kernel/tools/perf/util/evlist.h, branch linux-5.2.y

perf mmap: Implement dedicated memory buffer for data compression

2019-05-15T19:36:49Z

Implemented mmap data buffer that is used as the memory to operate on when compressing data in case of serial trace streaming. Signed-off-by: Alexey Budankov Reviewed-by: Jiri Olsa Cc: Alexander Shishkin Cc: Andi Kleen Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/49b31321-0f70-392b-9a4f-649d3affe090@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf record: Implement --mmap-flush= option

2019-04-01T18:18:10Z

Implement a --mmap-flush option that specifies minimal number of bytes that is extracted from mmaped kernel buffer to store into a trace. The default option value is 1 byte what means every time trace writing thread finds some new data in the mmaped buffer the data is extracted, possibly compressed and written to a trace. $ tools/perf/perf record --mmap-flush 1024 -e cycles -- matrix.gcc $ tools/perf/perf record --aio --mmap-flush 1K -e cycles -- matrix.gcc The option is independent from -z setting, doesn't vary with compression level and can serve two purposes. The first purpose is to increase the compression ratio of a trace data. Larger data chunks are compressed more effectively so the implemented option allows specifying data chunk size to compress. Also at some cases executing more write syscalls with smaller data size can take longer than executing less write syscalls with bigger data size due to syscall overhead so extracting bigger data chunks specified by the option value could additionally decrease runtime overhead. The second purpose is to avoid self monitoring live-lock issue in system wide (-a) profiling mode. Profiling in system wide mode with compression (-a -z) can additionally induce data into the kernel buffers along with the data from monitored processes. If performance data rate and volume from the monitored processes is high then trace streaming and compression activity in the tool is also high. High tool process activity can lead to subtle live-lock effect when compression of single new byte from some of mmaped kernel buffer leads to generation of the next single byte at some mmaped buffer. So perf tool process ends up in endless self monitoring. Implemented synch parameter is the mean to force data move independently from the specified flush threshold value. Despite the provided flush value the tool needs capability to unconditionally drain memory buffers, at least in the end of the collection. Committer testing: Running with the default value, i.e. as soon as there is something to read go on consuming, we first write the synthesized events, small chunks of about 128 bytes: # perf trace -m 2048 --call-graph dwarf -e write -- perf record 101.142 ( 0.004 ms): perf/25821 write(fd: 3, buf: 0x210db60, count: 120) = 120 __libc_write (/usr/lib64/libpthread-2.28.so) ion (/home/acme/bin/perf) record__write (inlined) process_synthesized_event (/home/acme/bin/perf) perf_tool__process_synth_event (inlined) perf_event__synthesize_mmap_events (/home/acme/bin/perf) Then we move to reading the mmap buffers consuming the events put there by the kernel perf infrastructure: 107.561 ( 0.005 ms): perf/25821 write(fd: 3, buf: 0x7f1befc02000, count: 336) = 336 __libc_write (/usr/lib64/libpthread-2.28.so) ion (/home/acme/bin/perf) record__write (inlined) record__pushfn (/home/acme/bin/perf) perf_mmap__push (/home/acme/bin/perf) record__mmap_read_evlist (inlined) record__mmap_read_all (inlined) __cmd_record (inlined) cmd_record (/home/acme/bin/perf) 12919.953 ( 0.136 ms): perf/25821 write(fd: 3, buf: 0x7f1befc83150, count: 184984) = 184984 12920.094 ( 0.155 ms): perf/25821 write(fd: 3, buf: 0x7f1befc02150, count: 261816) = 261816 12920.253 ( 0.093 ms): perf/25821 write(fd: 3, buf: 0x7f1befb81120, count: 170832) = 170832 If we limit it to write only when more than 16MB are available for reading, it throttles that to a quarter of the --mmap-pages set for 'perf record', which by default get to 528384 bytes, found out using 'record -v': mmap flush: 132096 mmap size 528384B With that in place all the writes coming from record__mmap_read_evlist(), i.e. from the mmap buffers setup by the kernel perf infrastructure were at least 132096 bytes long. Trying with a bigger mmap size: perf trace -e write perf record -v -m 2048 --mmap-flush 16M 74982.928 ( 2.471 ms): perf/26500 write(fd: 3, buf: 0x7ff94a6cc000, count: 3580888) = 3580888 74985.406 ( 2.353 ms): perf/26500 write(fd: 3, buf: 0x7ff949ecb000, count: 3453256) = 3453256 74987.764 ( 2.629 ms): perf/26500 write(fd: 3, buf: 0x7ff9496ca000, count: 3859232) = 3859232 74990.399 ( 2.341 ms): perf/26500 write(fd: 3, buf: 0x7ff948ec9000, count: 3769032) = 3769032 74992.744 ( 2.064 ms): perf/26500 write(fd: 3, buf: 0x7ff9486c8000, count: 3310520) = 3310520 74994.814 ( 2.619 ms): perf/26500 write(fd: 3, buf: 0x7ff947ec7000, count: 4194688) = 4194688 74997.439 ( 2.787 ms): perf/26500 write(fd: 3, buf: 0x7ff9476c6000, count: 4029760) = 4029760 Was again limited to a quarter of the mmap size: mmap flush: 2098176 mmap size 8392704B A warning about that would be good to have but can be added later, something like: "max flush is a quarter of the mmap size, if wanting to bump the mmap flush further, bump the mmap size as well using -m/--mmap-pages" Also rename the 'sync' parameters to 'synch' to keep tools/perf building with older glibcs: cc1: warnings being treated as errors builtin-record.c: In function 'record__mmap_read_evlist': builtin-record.c:775: warning: declaration of 'sync' shadows a global declaration /usr/include/unistd.h:933: warning: shadowed declaration is here builtin-record.c: In function 'record__mmap_read_all': builtin-record.c:856: warning: declaration of 'sync' shadows a global declaration /usr/include/unistd.h:933: warning: shadowed declaration is here Signed-off-by: Alexey Budankov Reviewed-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Alexander Shishkin Cc: Andi Kleen Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/f6600d72-ecfa-2eb7-7e51-f6954547d500@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf evsel: Fix max perf_event_attr.precise_ip detection

2019-03-28T17:31:56Z

After a discussion with Andi, move the perf_event_attr.precise_ip detection for maximum precise config (via :P modifier or for default cycles event) to perf_evsel__open(). The current detection in perf_event_attr__set_max_precise_ip() is tricky, because precise_ip config is specific for given event and it currently checks only hw cycles. We now check for valid precise_ip value right after failing sys_perf_event_open() for specific event, before any of the perf_event_attr fallback code gets executed. This way we get the proper config in perf_event_attr together with allowed precise_ip settings. We can see that code activity with -vv, like: $ perf record -vv ls ... ------------------------------------------------------------ perf_event_attr: size 112 { sample_period, sample_freq } 4000 ... precise_ip 3 sample_id_all 1 exclude_guest 1 mmap2 1 comm_exec 1 ksymbol 1 ------------------------------------------------------------ sys_perf_event_open: pid 9926 cpu 0 group_fd -1 flags 0x8 sys_perf_event_open failed, error -95 decreasing precise_ip by one (2) ------------------------------------------------------------ perf_event_attr: size 112 { sample_period, sample_freq } 4000 ... precise_ip 2 sample_id_all 1 exclude_guest 1 mmap2 1 comm_exec 1 ksymbol 1 ------------------------------------------------------------ sys_perf_event_open: pid 9926 cpu 0 group_fd -1 flags 0x8 = 4 ... Suggested-by: Andi Kleen Signed-off-by: Jiri Olsa Link: http://lkml.kernel.org/n/tip-dkvxxbeg7lu74155d4jhlmc9@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf evlist: Introduce side band thread

2019-03-21T14:27:03Z

This patch introduces side band thread that captures extended information for events like PERF_RECORD_BPF_EVENT. This new thread uses its own evlist that uses ring buffer with very low watermark for lower latency. To use side band thread, we need to: 1. add side band event(s) by calling perf_evlist__add_sb_event(); 2. calls perf_evlist__start_sb_thread(); 3. at the end of perf run, perf_evlist__stop_sb_thread(). In the next patch, we use this thread to handle PERF_RECORD_BPF_EVENT. Committer notes: Add fix by Jiri Olsa for when te sb_tread can't get started and then at the end the stop_sb_thread() segfaults when joining the (non-existing) thread. That can happen when running 'perf top' or 'perf record' as a normal user, for instance. Further checks need to be done on top of this to more graciously handle these possible failure scenarios. Signed-off-by: Song Liu Reviewed-by: Jiri Olsa Tested-by: Arnaldo Carvalho de Melo Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Namhyung Kim Cc: Peter Zijlstra Cc: Stanislav Fomichev Link: http://lkml.kernel.org/r/20190312053051.2690567-15-songliubraving@fb.com Signed-off-by: Arnaldo Carvalho de Melo

perf record: Allocate affinity masks

2019-02-06T13:00:39Z

Allocate affinity option and masks for mmap data buffers and record thread as well as initialize allocated objects. Signed-off-by: Alexey Budankov Reviewed-by: Jiri Olsa Cc: Alexander Shishkin Cc: Andi Kleen Cc: Namhyung Kim Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/526fa2b0-07de-6dbd-a7e9-26ba875593c9@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf report: Display arch specific diagnostic counter sets, starting with s390

2019-01-21T20:00:48Z

On s390 the event bc000 (also named CF_DIAG) extracts the CPU Measurement Facility diagnostic counter sets and displays them as counter number and counter value pairs sorted by counter set number. Output: [root@s35lp76 perf]# ./perf report -D --stdio [00000000] Counterset:0 Counters:6 Counter:000 Value:0x000000000085ec36 Counter:001 Value:0x0000000000796c94 Counter:002 Value:0x0000000000005ada Counter:003 Value:0x0000000000092460 Counter:004 Value:0x0000000000006073 Counter:005 Value:0x00000000001a9a73 [0x000038] Counterset:1 Counters:2 Counter:000 Value:0x000000000007c59f Counter:001 Value:0x000000000002fad6 [0x000050] Counterset:2 Counters:16 Counter:000 Value:000000000000000000 Counter:001 Value:000000000000000000 Counter:002 Value:000000000000000000 Counter:003 Value:000000000000000000 Counter:004 Value:000000000000000000 Counter:005 Value:000000000000000000 Counter:006 Value:000000000000000000 Counter:007 Value:000000000000000000 Counter:008 Value:000000000000000000 Counter:009 Value:000000000000000000 Counter:010 Value:000000000000000000 Counter:011 Value:000000000000000000 Counter:012 Value:000000000000000000 Counter:013 Value:000000000000000000 Counter:014 Value:000000000000000000 Counter:015 Value:000000000000000000 [0x0000d8] Counterset:3 Counters:128 Counter:000 Value:0x000000000000020f Counter:001 Value:0x00000000000001d8 Counter:002 Value:0x000000000000d7fa Counter:003 Value:0x000000000000008b ... The number in brackets is the offset into the raw data field of the sample. New functions trace_event_sample_raw__init() and s390_sample_raw() are introduced in the code path to enable interpretation on non s390 platforms. This event bc000 attached raw data is generated only on s390 platform. Correct display on other platforms requires correct endianness handling. Committer notes: Added a init function that sets up a evlist function pointer to avoid repeated tests on evlist->env and calls to perf_env__name() that involves normalizing, etc, for each PERF_RECORD_SAMPLE. Removed needless __maybe_unused from the trace_event_raw() prototype in session.h, move it to be an static function in evlist. The 'offset' variable is a size_t, not an u64, fix it to avoid this on some arches: CC /tmp/build/perf/util/s390-sample-raw.o util/s390-sample-raw.c: In function 's390_cpumcfdg_testctr': util/s390-sample-raw.c:77:4: error: format '%llx' expects argument of type 'long long unsigned int', but argument 4 has type 'size_t' [-Werror=format=] pr_err("Invalid counter set entry at %#" PRIx64 "\n", ^ cc1: all warnings being treated as errors Signed-off-by: Thomas Richter Reviewed-by: Hendrik Brueckner Link: https://lkml.kernel.org/r/9c856ac0-ef23-72b5-901d-a1f815508976@linux.ibm.com Cc: Martin Schwidefsky Cc: Heiko Carstens Link: https://lkml.kernel.org/n/tip-s3jhif06et9ug78qhclw41z1@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf record: Enable asynchronous trace writing

2018-12-17T17:55:08Z

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 Reviewed-by: Jiri Olsa Acked-by: Namhyung Kim Cc: Alexander Shishkin Cc: Andi Kleen Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/ce2d45e9-d236-871c-7c8f-1bed2d37e8ac@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo

perf evlist: Rename perf_evlist__set_filter* to perf_evlist__set_tp_filter*

2018-11-21T15:00:31Z

To better reflect that this is a tracepoint filter, as opposed, for instance to map based BPF filters. Cc: Adrian Hunter Cc: David Ahern Cc: Jiri Olsa Cc: Namhyung Kim Cc: Wang Nan Link: https://lkml.kernel.org/n/tip-9138svli6ddcphrr3ymy9oy3@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo

perf evlist: Move perf_evsel__reset_weak_group into evlist

2018-11-05T17:37:09Z

- Move the function from builtin-stat to evlist for reuse - Rename to evlist to match purpose better - Pass the evlist as first argument. - No functional changes Signed-off-by: Andi Kleen Acked-by: Jiri Olsa Link: http://lkml.kernel.org/r/20181001195927.14211-1-andi@firstfloor.org Signed-off-by: Arnaldo Carvalho de Melo

perf evlist: Introduce force_leader() method

2018-05-21T17:40:54Z

For non-explicit group (e.g. those created with -e '{eventA,eventB}'), 'perf report' supports a option '--group' which can enable group output. We also need to support 'perf annotate' with the same '--group'. Create a new function perf_evlist__force_leader() which contains common code to force setting the group leader. Signed-off-by: Jin Yao Tested-by: Arnaldo Carvalho de Melo Cc: Alexander Shishkin Cc: Andi Kleen Cc: Jiri Olsa Cc: Kan Liang Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/1526914666-31839-2-git-send-email-yao.jin@linux.intel.com Signed-off-by: Arnaldo Carvalho de Melo