Using the existing symbol_conf.pid_list_str and symbol_conf.tid_list_str
logic.
For example:
perf diff --tid 13965
It'll only diff the samples for thread 13965.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1551791143-10334-4-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To improve 'perf diff', implement a --cpu filter option.
Multiple CPUs can be provided as a comma-separated list with no space:
0,1. Ranges of CPUs are specified with -: 0-2. Default is to report
samples on all CPUs.
For example,
perf diff --cpu 0,1
It only diff the samples for CPU0 and CPU1.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1551791143-10334-3-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To improve 'perf diff', implement a --time filter option to diff the
samples within given time window.
It supports time percent with multiple time ranges. The time string
format is 'a%/n,b%/m,...' or 'a%-b%,c%-%d,...'.
For example:
Select the second 10% time slice to diff:
perf diff --time 10%/2
Select from 0% to 10% time slice to diff:
perf diff --time 0%-10%
Select the first and the second 10% time slices to diff:
perf diff --time 10%/1,10%/2
Select from 0% to 10% and 30% to 40% slices to diff:
perf diff --time 0%-10%,30%-40%
It also supports analysing samples within a given time window
<start>,<stop>.
Times have the format seconds.microseconds.
If 'start' is not given (i.e., time string is ',x.y') then analysis starts at
the beginning of the file.
If the stop time is not given (i.e, time string is 'x.y,') then analysis
goes to end of file.
Time string is 'a1.b1,c1.d1:a2.b2,c2.d2'. Use ':' to separate timestamps for
different perf.data files.
For example, we get the timestamp information from perf script.
perf script -i perf.data.old
mgen 13940 [000] 3946.361400: ...
perf script -i perf.data
mgen 13940 [000] 3971.150589 ...
perf diff --time 3946.361400,:3971.150589,
It analyzes the perf.data.old from the timestamp 3946.361400 to the end of
perf.data.old and analyzes the perf.data from the timestamp 3971.150589 to the
end of perf.data.
v4:
---
Update abstime_str_dup(), let it return error if strdup
is failed, and update __cmd_diff() accordingly.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1551791143-10334-2-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add support to add/remove fields for specific event types in -F option.
It's now possible to use '+-' after event type, like:
# cat > test.c
#include <stdio.h>
int main(void)
{
printf("Hello world\n");
while(1) {}
}
^D
# gcc -g -o test test.c
# perf probe -x test 'test.c:5'
# perf record -e '{cpu/cpu-cycles,period=10000/,probe_test:main}:S' ./test
...
# perf script -Ftrace:+period,-cpu
test 3859 396291.117343: 10275 cpu/cpu-cycles,period=10000/: 7f..
test 3859 396291.118234: 11041 cpu/cpu-cycles,period=10000/: ffffff..
test 3859 396291.118234: 1 probe_test:main:
test 3859 396291.118248: 8668 cpu/cpu-cycles,period=10000/: ffffff..
test 3859 396291.118263: 10139 cpu/cpu-cycles,period=10000/: ffffff..
Committer testing:
Couldn't make the test above work, but tested it with:
# perf probe -x hello main
Added new event:
probe_hello:main (on main in /home/acme/c/hello)
You can now use it in all perf tools, such as:
perf record -e probe_hello:main -aR sleep 1
# perf record -e probe_hello:main ./hello
hello, world
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.025 MB perf.data (1 samples) ]
# perf script
hello 21454 [002] 254116.874005: probe_hello:main: (401126)
#
# perf script -Ftrace:+period,-cpu
hello 21454 254116.874005: 1 probe_hello:main: (401126)
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
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: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20190220122800.864-4-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
According to the current documentation the flags section is placed after
the file header itself but the code assumes to find the flags section
after the data section. This change updates the documentation to that
assumption.
Signed-off-by: Jonas Rabenstein <jonas.rabenstein@studium.uni-erlangen.de>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Link: http://lkml.kernel.org/r/20190219154515.3954-2-jonas.rabenstein@studium.uni-erlangen.de
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The content of the HEADER_CMDLINE feature header is a perf_header_string_list
of the argument vector and not a perf_header_string of the commandline.
Signed-off-by: Jonas Rabenstein <jonas.rabenstein@studium.uni-erlangen.de>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Thomas Richter <tmricht@linux.ibm.com>
Link: http://lkml.kernel.org/r/20190219154515.3954-1-jonas.rabenstein@studium.uni-erlangen.de
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Implement --affinity=node|cpu option for the record mode defaulting
to system affinity mask bouncing.
Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/083f5422-ece9-10dd-8305-bf59c860f10f@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Hardware tracing:
Adrian Hunter:
- Handle calls optimized into jumps to a different symbol
in the thread stack routines used to process hardware traces (Adrian Hunter)
Intel PT:
Adrian Hunter:
- Fix overlap calculation for padding.
- Fix CYC timestamp calculation after OVF.
- Packet splitting can only happen in 32-bit.
- Add timestamp to auxtrace errors.
ARM CoreSight:
Leo Yan:
- Add last instruction information in packet
- Set sample flags for instruction range, exception and
return packets and for a trace discontinuity.
- Add exception number in exception packet
- Change tuple from traceID-CPU# to traceID-metadata
- Add traceID in packet
Mathieu Poirier:
- Add "sinks" group to PMU directory
- Use event attributes to send sink information to kernel
- Remove set_drv_config() API, no longer used.
perf annotate:
Jiri Olsa:
- Delay symbol annotation to the resort phase, speeding up 'perf report'
startup.
perf record:
Alexey Budankov:
- Allow binding userspace buffers to NUMA nodes.
Symbols:
Adrian Hunter:
- Fix calculating of symbol sizes when splitting kallsyms into
maps for kcore processing.
Vendor events:
William Cohen:
- Intel: Fix Load_Miss_Real_Latency on CLX
Misc:
Arnaldo Carvalho de Melo:
- Streamline headers, removing includes when all that is needed are
just forward declarations, fixup the fallout for cases where headers
should have been explicitely included but were instead obtained
indirectly, by sheer luck.
- Add fallback versions for CPU_{OR,EQUAL}(), so that code using it
continue to build on older systems where those were not yet introduced
or in systems using some other libc than the GNU one where those
helpers aren't present.
Documentation:
Changbin Du:
- Add documentation for BPF event selection.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQR2GiIUctdOfX2qHhGyPKLppCJ+JwUCXFsqugAKCRCyPKLppCJ+
JzpwAQDEh1mNZoxfdGZEi9d+8p2hnRlOs3GOUG4iGnqAYfae4QEAkMJ0V1wrmkdw
NXgV+PgWfDcgbD4Cn90eWA8M6KEcbgA=
=ogOF
-----END PGP SIGNATURE-----
Merge tag 'perf-core-for-mingo-5.1-20190206' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core
Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:
Hardware tracing:
Adrian Hunter:
- Handle calls optimized into jumps to a different symbol
in the thread stack routines used to process hardware traces (Adrian Hunter)
Intel PT:
Adrian Hunter:
- Fix overlap calculation for padding.
- Fix CYC timestamp calculation after OVF.
- Packet splitting can only happen in 32-bit.
- Add timestamp to auxtrace errors.
ARM CoreSight:
Leo Yan:
- Add last instruction information in packet
- Set sample flags for instruction range, exception and
return packets and for a trace discontinuity.
- Add exception number in exception packet
- Change tuple from traceID-CPU# to traceID-metadata
- Add traceID in packet
Mathieu Poirier:
- Add "sinks" group to PMU directory
- Use event attributes to send sink information to kernel
- Remove set_drv_config() API, no longer used.
perf annotate:
Jiri Olsa:
- Delay symbol annotation to the resort phase, speeding up 'perf report'
startup.
perf record:
Alexey Budankov:
- Allow binding userspace buffers to NUMA nodes.
Symbols:
Adrian Hunter:
- Fix calculating of symbol sizes when splitting kallsyms into
maps for kcore processing.
Vendor events:
William Cohen:
- Intel: Fix Load_Miss_Real_Latency on CLX
Misc:
Arnaldo Carvalho de Melo:
- Streamline headers, removing includes when all that is needed are
just forward declarations, fixup the fallout for cases where headers
should have been explicitely included but were instead obtained
indirectly, by sheer luck.
- Add fallback versions for CPU_{OR,EQUAL}(), so that code using it
continue to build on older systems where those were not yet introduced
or in systems using some other libc than the GNU one where those
helpers aren't present.
Documentation:
Changbin Du:
- Add documentation for BPF event selection.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add documentation for how to pass a BPF program as a perf event.
Signed-off-by: Changbin Du <changbin.du@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20190201134651.12373-1-changbin.du@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
PowerPC hardware does not have a builtin latency filter (--ldlat) for
the "mem-load" event and perf_mem_events by default includes
"/ldlat=30/" which is causing a failure on PowerPC. Refactor the code to
support "perf mem/c2c" on PowerPC.
This patch depends on kernel side changes done my Madhavan:
https://lists.ozlabs.org/pipermail/linuxppc-dev/2018-December/182596.html
Signed-off-by: Ravi Bangoria <ravi.bangoria@linux.ibm.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Dick Fowles <fowles@inreach.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Joe Mario <jmario@redhat.com>
Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: linuxppc-dev@lists.ozlabs.org
Link: http://lkml.kernel.org/r/20190129132412.771-1-ravi.bangoria@linux.ibm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
I.f. if children should inherit the parent perf_event configuration,
i.e. if we should trace children as well or just the parent.
The default is to follow children, to disable this and have a behaviour
similar to strace, set this config option or use the --no_inherit 'perf
trace' option.
E.g.:
Default:
# perf config trace.no_inherit
# trace -e clone,*sleep time sleep 1
0.000 time/21107 clone(clone_flags: CHILD_CLEARTID|CHILD_SETTID|0x11, newsp: 0, child_tidptr: 0x7f7b8f9ae810) = 21108 (time)
? time/21108 ... [continued]: clone()
0.691 sleep/21108 nanosleep(rqtp: 0x7ffed01d0540, rmtp: 0 ) = 0
0.00user 0.00system 0:01.00elapsed 0%CPU (0avgtext+0avgdata 1988maxresident)k
0inputs+0outputs (0major+76minor)pagefaults 0swaps
#
Disable it:
# trace -e clone,*sleep time sleep 1
0.000 clone(clone_flags: CHILD_CLEARTID|CHILD_SETTID|0x11, newsp: 0, child_tidptr: 0x7ff41e100810) = 21414 (time)
0.00user 0.00system 0:01.00elapsed 0%CPU (0avgtext+0avgdata 1964maxresident)k
0inputs+0outputs (0major+76minor)pagefaults 0swaps
#
Notice that since there is just one thread, the "comm/TID" column is
suppressed.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-thd8s16pagyza71ufi5vjlan@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The default so far, since we show argument names followed by its values,
was to make the output more compact by suppressing most zeroed args.
Make this configurable so that users can choose what best suit their
needs.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-q0gxws02ygodh94o0hzim5xd@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To add augmented_raw_syscalls to the events speficied by the user, or be
the only one if no events were specified by the user, one can add this
to perfconfig:
# cat ~/.perfconfig
[trace]
add_events = /home/acme/git/perf/tools/perf/examples/bpf/augmented_raw_syscalls.o
#
I.e. pre-compile the augmented_raw_syscalls.c BPF program and make it
always load, this way:
# perf trace -e open* cat /etc/passwd > /dev/null
0.000 ( 0.013 ms): cat/31557 openat(dfd: CWD, filename: /etc/ld.so.cache, flags: CLOEXEC) = 3
0.035 ( 0.007 ms): cat/31557 openat(dfd: CWD, filename: /lib64/libc.so.6, flags: CLOEXEC) = 3
0.353 ( 0.009 ms): cat/31557 openat(dfd: CWD, filename: /usr/lib/locale/locale-archive, flags: CLOEXEC) = 3
0.424 ( 0.006 ms): cat/31557 openat(dfd: CWD, filename: /etc/passwd) = 3
#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Luis Cláudio Gonçalves <lclaudio@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-0lgj7vh64hg3ce44gsmvj7ud@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
I was trigger happy on this one, as using ordered_events as implemented
by Jiri for use with the --block code under discussion on lkml incurs
in delaying processing to form batches that then get ordered and then
printed.
With 'perf trace' we want to process the events as they go, without that
delay, and doing it that way works well for the common case which is to
trace a thread or a workload started by 'perf trace'.
So revert back to not using ordered_events but add an option to select
that mode so that users can experiment with their particular use case to
see if works better, i.e. if the added delay is not a problem and the
ordering helps.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-8ki7sld6rusnjhhtaly26i5o@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When looking at PT or brstackinsn traces with 'perf script' it can be
very useful to see the source code. This adds a simple facility to print
them with 'perf script', if the information is available through dwarf
% perf record ...
% perf script -F insn,ip,sym,srccode
...
4004c6 main
5 for (i = 0; i < 10000000; i++)
4004cd main
5 for (i = 0; i < 10000000; i++)
4004c6 main
5 for (i = 0; i < 10000000; i++)
4004cd main
5 for (i = 0; i < 10000000; i++)
4004cd main
5 for (i = 0; i < 10000000; i++)
4004cd main
5 for (i = 0; i < 10000000; i++)
4004cd main
5 for (i = 0; i < 10000000; i++)
4004cd main
5 for (i = 0; i < 10000000; i++)
4004b3 main
6 v++;
% perf record -b ...
% perf script -F insn,ip,sym,srccode,brstackinsn
...
main+22:
0000000000400543 insn: e8 ca ff ff ff # PRED
|18 f1();
f1:
0000000000400512 insn: 55
|10 {
0000000000400513 insn: 48 89 e5
0000000000400516 insn: b8 00 00 00 00
|11 f2();
000000000040051b insn: e8 d6 ff ff ff # PRED
f2:
00000000004004f6 insn: 55
|5 {
00000000004004f7 insn: 48 89 e5
00000000004004fa insn: 8b 05 2c 0b 20 00
|6 c = a / b;
0000000000400500 insn: 8b 0d 2a 0b 20 00
0000000000400506 insn: 99
0000000000400507 insn: f7 f9
0000000000400509 insn: 89 05 29 0b 20 00
000000000040050f insn: 90
|7 }
0000000000400510 insn: 5d
0000000000400511 insn: c3 # PRED
f1+14:
0000000000400520 insn: b8 00 00 00 00
|12 f2();
0000000000400525 insn: e8 cc ff ff ff # PRED
f2:
00000000004004f6 insn: 55
|5 {
00000000004004f7 insn: 48 89 e5
00000000004004fa insn: 8b 05 2c 0b 20 00
|6 c = a / b;
Not supported for callchains currently, would need some layout changes
there.
Committer notes:
Fixed the build on Alpine Linux (3.4 .. 3.8) by addressing this
warning:
In file included from util/srccode.c:19:0:
/usr/include/sys/fcntl.h:1:2: error: #warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h> [-Werror=cpp]
#warning redirecting incorrect #include <sys/fcntl.h> to <fcntl.h>
^~~~~~~
cc1: all warnings being treated as errors
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Tested-by: Jiri Olsa <jolsa@kernel.org>
Link: http://lkml.kernel.org/r/20181204001848.24769-1-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The default timeout of 500ms for parsing /proc/<pid>/maps files is too
short for profiling many of our services.
This can be overridden by passing --proc-map-timeout to the relevant
command but it'd be nice to globally increase our default value.
This patch permits setting a different default with the
core.proc-map-timeout config file parameter.
Signed-off-by: Mark Drayton <mbd@fb.com>
Acked-by: Song Liu <songliubraving@fb.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20181204203420.1683114-1-mbd@fb.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Go over the tools/ files that are maintained in Arnaldo's tree and
fix common typos: half of them were in comments, the other half
in JSON files.
No change in functionality intended.
Committer notes:
This was split from a larger patch as there are code that is,
additionally, maintained outside the kernel tree, so to ease cherry
picking and/or backporting, split this into multiple patches.
In this particular case, it affects documentation, so may be interesting
to cherry pick as it is information that is presented to the user.
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20181203102200.GA104797@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add explanations for new columns "IPC" and "IPC coverage" in perf
documentation.
v5:
---
Update the description according to Ingo's comments.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1543586097-27632-5-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Multi AIO trace writing allows caching more kernel data into userspace
memory postponing trace writing for the sake of overall profiling data
thruput increase. It could be seen as kernel data buffer extension into
userspace memory.
With an --aio option value different from 0 (default value is 1) the
tool has capability to cache more and more data into user space along
with delegating spill to AIO.
That allows avoiding to suspend at record__aio_sync() between calls of
record__mmap_read_evlist() and increases profiling data thruput at the
cost of userspace memory.
Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/050bb053-e7f3-aa83-fde7-f27ff90be7f6@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The trace file offset is read once before mmaps iterating loop and
written back after all performance data is enqueued for aio writing.
The trace file offset is incremented linearly after every successful aio
write operation.
record__aio_sync() blocks till completion of the started AIO operation
and then proceeds.
record__aio_mmap_read_sync() implements a barrier for all incomplete
aio write requests.
Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com>
Reviewed-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/ce2d45e9-d236-871c-7c8f-1bed2d37e8ac@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This basically replicates what was done for 'perf report' in:
b226a5a729 ("perf report: Allow user to specify path to kallsyms file")
This should help with resolving eBPF symbols, that are in kallsyms but,
of course, not in vmlinux.
Reported-by: Ivan Babrou <ibobrik@gmail.com>
Tested-by: Ivan Babrou <ibobrik@gmail.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexei Starovoitov <ast@kernel.org>
Cc: Daniel Borkmann <daniel@iogearbox.net>
Cc: David Ahern <dsahern@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-x52mx1ybq8128rtg9hjrj5qk@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
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>
Implement a weak group fallback for 'perf record', similar to the
existing 'perf stat' support. This allows to use groups that might be
longer than the available counters without failing.
Before:
$ perf record -e '{cycles,cache-misses,cache-references,cpu_clk_unhalted.thread,cycles,cycles,cycles}' -a sleep 1
Error:
The sys_perf_event_open() syscall returned with 22 (Invalid argument) for event (cycles).
/bin/dmesg | grep -i perf may provide additional information.
After:
$ ./perf record -e '{cycles,cache-misses,cache-references,cpu_clk_unhalted.thread,cycles,cycles,cycles}:W' -a sleep 1
WARNING: No sample_id_all support, falling back to unordered processing
[ perf record: Woken up 3 times to write data ]
[ perf record: Captured and wrote 8.136 MB perf.data (134069 samples) ]
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Link: http://lkml.kernel.org/r/20181001195927.14211-2-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Enabling --overwrite mode allows us to to use just the most recent
records, which helps in high core count machines such as Knights
Landing/Mill, but right now is being disabled by default as the pausing
used in this technique is leading to loss of metadata events such as
PERF_RECORD_MMAP which makes 'perf top' unable to resolve samples,
leading to lots of unknown samples appearing on the UI.
Enabling this may be useful if you are in such machines and profiling a
workload that doesn't creates short lived threads and/or doesn't uses
many executable mmap operations.
Work is being planed to solve this situation, till then, this will
remain disabled by default.
Reported-by: David Miller <davem@davemloft.net>
Acked-by: Kan Liang <kan.liang@intel.com>
Link: https://lkml.kernel.org/r/4f84468f-37d9-cf1b-12c1-514ef74b6a48@linux.intel.com
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: ebebbf0823 ("perf top: Switch default mode to overwrite mode")
Link: https://lkml.kernel.org/n/tip-ehvf77vi1si9409r7p4wx788@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In ebebbf0823 ("perf top: Switch default mode to overwrite mode") we
forgot to leave a way to disable that new default, add a --overwrite
option that can be disabled using --no-overwrite, since the code already
in such a way that we can readily disable this mode.
This is useful when investigating bugs with this mode like the recent
report from David Miller where lots of unknown symbols appear due to
disabling the events while processing them which disables all record
types, not just PERF_RECORD_SAMPLE, which makes it impossible to resolve
maps when we lose PERF_RECORD_MMAP records.
This can be easily seen while building a kernel, when there are lots of
short lived processes.
Reported-by: David Miller <davem@davemloft.net>
Acked-by: Kan Liang <kan.liang@intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: ebebbf0823 ("perf top: Switch default mode to overwrite mode")
Link: https://lkml.kernel.org/n/tip-oqgsz2bq4kgrnnajrafcdhie@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
By default 'perf script' for itrace outputs sampled instructions or
branches. In my experience this is confusing to users because it's hard
to correlate with real program behavior. The sampling makes sense for
tools like 'perf report' that actually sample to reduce the run time,
but run time is normally not a problem for 'perf script'. It's better
to give an accurate representation of the program flow.
Default 'perf script' to output all calls for itrace. That's a much saner
default. The old behavior can be still requested with 'perf script'
--itrace=ibxwpe100000
v2: Fix ETM build failure
v3: Really fix ETM build failure (Kim Phillips)
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Kim Phillips <kim.phillips@arm.com>
Cc: Leo Yan <leo.yan@linaro.org>
Link: http://lkml.kernel.org/r/20180920180540.14039-3-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Additional reports will be added to the script so rename to reflect the
more general purpose.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: http://lkml.kernel.org/r/20181001062853.28285-13-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The dependency for the man page rule using asciidoctor incorrectly
specifies a source file in $(OUTPUT). When building out-of-tree, the
source file is not found, resulting in a fall-back to the following rule
which uses xmlto.
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20180916151704.GF4765@decadent.org.uk
Fixes: ffef80ecf8 ("perf Documentation: Support for asciidoctor")
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Set annotation percent type from following choices:
global-period, local-period, global-hits, local-hits
With following report option setup the percent type will be passed to
annotation browser:
$ perf report --percent-type period-local
The local/global keywords set if the percentage is computed in the scope
of the function (local) or the whole data (global). The period/hits
keywords set the base the percentage is computed on - the samples period
or the number of samples (hits).
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/20180804130521.11408-21-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add --percent-type option to set annotation percent type from following
choices:
global-period, local-period, global-hits, local-hits
Examples:
$ perf annotate --percent-type period-local --stdio | head -1
Percent | Source code ... es, percent: local period)
$ perf annotate --percent-type hits-local --stdio | head -1
Percent | Source code ... es, percent: local hits)
$ perf annotate --percent-type hits-global --stdio | head -1
Percent | Source code ... es, percent: global hits)
$ perf annotate --percent-type period-global --stdio | head -1
Percent | Source code ... es, percent: global period)
The local/global keywords set if the percentage is computed in the scope
of the function (local) or the whole data (global).
The period/hits keywords set the base the percentage is computed on -
the samples period or the number of samples (hits).
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/20180804130521.11408-20-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add missing documentation for --desc and --debug options to the 'perf
list' man page.
Signed-off-by: Sangwon Hong <qpakzk@gmail.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/20180717110738.10779-1-qpakzk@gmail.com
[ Clarify that --desc is by default active ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Some of the comments in the perf events code use articles incorrectly,
using 'a' for words beginning with a vowel sound, where 'an' should be
used.
Signed-off-by: Tobias Tefke <tobias.tefke@tutanota.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: acme@kernel.org
Cc: alexander.shishkin@linux.intel.com
Cc: jolsa@redhat.com
Cc: namhyung@kernel.org
Link: http://lkml.kernel.org/r/20180709105715.22938-1-tobias.tefke@tutanota.com
[ Fix a few more perf related 'a event' typo fixes from all around the kernel and tooling tree. ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Adding --interval-clear option to clear the screen before next interval.
Committer testing:
# perf stat -I 1000 --interval-clear
And, as expected, it behaves almost like:
# watch -n 0 perf stat -a sleep 1
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <frederic@kernel.org>
Cc: Milian Wolff <milian.wolff@kdab.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/20180606221513.11302-4-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add a brief introduction about fields to perf-script-python.txt.
It should help python script developers in easily finding what fields
are supported.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Reviewed-by: Andi Kleen <ak@linux.intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Jin Yao <yao.jin@intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1527843663-32288-4-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Adding the support to read rusage data once the workload is finished and
display the system/user time values:
$ perf stat --null perf bench sched pipe
...
Performance counter stats for 'perf bench sched pipe':
5.342599256 seconds time elapsed
2.544434000 seconds user
4.549691000 seconds sys
It works only in non -r mode and only for workload target.
So as of now, for workload targets, we display 3 types of timings. The
time we meassure in perf stat from enable to disable+period:
5.342599256 seconds time elapsed
The time spent in user and system lands, displayed only for workload
session/target:
2.544434000 seconds user
4.549691000 seconds sys
Those times are the very same displayed by 'time' tool. They are
returned by wait4 call via the getrusage struct interface.
Committer notes:
Had to rename some variables to avoid this on older systems such as
centos:6:
builtin-stat.c: In function 'print_footer':
builtin-stat.c:1831: warning: declaration of 'stime' shadows a global declaration
/usr/include/time.h:297: warning: shadowed declaration is here
Committer testing:
# perf stat --null time perf bench sched pipe
# Running 'sched/pipe' benchmark:
# Executed 1000000 pipe operations between two processes
Total time: 5.526 [sec]
5.526534 usecs/op
180945 ops/sec
1.00user 6.25system 0:05.52elapsed 131%CPU (0avgtext+0avgdata 8056maxresident)k
0inputs+0outputs (0major+606minor)pagefaults 0swaps
Performance counter stats for 'time perf bench sched pipe':
5.530978744 seconds time elapsed
1.004037000 seconds user
6.259937000 seconds sys
#
Suggested-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20180605121313.31337-1-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Enable complex event names containing [.:=,] symbols to be encoded into Perf
trace using name= modifier e.g. like this:
perf record -e cpu/name=\'OFFCORE_RESPONSE:request=DEMAND_RFO:response=L3_HIT.SNOOP_HITM\',\
period=0x3567e0,event=0x3c,cmask=0x1/Duk ./futex
Below is how it looks like in the report output. Please note explicit escaped
quoting at cmdline string in the header so that thestring can be directly reused
for another collection in shell:
perf report --header
# ========
...
# cmdline : /root/abudanko/kernel/tip/tools/perf/perf record -v -e cpu/name=\'OFFCORE_RESPONSE:request=DEMAND_RFO:response=L3_HIT.SNOOP_HITM\',period=0x3567e0,event=0x3c,cmask=0x1/Duk ./futex
# event : name = OFFCORE_RESPONSE:request=DEMAND_RFO:response=L3_HIT.SNOOP_HITM, , type = 4, size = 112, config = 0x100003c, { sample_period, sample_freq } = 3500000, sample_type = IP|TID|TIME, disabled = 1, inh
...
# ========
#
#
# Total Lost Samples: 0
#
# Samples: 24K of event 'OFFCORE_RESPONSE:request=DEMAND_RFO:response=L3_HIT.SNOOP_HITM'
# Event count (approx.): 86492000000
#
# Overhead Command Shared Object Symbol
# ........ ....... ................ ..............................................
#
14.75% futex [kernel.vmlinux] [k] __entry_trampoline_start
...
perf stat -e cpu/name=\'CPU_CLK_UNHALTED.THREAD:cmask=0x1\',period=0x3567e0,event=0x3c,cmask=0x1/Duk ./futex
10000000 process context switches in 16678890291ns (1667.9ns/ctxsw)
Performance counter stats for './futex':
88,095,770,571 CPU_CLK_UNHALTED.THREAD:cmask=0x1
16.679542407 seconds time elapsed
Signed-off-by: Alexey Budankov <alexey.budankov@linux.intel.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/c194b060-761d-0d50-3b21-bb4ed680002d@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In the perf.data HEADER_CPUDESC feadure header we store first the number
of available CPUs in the system, then the number of CPUs at the time of
writing the header, not the other way around.
Reported-by: Thomas-Mich Richter <tmricht@linux.ibm.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: He Kuang <hekuang@huawei.com>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Jin Yao <yao.jin@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kim Phillips <kim.phillips@arm.com>
Cc: Lakshman Annadorai <lakshmana@google.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Simon Que <sque@chromium.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-j7o92acm2vnxjv70y4o3swoc@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
