- 13 Nov, 2021 36 commits
-
-
Namhyung Kim authored
When perf report synthesize events from ARM SPE data, it refers to current cpu, pid and tid in the machine. But there's no place to set them in the ARM SPE decoder. I'm seeing all pid/tid is set to -1 and user symbols are not resolved in the output. # perf record -a -e arm_spe_0/ts_enable=1/ sleep 1 # perf report -q | head 8.77% 8.77% :-1 [kernel.kallsyms] [k] format_decode 7.02% 7.02% :-1 [kernel.kallsyms] [k] seq_printf 7.02% 7.02% :-1 [unknown] [.] 0x0000ffff9f687c34 5.26% 5.26% :-1 [kernel.kallsyms] [k] vsnprintf 3.51% 3.51% :-1 [kernel.kallsyms] [k] string 3.51% 3.51% :-1 [unknown] [.] 0x0000ffff9f66ae20 3.51% 3.51% :-1 [unknown] [.] 0x0000ffff9f670b3c 3.51% 3.51% :-1 [unknown] [.] 0x0000ffff9f67c040 1.75% 1.75% :-1 [kernel.kallsyms] [k] ___cache_free 1.75% 1.75% :-1 [kernel.kallsyms] [k] __count_memcg_events Like Intel PT, add context switch records to track task info. As ARM SPE support was added later than PERF_RECORD_SWITCH_CPU_WIDE, I think we can safely set the attr.context_switch bit and use it. Reviewed-by:Leo Yan <leo.yan@linaro.org> Signed-off-by:
German Gomez <german.gomez@arm.com> Signed-off-by:
Namhyung Kim <namhyung@kernel.org> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: John Garry <john.garry@huawei.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mathieu Poirier <mathieu.poirier@linaro.org> Cc: Will Deacon <will@kernel.org> Cc: linux-arm-kernel@lists.infradead.org Link: https://lore.kernel.org/r/20211111133625.193568-2-german.gomez@arm.comSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Kajol Jain authored
Add PMU metric JSON file for power10 platform. Signed-off-by:
Kajol Jain <kjain@linux.ibm.com> Reviewed-by:
Paul Clarke <pc@us.ibm.com> Cc: Athira Jajeev <atrajeev@linux.vnet.ibm.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: Madhavan Srinivasan <maddy@linux.vnet.ibm.com> Cc: Nageswara R Sastry <rnsastry@linux.ibm.com> Cc: linuxppc-dev@lists.ozlabs.org Link: http://lore.kernel.org/lkml/20211108060010.177517-1-kjain@linux.ibm.comSigned-off-by:
-
Like Xu authored
We're not surprised that there are tons of Linux users who only read the documentation to learn about the kernel. Let's update the perf part for common hardware events since three new *generic* hardware events were added. Signed-off-by:
Like Xu <likexu@tencent.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> Link: http://lore.kernel.org/lkml/20211109090147.56978-1-likexu@tencent.comSigned-off-by:
-
Andrew Kilroy authored
Since the size is already printed earlier in hex, print the same data using the same format, in hex. Reviewed-by:
James Clark <james.clark@arm.com> Reviewed-by:
Andrew Kilroy <andrew.kilroy@arm.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Jiri Olsa <jolsa@redhat.com> Cc: John Garry <john.garry@huawei.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Mathieu Poirier <mathieu.poirier@linaro.org> Cc: Mike Leach <mike.leach@linaro.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Will Deacon <will@kernel.org> Cc: coresight@lists.linaro.org Cc: linux-arm-kernel@lists.infradead.org Link: https://lore.kernel.org/r/20211109142153.56546-3-german.gomez@arm.comSigned-off-by:
-
Andrew Kilroy authored
Since the size is already printed earlier in hex, print the same data using the same format, in hex. Reviewed-by:
Mathieu Poirier <mathieu.poirier@linaro.org> Signed-off-by:
-
German Gomez authored
Shell script test_arm_spe.sh has been added to test the recording of SPE tracing events in snapshot mode. Reviewed-by:
-
German Gomez authored
The head pointer of the AUX buffer managed by the arm_spe_pmu.c driver is not monotonically increasing, therefore the find_snapshot callback is needed in order to find the trace data within the AUX buffer and avoid wasting space in the perf.data file. The pointer is assumed to have wrapped if the buffer contains non-zero data at the end. If it has wrapped, the entire contents of the AUX buffer are stored in the perf.data file. Otherwise only the data up to the head pointer is stored. Reviewed-by:
-
German Gomez authored
This patch enables support for snapshot mode of arm_spe events, including the implementation of the necessary callbacks (excluding find_snapshot, which is to be included in a followup commit). Reviewed-by:
-
Ian Rogers authored
Events like uncore_imc/cas_count_read/ on Skylake open multiple events and then aggregate in the metric leader. To determine the average value per event the number of these events is needed. Add a source_count function that returns this value by counting the number of events with the given metric leader. For most events the value is 1 but for uncore_imc/cas_count_read/ it can yield values like 6. Add a generic test, but manually tested with a test metric that uses the function. Signed-off-by:
Ian Rogers <irogers@google.com> Acked-by:
Jiri Olsa <jolsa@redhat.com> Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com> Cc: Andi Kleen <ak@linux.intel.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: John Garry <john.garry@huawei.com> Cc: Kajol Jain <kjain@linux.ibm.com> Cc: Kan Liang <kan.liang@linux.intel.com> Cc: Madhavan Srinivasan <maddy@linux.ibm.com> Cc: Mark Rutland <mark.rutland@arm.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Paul A . Clarke <pc@us.ibm.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Riccardo Mancini <rickyman7@gmail.com> Cc: Song Liu <song@kernel.org> Cc: Wan Jiabing <wanjiabing@vivo.com> Cc: Yury Norov <yury.norov@gmail.com> Link: https://lore.kernel.org/r/20211111002109.194172-9-irogers@google.comSigned-off-by:
-
Ian Rogers authored
This will facilitate sharing in a follow-on change. Signed-off-by:
-
Ian Rogers authored
Allow the number of cpus, cores, dies and packages to be queried by a metric expression. Signed-off-by:
-
Ian Rogers authored
It is useful to have literal values for constants relating to topologies, SMT, etc. Make the parsing of literals shared code and add a lookup function. Move #smt_on to this function. Signed-off-by:
-
Ian Rogers authored
The topology name for thread_siblings is core_cpus_list, use this for consistency and add documentation. Signed-off-by:
-
Ian Rogers authored
The topology name for die_siblings is die_cpus_list, use this for consistency and add documentation. Signed-off-by:
-
Ian Rogers authored
core_siblings_list is the deprecated topology name for package_cpus_list, update the code to try the non-deprecated path first. Adjust variable names to match topology name. Signed-off-by:
-
Ian Rogers authored
An example of such an event is topdown-fe-bound. Signed-off-by:
-
Ian Rogers authored
Remove optionality, always run tests in a suite even if one fails. This brings perf's test more inline with kunit that lacks this notion. Signed-off-by:
Sohaib Mohamed <sohaib.amhmd@gmail.com> Acked-by:
-
Ian Rogers authored
All tests now return TEST_SKIP if not supported. Removing this function brings perf's test_suite struct more inline with kunit. Signed-off-by:
-
Ian Rogers authored
Migrate the is_supported functionality to returning TEST_SKIP. Motivation is kunit has no is_supported function. Signed-off-by:
-
Ian Rogers authored
Migrate the is_supported functionality to returning TEST_SKIP. Motivation is kunit has no is_supported function. Signed-off-by:
-
Ian Rogers authored
Convert shell tests to also run using test case style. Signed-off-by:
-
Ian Rogers authored
Migration toward kunit style test cases. Signed-off-by:
-
Ian Rogers authored
Migration toward kunit style test cases. Signed-off-by:
-
Ian Rogers authored
Replaced by null terminated test case array. Signed-off-by:
-
Ian Rogers authored
Use null terminated array of test cases rather than the previous sub test functions. Signed-off-by:
-
Ian Rogers authored
Use null terminated array of test cases rather than the previous sub test functions. Signed-off-by:
-
Ian Rogers authored
Use null terminated array of test cases rather than the previous sub test functions. Signed-off-by:
-
Ian Rogers authored
Use null terminated array of test cases rather than the previous sub test functions. Committer notes: On s/390x we don't use __event(), so wrap it with __s390x__ Signed-off-by:
-
Ian Rogers authored
Use null terminated array of test cases rather than the previous sub test functions. Signed-off-by:
-
Ian Rogers authored
Use null terminated array of test cases rather than the previous sub test functions. Signed-off-by:
-
Ian Rogers authored
This doesn't exist in kunit, but will ease the transition from perf tests. Signed-off-by:
-
Ian Rogers authored
Add a test case struct mirroring the 'struct kunit_case'. Use the struct with the DEFINE_SUITE macro, where the single test is turned into a test case. Update the helpers in builtin-test to handle test cases. Signed-off-by:
-
Ian Rogers authored
Abstract certain test features so that they can be refactored in later changes. No functional change. Signed-off-by:
-
Ian Rogers authored
This is to align with kunit's terminology. Signed-off-by:
-
Ian Rogers authored
Rather than export test functions, export the test struct. Rename with a suite__ prefix to avoid name collisions. Committer notes: Its '&suite__vectors_page', not '&suite__vectors_pages', noticed when cross building to arm (32-bit). Signed-off-by:
-
Ian Rogers authored
By switching to an array of pointers to tests (later to be suites) the definition of the tests can be moved to the file containing the tests. Committer notes: It's "&vectors_page", not "&vectors_pages", noticed when cross building to 32-bit ARM. Also the DEFINE_SUITE(vectors_page) should be done where its function is implemented, in tools/perf/arch/arm/tests/vectors-page.c, so that we can make it static, as we don't have anymore its declaration in tests.h. Signed-off-by:
-
- 12 Nov, 2021 4 commits
-
-
Ian Rogers authored
Add a macro to simplify later refactoring. No functional change. Signed-off-by:
-
Ian Rogers authored
Currently tests are setup in builtin-test with function pointers. Kunit exposes tests as a kunit_suite with a null terminated array of test cases. Use a macro to aid transition from one to the other in later changes. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
SOL_SOCKET has a different value according to the architecture, some have it as 0xffff while all the others have it as 1, so a simple string array isn't usable, add a scnprintf routine that treats it as a special case, using the array for other values. Signed-off-by: -
Arnaldo Carvalho de Melo authored
# perf trace -e setsockopt 0.000 ( 0.019 ms): systemd-resolv/1121 setsockopt(fd: 22, level: IP, optname: 50, optval: 0x7ffee2c0c134, optlen: 4) = 0 0.022 ( 0.003 ms): systemd-resolv/1121 setsockopt(fd: 22, level: IP, optname: 11, optval: 0x7ffee2c0c114, optlen: 4) = 0 0.027 ( 0.003 ms): systemd-resolv/1121 setsockopt(fd: 22, level: IP, optname: 8, optval: 0x7ffee2c0c134, optlen: 4) = 0 0.032 ( 0.002 ms): systemd-resolv/1121 setsockopt(fd: 22, level: IP, optname: 10, optval: 0x7ffee2c0c134, optlen: 4) = 0 0.036 ( 0.002 ms): systemd-resolv/1121 setsockopt(fd: 22, level: IP, optname: 25, optval: 0x7ffee2c0c114, optlen: 4) = 0 0.043 ( 0.003 ms): systemd-resolv/1121 setsockopt(fd: 22, level: 1, optname: 62, optval: 0x7ffee2c0c0fc, optlen: 4) = 0 0.055 ( 0.003 ms): systemd-resolv/1121 setsockopt(fd: 22, level: 1, optname: 25) ^C# So the simple straight STRARRAY method is not enough as SOL_SOCKET is '1' in most architectures but some use 0xffff (alpha, mips, parisc and sparc), so a followup patch will create a specialized scnprintf to cover that. Signed-off-by:
-
