Age | Commit message (Collapse) | Author |
|
|
|
|
|
|
|
Make below changes to better suppport inferno:
1. Save product properties of device in perf.data.
2. Add a python/c api GetFeatureSection. It is used to support
reading record cmd and meta info from perf.data.
3. Remove old api GetNextMetaInfo, because meta info can be
exported by GetFeatureSection more easily.
4. When reading perf.data in record_file_reader.cpp, remove
callchain at and below ip == 0 to avoid caller's effort.
Also move to use android-base/properties.h.
Bug: http://b/64035530
Test: run simpleperf_unit_test and test.py.
Change-Id: Ib6743a09167e2b7cd65a12f17d991bc1ac628588
|
|
Bug: None
Test: None
Change-Id: If06288b0b5b2b2ec20a7fc4f930c5ca3d0198978
|
|
|
|
The warning of missing symbol for each lib is shown twice,
once in recording, and once in reporting. This is redundant and
maybe misleading. Because when reporting we actually don't know
if we have collected enough symbols in perf.data. So move warning
of missing symbols to debug info if we have symbols in perf.data.
Use customized StderrLogger, because simpleperf doesn't need
time,pid,tid information in log.
Bug: http://b/29574526
Test: run simpleperf manually and run simpleperf_unit_test.
Change-Id: I9baf6d8fdcd63907681f2daa45b8fad6bf7e2516
|
|
Format python code using autopep8.
Change Javascript code to ES6, and format it using eslint.
Bug: http://b/64035530
Test: run inferno.
Change-Id: I8b3553bf49423275da8eb3ad6fe3d8bb8dcf0a5b
|
|
|
|
|
|
|
|
Previous CL wrongly makes kernel dump samples for stat cmd.
Bug: http://b/64489160
Test: run simpleperf_unit_test.
Change-Id: I4f5c08839e283b2361e47d61310e5161433824bb
|
|
This makes it optional whether simpleperf stops recording
when the parent dies.
Bug: http://b/64391339
Test: run simpleperf_unit_test.
Change-Id: I0827a17d6fedbb9763347df76da9dc981fea6312
|
|
Also add tests to check if we can run inferno successfully.
More tests will be added to check report results of inferno.
Bug: http://b/64035530
Test: run test.py.
Change-Id: Ie5dd1f5cb47c8c7a2ae68f8614cf81b68e3c7dbf
|
|
|
|
|
|
Currently record command interface only allows one sample freq or sample period
for all events. This is not convenient when recording both non tracepoint events
and tracepoint events. This CL allows setting different sample speed for
different events. For example, for "-f 1000 -e cpu-cycles -c 1 sched:sched_switch",
"-f 1000" applies to cpu-cycles, and "-c 1" applies to sched:sched_switch.
It also fixes a bug about trace-offcpu: if "-f 1000 --trace-offcpu" is used,
the sched:sched_switch is be samples with "-f 1000". But we want to sample it
with "-c 1".
Also change the order of options in the help msg of record cmd to make it more
readable. Remove -F option. Because adding it seems not useful.
Bug: http://b/37572306
Test: run simpleperf_unit_test.
Change-Id: Ifdbd27c8f9fec49aade0e0e6ce624d8114042020
|
|
Bug: N/A
Test: ran tests
Change-Id: I10bdeeb38404583560685bf47c58f6770fc9050b
Signed-off-by: Ivan Maidanski <i.maidanski@samsung.com>
|
|
|
|
|
|
To support perf.data generated with --trace-offcpu option,
update interface in report_sample.proto:
1. Add meta info to show all event types.
2. Add event_type_id in each sample to show which even type it belongs to.
Bug: http://b/37572306
Test: run simpleperf_unit_test.
Change-Id: I2878979ec2023904df1006ce353dcf233b6a2642
|
|
|
|
This is a preparation of using app_profiler.py in inferno.
In app_profiler.py:
Add -np option to profile native programs like surfaceflinger.
Add -cmd option to profile command lines.
Add --arch option to set the app arch, because we don't guess the app arch
when profiling command lines.
Add corresponding tests in test.py.
Also improve the way of detecting whether the device supports trace-offcpu.
Bug: http://b/63006886
Test: run test.py.
Change-Id: Id899063d9a94beec67b22fd7710cda7a438557fb
|
|
* Use const reference parameter type to avoid unnecessary copy.
* Use more efficient overloaded string methods.
Bug: 30407689
Bug: 30411878
Test: build with WITH_TIDY=1
Change-Id: I558d482910c8a53c042d876848e35cdce8b8c15b
|
|
|
|
Also fix a problem that binary_cache_builder.py tries to pull /dev/zero.
Bug: None.
Test: run test.py TestExamplePureJava.test_app_profiler_with_ctrl_c.
Change-Id: Ie99af6795bb1e81ae6e93f8b7a8907d49c048694
|
|
|
|
Taken from branch aosp-master.
Test: None.
Change-Id: Idcf75340fb9c0d7b8c81f64eeb2cb94a4c432224
|
|
|
|
Move README.md to doc directory, split inferno doc into inferno.md,
move pictures from scripts/inferno/ to doc/.
Improve the content about showing flamegraphs in README.md.
Bug: http://b/64123842
Test: No.
Change-Id: Ia3ebd0bbbda21fa816b94cd349c901b5c7713564
|
|
|
|
|
|
It is used to show features supported on the device.
Bug: http://b/64147273
Test: run simpleperf_unit_test.
Change-Id: Idb7821e74d1a23f8988ef2207696114498713f47
|
|
|
|
Add tests for Trace offcpu.
Add test for jni call.
Build testdata when running test.py.
Bug: http://b/63006886
Test: run test.py.
Change-Id: I5a87b8d2a8f59ead858c1b78a78c7e78847854f4
|
|
Modify the host side linux byteorder.h to set the defines for little
endian. The new kernel headers require the defines, or there will be
compile errors.
Test: Ran all host side unit tests.
Change-Id: I05f0ea0695a56525af3cd808f2ae9a8d6b5c49db
|
|
|
|
Some devices have very slow storage media. This feature can speed
up filesystem creation time noticeably.
Bug: 63809106
Bug: 64101366
Change-Id: Id4886822035808bcb279dbf6f81989fee9cc0ace
(cherry picked from commit 794223c633652b714db804729b5f57d3689a1f91)
|
|
|
|
|
|
Taken from branch aosp-master.
Test: none.
Change-Id: I64c9a60b4251549a22d44e9444824a439d5f4cc1
|
|
Bug: http://b/63006886
Test: none.
Change-Id: If27d2f0d91e2121c3d58c0b5451dee98934f2d40
|
|
|
|
This fixes the following memory leak:
system/extras/tests/tcp_nuke_addr/tcp_nuke_addr_test.cpp:146:20:
warning: Potential memory leak [clang-analyzer-cplusplus.NewDeleteLeaks]
Namely, clang was upset that we didn't free the std::threads we new'ed
up. Immediately detach()ing them gets us the same effect.
(Context: We're trying to clean up scary-sounding warnings like this in
Android. I realize that, in this case, a leak is harmless, but...)
Bug: 27101951
Test: mma. Memory leak warning is gone. Ran the test on bullhead for
60s; seemed to work (last line of output is "60s: 246 cps, total 25432")
Change-Id: I4b6b5679430968e01ffea5bc10309a4ced04ba4e
|
|
|
|
simpleperf_report_lib."
|
|
This is a preparation for integrating inferno with python scripts under scripts/.
Bug: http://b/64035530
Test: run inferno.py.
Change-Id: I60ee1c30f650307aa58a038448cddb4eef43acb5
|
|
|
|
Export python interface for reading meta info.
Change sample.period to sample time difference when trace_offcpu is true.
Add unit tests.
Bug: http://b/37572306
Test: run python test.py.
Change-Id: Ic144314184d115cc55630d3c69b66a0d8594721a
|
|
|