- 741.734 usec)
Average test_bit loop took: 26125.040 usec (+- 274.127 usec)
100000 operations 32 bits set of 128 bits
Average for_each_set_bit took: 7626.200 usec (+- 1404.950 usec)
Average test_bit loop took: 27038.867 usec (+- 442.554 usec)
100000 operations 64 bits set of 128 bits
Average for_each_set_bit took: 13343.371 usec (+- 2686.460 usec)
Average test_bit loop took: 28936.543 usec (+- 883.257 usec)
100000 operations 128 bits set of 128 bits
Average for_each_set_bit took: 23442.950 usec (+- 4880.541 usec)
Average test_bit loop took: 32484.125 usec (+- 1691.931 usec)
100000 operations 1 bits set of 256 bits
Average for_each_set_bit took: 1183.000 usec (+- 32.073 usec)
Average test_bit loop took: 50114.600 usec (+- 198.880 usec)
100000 operations 2 bits set of 256 bits
Average for_each_set_bit took: 1550.000 usec (+- 124.550 usec)
Average test_bit loop took: 50334.200 usec (+- 128.425 usec)
100000 operations 4 bits set of 256 bits
Average for_each_set_bit took: 2164.333 usec (+- 246.359 usec)
Average test_bit loop took: 49959.867 usec (+- 188.035 usec)
100000 operations 8 bits set of 256 bits
Average for_each_set_bit took: 3211.200 usec (+- 454.829 usec)
Average test_bit loop took: 50140.850 usec (+- 176.046 usec)
100000 operations 16 bits set of 256 bits
Average for_each_set_bit took: 5181.640 usec (+- 882.726 usec)
Average test_bit loop took: 51003.160 usec (+- 419.601 usec)
100000 operations 32 bits set of 256 bits
Average for_each_set_bit took: 8369.333 usec (+- 1513.150 usec)
Average test_bit loop took: 52096.700 usec (+- 573.022 usec)
100000 operations 64 bits set of 256 bits
Average for_each_set_bit took: 13866.857 usec (+- 2649.393 usec)
Average test_bit loop took: 53989.600 usec (+- 938.808 usec)
100000 operations 128 bits set of 256 bits
Average for_each_set_bit took: 23588.350 usec (+- 4724.222 usec)
Average test_bit loop took: 57300.625 usec (+- 1625.962 usec)
100000 operations 256 bits set of 256 bits
Average for_each_set_bit took: 42752.200 usec (+- 9202.084 usec)
Average test_bit loop took: 64426.933 usec (+- 3402.326 usec)
100000 operations 1 bits set of 512 bits
Average for_each_set_bit took: 1632.000 usec (+- 229.954 usec)
Average test_bit loop took: 98090.000 usec (+- 1120.435 usec)
100000 operations 2 bits set of 512 bits
Average for_each_set_bit took: 1937.700 usec (+- 148.902 usec)
Average test_bit loop took: 100364.100 usec (+- 1433.219 usec)
100000 operations 4 bits set of 512 bits
Average for_each_set_bit took: 2528.000 usec (+- 243.654 usec)
Average test_bit loop took: 99932.067 usec (+- 955.868 usec)
100000 operations 8 bits set of 512 bits
Average for_each_set_bit took: 3734.100 usec (+- 512.359 usec)
Average test_bit loop took: 98944.750 usec (+- 812.070 usec)
100000 operations 16 bits set of 512 bits
Average for_each_set_bit took: 5551.400 usec (+- 846.605 usec)
Average test_bit loop took: 98691.600 usec (+- 654.753 usec)
100000 operations 32 bits set of 512 bits
Average for_each_set_bit took: 8594.500 usec (+- 1446.072 usec)
Average test_bit loop took: 99176.867 usec (+- 579.990 usec)
100000 operations 64 bits set of 512 bits
Average for_each_set_bit took: 13840.743 usec (+- 2527.055 usec)
Average test_bit loop took: 100758.743 usec (+- 833.865 usec)
100000 operations 128 bits set of 512 bits
Average for_each_set_bit took: 23185.925 usec (+- 4532.910 usec)
Average test_bit loop took: 103786.700 usec (+- 1475.276 usec)
100000 operations 256 bits set of 512 bits
Average for_each_set_bit took: 40322.400 usec (+- 8341.802 usec)
Average test_bit loop took: 109433.378 usec (+- 2742.615 usec)
100000 operations 512 bits set of 512 bits
Average for_each_set_bit took: 71804.540 usec (+- 15436.546 usec)
Average test_bit loop took: 120255.440 usec (+- 5252.777 usec)
100000 operations 1 bits set of 1024 bits
Average for_each_set_bit took: 1859.600 usec (+- 27.969 usec)
Average test_bit loop took: 187676.000 usec (+- 1337.770 usec)
100000 operations 2 bits set of 1024 bits
Average for_each_set_bit took: 2273.600 usec (+- 139.420 usec)
Average test_bit loop took: 188176.000 usec (+- 684.357 usec)
100000 operations 4 bits set of 1024 bits
Average for_each_set_bit took: 2940.400 usec (+- 268.213 usec)
Average test_bit loop took: 189172.600 usec (+- 593.295 usec)
100000 operations 8 bits set of 1024 bits
Average for_each_set_bit took: 4224.200 usec (+- 547.933 usec)
Average test_bit loop took: 190257.250 usec (+- 621.021 usec)
100000 operations 16 bits set of 1024 bits
Average for_each_set_bit took: 6090.560 usec (+- 877.975 usec)
Average test_bit loop took: 190143.880 usec (+- 503.753 usec)
100000 operations 32 bits set of 1024 bits
Average for_each_set_bit took: 9178.800 usec (+- 1475.136 usec)
Average test_bit loop took: 190757.100 usec (+- 494.757 usec)
100000 operations 64 bits set of 1024 bits
Average for_each_set_bit took: 14441.457 usec (+- 2545.497 usec)
Average test_bit loop took: 192299.486 usec (+- 795.251 usec)
100000 operations 128 bits set of 1024 bits
Average for_each_set_bit took: 23623.825 usec (+- 4481.182 usec)
Average test_bit loop took: 194885.550 usec (+- 1300.817 usec)
100000 operations 256 bits set of 1024 bits
Average for_each_set_bit took: 40194.956 usec (+- 8109.056 usec)
Average test_bit loop took: 200259.311 usec (+- 2566.085 usec)
100000 operations 512 bits set of 1024 bits
Average for_each_set_bit took: 70983.560 usec (+- 15074.982 usec)
Average test_bit loop took: 210527.460 usec (+- 4968.980 usec)
100000 operations 1024 bits set of 1024 bits
Average for_each_set_bit took: 136530.345 usec (+- 31584.400 usec)
Average test_bit loop took: 233329.691 usec (+- 10814.036 usec)
100000 operations 1 bits set of 2048 bits
Average for_each_set_bit took: 3077.600 usec (+- 76.376 usec)
Average test_bit loop took: 402154.400 usec (+- 518.571 usec)
100000 operations 2 bits set of 2048 bits
Average for_each_set_bit took: 3508.600 usec (+- 148.350 usec)
Average test_bit loop took: 403814.500 usec (+- 1133.027 usec)
100000 operations 4 bits set of 2048 bits
Average for_each_set_bit took: 4219.333 usec (+- 285.844 usec)
Average test_bit loop took: 404312.533 usec (+- 985.751 usec)
100000 operations 8 bits set of 2048 bits
Average for_each_set_bit took: 5670.550 usec (+- 615.238 usec)
Average test_bit loop took: 405321.800 usec (+- 1038.487 usec)
100000 operations 16 bits set of 2048 bits
Average for_each_set_bit took: 7785.080 usec (+- 992.522 usec)
Average test_bit loop took: 406746.160 usec (+- 1015.478 usec)
100000 operations 32 bits set of 2048 bits
Average for_each_set_bit took: 11163.800 usec (+- 1627.320 usec)
Average test_bit loop took: 406124.267 usec (+- 898.785 usec)
100000 operations 64 bits set of 2048 bits
Average for_each_set_bit took: 16964.629 usec (+- 2806.130 usec)
Average test_bit loop took: 406618.514 usec (+- 798.356 usec)
100000 operations 128 bits set of 2048 bits
Average for_each_set_bit took: 27219.625 usec (+- 4988.458 usec)
Average test_bit loop took: 410149.325 usec (+- 1705.641 usec)
100000 operations 256 bits set of 2048 bits
Average for_each_set_bit took: 45138.578 usec (+- 8831.021 usec)
Average test_bit loop took: 415462.467 usec (+- 2725.418 usec)
100000 operations 512 bits set of 2048 bits
Average for_each_set_bit took: 77450.540 usec (+- 15962.238 usec)
Average test_bit loop took: 426089.180 usec (+- 5171.788 usec)
100000 operations 1024 bits set of 2048 bits
Average for_each_set_bit took: 138023.636 usec (+- 29826.959 usec)
Average test_bit loop took: 446346.636 usec (+- 9904.417 usec)
100000 operations 2048 bits set of 2048 bits
Average for_each_set_bit took: 251072.600 usec (+- 55947.692 usec)
Average test_bit loop took: 484855.983 usec (+- 18970.431 usec)
#
Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lore.kernel.org/lkml/20200729220034.1337168-1-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
Davidlohr Bueso
|
c2a0820305 |
perf bench: Add basic syscall benchmark
The usefulness of having a standard way of testing syscall performance has come up from time to time[0]. Furthermore, some of our testing machinery (such as 'mmtests') already makes use of a simplified version of the microbenchmark. This patch mainly takes the same idea to measure syscall throughput compatible with 'perf-bench' via getppid(2), yet without any of the additional template stuff from Ingo's version (based on numa.c). The code is identical to what mmtests uses. [0] https://lore.kernel.org/lkml/20160201074156.GA27156@gmail.com/ Committer notes: Add mising stdlib.h and unistd.h to get the prototypes for exit() and getppid(). Committer testing: $ perf bench Usage: perf bench [<common options>] <collection> <benchmark> [<options>] # List of all available benchmark collections: sched: Scheduler and IPC benchmarks syscall: System call benchmarks mem: Memory access benchmarks numa: NUMA scheduling and MM benchmarks futex: Futex stressing benchmarks epoll: Epoll stressing benchmarks internals: Perf-internals benchmarks all: All benchmarks $ $ perf bench syscall # List of available benchmarks for collection 'syscall': basic: Benchmark for basic getppid(2) calls all: Run all syscall benchmarks $ perf bench syscall basic # Running 'syscall/basic' benchmark: # Executed 10000000 getppid() calls Total time: 3.679 [sec] 0.367957 usecs/op 2717708 ops/sec $ perf bench syscall all # Running syscall/basic benchmark... # Executed 10000000 getppid() calls Total time: 3.644 [sec] 0.364456 usecs/op 2743815 ops/sec $ Signed-off-by: Davidlohr Bueso <dbueso@suse.de> Acked-by: Josh Poimboeuf <jpoimboe@redhat.com> Acked-by: Mel Gorman <mgorman@techsingularity.net> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Link: http://lore.kernel.org/lkml/20190308181747.l36zqz2avtivrr3c@linux-r8p5 Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Ian Rogers
|
51876bd452 |
perf bench: Add kallsyms parsing
Add a benchmark for kallsyms parsing. Example output:
Running 'internals/kallsyms-parse' benchmark:
Average kallsyms__parse took: 103.971 ms (+- 0.121 ms)
Committer testing:
Test Machine: AMD Ryzen 5 3600X 6-Core Processor
[root@five ~]# perf bench internals kallsyms-parse
# Running 'internals/kallsyms-parse' benchmark:
Average kallsyms__parse took: 79.692 ms (+- 0.101 ms)
[root@five ~]# perf stat -r5 perf bench internals kallsyms-parse
# Running 'internals/kallsyms-parse' benchmark:
Average kallsyms__parse took: 80.563 ms (+- 0.079 ms)
# Running 'internals/kallsyms-parse' benchmark:
Average kallsyms__parse took: 81.046 ms (+- 0.155 ms)
# Running 'internals/kallsyms-parse' benchmark:
Average kallsyms__parse took: 80.874 ms (+- 0.104 ms)
# Running 'internals/kallsyms-parse' benchmark:
Average kallsyms__parse took: 81.173 ms (+- 0.133 ms)
# Running 'internals/kallsyms-parse' benchmark:
Average kallsyms__parse took: 81.169 ms (+- 0.074 ms)
Performance counter stats for 'perf bench internals kallsyms-parse' (5 runs):
8,093.54 msec task-clock # 0.999 CPUs utilized ( +- 0.14% )
3,165 context-switches # 0.391 K/sec ( +- 0.18% )
10 cpu-migrations # 0.001 K/sec ( +- 23.13% )
744 page-faults # 0.092 K/sec ( +- 0.21% )
34,551,564,954 cycles # 4.269 GHz ( +- 0.05% ) (83.33%)
1,160,584,308 stalled-cycles-frontend # 3.36% frontend cycles idle ( +- 1.60% ) (83.33%)
14,974,323,985 stalled-cycles-backend # 43.34% backend cycles idle ( +- 0.24% ) (83.33%)
58,712,905,705 instructions # 1.70 insn per cycle
# 0.26 stalled cycles per insn ( +- 0.01% ) (83.34%)
14,136,433,778 branches # 1746.632 M/sec ( +- 0.01% ) (83.33%)
141,943,217 branch-misses # 1.00% of all branches ( +- 0.04% ) (83.33%)
8.1040 +- 0.0115 seconds time elapsed ( +- 0.14% )
[root@five ~]#
Signed-off-by: Ian Rogers <irogers@google.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lore.kernel.org/lkml/20200501221315.54715-2-irogers@google.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Ian Rogers
|
2a4b51666a |
perf bench: Add event synthesis benchmark
Event synthesis may occur at the start or end (tail) of a perf command. In system-wide mode it can scan every process in /proc, which may add seconds of latency before event recording. Add a new benchmark that times how long event synthesis takes with and without data synthesis. An example execution looks like: $ perf bench internals synthesize # Running 'internals/synthesize' benchmark: Average synthesis took: 168.253800 usec Average data synthesis took: 208.104700 usec Signed-off-by: Ian Rogers <irogers@google.com> Acked-by: Jiri Olsa <jolsa@redhat.com> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andrey Zhizhikin <andrey.z@gmail.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Kefeng Wang <wangkefeng.wang@huawei.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Petr Mladek <pmladek@suse.com> Cc: Stephane Eranian <eranian@google.com> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lore.kernel.org/lkml/20200402154357.107873-2-irogers@google.com Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Davidlohr Bueso
|
231457ec70 |
perf bench: Add epoll_ctl(2) benchmark
Benchmark the various operations allowed for epoll_ctl(2). The idea is
to concurrently stress a single epoll instance doing add/mod/del
operations.
Committer testing:
# perf bench epoll ctl
# Running 'epoll/ctl' benchmark:
Run summary [PID 20344]: 4 threads doing epoll_ctl ops 64 file-descriptors for 8 secs.
[thread 0] fdmap: 0x21a46b0 ... 0x21a47ac [ add: 1680960 ops; mod: 1680960 ops; del: 1680960 ops ]
[thread 1] fdmap: 0x21a4960 ... 0x21a4a5c [ add: 1685440 ops; mod: 1685440 ops; del: 1685440 ops ]
[thread 2] fdmap: 0x21a4c10 ... 0x21a4d0c [ add: 1674368 ops; mod: 1674368 ops; del: 1674368 ops ]
[thread 3] fdmap: 0x21a4ec0 ... 0x21a4fbc [ add: 1677568 ops; mod: 1677568 ops; del: 1677568 ops ]
Averaged 1679584 ADD operations (+- 0.14%)
Averaged 1679584 MOD operations (+- 0.14%)
Averaged 1679584 DEL operations (+- 0.14%)
#
Lets measure those calls with 'perf trace' to get a glympse at what this
benchmark is doing in terms of syscalls:
# perf trace -m32768 -s perf bench epoll ctl
# Running 'epoll/ctl' benchmark:
Run summary [PID 20405]: 4 threads doing epoll_ctl ops 64 file-descriptors for 8 secs.
[thread 0] fdmap: 0x21764e0 ... 0x21765dc [ add: 1100480 ops; mod: 1100480 ops; del: 1100480 ops ]
[thread 1] fdmap: 0x2176790 ... 0x217688c [ add: 1250176 ops; mod: 1250176 ops; del: 1250176 ops ]
[thread 2] fdmap: 0x2176a40 ... 0x2176b3c [ add: 1022464 ops; mod: 1022464 ops; del: 1022464 ops ]
[thread 3] fdmap: 0x2176cf0 ... 0x2176dec [ add: 705472 ops; mod: 705472 ops; del: 705472 ops ]
Averaged 1019648 ADD operations (+- 11.27%)
Averaged 1019648 MOD operations (+- 11.27%)
Averaged 1019648 DEL operations (+- 11.27%)
Summary of events:
epoll-ctl (20405), 1264 events, 0.0%
syscall calls total min avg max stddev
(msec) (msec) (msec) (msec) (%)
--------------- -------- --------- --------- --------- --------- ------
eventfd2 256 9.514 0.001 0.037 5.243 68.00%
clone 4 1.245 0.204 0.311 0.531 24.13%
mprotect 66 0.345 0.002 0.005 0.021 7.43%
openat 45 0.313 0.004 0.007 0.073 21.93%
mmap 88 0.302 0.002 0.003 0.013 5.02%
futex 4 0.160 0.002 0.040 0.140 83.43%
sched_setaffinity 4 0.124 0.005 0.031 0.070 49.39%
read 44 0.103 0.001 0.002 0.013 15.54%
fstat 40 0.052 0.001 0.001 0.003 5.43%
close 39 0.039 0.001 0.001 0.001 1.48%
stat 9 0.034 0.003 0.004 0.006 7.30%
access 3 0.023 0.007 0.008 0.008 4.25%
open 2 0.021 0.008 0.011 0.013 22.60%
getdents 4 0.019 0.001 0.005 0.009 37.15%
write 2 0.013 0.004 0.007 0.009 38.48%
munmap 1 0.010 0.010 0.010 0.010 0.00%
brk 3 0.006 0.001 0.002 0.003 26.34%
rt_sigprocmask 2 0.004 0.001 0.002 0.003 43.95%
rt_sigaction 3 0.004 0.001 0.001 0.002 16.07%
prlimit64 3 0.004 0.001 0.001 0.001 5.39%
prctl 1 0.003 0.003 0.003 0.003 0.00%
epoll_create 1 0.003 0.003 0.003 0.003 0.00%
lseek 2 0.002 0.001 0.001 0.001 11.42%
sched_getaffinity 1 0.002 0.002 0.002 0.002 0.00%
arch_prctl 1 0.002 0.002 0.002 0.002 0.00%
set_tid_address 1 0.001 0.001 0.001 0.001 0.00%
getpid 1 0.001 0.001 0.001 0.001 0.00%
set_robust_list 1 0.001 0.001 0.001 0.001 0.00%
execve 1 0.000 0.000 0.000 0.000 0.00%
epoll-ctl (20406), 1245480 events, 14.6%
syscall calls total min avg max stddev
(msec) (msec) (msec) (msec) (%)
--------------- -------- --------- --------- --------- --------- ------
epoll_ctl 619511 1034.927 0.001 0.002 6.691 0.67%
nanosleep 3226 616.114 0.006 0.191 10.376 7.57%
futex 2 11.336 0.002 5.668 11.334 99.97%
set_robust_list 1 0.001 0.001 0.001 0.001 0.00%
clone 1 0.000 0.000 0.000 0.000 0.00%
epoll-ctl (20407), 1243151 events, 14.5%
syscall calls total min avg max stddev
(msec) (msec) (msec) (msec) (%)
--------------- -------- --------- --------- --------- --------- ------
epoll_ctl 618350 1042.181 0.001 0.002 2.512 0.40%
nanosleep 3220 366.261 0.012 0.114 18.162 9.59%
futex 4 5.463 0.001 1.366 5.427 99.12%
set_robust_list 1 0.002 0.002 0.002 0.002 0.00%
epoll-ctl (20408), 1801690 events, 21.1%
syscall calls total min avg max stddev
(msec) (msec) (msec) (msec) (%)
--------------- -------- --------- --------- --------- --------- ------
epoll_ctl 896174 1540.581 0.001 0.002 6.987 0.74%
nanosleep 4667 783.393 0.006 0.168 10.419 7.10%
futex 2 4.682 0.002 2.341 4.681 99.93%
set_robust_list 1 0.002 0.002 0.002 0.002 0.00%
clone 1 0.000 0.000 0.000 0.000 0.00%
epoll-ctl (20409), 4254890 events, 49.8%
syscall calls total min avg max stddev
(msec) (msec) (msec) (msec) (%)
--------------- -------- --------- --------- --------- --------- ------
epoll_ctl 2116416 3768.097 0.001 0.002 9.956 0.41%
nanosleep 11023 1141.778 0.006 0.104 9.447 4.95%
futex 3 0.037 0.002 0.012 0.029 70.50%
set_robust_list 1 0.008 0.008 0.008 0.008 0.00%
madvise 1 0.005 0.005 0.005 0.005 0.00%
clone 1 0.000 0.000 0.000 0.000 0.00%
#
Committer notes:
Fix build on fedora:24-x-ARC-uClibc, debian:experimental-x-mips,
debian:experimental-x-mipsel, ubuntu:16.04-x-arm and ubuntu:16.04-x-powerpc
CC /tmp/build/perf/bench/epoll-ctl.o
bench/epoll-ctl.c: In function 'init_fdmaps':
bench/epoll-ctl.c:214:16: error: comparison between signed and unsigned integer expressions [-Werror=sign-compare]
for (i = 0; i < nfds; i+=inc) {
^
bench/epoll-ctl.c: In function 'bench_epoll_ctl':
bench/epoll-ctl.c:377:16: error: comparison between signed and unsigned integer expressions [-Werror=sign-compare]
for (i = 0; i < nthreads; i++) {
^
bench/epoll-ctl.c:388:16: error: comparison between signed and unsigned integer expressions [-Werror=sign-compare]
for (i = 0; i < nthreads; i++) {
^
cc1: all warnings being treated as errors
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Davidlohr Bueso <dbueso@suse.de>
Cc: Jason Baron <jbaron@akamai.com>
Link: http://lkml.kernel.org/r/20181106152226.20883-3-dave@stgolabs.net
[ Use inttypes.h to print rlim_t fields, fixing the build on Alpine Linux / musl libc ]
[ Check if eventfd() is available, i.e. if HAVE_EVENTFD is defined ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Davidlohr Bueso
|
121dd9ea01 |
perf bench: Add epoll parallel epoll_wait benchmark
This program benchmarks concurrent epoll_wait(2) for file descriptors
that are monitored with with EPOLLIN along various semantics, by a
single epoll instance. Such conditions can be found when using
single/combined or multiple queuing when load balancing.
Each thread has a number of private, nonblocking file descriptors,
referred to as fdmap. A writer thread will constantly be writing to the
fdmaps of all threads, minimizing each threads's chances of epoll_wait
not finding any ready read events and blocking as this is not what we
want to stress. Full details in the start of the C file.
Committer testing:
# perf bench
Usage:
perf bench [<common options>] <collection> <benchmark> [<options>]
# List of all available benchmark collections:
sched: Scheduler and IPC benchmarks
mem: Memory access benchmarks
numa: NUMA scheduling and MM benchmarks
futex: Futex stressing benchmarks
epoll: Epoll stressing benchmarks
all: All benchmarks
# perf bench epoll
# List of available benchmarks for collection 'epoll':
wait: Benchmark epoll concurrent epoll_waits
all: Run all futex benchmarks
# perf bench epoll wait
# Running 'epoll/wait' benchmark:
Run summary [PID 19295]: 3 threads monitoring on 64 file-descriptors for 8 secs.
[thread 0] fdmap: 0xdaa650 ... 0xdaa74c [ 328241 ops/sec ]
[thread 1] fdmap: 0xdaa900 ... 0xdaa9fc [ 351695 ops/sec ]
[thread 2] fdmap: 0xdaabb0 ... 0xdaacac [ 381423 ops/sec ]
Averaged 353786 operations/sec (+- 4.35%), total secs = 8
#
Committer notes:
Fix the build on debian:experimental-x-mips, debian:experimental-x-mipsel
and others:
CC /tmp/build/perf/bench/epoll-wait.o
bench/epoll-wait.c: In function 'writerfn':
bench/epoll-wait.c:399:12: error: format '%ld' expects argument of type 'long int', but argument 2 has type 'size_t' {aka 'unsigned int'} [-Werror=format=]
printinfo("exiting writer-thread (total full-loops: %ld)\n", iter);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~
bench/epoll-wait.c:86:31: note: in definition of macro 'printinfo'
do { if (__verbose) { printf(fmt, ## arg); fflush(stdout); } } while (0)
^~~
cc1: all warnings being treated as errors
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Davidlohr Bueso <dbueso@suse.de>
Cc: Jason Baron <jbaron@akamai.com> <jbaron@akamai.com>
Link: http://lkml.kernel.org/r/20181106152226.20883-2-dave@stgolabs.net
Link: http://lkml.kernel.org/r/20181106182349.thdkpvshkna5vd7o@linux-r8p5>
[ Applied above fixup as per Davidlohr's request ]
[ Use inttypes.h to print rlim_t fields, fixing the build on Alpine Linux / musl libc ]
[ Check if eventfd() is available, i.e. if HAVE_EVENTFD is defined ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
|
Arnaldo Carvalho de Melo
|
1f27a050fc |
tools arch: Update arch/x86/lib/memcpy_64.S copy used in 'perf bench mem memcpy'
To cope with the changes in:
12c89130a56a ("x86/asm/memcpy_mcsafe: Add write-protection-fault handling")
60622d68227d ("x86/asm/memcpy_mcsafe: Return bytes remaining")
bd131544aa7e ("x86/asm/memcpy_mcsafe: Add labels for __memcpy_mcsafe() write fault handling")
da7bc9c57eb0 ("x86/asm/memcpy_mcsafe: Remove loop unrolling")
This needed introducing a file with a copy of the mcsafe_handle_tail()
function, that is used in the new memcpy_64.S file, as well as a dummy
mcsafe_test.h header.
Testing it:
$ nm ~/bin/perf | grep mcsafe
0000000000484130 T mcsafe_handle_tail
0000000000484300 T __memcpy_mcsafe
$
$ perf bench mem memcpy
# Running 'mem/memcpy' benchmark:
# function 'default' (Default memcpy() provided by glibc)
# Copying 1MB bytes ...
44.389205 GB/sec
# function 'x86-64-unrolled' (unrolled memcpy() in arch/x86/lib/memcpy_64.S)
# Copying 1MB bytes ...
22.710756 GB/sec
# function 'x86-64-movsq' (movsq-based memcpy() in arch/x86/lib/memcpy_64.S)
# Copying 1MB bytes ...
42.459239 GB/sec
# function 'x86-64-movsb' (movsb-based memcpy() in arch/x86/lib/memcpy_64.S)
# Copying 1MB bytes ...
42.459239 GB/sec
$
This silences this perf tools build warning:
Warning: Kernel ABI header at 'tools/arch/x86/lib/memcpy_64.S' differs from latest version at 'arch/x86/lib/memcpy_64.S'
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mika Penttilä <mika.penttila@nextfour.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-igdpciheradk3gb3qqal52d0@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
|
||
Ingo Molnar
|
9b2fa7f3e7 |
perf bench: Rename 'mem-memcpy.c' => 'mem-functions.c'
So mem-memcpy.c started out as a simple memcpy() benchmark, then it grew memset() functionality and now I plan to add string copy benchmarks as well. This makes the file name a misnomer: rename it to the more generic mem-functions.c name. Signed-off-by: Ingo Molnar <mingo@kernel.org> Cc: David Ahern <dsahern@gmail.com> Cc: Hitoshi Mitake <mitake@dcl.info.waseda.ac.jp> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Link: http://lkml.kernel.org/r/1445241870-24854-5-git-send-email-mingo@kernel.org [ The "rename" was introducing __unused, wasn't removing the old file, and didn't update tools/perf/bench/Build, fix it ] Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Davidlohr Bueso
|
d2f3f5d2e9 |
perf bench futex: Add lock_pi stresser
Allows a way of measuring low level kernel implementation of FUTEX_LOCK_PI and FUTEX_UNLOCK_PI. The program comes in two flavors: (i) single futex (default), all threads contend on the same uaddr. For the sake of the benchmark, we call into kernel space even when the lock is uncontended. The kernel will set it to TID, any waters that come in and contend for the pi futex will be handled respectively by the kernel. (ii) -M option for multiple futexes, each thread deals with its own futex. This is a trivial scenario and only measures kernel handling of 0->TID transition. Signed-off-by: Davidlohr Bueso <dbueso@suse.de> Cc: Mel Gorman <mgorman@suse.de> Link: http://lkml.kernel.org/r/1436259353.12255.78.camel@stgolabs.net Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
|
Davidlohr Bueso
|
d65817b4e7 |
perf bench futex: Support parallel waker threads
The futex-wake benchmark only measures wakeups done within a single process. While this has value in its own, it does not really generate any hb->lock contention. A new benchmark 'wake-parallel' is added, by extending the futex-wake code such that we can measure parallel waker threads. The program output shows the avg per-thread latency in order to complete its share of wakeups: Run summary [PID 13474]: blocking on 512 threads (at [private] futex 0xa88668), 8 threads waking up 64 at a time. [Run 1]: Avg per-thread latency (waking 64/512 threads) in 0.6230 ms (+-15.31%) [Run 2]: Avg per-thread latency (waking 64/512 threads) in 0.5175 ms (+-29.95%) [Run 3]: Avg per-thread latency (waking 64/512 threads) in 0.7578 ms (+-18.03%) [Run 4]: Avg per-thread latency (waking 64/512 threads) in 0.8944 ms (+-12.54%) [Run 5]: Avg per-thread latency (waking 64/512 threads) in 1.1204 ms (+-23.85%) Avg per-thread latency (waking 64/512 threads) in 0.7826 ms (+-9.91%) Naturally, different combinations of numbers of blocking and waker threads will exhibit different information. Signed-off-by: Davidlohr Bueso <dbueso@suse.de> Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com> Cc: Davidlohr Bueso <dbueso@suse.de> Link: http://lkml.kernel.org/r/1431110280-20231-1-git-send-email-dave@stgolabs.net Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |
||
Jiri Olsa
|
72965b87c5 |
perf build: Add bench objects building
Move bench objects building under build framework and enable perf-in.o rule. Signed-off-by: Jiri Olsa <jolsa@kernel.org> Tested-by: Sukadev Bhattiprolu <sukadev@linux.vnet.ibm.com> Tested-by: Will Deacon <will.deacon@arm.com> Cc: Alexis Berlemont <alexis.berlemont@gmail.com> Cc: Borislav Petkov <bp@alien8.de> Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Stephane Eranian <eranian@google.com> Link: http://lkml.kernel.org/n/tip-b0gxubmn3qjabaq0lune53y3@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com> |