perf/core improvements and fixes
Arnaldo Carvalho de Melo
maintainerHi Ingo/Thomas,
Please consider pulling,
Best regards,
- Arnaldo
Test results at the end of this message, as usual.
The following changes since commit 2b32769700f857a8e608a8ee24080833889965b9:
Merge tag 'perf-urgent-for-mingo-5.4-20190921' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/urgent (2019-09-22 12:45:11 +0200)
are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git tags/perf-core-for-mingo-5.5-20190925
for you to fetch changes up to d6840d87b2d148e19e244ad2b44d28ba07f437a0:
perf parser: Remove needless include directives (2019-09-25 16:26:41 -0300)
----------------------------------------------------------------
perf/core improvements and fixes:
perf record:
Stephane Eranian:
- Fix priv level with branch sampling for paranoid=2, i.e. the kernel checks if perf_event_attr_attr.exclude_hv is set in addition to .exclude_kernel, so reset both to zero.
Arnaldo Carvalho de Melo:
- Don't warn about not being able to read kernel maps (kallsyms, etc) when kernel samples aren't being collected.
perf list:
Kim Phillips:
- Allow plurals for metric, metricgroup., i.e.:
$ perf list metrics
was showing nothing, which is very confusing, make it work like:
$ perf stat metric
perf stat:
Andi Kleen:
- Free memory access/leaks detected via valgrind, related to metrics.
Libraries:
libperf:
Jiri Olsa:
- Move more stuff from tools/perf, this time a first stab at moving perf_mmap methods.
libtracevent:
Steven Rostedt (VMware):
- Round up in tep_print_event() time precision.
Tzvetomir Stoyanov (VMware):
- Man pages for event print and related and plugins APIs.
- Move traceevent plugins in its own subdirectory.
Feature detection:
Thomas Richter:
- Add detection of java-11-openjdk-devel package, in addition to the older versions supported.
Architecture specific:
S/390:
Thomas Richter (2):
- Include JVMTI support for s390
Vendor events:
AMD:
Kim Phillips:
- Add L3 cache events for Family 17h.
- Remove redundant '['.
PowerPC:
Mamatha Inamdar:
- Remove P8 HW events which are not supported.
Cleanups:
Arnaldo Carvalho de Melo:
- Remove needless headers, add needed ones, move things around to reduce the headers dependency tree, speeding up builds by not doing needless compiles when unrelated stuff gets changed.
- Ditch unused code that was dragging headers.
----------------------------------------------------------------
Andi Kleen (2):
perf stat: Fix free memory access / memory leaks in metrics perf evlist: Fix access of freed id arrays
Arnaldo Carvalho de Melo (12):
perf record: Move restricted maps check to after a possible fallback to not collect kernel samples perf evlist: Adopt backwards ring buffer state enum
libperf: Add missing 'struct xyarray' forward declaration
perf tools: No need to include internal/lib.h from util/util.h
libperf: Use sys/types.h to get ssize_t, not unistd.h
perf copyfile: Move copyfile routines to separate files
perf evsel: Remove need for symbol_conf in evsel_fprintf.c
perf evsel: Introduce evsel_fprintf.h
perf evlist: Remove unused perf_evlist__fprintf() method
perf evsel: Move config terms to a separate header
perf tools: Replace needless mmap.h with what is needed, event.h
perf parser: Remove needless include directives
Jiri Olsa (37):
tools: Add missing stdio.h include to asm/bug.h header perf tools: Rename 'struct perf_mmap' to 'struct mmap' perf tools: Rename perf_evlist__mmap() to evlist__mmap() perf tools: Rename perf_evlist__munmap() to evlist__munmap() perf tools: Rename perf_evlist__alloc_mmap() to evlist__alloc_mmap() perf tools: Rename perf_evlist__exit() to evlist__exit() perf tools: Rename perf_evlist__purge() to evlist__purge()
libperf: Link libapi.a in libperf.so
libperf: Add perf_mmap struct
libperf: Add 'mask' to struct perf_mmap
libperf: Add 'fd' to struct perf_mmap
libperf: Add 'cpu' to struct perf_mmap
libperf: Add 'refcnt' to struct perf_mmap
libperf: Add prev/start/end to struct perf_mmap
libperf: Add 'overwrite' to 'struct perf_mmap'
libperf: Add 'event_copy' to 'struct perf_mmap'
libperf: Add 'flush' to 'struct perf_mmap'
libperf: Move 'system_wide' from 'struct evsel' to 'struct perf_evsel'
libperf: Move 'nr_mmaps' from 'struct evlist' to 'struct perf_evlist'
libperf: Move 'mmap_len' from 'struct evlist' to 'struct perf_evlist'
libperf: Move 'pollfd' from 'struct evlist' to 'struct perf_evlist'
libperf: Move 'sample_id' from 'struct evsel' to 'struct perf_evsel'
libperf: Move 'id' from 'struct evsel' to 'struct perf_evsel'
libperf: Move 'ids' from 'struct evsel' to 'struct perf_evsel'
libperf: Move 'heads' from 'struct evlist' to 'struct perf_evlist'
libperf: Add perf_evsel__alloc_id/perf_evsel__free_id functions
libperf: Add perf_evlist__first()/last() functions
libperf: Add perf_evlist__read_format() function
libperf: Add perf_evlist__id_add() function
libperf: Add perf_evlist__id_add_fd() function
libperf: Move 'page_size' global variable to libperf
libperf: Add libperf dependency for tests targets
libperf: Merge libperf_set_print() into libperf_init()
libperf: Add libperf_init() call to the tests
libperf: Add perf_evlist__alloc_pollfd() function
libperf: Add perf_evlist__add_pollfd() function
libperf: Add perf_evlist__poll() function
Kim Phillips (4):
perf vendor events amd: Add L3 cache events for Family 17h perf vendor events amd: Remove redundant '[' perf vendor events: Minor fixes to the README perf list: Allow plurals for metric, metricgroup
Mamatha Inamdar (1):
perf vendor events: Remove P8 HW events which are not supported
Stephane Eranian (1):
perf record: Fix priv level with branch sampling for paranoid=2
Steven Rostedt (VMware) (1):
libtraceevent: Round up in tep_print_event() time precision
Thomas Richter (2):
perf jvmti: Include JVMTI support for s390
perf build: Add detection of java-11-openjdk-devel package
Tzvetomir Stoyanov (2):
libtraceevent: Man pages for libtraceevent event print related API
libtraceevent: Man pages for tep plugins APIs
Tzvetomir Stoyanov (VMware) (4):
libtraceevent: Man pages fix, rename tep_ref_get() to tep_get_ref()
libtraceevent: Man pages fix, changes in event printing APIs
libtraceevent: Add tep_get_event() in event-parse.h
libtraceevent: Move traceevent plugins in its own subdirectory
tools/include/asm/bug.h | 1 +
tools/lib/traceevent/Build | 11 -
.../Documentation/libtraceevent-event_print.txt | 130 +++++++++
.../Documentation/libtraceevent-handle.txt | 8 +-
.../Documentation/libtraceevent-plugins.txt | 99 +++++++
.../lib/traceevent/Documentation/libtraceevent.txt | 15 +-
tools/lib/traceevent/Makefile | 94 ++-----
tools/lib/traceevent/event-parse.c | 4 +-
tools/lib/traceevent/event-parse.h | 2 +
tools/lib/traceevent/plugins/Build | 10 +
tools/lib/traceevent/plugins/Makefile | 222 ++++++++++++++++
.../lib/traceevent/{ => plugins}/plugin_cfg80211.c | 0
.../lib/traceevent/{ => plugins}/plugin_function.c | 0
.../lib/traceevent/{ => plugins}/plugin_hrtimer.c | 0
tools/lib/traceevent/{ => plugins}/plugin_jbd2.c | 0
tools/lib/traceevent/{ => plugins}/plugin_kmem.c | 0
tools/lib/traceevent/{ => plugins}/plugin_kvm.c | 0
.../lib/traceevent/{ => plugins}/plugin_mac80211.c | 0
.../traceevent/{ => plugins}/plugin_sched_switch.c | 0
tools/lib/traceevent/{ => plugins}/plugin_scsi.c | 0
tools/lib/traceevent/{ => plugins}/plugin_xen.c | 0
tools/perf/Makefile.config | 2 +-
tools/perf/Makefile.perf | 4 +-
tools/perf/arch/arm/util/cs-etm.c | 7 +-
tools/perf/arch/arm64/util/arm-spe.c | 6 +-
tools/perf/arch/s390/Makefile | 1 +
tools/perf/arch/s390/util/auxtrace.c | 1 +
tools/perf/arch/s390/util/machine.c | 2 +-
tools/perf/arch/x86/tests/intel-cqm.c | 5 +-
tools/perf/arch/x86/tests/perf-time-to-tsc.c | 11 +-
tools/perf/arch/x86/tests/rdpmc.c | 2 +-
tools/perf/arch/x86/util/intel-bts.c | 9 +-
tools/perf/arch/x86/util/intel-pt.c | 17 +-
tools/perf/arch/x86/util/machine.c | 2 +-
tools/perf/builtin-evlist.c | 1 +
tools/perf/builtin-kvm.c | 13 +-
tools/perf/builtin-list.c | 4 +-
tools/perf/builtin-record.c | 102 +++----
tools/perf/builtin-sched.c | 3 +-
tools/perf/builtin-script.c | 11 +-
tools/perf/builtin-stat.c | 6 +-
tools/perf/builtin-top.c | 22 +-
tools/perf/builtin-trace.c | 17 +-
tools/perf/lib/Makefile | 35 ++-
tools/perf/lib/core.c | 13 +-
tools/perf/lib/evlist.c | 124 +++++++++
tools/perf/lib/evsel.c | 30 +++
tools/perf/lib/include/internal/evlist.h | 33 +++
tools/perf/lib/include/internal/evsel.h | 33 +++
tools/perf/lib/include/internal/lib.h | 4 +-
tools/perf/lib/include/internal/mmap.h | 32 +++
tools/perf/lib/include/perf/core.h | 2 +-
tools/perf/lib/include/perf/evlist.h | 1 +
tools/perf/lib/lib.c | 2 +
tools/perf/lib/libperf.map | 3 +-
tools/perf/lib/tests/test-cpumap.c | 10 +
tools/perf/lib/tests/test-evlist.c | 10 +
tools/perf/lib/tests/test-evsel.c | 10 +
tools/perf/lib/tests/test-threadmap.c | 10 +
tools/perf/perf.c | 13 +-
tools/perf/pmu-events/README | 22 +-
.../perf/pmu-events/arch/powerpc/power8/other.json | 24 --
.../perf/pmu-events/arch/x86/amdfam17h/cache.json | 42 +++
tools/perf/pmu-events/arch/x86/amdfam17h/core.json | 2 +-
tools/perf/pmu-events/jevents.c | 1 +
tools/perf/tests/backward-ring-buffer.c | 11 +-
tools/perf/tests/bpf.c | 9 +-
tools/perf/tests/code-reading.c | 11 +-
tools/perf/tests/event-times.c | 14 +-
tools/perf/tests/event_update.c | 6 +-
tools/perf/tests/evsel-roundtrip-name.c | 2 +-
tools/perf/tests/hists_cumulate.c | 2 +-
tools/perf/tests/hists_link.c | 5 +-
tools/perf/tests/hists_output.c | 2 +-
tools/perf/tests/keep-tracking.c | 11 +-
tools/perf/tests/mmap-basic.c | 5 +-
tools/perf/tests/mmap-thread-lookup.c | 2 +-
tools/perf/tests/openat-syscall-tp-fields.c | 11 +-
tools/perf/tests/parse-events.c | 116 ++++----
tools/perf/tests/perf-record.c | 13 +-
tools/perf/tests/sdt.c | 1 +
tools/perf/tests/sw-clock.c | 5 +-
tools/perf/tests/switch-tracking.c | 29 +-
tools/perf/tests/task-exit.c | 9 +-
tools/perf/tests/vmlinux-kallsyms.c | 2 +-
tools/perf/ui/browsers/hists.c | 6 +-
tools/perf/ui/gtk/hists.c | 1 +
tools/perf/util/Build | 2 +
tools/perf/util/annotate.c | 1 +
tools/perf/util/auxtrace.c | 8 +-
tools/perf/util/auxtrace.h | 8 +-
tools/perf/util/bpf-loader.c | 2 +-
tools/perf/util/build-id.c | 3 +-
tools/perf/util/copyfile.c | 144 ++++++++++
tools/perf/util/copyfile.h | 16 ++
tools/perf/util/cs-etm.c | 2 +-
tools/perf/util/evlist.c | 295 ++++++---------------
tools/perf/util/evlist.h | 81 +++---
tools/perf/util/evsel.c | 204 ++------------
tools/perf/util/evsel.h | 121 +--------
tools/perf/util/evsel_config.h | 50 ++++
tools/perf/util/evsel_fprintf.c | 15 +-
tools/perf/util/evsel_fprintf.h | 50 ++++
tools/perf/util/genelf.h | 3 +
tools/perf/util/header.c | 29 +-
tools/perf/util/intel-bts.c | 4 +-
tools/perf/util/intel-pt.c | 10 +-
tools/perf/util/jitdump.c | 2 +-
tools/perf/util/machine.c | 1 +
tools/perf/util/mmap.c | 185 ++++++-------
tools/perf/util/mmap.h | 77 ++----
tools/perf/util/parse-events.c | 8 +-
tools/perf/util/parse-events.y | 4 +-
tools/perf/util/perf_event_attr_fprintf.c | 148 +++++++++++
tools/perf/util/python-ext-sources | 1 +
tools/perf/util/python.c | 24 +-
tools/perf/util/record.c | 6 +-
tools/perf/util/session.c | 5 +-
tools/perf/util/sort.c | 2 +-
tools/perf/util/srccode.c | 2 +-
tools/perf/util/stat-shadow.c | 4 +-
tools/perf/util/stat.c | 2 +-
tools/perf/util/symbol-elf.c | 2 +-
tools/perf/util/synthetic-events.c | 20 +-
tools/perf/util/top.c | 2 +-
tools/perf/util/trace-event-info.c | 2 +-
tools/perf/util/util.c | 136 ----------
tools/perf/util/util.h | 8 -
128 files changed, 1941 insertions(+), 1321 deletions(-)
create mode 100644 tools/lib/traceevent/Documentation/libtraceevent-event_print.txt
create mode 100644 tools/lib/traceevent/Documentation/libtraceevent-plugins.txt
create mode 100644 tools/lib/traceevent/plugins/Build
create mode 100644 tools/lib/traceevent/plugins/Makefile
rename tools/lib/traceevent/{ => plugins}/plugin_cfg80211.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_function.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_hrtimer.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_jbd2.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_kmem.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_kvm.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_mac80211.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_sched_switch.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_scsi.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_xen.c (100%)
create mode 100644 tools/perf/lib/include/internal/mmap.h
create mode 100644 tools/perf/util/copyfile.c
create mode 100644 tools/perf/util/copyfile.h
create mode 100644 tools/perf/util/evsel_config.h
create mode 100644 tools/perf/util/evsel_fprintf.h
create mode 100644 tools/perf/util/perf_event_attr_fprintf.cTest 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.
Clearlinux is failing when building with libpython, but that is not a perf regression, will try to remove one compiler warning that is causing the problem when building some of the glue code files in the python files, outside perf.
# export PERF_TARBALL=http://192.168.124.1/perf/perf-5.3.0.tar.xz # dm 1 alpine:3.4 : Ok gcc (Alpine 5.3.0) 5.3.0, clang version 3.8.0 (tags/RELEASE_380/final) 2 alpine:3.5 : Ok gcc (Alpine 6.2.1) 6.2.1 20160822, clang version 3.8.1 (tags/RELEASE_381/final) 3 alpine:3.6 : Ok gcc (Alpine 6.3.0) 6.3.0, clang version 4.0.0 (tags/RELEASE_400/final) 4 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 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 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 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 alpine:edge : Ok gcc (Alpine 8.3.0) 8.3.0, Alpine clang version 8.0.1 (tags/RELEASE_801/final) (based on LLVM 8.0.1) 9 amazonlinux:1 : Ok gcc (GCC) 7.2.1 20170915 (Red Hat 7.2.1-2), clang version 3.6.2 (tags/RELEASE_362/final) 10 amazonlinux:2 : Ok gcc (GCC) 7.3.1 20180303 (Red Hat 7.3.1-5), clang version 7.0.1 (Amazon Linux 2 7.0.1-1.amzn2.0.2) 11 android-ndk:r12b-arm : Ok arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease) 12 android-ndk:r15c-arm : Ok arm-linux-androideabi-gcc (GCC) 4.9.x 20150123 (prerelease) 13 centos:5 : Ok gcc (GCC) 4.1.2 20080704 (Red Hat 4.1.2-55) 14 centos:6 : Ok gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-23) 15 centos:7 : Ok gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39), clang version 3.4.2 (tags/RELEASE_34/dot2-final) 16 clearlinux:latest : Ok gcc (Clear Linux OS for Intel Architecture) 9.2.1 20190908 gcc-9-branch@275492, clang version 8.0.0 (tags/RELEASE_800/final) 17 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) 18 debian:9 : Ok gcc (Debian 6.3.0-18+deb9u1) 6.3.0 20170516, clang version 3.8.1-24 (tags/RELEASE_381/final) 19 debian:10 : Ok gcc (Debian 8.3.0-6) 8.3.0, clang version 7.0.1-8 (tags/RELEASE_701/final) 20 debian:experimental : Ok gcc (Debian 9.2.1-8) 9.2.1 20190909, clang version 8.0.1-3+b1 (tags/RELEASE_801/final) 21 debian:experimental-x-arm64 : Ok aarch64-linux-gnu-gcc (Debian 8.3.0-19) 8.3.0 22 debian:experimental-x-mips : Ok mips-linux-gnu-gcc (Debian 8.3.0-19) 8.3.0 23 debian:experimental-x-mips64 : Ok mips64-linux-gnuabi64-gcc (Debian 8.3.0-19) 8.3.0 24 debian:experimental-x-mipsel : Ok mipsel-linux-gnu-gcc (Debian 8.3.0-19) 8.3.0 25 fedora:20 : Ok gcc (GCC) 4.8.3 20140911 (Red Hat 4.8.3-7), clang version 3.4.2 (tags/RELEASE_34/dot2-final) 26 fedora:22 : Ok gcc (GCC) 5.3.1 20160406 (Red Hat 5.3.1-6), clang version 3.5.0 (tags/RELEASE_350/final) 27 fedora:23 : Ok gcc (GCC) 5.3.1 20160406 (Red Hat 5.3.1-6), clang version 3.7.0 (tags/RELEASE_370/final) 28 fedora:24 : Ok gcc (GCC) 6.3.1 20161221 (Red Hat 6.3.1-1), clang version 3.8.1 (tags/RELEASE_381/final) 29 fedora:24-x-ARC-uClibc : Ok arc-linux-gcc (ARCompact ISA Linux uClibc toolchain 2017.09-rc2) 7.1.1 20170710 30 fedora:25 : Ok gcc (GCC) 6.4.1 20170727 (Red Hat 6.4.1-1), clang version 3.9.1 (tags/RELEASE_391/final) 31 fedora:26 : Ok gcc (GCC) 7.3.1 20180130 (Red Hat 7.3.1-2), clang version 4.0.1 (tags/RELEASE_401/final) 32 fedora:27 : Ok gcc (GCC) 7.3.1 20180712 (Red Hat 7.3.1-6), clang version 5.0.2 (tags/RELEASE_502/final) 33 fedora:28 : Ok gcc (GCC) 8.3.1 20190223 (Red Hat 8.3.1-2), clang version 6.0.1 (tags/RELEASE_601/final) 34 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) 35 fedora:30 : Ok gcc (GCC) 9.2.1 20190827 (Red Hat 9.2.1-1), clang version 8.0.0 (Fedora 8.0.0-1.fc30) 36 fedora:30-x-ARC-glibc : Ok arc-linux-gcc (ARC HS GNU/Linux glibc toolchain 2019.03-rc1) 8.3.1 20190225 37 fedora:30-x-ARC-uClibc : Ok arc-linux-gcc (ARCv2 ISA Linux uClibc toolchain 2019.03-rc1) 8.3.1 20190225 38 fedora:31 : Ok gcc (GCC) 9.2.1 20190827 (Red Hat 9.2.1-1), clang version 8.0.0 (Fedora 8.0.0-3.fc31.1) 39 fedora:rawhide : Ok gcc (GCC) 9.2.1 20190827 (Red Hat 9.2.1-1), clang version 9.0.0 (Fedora 9.0.0-0.2.rc3.fc31) 40 gentoo-stage3-amd64:latest : Ok gcc (Gentoo 8.3.0-r1 p1.1) 8.3.0 41 mageia:5 : Ok gcc (GCC) 4.9.2, clang version 3.5.2 (tags/RELEASE_352/final) 42 mageia:6 : Ok gcc (Mageia 5.5.0-1.mga6) 5.5.0, clang version 3.9.1 (tags/RELEASE_391/final) 43 mageia:7 : Ok gcc (Mageia 8.3.1-0.20190524.1.mga7) 8.3.1 20190524, clang version 8.0.0 (Mageia 8.0.0-1.mga7) 44 manjaro:latest : Ok gcc (GCC) 9.1.0, clang version 8.0.1 (tags/RELEASE_801/final) 45 opensuse:15.0 : Ok gcc (SUSE Linux) 7.4.1 20190424 [gcc-7-branch revision 270538], clang version 5.0.1 (tags/RELEASE_501/final 312548) 46 opensuse:15.1 : Ok gcc (SUSE Linux) 7.4.1 20190424 [gcc-7-branch revision 270538], clang version 7.0.1 (tags/RELEASE_701/final 349238) 47 opensuse:42.3 : Ok gcc (SUSE Linux) 4.8.5, clang version 3.8.0 (tags/RELEASE_380/final 262553) 48 opensuse:tumbleweed : Ok gcc (SUSE Linux) 9.2.1 20190820 [gcc-9-branch revision 274748], clang version 8.0.1 (tags/RELEASE_801/final 366581) 49 oraclelinux:6 : Ok gcc (GCC) 4.4.7 20120313 (Red Hat 4.4.7-23.0.1) 50 oraclelinux:7 : Ok gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-39.0.1), clang version 3.4.2 (tags/RELEASE_34/dot2-final) 51 oraclelinux:8 : Ok gcc (GCC) 8.2.1 20180905 (Red Hat 8.2.1-3.0.1), clang version 7.0.1 (tags/RELEASE_701/final) 52 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) 53 ubuntu:14.04 : Ok gcc (Ubuntu 4.8.4-2ubuntu1~14.04.4) 4.8.4, Ubuntu clang version 3.4-1ubuntu3 (tags/RELEASE_34/final) (based on LLVM 3.4) 54 ubuntu:16.04 : Ok gcc (Ubuntu 5.4.0-6ubuntu1~16.04.11) 5.4.0 20160609, clang version 3.8.0-2ubuntu4 (tags/RELEASE_380/final) 55 ubuntu:16.04-x-arm : Ok arm-linux-gnueabihf-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609 56 ubuntu:16.04-x-arm64 : Ok aarch64-linux-gnu-gcc (Ubuntu/Linaro 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609 57 ubuntu:16.04-x-powerpc : Ok powerpc-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609 58 ubuntu:16.04-x-powerpc64 : Ok powerpc64-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609 59 ubuntu:16.04-x-powerpc64el : Ok powerpc64le-linux-gnu-gcc (Ubuntu/IBM 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609 60 ubuntu:16.04-x-s390 : Ok s390x-linux-gnu-gcc (Ubuntu 5.4.0-6ubuntu1~16.04.9) 5.4.0 20160609 61 ubuntu:18.04 : Ok gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0, clang version 6.0.0-1ubuntu2 (tags/RELEASE_600/final) 62 ubuntu:18.04-x-arm : Ok arm-linux-gnueabihf-gcc (Ubuntu/Linaro 7.4.0-1ubuntu1~18.04.1) 7.4.0 63 ubuntu:18.04-x-arm64 : Ok aarch64-linux-gnu-gcc (Ubuntu/Linaro 7.4.0-1ubuntu1~18.04.1) 7.4.0 64 ubuntu:18.04-x-m68k : Ok m68k-linux-gnu-gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0 65 ubuntu:18.04-x-powerpc : Ok powerpc-linux-gnu-gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0 66 ubuntu:18.04-x-powerpc64 : Ok powerpc64-linux-gnu-gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0 67 ubuntu:18.04-x-powerpc64el : Ok powerpc64le-linux-gnu-gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0 68 ubuntu:18.04-x-riscv64 : Ok riscv64-linux-gnu-gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0 69 ubuntu:18.04-x-s390 : Ok s390x-linux-gnu-gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0 70 ubuntu:18.04-x-sh4 : Ok sh4-linux-gnu-gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0 71 ubuntu:18.04-x-sparc64 : Ok sparc64-linux-gnu-gcc (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0 72 ubuntu:18.10 : Ok gcc (Ubuntu 8.3.0-6ubuntu1~18.10.1) 8.3.0, clang version 7.0.0-3 (tags/RELEASE_700/final) 73 ubuntu:19.04 : Ok gcc (Ubuntu 8.3.0-6ubuntu1) 8.3.0, clang version 8.0.0-3 (tags/RELEASE_800/final) 74 ubuntu:19.04-x-alpha : Ok alpha-linux-gnu-gcc (Ubuntu 8.3.0-6ubuntu1) 8.3.0 75 ubuntu:19.04-x-arm64 : Ok aarch64-linux-gnu-gcc (Ubuntu/Linaro 8.3.0-6ubuntu1) 8.3.0 76 ubuntu:19.04-x-hppa : Ok hppa-linux-gnu-gcc (Ubuntu 8.3.0-6ubuntu1) 8.3.0 77 ubuntu:19.10 : Ok gcc (Ubuntu 9.2.1-8ubuntu1) 9.2.1 20190909, clang version 9.0.0-+rc5-1~exp1 (tags/RELEASE_900/rc5) #
# uname -a
Linux quaco 5.2.17-200.fc30.x86_64 #1 SMP Mon Sep 23 13:42:32 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux
# git log --oneline -1
d6840d87b2d1 perf parser: Remove needless include directives
# perf version --build-options
perf version 5.3.gd6840d87b2d1
dwarf: [ on ] # HAVE_DWARF_SUPPORT
dwarf_getlocations: [ on ] # HAVE_DWARF_GETLOCATIONS_SUPPORT
glibc: [ on ] # HAVE_GLIBC_SUPPORT
gtk2: [ on ] # HAVE_GTK2_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: DSO data read : Ok
11: DSO data cache : Ok
12: DSO data reopen : Ok
13: Roundtrip evsel->name : Ok
14: Parse sched tracepoints fields : Ok
15: syscalls:sys_enter_openat event fields : Ok
16: Setup struct perf_event_attr : Ok
17: Match and link multiple hists : Ok
18: 'import perf' in python : Ok
19: Breakpoint overflow signal handler : Ok
20: Breakpoint overflow sampling : Ok
21: Breakpoint accounting : Ok
22: Watchpoint :
22.1: Read Only Watchpoint : Skip
22.2: Write Only Watchpoint : Ok
22.3: Read / Write Watchpoint : Ok
22.4: Modify Watchpoint : Ok
23: Number of exit events of a simple workload : Ok
24: Software clock events period values : Ok
25: Object code reading : Ok
26: Sample parsing : Ok
27: Use a dummy software event to keep tracking : Ok
28: Parse with no sample_id_all bit set : Ok
29: Filter hist entries : Ok
30: Lookup mmap thread : Ok
31: Share thread mg : Ok
32: Sort output of hist entries : Ok
33: Cumulate child hist entries : Ok
34: Track with sched_switch : Ok
35: Filter fds with revents mask in a fdarray : Ok
36: Add fd to a fdarray, making it autogrow : Ok
37: kmod_path__parse : Ok
38: Thread map : Ok
39: LLVM search and compile :
39.1: Basic BPF llvm compile : Ok
39.2: kbuild searching : Ok
39.3: Compile source for BPF prologue generation : Ok
39.4: Compile source for BPF relocation : Ok
40: Session topology : Ok
41: BPF filter :
41.1: Basic BPF filtering : Ok
41.2: BPF pinning : Ok
41.3: BPF prologue generation : Ok
41.4: BPF relocation checker : Ok
42: Synthesize thread map : Ok
43: Remove thread map : Ok
44: Synthesize cpu map : Ok
45: Synthesize stat config : Ok
46: Synthesize stat : Ok
47: Synthesize stat round : Ok
48: Synthesize attr update : Ok
49: Event times : Ok
50: Read backward ring buffer : Ok
51: Print cpu map : Ok
52: Probe SDT events : Ok
53: is_printable_array : Ok
54: Print bitmap : Ok
55: perf hooks : Ok
56: builtin clang support : Skip (not compiled in)
57: unit_number__scnprintf : Ok
58: mem2node : Ok
59: time utils : Ok
60: map_groups__merge_in : Ok
61: x86 rdpmc : Ok
62: Convert perf time to TSC : Ok
63: DWARF unwind : Ok
64: x86 instruction decoder - new instructions : Ok
65: Intel PT packet decoder : Ok
66: x86 bp modify : Ok
67: probe libc's inet_pton & backtrace it with ping : Ok
68: Use vfs_getname probe to get syscall args filenames : Ok
69: Add vfs_getname probe to get syscall args filenames : Ok
70: Check open filename arg using perf trace + vfs_getname: Ok
71: Zstd perf.data compression/decompression : Ok
# $ make -C tools/perf build-test | tee /wb/build-test
make: Entering directory '/home/acme/git/perf/tools/perf'
- tarpkg: ./tests/perf-targz-src-pkg .
make_no_ui_O: make NO_NEWT=1 NO_SLANG=1 NO_GTK2=1
make_no_scripts_O: make NO_LIBPYTHON=1 NO_LIBPERL=1
make_no_libdw_dwarf_unwind_O: make NO_LIBDW_DWARF_UNWIND=1
make_no_demangle_O: make NO_DEMANGLE=1
make_no_libbpf_O: make NO_LIBBPF=1
make_install_O: make install
make_cscope_O: make cscope
make_no_auxtrace_O: make NO_AUXTRACE=1
make_no_libelf_O: make NO_LIBELF=1
make_perf_o_O: make perf.o
make_no_libpython_O: make NO_LIBPYTHON=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
make_no_slang_O: make NO_SLANG=1
make_no_gtk2_O: make NO_GTK2=1
make_tags_O: make tags
make_pure_O: make
make_util_map_o_O: make util/map.o
make_help_O: make help
make_no_libnuma_O: make NO_LIBNUMA=1
make_install_prefix_O: make install prefix=/tmp/krava
make_with_babeltrace_O: make LIBBABELTRACE=1
make_clean_all_O: make clean all
make_install_prefix_slash_O: make install prefix=/tmp/krava/
make_no_newt_O: make NO_NEWT=1
make_static_O: make LDFLAGS=-static NO_PERF_READ_VDSO32=1 NO_PERF_READ_VDSOX32=1 NO_JVMTI=1
make_with_clangllvm_O: make LIBCLANGLLVM=1
make_util_pmu_bison_o_O: make util/pmu-bison.o
make_doc_O: make doc
make_no_libperl_O: make NO_LIBPERL=1
make_install_bin_O: make install-bin
make_no_libunwind_O: make NO_LIBUNWIND=1
make_no_backtrace_O: make NO_BACKTRACE=1
make_no_libbionic_O: make NO_LIBBIONIC=1
make_debug_O: make DEBUG=1
make_no_libaudit_O: make NO_LIBAUDIT=1
OK
make: Leaving directory '/home/acme/git/perf/tools/perf'
$ Arnaldo Carvalho de Melo
maintainerPATCH 1/66 perf record: Fix priv level with branch sampling for paranoid=2
From: Stephane Eranian <[email protected]>
Now that the default perf_events paranoid level is set to 2, a regular user cannot monitor kernel level activity anymore. As such, with the following cmdline:
$ perf record -e cycles date
The perf tool first tries cycles:uk but then falls back to cycles:u as can be seen in the perf report --header-only output:
cmdline : /export/hda3/tmp/perf.tip record -e cycles ls event : name = cycles:u, , id = { 436186, ... }
This is okay as long as there is way to learn the priv level was changed internally by the tool.
But consider a similar example:
$ perf record -b -e cycles date Error: You may not have permission to collect stats.
Consider tweaking /proc/sys/kernel/perf_event_paranoid, which controls use of the performance events system by unprivileged users (without CAP_SYS_ADMIN).
...
Why is that treated differently given that the branch sampling inherits the priv level of the first event in this case, i.e., cycles:u? It turns out that the branch sampling code is more picky and also checks exclude_hv.
In the fallback path, perf record is setting exclude_kernel = 1, but it does not change exclude_hv. This does not seem to match the restriction imposed by paranoid = 2.
This patch fixes the problem by forcing exclude_hv = 1 in the fallback for paranoid=2. With this in place:
$ perf record -b -e cycles date
cmdline : /export/hda3/tmp/perf.tip record -b -e cycles ls
event : name = cycles:u, , id = { 436847, ... }And the command succeeds as expected.
V2 fix a white space.
Committer testing:
After aplying the patch we get:
[acme@quaco ~]$ perf record -b -e cycles date
WARNING: Kernel address maps (/proc/{kallsyms,modules}) are restricted, check /proc/sys/kernel/kptr_restrict and /proc/sys/kernel/perf_event_paranoid.
Samples in kernel functions may not be resolved if a suitable vmlinux file is not found in the buildid cache or in the vmlinux path.
Samples in kernel modules won't be resolved at all.
If some relocation was applied (e.g. kexec) symbols may be misresolved even with a suitable vmlinux or kallsyms file.
Mon 23 Sep 2019 11:00:59 AM -03
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.005 MB perf.data (14 samples) ]
[acme@quaco ~]$ perf evlist -v
cycles:u: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|PERIOD|BRANCH_STACK, read_format: ID, disabled: 1, inherit: 1, exclude_kernel: 1, exclude_hv: 1, mmap: 1, comm: 1, freq: 1, enable_on_exec: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1, branch_sample_type: ANY [acme@quaco ~]$
That warning about restricted kernel maps will be suppressed in a follow up patch, as perf_event_attr.exclude_kernel is set, i.e. no samples for the kernel will be taken and thus no need for those maps.
--- tools/perf/util/evsel.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
tools/perf/util/evsel.c +4 -2
@@ -2535,9 +2535,11 @@ bool perf_evsel__fallback(struct evsel *evsel, int err,if (evsel->name)free (evsel->name); evsel->name = new_name; -scnprintf (msg, msgsize, -"kernel.perf_event_paranoid=%d, trying to fall back to excluding kernel samples" , paranoid); +scnprintf (msg, msgsize,"kernel.perf_event_paranoid=%d, trying " +"to fall back to excluding kernel and hypervisor " +" samples" , paranoid); evsel->core.attr.exclude_kernel =1 ; + evsel->core.attr.exclude_hv =1 ;return true; }
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 2/66 perf record: Move restricted maps check to after a possible fallback to not collect kernel samples
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
Before:
[acme@quaco ~]$ perf record -b -e cycles date
WARNING: Kernel address maps (/proc/{kallsyms,modules}) are restricted, check /proc/sys/kernel/kptr_restrict and /proc/sys/kernel/perf_event_paranoid.
Samples in kernel functions may not be resolved if a suitable vmlinux file is not found in the buildid cache or in the vmlinux path.
Samples in kernel modules won't be resolved at all.
If some relocation was applied (e.g. kexec) symbols may be misresolved even with a suitable vmlinux or kallsyms file.
Mon 23 Sep 2019 11:00:59 AM -03
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.005 MB perf.data (14 samples) ] [acme@quaco ~]$
But we did a fallback and exclude_kernel was set, so no need for resolving kernel symbols:
$ perf evlist -v
cycles:u: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|PERIOD|BRANCH_STACK, read_format: ID, disabled: 1, inherit: 1, exclude_kernel: 1, exclude_hv: 1, mmap: 1, comm: 1, freq: 1, enable_on_exec: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1, branch_sample_type: ANY
$After:
[acme@quaco ~]$ perf record -b -e cycles date
Mon 23 Sep 2019 11:07:18 AM -03
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.007 MB perf.data (16 samples) ]
[acme@quaco ~]$ perf evlist -v
cycles:u: size: 112, { sample_period, sample_freq }: 4000, sample_type: IP|TID|TIME|PERIOD|BRANCH_STACK, read_format: ID, disabled: 1, inherit: 1, exclude_kernel: 1, exclude_hv: 1, mmap: 1, comm: 1, freq: 1, enable_on_exec: 1, task: 1, sample_id_all: 1, exclude_guest: 1, mmap2: 1, comm_exec: 1, ksymbol: 1, bpf_event: 1, branch_sample_type: ANY [acme@quaco ~]$
No needless warning is emitted.
--- tools/perf/builtin-record.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-)
tools/perf/builtin-record.c +11 -10
@@ -788,6 +788,17 @@ static int record__open(struct record *rec) pos->supported = true; } +if (symbol_conf.kptr_restrict && !perf_evlist__exclude_kernel (evlist)) { +pr_warning ( +"WARNING: Kernel address maps (/proc/{kallsyms,modules}) are restricted,\n" +"check /proc/sys/kernel/kptr_restrict and /proc/sys/kernel/perf_event_paranoid.\n\n" +"Samples in kernel functions may not be resolved if a suitable vmlinux\n" +"file is not found in the buildid cache or in the vmlinux path.\n\n" +"Samples in kernel modules won't be resolved at all.\n\n" +"If some relocation was applied (e.g. kexec) symbols may be misresolved\n" +"even with a suitable vmlinux or kallsyms file.\n\n" ); + } +if (perf_evlist__apply_filters (evlist, &pos)) { pr_err("failed to set filter \"%s\" on event %s with %d (%s)\n", pos->filter, perf_evsel__name(pos), errno,
@@ -2364,16 +2375,6 @@ int cmd_record(int argc, const char **argv) err = -ENOMEM ; -if (symbol_conf.kptr_restrict && !perf_evlist__exclude_kernel (rec->evlist)) -pr_warning ( -"WARNING: Kernel address maps (/proc/{kallsyms,modules}) are restricted,\n" -"check /proc/sys/kernel/kptr_restrict and /proc/sys/kernel/perf_event_paranoid.\n\n" -"Samples in kernel functions may not be resolved if a suitable vmlinux\n" -"file is not found in the buildid cache or in the vmlinux path.\n\n" -"Samples in kernel modules won't be resolved at all.\n\n" -"If some relocation was applied (e.g. kexec) symbols may be misresolved\n" -"even with a suitable vmlinux or kallsyms file.\n\n" ); -if (rec->no_buildid_cache || rec->no_buildid) {disable_buildid_cache (); }else if (rec->switch_output.enabled) {
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 3/66 perf vendor events amd: Add L3 cache events for Family 17h
In reply to Arnaldo Carvalho de MeloFrom: Kim Phillips <[email protected]>
Allow users to symbolically specify L3 events for Family 17h processors using the existing AMD Uncore driver.
Source of events descriptions are from section 2.1.15.4.1 "L3 Cache PMC Events" of the latest Family 17h PPR, available here:
https://www.amd.com/system/files/TechDocs/55570-B1_PUB.zip
Opnly BriefDescriptions added, since they show with and without the -v and --details flags.
Tested with:
# perf stat -e l3_request_g1.caching_l3_cache_accesses,amd_l3/event=0x01,umask=0x80/,l3_comb_clstr_state.request_miss,amd_l3/event=0x06,umask=0x01/ perf bench mem memcpy -s 4mb -l 100 -f default
...
7,006,831 l3_request_g1.caching_l3_cache_accesses
7,006,830 amd_l3/event=0x01,umask=0x80/
366,530 l3_comb_clstr_state.request_miss
366,568 amd_l3/event=0x06,umask=0x01/--- .../pmu-events/arch/x86/amdfam17h/cache.json | 42 +++++++++++++++++++ tools/perf/pmu-events/jevents.c | 1 + 2 files changed, 43 insertions(+)
tools/perf/pmu-events/arch/x86/amdfam17h/cache.json +42 -0
@@ -283,5 +283,47 @@ "BriefDescription": "Total cycles spent with one or more fill requests in flight from L2.", "PublicDescription": "Total cycles spent with one or more fill requests in flight from L2.", "UMask": "0x1" + }, + { + "EventName": "l3_request_g1.caching_l3_cache_accesses", + "EventCode": "0x01", + "BriefDescription": "Caching: L3 cache accesses", + "UMask": "0x80", + "Unit": "L3PMC" + }, + { + "EventName": "l3_lookup_state.all_l3_req_typs", + "EventCode": "0x04", + "BriefDescription": "All L3 Request Types", + "UMask": "0xff", + "Unit": "L3PMC" + }, + { + "EventName": "l3_comb_clstr_state.other_l3_miss_typs", + "EventCode": "0x06", + "BriefDescription": "Other L3 Miss Request Types", + "UMask": "0xfe", + "Unit": "L3PMC" + }, + { + "EventName": "l3_comb_clstr_state.request_miss", + "EventCode": "0x06", + "BriefDescription": "L3 cache misses", + "UMask": "0x01", + "Unit": "L3PMC" + }, + { + "EventName": "xi_sys_fill_latency", + "EventCode": "0x90", + "BriefDescription": "L3 Cache Miss Latency. Total cycles for all transactions divided by 16. Ignores SliceMask and ThreadMask.", + "UMask": "0x00", + "Unit": "L3PMC" + }, + { + "EventName": "xi_ccx_sdp_req1.all_l3_miss_req_typs", + "EventCode": "0x9a", + "BriefDescription": "All L3 Miss Request Types. Ignores SliceMask and ThreadMask.", + "UMask": "0x3f", + "Unit": "L3PMC" } ]
tools/perf/pmu-events/jevents.c +1 -0
@@ -239,6 +239,7 @@ static struct map { { "hisi_sccl,ddrc", "hisi_sccl,ddrc" }, { "hisi_sccl,hha", "hisi_sccl,hha" }, { "hisi_sccl,l3c", "hisi_sccl,l3c" }, + { "L3PMC", "amd_l3" }, {} };
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 4/66 perf vendor events amd: Remove redundant '['
In reply to Arnaldo Carvalho de MeloFrom: Kim Phillips <[email protected]>
Remove the redundant '['.
'perf list' output before:
ex_ret_brn
[[Retired Branch Instructions]
'perf list' output after:
ex_ret_brn
[Retired Branch Instructions]
--- tools/perf/pmu-events/arch/x86/amdfam17h/core.json | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/pmu-events/arch/x86/amdfam17h/core.json +1 -1
@@ -13,7 +13,7 @@ { "EventName": "ex_ret_brn", "EventCode": "0xc2", - "BriefDescription": "[Retired Branch Instructions.", + "BriefDescription": "Retired Branch Instructions.", "PublicDescription": "The number of branch instructions retired. This includes all types of architectural control flow changes, including exceptions and interrupts." }, {
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 5/66 perf vendor events: Minor fixes to the README
In reply to Arnaldo Carvalho de MeloFrom: Kim Phillips <[email protected]>
Some grammatical fixes, and updates to some path references that have since changed.
--- tools/perf/pmu-events/README | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-)
tools/perf/pmu-events/README +11 -11
@@ -30,9 +30,9 @@ the topic. Eg: "Floating-point.json". All the topic JSON files for a CPU model/family should be in a separate sub directory. Thus for the Silvermont X86 CPU: - $ ls tools/perf/pmu-events/arch/x86/Silvermont_core - Cache.json Memory.json Virtual-Memory.json - Frontend.json Pipeline.json + $ ls tools/perf/pmu-events/arch/x86/silvermont + cache.json memory.json virtual-memory.json + frontend.json pipeline.json The JSONs folder for a CPU model/family may be placed in the root arch folder, or may be placed in a vendor sub-folder under the arch folder
@@ -94,7 +94,7 @@ users to specify events by their name: where 'pm_1plus_ppc_cmpl' is a Power8 PMU event. -However some errors in processing may cause the perf build to fail. +However some errors in processing may cause the alias build to fail. Mapfile format ===============
@@ -119,7 +119,7 @@ where: Header line The header line is the first line in the file, which is - always _IGNORED_. It can empty. + always _IGNORED_. It can be empty. CPUID: CPUID is an arch-specific char string, that can be used
@@ -138,15 +138,15 @@ where: files, relative to the directory containing the mapfile.csv Type: - indicates whether the events or "core" or "uncore" events. + indicates whether the events are "core" or "uncore" events. Eg: - $ grep Silvermont tools/perf/pmu-events/arch/x86/mapfile.csv - GenuineIntel-6-37,V13,Silvermont_core,core - GenuineIntel-6-4D,V13,Silvermont_core,core - GenuineIntel-6-4C,V13,Silvermont_core,core + $ grep silvermont tools/perf/pmu-events/arch/x86/mapfile.csv + GenuineIntel-6-37,v13,silvermont,core + GenuineIntel-6-4D,v13,silvermont,core + GenuineIntel-6-4C,v13,silvermont,core i.e the three CPU models use the JSON files (i.e PMU events) listed - in the directory 'tools/perf/pmu-events/arch/x86/Silvermont_core'. + in the directory 'tools/perf/pmu-events/arch/x86/silvermont'.
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 6/66 perf list: Allow plurals for metric, metricgroup
In reply to Arnaldo Carvalho de MeloFrom: Kim Phillips <[email protected]>
Enhance usability by allowing the same plurality used in the output title, for the command line parameter.
BEFORE, perf deceitfully acts as if there are no metrics to be had:
$ perf list metrics
List of pre-defined events (to be used in -e):
Metric Groups:
$
But singular 'metric' shows a list of metrics:
$ perf list metric
List of pre-defined events (to be used in -e):
Metrics:
IPC
[Instructions Per Cycle (per logical thread)]
UPI
[Uops Per Instruction]
AFTER, when asking for 'metrics', we actually see the metrics get listed:
$ perf list metrics
List of pre-defined events (to be used in -e):
Metrics:
IPC
[Instructions Per Cycle (per logical thread)]
UPI
[Uops Per Instruction]
--- tools/perf/builtin-list.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/builtin-list.c +2 -2
@@ -81,9 +81,9 @@ int cmd_list(int argc, const char **argv) long_desc_flag, details_flag); else if (strcmp(argv[i], "sdt") == 0) print_sdt_events(NULL, NULL, raw_dump); - else if (strcmp(argv[i], "metric") == 0) + else if (strcmp(argv[i], "metric") == 0 || strcmp(argv[i], "metrics") == 0) metricgroup__print(true, false,NULL , raw_dump, details_flag); - else if (strcmp(argv[i], "metricgroup") == 0) + else if (strcmp(argv[i], "metricgroup") == 0 || strcmp(argv[i], "metricgroups") == 0) metricgroup__print(false, true,NULL , raw_dump, details_flag); else if ((sep = strchr(argv[i], ':')) != NULL) {int sep_idx;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 7/66 libtraceevent: Round up in tep_print_event() time precision
In reply to Arnaldo Carvalho de MeloFrom: "Steven Rostedt (VMware)" <[email protected]>
When testing the output of the old trace-cmd compared to the one that uses the updated tep_print_event() logic, it was different in that the time stamp precision in the old format would round up to the nearest precision, where as the new logic truncates. Bring back the old method of rounding up.
--- tools/lib/traceevent/event-parse.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
tools/lib/traceevent/event-parse.c +3 -1
@@ -5527,8 +5527,10 @@ static void print_event_time(struct tep_handle *tep, struct trace_seq *s,if (divstr &&isdigit (*(divstr +1 ))) div =atoi (divstr +1 ); time = record->ts; -if (div) +if (div) { + time += div /2 ; time /= div; + } pr = prec;while (pr--) p10 *=10 ;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 8/66 libtraceevent: Man pages for libtraceevent event print related API
In reply to Arnaldo Carvalho de MeloFrom: Tzvetomir Stoyanov <[email protected]>
Added new man page, describing tep_print_event() libtraceevent API.
--- .../libtraceevent-event_print.txt | 130 ++++++++++++++++++ 1 file changed, 130 insertions(+) create mode 100644 tools/lib/traceevent/Documentation/libtraceevent-event_print.txt
tools/lib/traceevent/Documentation/libtraceevent-event_print.txt new file +130 -0
@@ -0,0 +1,130 @@ +libtraceevent(3) +================ + +NAME +---- +tep_print_event - Writes event information into a trace sequence. + +SYNOPSIS +-------- +[verse] +-- +*#include <event-parse.h>* +*#include <trace-seq.h>* + +void *tep_print_event*(struct tep_handle pass:[*]_tep_, struct trace_seqpass:[*]_s_, struct tep_record pass:[*]_record_, const char pass:[*]_fmt_, _..._) +-- + +DESCRIPTION +----------- + +The _tep_print_event()_ function parses the event information of the given +_record_ and writes it into the trace sequence _s_, according to the format +string _fmt_. The desired information is specified after the format string. +The _fmt_ is printf-like format string, following arguments are supported: +[verse] +-- + TEP_PRINT_PID, "%d" - PID of the event. + TEP_PRINT_CPU, "%d" - Event CPU. + TEP_PRINT_COMM, "%s" - Event command string. + TEP_PRINT_NAME, "%s" - Event name. + TEP_PRINT_LATENCY, "%s" - Latency of the event. It prints 4 or more + fields - interrupt state, scheduling state, + current context, and preemption count. + Field 1 is the interrupt enabled state: + d : Interrupts are disabled + . : Interrupts are enabled + X : The architecture does not support this + information + Field 2 is the "need resched" state. + N : The task is set to call the scheduler when + possible, as another higher priority task + may need to be scheduled in. + . : The task is not set to call the scheduler. + Field 3 is the context state. + . : Normal context + s : Soft interrupt context + h : Hard interrupt context + H : Hard interrupt context which triggered + during soft interrupt context. + z : NMI context + Z : NMI context which triggered during hard + interrupt context + Field 4 is the preemption count. + . : The preempt count is zero. + On preemptible kernels (where the task can be scheduled + out in arbitrary locations while in kernel context), the + preempt count, when non zero, will prevent the kernel + from scheduling out the current task. The preempt count + number is displayed when it is not zero. + Depending on the kernel, it may show other fields + (lock depth, or migration disabled, which are unique to + specialized kernels). + TEP_PRINT_TIME, %d - event time stamp. A divisor and precision can be + specified as part of this format string: + "%precision.divisord". Example: + "%3.1000d" - divide the time by 1000 and print the first + 3 digits before the dot. Thus, the time stamp + "123456000" will be printed as "123.456" + TEP_PRINT_INFO, "%s" - event information. + TEP_PRINT_INFO_RAW, "%s" - event information, in raw format. + +-- +EXAMPLE +------- +[source,c] +-- +#include <event-parse.h> +#include <trace-seq.h> +... +struct trace_seq seq; +trace_seq_init(&seq); +struct tep_handle *tep = tep_alloc(); +... +void print_my_event(struct tep_record *record) +{ + trace_seq_reset(&seq); + tep_print_event(tep, s, record, "%16s-%-5d [%03d] %s %6.1000d %s %s", + TEP_PRINT_COMM, TEP_PRINT_PID, TEP_PRINT_CPU, + TEP_PRINT_LATENCY, TEP_PRINT_TIME, TEP_PRINT_NAME, + TEP_PRINT_INFO); +} +... +-- + +FILES +----- +[verse] +-- +*event-parse.h* + Header file to include in order to have access to the library APIs. +*trace-seq.h* + Header file to include in order to have access to trace sequences related APIs. + Trace sequences are used to allow a function to call several other functions + to create a string of data to use. +*-ltraceevent* + Linker switch to add when building a program that uses the library. +-- + +SEE ALSO +-------- +_libtraceevent(3)_, _trace-cmd(1)_ + +AUTHOR +------ +[verse] +-- +*Steven Rostedt* <[email protected]>, author of *libtraceevent*. +*Tzvetomir Stoyanov* <[email protected]>, author of this man page. +-- +REPORTING BUGS +-------------- +Report bugs to <[email protected]> + +LICENSE +------- +libtraceevent is Free Software licensed under the GNU LGPL 2.1 + +RESOURCES +--------- +https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 9/66 libtraceevent: Man pages fix, rename tep_ref_get() to tep_get_ref()
In reply to Arnaldo Carvalho de MeloFrom: "Tzvetomir Stoyanov (VMware)" <[email protected]>
The tep_ref_get() was renamed to tep_get_ref(), to be more consistent with the other tep_ref_* APIs. However, in the man pages the API is still with the old name. The documentation is fixed to reflect the actual name of the API.
--- .../lib/traceevent/Documentation/libtraceevent-handle.txt | 8 ++++---- tools/lib/traceevent/Documentation/libtraceevent.txt | 2 +- 2 files changed, 5 insertions(+), 5 deletions(-)
tools/lib/traceevent/Documentation/libtraceevent-handle.txt +4 -4
@@ -3,7 +3,7 @@ libtraceevent(3) NAME ---- -tep_alloc, tep_free,tep_ref, tep_unref,tep_ref_get - Create, destroy, manage +tep_alloc, tep_free,tep_ref, tep_unref,tep_get_ref - Create, destroy, manage references of trace event parser context. SYNOPSIS
@@ -16,7 +16,7 @@ struct tep_handle pass:[*]*tep_alloc*(void); void *tep_free*(struct tep_handle pass:[*]_tep_); void *tep_ref*(struct tep_handle pass:[*]_tep_); void *tep_unref*(struct tep_handle pass:[*]_tep_); -int *tep_ref_get*(struct tep_handle pass:[*]_tep_); +int *tep_get_ref*(struct tep_handle pass:[*]_tep_); -- DESCRIPTION
@@ -57,9 +57,9 @@ EXAMPLE ... struct tep_handle *tep = tep_alloc(); ... -int ref = tep_ref_get(tep); +int ref = tep_get_ref(tep); tep_ref(tep); -if ( (ref+1) != tep_ref_get(tep)) { +if ( (ref+1) != tep_get_ref(tep)) { /* Something wrong happened, the counter is not incremented by 1 */ } tep_unref(tep);
tools/lib/traceevent/Documentation/libtraceevent.txt +1 -1
@@ -16,7 +16,7 @@ Management of tep handler data structure and access of its members: void *tep_free*(struct tep_handle pass:[*]_tep_); void *tep_ref*(struct tep_handle pass:[*]_tep_); void *tep_unref*(struct tep_handle pass:[*]_tep_); - int *tep_ref_get*(struct tep_handle pass:[*]_tep_); + int *tep_get_ref*(struct tep_handle pass:[*]_tep_); void *tep_set_flag*(struct tep_handle pass:[*]_tep_, enum tep_flag _flag_); void *tep_clear_flag*(struct tep_handle pass:[*]_tep_, enum tep_flag _flag_); bool *tep_test_flag*(struct tep_handle pass:[*]_tep_, enum tep_flag _flags_);
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 10/66 libtraceevent: Man pages fix, changes in event printing APIs
In reply to Arnaldo Carvalho de MeloFrom: "Tzvetomir Stoyanov (VMware)" <[email protected]>
APIs for printing various trace event information were redesigned to be more simple. However, the main libtraceevent man page was not updated with those changes. The documentation is updated to describe the new event print API.
--- .../lib/traceevent/Documentation/libtraceevent.txt | 13 +------------ 1 file changed, 1 insertion(+), 12 deletions(-)
tools/lib/traceevent/Documentation/libtraceevent.txt +1 -12
@@ -26,15 +26,12 @@ Management of tep handler data structure and access of its members: void *tep_set_long_size*(struct tep_handle pass:[*]_tep_, int _long_size_); int *tep_get_page_size*(struct tep_handle pass:[*]_tep_); void *tep_set_page_size*(struct tep_handle pass:[*]_tep_, int _page_size_); - bool *tep_is_latency_format*(struct tep_handle pass:[*]_tep_); - void *tep_set_latency_format*(struct tep_handle pass:[*]_tep_, int _lat_); int *tep_get_header_page_size*(struct tep_handle pass:[*]_tep_); int *tep_get_header_timestamp_size*(struct tep_handle pass:[*]_tep_); bool *tep_is_old_format*(struct tep_handle pass:[*]_tep_); int *tep_strerror*(struct tep_handle pass:[*]_tep_, enum tep_errno _errnum_, char pass:[*]_buf_, size_t _buflen_); Register / unregister APIs: - int *tep_register_trace_clock*(struct tep_handle pass:[*]_tep_, const char pass:[*]_trace_clock_); int *tep_register_function*(struct tep_handle pass:[*]_tep_, char pass:[*]_name_, unsigned long long _addr_, char pass:[*]_mod_); int *tep_register_event_handler*(struct tep_handle pass:[*]_tep_, int _id_, const char pass:[*]_sys_name_, const char pass:[*]_event_name_, tep_event_handler_func _func_, void pass:[*]_context_); int *tep_unregister_event_handler*(struct tep_handle pass:[*]tep, int id, const char pass:[*]sys_name, const char pass:[*]event_name, tep_event_handler_func func, void pass:[*]_context_);
@@ -57,14 +54,7 @@ Event related APIs: int *tep_get_events_count*(struct tep_handle pass:[*]_tep_); struct tep_event pass:[*]pass:[*]*tep_list_events*(struct tep_handle pass:[*]_tep_, enum tep_event_sort_type _sort_type_); struct tep_event pass:[*]pass:[*]*tep_list_events_copy*(struct tep_handle pass:[*]_tep_, enum tep_event_sort_type _sort_type_); - -Event printing: - void *tep_print_event*(struct tep_handle pass:[*]_tep_, struct trace_seq pass:[*]_s_, struct tep_record pass:[*]_record_, bool _use_trace_clock_); - void *tep_print_event_data*(struct tep_handle pass:[*]_tep_, struct trace_seq pass:[*]_s_, struct tep_event pass:[*]_event_, struct tep_record pass:[*]_record_); - void *tep_event_info*(struct trace_seq pass:[*]_s_, struct tep_event pass:[*]_event_, struct tep_record pass:[*]_record_); - void *tep_print_event_task*(struct tep_handle pass:[*]_tep_, struct trace_seq pass:[*]_s_, struct tep_event pass:[*]_event_, struct tep_record pass:[*]_record_); - void *tep_print_event_time*(struct tep_handle pass:[*]_tep_, struct trace_seq pass:[*]_s_, struct tep_event pass:[*]_event_, struct tep_record pass:[*]record, bool _use_trace_clock_); - void *tep_set_print_raw*(struct tep_handle pass:[*]_tep_, int _print_raw_); + void *tep_print_event*(struct tep_handle pass:[*]_tep_, struct trace_seq pass:[*]_s_, struct tep_record pass:[*]_record_, const char pass:[*]_fmt_, _..._); Event finding: struct tep_event pass:[*]*tep_find_event*(struct tep_handle pass:[*]_tep_, int _id_);
@@ -116,7 +106,6 @@ Filter management: int *tep_filter_compare*(struct tep_event_filter pass:[*]_filter1_, struct tep_event_filter pass:[*]_filter2_); Parsing various data from the records: - void *tep_data_latency_format*(struct tep_handle pass:[*]_tep_, struct trace_seq pass:[*]_s_, struct tep_record pass:[*]_record_); int *tep_data_type*(struct tep_handle pass:[*]_tep_, struct tep_record pass:[*]_rec_); int *tep_data_pid*(struct tep_handle pass:[*]_tep_, struct tep_record pass:[*]_rec_); int *tep_data_preempt_count*(struct tep_handle pass:[*]_tep_, struct tep_record pass:[*]_rec_);
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 11/66 libtraceevent: Add tep_get_event() in event-parse.h
In reply to Arnaldo Carvalho de MeloFrom: "Tzvetomir Stoyanov (VMware)" <[email protected]>
The tep_get_event() function is an official libtracevent API, described in the library man pages. However, it cannot be used by the library users because it is not declared in the event-parse.h file, where all libtracevent APIs are. The function declaration is added in event-parse.h file.
--- tools/lib/traceevent/event-parse.h | 2 ++ 1 file changed, 2 insertions(+)
tools/lib/traceevent/event-parse.h +2 -0
@@ -441,6 +441,8 @@ int tep_register_print_string(struct tep_handle *tep, const char *fmt, unsigned long long addr);bool tep_is_pid_registered (struct tep_handle *tep,int pid); +struct tep_event *tep_get_event (struct tep_handle *tep,int index); +#define TEP_PRINT_INFO "INFO"#define TEP_PRINT_INFO_RAW "INFO_RAW"#define TEP_PRINT_COMM "COMM"
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 12/66 libtraceevent: Move traceevent plugins in its own subdirectory
In reply to Arnaldo Carvalho de MeloFrom: "Tzvetomir Stoyanov (VMware)" <[email protected]>
All traceevent plugins code is moved to tools/lib/traceevent/plugins subdirectory. It makes traceevent implementation in trace-cmd and in kernel tree consistent. There is no changes in the way libtraceevent and plugins are compiled and installed.
Committer notes:
Applied fixup provided by Steven, fixing the tools/perf/Makefile.perf target for the plugin dynamic list file. Problem noticed when cross building to aarch64 from a Ubuntu 19.04 container.
---
tools/lib/traceevent/Build | 11 -
tools/lib/traceevent/Makefile | 94 ++------
tools/lib/traceevent/plugins/Build | 10 +
tools/lib/traceevent/plugins/Makefile | 222 ++++++++++++++++++
.../{ => plugins}/plugin_cfg80211.c | 0
.../{ => plugins}/plugin_function.c | 0
.../traceevent/{ => plugins}/plugin_hrtimer.c | 0
.../traceevent/{ => plugins}/plugin_jbd2.c | 0
.../traceevent/{ => plugins}/plugin_kmem.c | 0
.../lib/traceevent/{ => plugins}/plugin_kvm.c | 0
.../{ => plugins}/plugin_mac80211.c | 0
.../{ => plugins}/plugin_sched_switch.c | 0
.../traceevent/{ => plugins}/plugin_scsi.c | 0
.../lib/traceevent/{ => plugins}/plugin_xen.c | 0
tools/perf/Makefile.perf | 4 +-
15 files changed, 250 insertions(+), 91 deletions(-)
create mode 100644 tools/lib/traceevent/plugins/Build
create mode 100644 tools/lib/traceevent/plugins/Makefile
rename tools/lib/traceevent/{ => plugins}/plugin_cfg80211.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_function.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_hrtimer.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_jbd2.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_kmem.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_kvm.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_mac80211.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_sched_switch.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_scsi.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_xen.c (100%)tools/lib/traceevent/Build +0 -11
@@ -6,14 +6,3 @@ libtraceevent-y += parse-utils.o libtraceevent-y += kbuffer-parse.o libtraceevent-y += tep_strerror.o libtraceevent-y += event-parse-api.o - -plugin_jbd2-y += plugin_jbd2.o -plugin_hrtimer-y += plugin_hrtimer.o -plugin_kmem-y += plugin_kmem.o -plugin_kvm-y += plugin_kvm.o -plugin_mac80211-y += plugin_mac80211.o -plugin_sched_switch-y += plugin_sched_switch.o -plugin_function-y += plugin_function.o -plugin_xen-y += plugin_xen.o -plugin_scsi-y += plugin_scsi.o -plugin_cfg80211-y += plugin_cfg80211.o
tools/lib/traceevent/Makefile +16 -78
148 lines not on this page · raw message
tools/lib/traceevent/plugins/Build new file +10 -0
@@ -0,0 +1,10 @@ +plugin_jbd2-y += plugin_jbd2.o +plugin_hrtimer-y += plugin_hrtimer.o +plugin_kmem-y += plugin_kmem.o +plugin_kvm-y += plugin_kvm.o +plugin_mac80211-y += plugin_mac80211.o +plugin_sched_switch-y += plugin_sched_switch.o +plugin_function-y += plugin_function.o +plugin_xen-y += plugin_xen.o +plugin_scsi-y += plugin_scsi.o +plugin_cfg80211-y += plugin_cfg80211.o
tools/lib/traceevent/plugins/Makefile new file +222 -0
222 lines not on this page · raw message
tools/lib/traceevent/plugins/plugin_cfg80211.c renamed from tools/lib/traceevent/plugin_cfg80211.c +0 -0
tools/lib/traceevent/plugins/plugin_function.c renamed from tools/lib/traceevent/plugin_function.c +0 -0
tools/lib/traceevent/plugins/plugin_hrtimer.c renamed from tools/lib/traceevent/plugin_hrtimer.c +0 -0
tools/lib/traceevent/plugins/plugin_jbd2.c renamed from tools/lib/traceevent/plugin_jbd2.c +0 -0
tools/lib/traceevent/plugins/plugin_kmem.c renamed from tools/lib/traceevent/plugin_kmem.c +0 -0
tools/lib/traceevent/plugins/plugin_kvm.c renamed from tools/lib/traceevent/plugin_kvm.c +0 -0
tools/lib/traceevent/plugins/plugin_mac80211.c renamed from tools/lib/traceevent/plugin_mac80211.c +0 -0
tools/lib/traceevent/plugins/plugin_sched_switch.c renamed from tools/lib/traceevent/plugin_sched_switch.c +0 -0
tools/lib/traceevent/plugins/plugin_scsi.c renamed from tools/lib/traceevent/plugin_scsi.c +0 -0
tools/lib/traceevent/plugins/plugin_xen.c renamed from tools/lib/traceevent/plugin_xen.c +0 -0
tools/perf/Makefile.perf +2 -2
@@ -292,7 +292,7 @@ endif LIBTRACEEVENT = $(TE_PATH)libtraceevent.a export LIBTRACEEVENT -LIBTRACEEVENT_DYNAMIC_LIST = $(TE_PATH)libtraceevent-dynamic-list +LIBTRACEEVENT_DYNAMIC_LIST = $(TE_PATH)plugins/libtraceevent-dynamic-list # # The static build has no dynsym table, so this does not work for
@@ -737,7 +737,7 @@ libtraceevent_plugins: FORCE $(Q)$(MAKE) -C $(TRACE_EVENT_DIR) $(LIBTRACEEVENT_FLAGS) O=$(OUTPUT) plugins $(LIBTRACEEVENT_DYNAMIC_LIST): libtraceevent_plugins - $(Q)$(MAKE) -C $(TRACE_EVENT_DIR) $(LIBTRACEEVENT_FLAGS) O=$(OUTPUT) $(OUTPUT)libtraceevent-dynamic-list + $(Q)$(MAKE) -C $(TRACE_EVENT_DIR) $(LIBTRACEEVENT_FLAGS) O=$(OUTPUT) $(OUTPUT)plugins/libtraceevent-dynamic-list $(LIBTRACEEVENT)-clean: $(call QUIET_CLEAN, libtraceevent)
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 13/66 libtraceevent: Man pages for tep plugins APIs
In reply to Arnaldo Carvalho de MeloFrom: Tzvetomir Stoyanov <[email protected]>
Create man pages for libtraceevent APIs:
tep_load_plugins(),
tep_unload_plugin()
--- .../Documentation/libtraceevent-plugins.txt | 99 +++++++++++++++++++ 1 file changed, 99 insertions(+) create mode 100644 tools/lib/traceevent/Documentation/libtraceevent-plugins.txt
tools/lib/traceevent/Documentation/libtraceevent-plugins.txt new file +99 -0
@@ -0,0 +1,99 @@ +libtraceevent(3) +================ + +NAME +---- +tep_load_plugins, tep_unload_plugins - Load / unload traceevent plugins. + +SYNOPSIS +-------- +[verse] +-- +*#include <event-parse.h>* + +struct tep_plugin_list pass:[*]*tep_load_plugins*(struct tep_handle pass:[*]_tep_); +void *tep_unload_plugins*(struct tep_plugin_list pass:[*]_plugin_list_, struct tep_handle pass:[*]_tep_); +-- + +DESCRIPTION +----------- +The _tep_load_plugins()_ function loads all plugins, located in the plugin +directories. The _tep_ argument is trace event parser context. +The plugin directories are : +[verse] +-- + - System's plugin directory, defined at the library compile time. It + depends on the library installation prefix and usually is + _(install_preffix)/lib/traceevent/plugins_ + - Directory, defined by the environment variable _TRACEEVENT_PLUGIN_DIR_ + - User's plugin directory, located at _~/.local/lib/traceevent/plugins_ +-- +Loading of plugins can be controlled by the _tep_flags_, using the +_tep_set_flag()_ API: +[verse] +-- + _TEP_DISABLE_SYS_PLUGINS_ - do not load plugins, located in + the system's plugin directory. + _TEP_DISABLE_PLUGINS_ - do not load any plugins. +-- +The _tep_set_flag()_ API needs to be called before _tep_load_plugins()_, if +loading of all plugins is not the desired case. + +The _tep_unload_plugins()_ function unloads the plugins, previously loaded by +_tep_load_plugins()_. The _tep_ argument is trace event parser context. The +_plugin_list_ is the list of loaded plugins, returned by +the _tep_load_plugins()_ function. + +RETURN VALUE +------------ +The _tep_load_plugins()_ function returns a list of successfully loaded plugins, +or NULL in case no plugins are loaded. + +EXAMPLE +------- +[source,c] +-- +#include <event-parse.h> +... +struct tep_handle *tep = tep_alloc(); +... +struct tep_plugin_list *plugins = tep_load_plugins(tep); +if (plugins == NULL) { + /* no plugins are loaded */ +} +... +tep_unload_plugins(plugins, tep); +-- + +FILES +----- +[verse] +-- +*event-parse.h* + Header file to include in order to have access to the library APIs. +*-ltraceevent* + Linker switch to add when building a program that uses the library. +-- + +SEE ALSO +-------- +_libtraceevent(3)_, _trace-cmd(1)_, _tep_set_flag(3)_ + +AUTHOR +------ +[verse] +-- +*Steven Rostedt* <[email protected]>, author of *libtraceevent*. +*Tzvetomir Stoyanov* <[email protected]>, author of this man page. +-- +REPORTING BUGS +-------------- +Report bugs to <[email protected]> + +LICENSE +------- +libtraceevent is Free Software licensed under the GNU LGPL 2.1 + +RESOURCES +--------- +https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 14/66 tools: Add missing stdio.h include to asm/bug.h header
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
We have a direct fprintf() call in the header, so we need stdio.h include, otherwise it could fail compilation if there's no prior stdio.h include directive.
--- tools/include/asm/bug.h | 1 + 1 file changed, 1 insertion(+)
tools/include/asm/bug.h +1 -0
@@ -3,6 +3,7 @@#define _TOOLS_ASM_BUG_H#include <linux/compiler.h> +#include <stdio.h> #define __WARN_printf(arg...) do { fprintf(stderr, arg); } while (0)
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 15/66 perf tools: Rename 'struct perf_mmap' to 'struct mmap'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Rename 'struct perf_evlist' to 'struct evlist', so we don't have a name clash when we add 'struct perf_mmap' to libperf.
--- tools/perf/arch/x86/tests/perf-time-to-tsc.c | 2 +- tools/perf/builtin-kvm.c | 2 +- tools/perf/builtin-record.c | 34 ++++++------ tools/perf/builtin-top.c | 2 +- tools/perf/builtin-trace.c | 2 +- tools/perf/tests/backward-ring-buffer.c | 2 +- tools/perf/tests/bpf.c | 2 +- tools/perf/tests/code-reading.c | 2 +- tools/perf/tests/keep-tracking.c | 2 +- tools/perf/tests/mmap-basic.c | 2 +- tools/perf/tests/openat-syscall-tp-fields.c | 2 +- tools/perf/tests/perf-record.c | 2 +- tools/perf/tests/sw-clock.c | 2 +- tools/perf/tests/switch-tracking.c | 2 +- tools/perf/tests/task-exit.c | 2 +- tools/perf/util/auxtrace.c | 6 +-- tools/perf/util/auxtrace.h | 8 +-- tools/perf/util/evlist.c | 14 ++--- tools/perf/util/evlist.h | 4 +- tools/perf/util/mmap.c | 54 ++++++++++---------- tools/perf/util/mmap.h | 32 ++++++------ tools/perf/util/python.c | 6 +-- 22 files changed, 93 insertions(+), 93 deletions(-)
tools/perf/arch/x86/tests/perf-time-to-tsc.c +1 -1
@@ -65,7 +65,7 @@ int test__perf_time_to_tsc(struct test *test __maybe_unused, int subtest __maybeunion perf_event *event;u64 test_tsc, comm1_tsc, comm2_tsc;u64 test_time, comm1_time =0 , comm2_time =0 ; -struct perf_mmap *md; +struct mmap *md; threads =thread_map__new (-1 ,getpid (),UINT_MAX );CHECK_NOT_NULL__ (threads);
tools/perf/builtin-kvm.c +1 -1
@@ -750,7 +750,7 @@ static s64 perf_kvm__mmap_read_idx(struct perf_kvm_stat *kvm, int idx, {struct evlist *evlist = kvm->evlist;union perf_event *event; -struct perf_mmap *md; +struct mmap *md;u64 timestamp;s64 n =0 ;int err;
tools/perf/builtin-record.c +17 -17
@@ -119,7 +119,7 @@ static bool switch_output_time(struct record *rec)trigger_is_ready (&switch_output_trigger); } -static int record__write (struct record *rec,struct perf_mmap *map__maybe_unused , +static int record__write (struct record *rec,struct mmap *map__maybe_unused ,void *bf,size_t size) {struct perf_data_file *file = &rec->session->data->file;
@@ -168,7 +168,7 @@ static int record__aio_write(struct aiocb *cblock, int trace_fd,return rc; } -static int record__aio_complete (struct perf_mmap *md,struct aiocb *cblock) +static int record__aio_complete (struct mmap *md,struct aiocb *cblock) {void *rem_buf;off_t rem_off;
@@ -214,7 +214,7 @@ static int record__aio_complete(struct perf_mmap *md, struct aiocb *cblock)return rc; } -static int record__aio_sync (struct perf_mmap *md,bool sync_all) +static int record__aio_sync (struct mmap *md,bool sync_all) {struct aiocb **aiocb = md->aio.aiocb;struct aiocb *cblocks = md->aio.cblocks;
@@ -255,7 +255,7 @@ struct record_aio {size_t size; }; -static int record__aio_pushfn (struct perf_mmap *map,void *to,void *buf,size_t size) +static int record__aio_pushfn (struct mmap *map,void *to,void *buf,size_t size) {struct record_aio *aio = to;
@@ -300,7 +300,7 @@ static int record__aio_pushfn(struct perf_mmap *map, void *to, void *buf, size_treturn size; } -static int record__aio_push (struct record *rec,struct perf_mmap *map,off_t *off) +static int record__aio_push (struct record *rec,struct mmap *map,off_t *off) {int ret, idx;int trace_fd = rec->session->data->file.fd;
@@ -351,13 +351,13 @@ static void record__aio_mmap_read_sync(struct record *rec) {int i;struct evlist *evlist = rec->evlist; -struct perf_mmap *maps = evlist->mmap; +struct mmap *maps = evlist->mmap;if (!record__aio_enabled (rec))return ;for (i =0 ; i < evlist->nr_mmaps; i++) { -struct perf_mmap *map = &maps[i]; +struct mmap *map = &maps[i];if (map->base)record__aio_sync (map, true);
@@ -387,7 +387,7 @@ static int record__aio_parse(const struct option *opt,#else /* HAVE_AIO_SUPPORT */ static int nr_cblocks_max =0 ; -static int record__aio_push (struct record *rec__maybe_unused ,struct perf_mmap *map__maybe_unused , +static int record__aio_push (struct record *rec__maybe_unused ,struct mmap *map__maybe_unused ,off_t *off__maybe_unused ) {return -1 ;
@@ -482,7 +482,7 @@ static int process_synthesized_event(struct perf_tool *tool,return record__write (rec,NULL , event, event->header.size); } -static int record__pushfn (struct perf_mmap *map,void *to,void *bf,size_t size) +static int record__pushfn (struct mmap *map,void *to,void *bf,size_t size) {struct record *rec = to;
@@ -527,7 +527,7 @@ static void record__sig_exit(void)#ifdef HAVE_AUXTRACE_SUPPORT static int record__process_auxtrace (struct perf_tool *tool, -struct perf_mmap *map, +struct mmap *map,union perf_event *event,void *data1,size_t len1,void *data2,size_t len2) {
@@ -565,7 +565,7 @@ static int record__process_auxtrace(struct perf_tool *tool, }static int record__auxtrace_mmap_read (struct record *rec, -struct perf_mmap *map) +struct mmap *map) {int ret;
@@ -581,7 +581,7 @@ static int record__auxtrace_mmap_read(struct record *rec, }static int record__auxtrace_mmap_read_snapshot (struct record *rec, -struct perf_mmap *map) +struct mmap *map) {int ret;
@@ -603,7 +603,7 @@ static int record__auxtrace_read_snapshot_all(struct record *rec)int rc =0 ;for (i =0 ; i < rec->evlist->nr_mmaps; i++) { -struct perf_mmap *map = &rec->evlist->mmap[i]; +struct mmap *map = &rec->evlist->mmap[i];if (!map->auxtrace_mmap.base)continue ;
@@ -668,7 +668,7 @@ static int record__auxtrace_init(struct record *rec)static inline int record__auxtrace_mmap_read (struct record *rec__maybe_unused , -struct perf_mmap *map__maybe_unused ) +struct mmap *map__maybe_unused ) {return 0 ; }
@@ -901,7 +901,7 @@ static struct perf_event_header finished_round_event = { .type = PERF_RECORD_FINISHED_ROUND, }; -static void record__adjust_affinity (struct record *rec,struct perf_mmap *map) +static void record__adjust_affinity (struct record *rec,struct mmap *map) {if (rec->opts.affinity !=PERF_AFFINITY_SYS && !CPU_EQUAL (&rec->affinity_mask, &map->affinity_mask)) {
@@ -948,7 +948,7 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,u64 bytes_written = rec->bytes_written;int i;int rc =0 ; -struct perf_mmap *maps; +struct mmap *maps;int trace_fd = rec->data.file.fd;off_t off =0 ;
@@ -967,7 +967,7 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,for (i =0 ; i < evlist->nr_mmaps; i++) {u64 flush =0 ; -struct perf_mmap *map = &maps[i]; +struct mmap *map = &maps[i];if (map->base) {record__adjust_affinity (rec, map);
tools/perf/builtin-top.c +1 -1
@@ -863,7 +863,7 @@ static void perf_top__mmap_read_idx(struct perf_top *top, int idx) {struct record_opts *opts = &top->record_opts;struct evlist *evlist = top->evlist; -struct perf_mmap *md; +struct mmap *md;union perf_event *event; md = opts->overwrite ? &evlist->overwrite_mmap[idx] : &evlist->mmap[idx];
tools/perf/builtin-trace.c +1 -1
@@ -3444,7 +3444,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)for (i =0 ; i < evlist->nr_mmaps; i++) {union perf_event *event; -struct perf_mmap *md; +struct mmap *md; md = &evlist->mmap[i];if (perf_mmap__read_init (md) <0 )
tools/perf/tests/backward-ring-buffer.c +1 -1
@@ -33,7 +33,7 @@ static int count_samples(struct evlist *evlist, int *sample_count,int i;for (i =0 ; i < evlist->nr_mmaps; i++) { -struct perf_mmap *map = &evlist->overwrite_mmap[i]; +struct mmap *map = &evlist->overwrite_mmap[i];union perf_event *event;perf_mmap__read_init (map);
tools/perf/tests/bpf.c +1 -1
@@ -180,7 +180,7 @@ static int do_test(struct bpf_object *obj, int (*func)(void),for (i =0 ; i < evlist->nr_mmaps; i++) {union perf_event *event; -struct perf_mmap *md; +struct mmap *md; md = &evlist->mmap[i];if (perf_mmap__read_init (md) <0 )
tools/perf/tests/code-reading.c +1 -1
@@ -419,7 +419,7 @@ static int process_events(struct machine *machine, struct evlist *evlist, struct state *state) {union perf_event *event; -struct perf_mmap *md; +struct mmap *md;int i, ret;for (i =0 ; i < evlist->nr_mmaps; i++) {
tools/perf/tests/keep-tracking.c +1 -1
@@ -31,7 +31,7 @@static int find_comm (struct evlist *evlist,const char *comm) {union perf_event *event; -struct perf_mmap *md; +struct mmap *md;int i, found; found =0 ;
tools/perf/tests/mmap-basic.c +1 -1
@@ -42,7 +42,7 @@ int test__basic_mmap(struct test *test __maybe_unused, int subtest __maybe_unuse expected_nr_events[nsyscalls], i, j;struct evsel *evsels[nsyscalls], *evsel;char sbuf[STRERR_BUFSIZE ]; -struct perf_mmap *md; +struct mmap *md; threads =thread_map__new (-1 ,getpid (),UINT_MAX );if (threads ==NULL ) {
tools/perf/tests/openat-syscall-tp-fields.c +1 -1
@@ -88,7 +88,7 @@ int test__syscall_openat_tp_fields(struct test *test __maybe_unused, int subtestfor (i =0 ; i < evlist->nr_mmaps; i++) {union perf_event *event; -struct perf_mmap *md; +struct mmap *md; md = &evlist->mmap[i];if (perf_mmap__read_init (md) <0 )
tools/perf/tests/perf-record.c +1 -1
@@ -166,7 +166,7 @@ int test__PERF_RECORD(struct test *test __maybe_unused, int subtest __maybe_unusfor (i =0 ; i < evlist->nr_mmaps; i++) {union perf_event *event; -struct perf_mmap *md; +struct mmap *md; md = &evlist->mmap[i];if (perf_mmap__read_init (md) <0 )
tools/perf/tests/sw-clock.c +1 -1
@@ -42,7 +42,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) };struct perf_cpu_map *cpus;struct perf_thread_map *threads; -struct perf_mmap *md; +struct mmap *md; attr.sample_freq =500 ;
tools/perf/tests/switch-tracking.c +1 -1
@@ -263,7 +263,7 @@ static int process_events(struct evlist *evlist,unsigned pos, cnt =0 ;LIST_HEAD (events);struct event_node *events_array, *node; -struct perf_mmap *md; +struct mmap *md;int i, ret;for (i =0 ; i < evlist->nr_mmaps; i++) {
tools/perf/tests/task-exit.c +1 -1
@@ -51,7 +51,7 @@ int test__task_exit(struct test *test __maybe_unused, int subtest __maybe_unusedchar sbuf[STRERR_BUFSIZE ];struct perf_cpu_map *cpus;struct perf_thread_map *threads; -struct perf_mmap *md; +struct mmap *md;signal (SIGCHLD , sig_handler);
tools/perf/util/auxtrace.c +3 -3
@@ -1228,7 +1228,7 @@ int perf_event__process_auxtrace_error(struct perf_session *session,return 0 ; } -static int __auxtrace_mmap__read (struct perf_mmap *map, +static int __auxtrace_mmap__read (struct mmap *map,struct auxtrace_record *itr,struct perf_tool *tool,process_auxtrace_t fn,bool snapshot,size_t snapshot_size)
@@ -1339,13 +1339,13 @@ static int __auxtrace_mmap__read(struct perf_mmap *map,return 1 ; } -int auxtrace_mmap__read (struct perf_mmap *map,struct auxtrace_record *itr, +int auxtrace_mmap__read (struct mmap *map,struct auxtrace_record *itr,struct perf_tool *tool,process_auxtrace_t fn) {return __auxtrace_mmap__read (map, itr, tool, fn, false,0 ); } -int auxtrace_mmap__read_snapshot (struct perf_mmap *map, +int auxtrace_mmap__read_snapshot (struct mmap *map,struct auxtrace_record *itr,struct perf_tool *tool,process_auxtrace_t fn,size_t snapshot_size)
tools/perf/util/auxtrace.h +4 -4
@@ -22,7 +22,7 @@ union perf_event;struct perf_session ;struct evlist ;struct perf_tool ; -struct perf_mmap ; +struct mmap ;struct perf_sample ;struct option ;struct record_opts ;
@@ -445,14 +445,14 @@ void auxtrace_mmap_params__set_idx(struct auxtrace_mmap_params *mp, bool per_cpu);typedef int (*process_auxtrace_t )(struct perf_tool *tool, -struct perf_mmap *map, +struct mmap *map,union perf_event *event,void *data1,size_t len1,void *data2,size_t len2); -int auxtrace_mmap__read (struct perf_mmap *map,struct auxtrace_record *itr, +int auxtrace_mmap__read (struct mmap *map,struct auxtrace_record *itr,struct perf_tool *tool,process_auxtrace_t fn); -int auxtrace_mmap__read_snapshot (struct perf_mmap *map, +int auxtrace_mmap__read_snapshot (struct mmap *map,struct auxtrace_record *itr,struct perf_tool *tool,process_auxtrace_t fn,size_t snapshot_size);
tools/perf/util/evlist.c +7 -7
@@ -423,7 +423,7 @@ int perf_evlist__alloc_pollfd(struct evlist *evlist) }static int __perf_evlist__add_pollfd (struct evlist *evlist,int fd, -struct perf_mmap *map,short revent) +struct mmap *map,short revent) {int pos =fdarray__add (&evlist->pollfd, fd, revent |POLLERR |POLLHUP );/*
@@ -447,7 +447,7 @@ int perf_evlist__add_pollfd(struct evlist *evlist, int fd)static void perf_evlist__munmap_filtered (struct fdarray *fda,int fd,void *arg__maybe_unused ) { -struct perf_mmap *map = fda->priv[fd].ptr; +struct mmap *map = fda->priv[fd].ptr;if (map)perf_mmap__put (map);
@@ -693,16 +693,16 @@ void perf_evlist__munmap(struct evlist *evlist)zfree (&evlist->overwrite_mmap); } -static struct perf_mmap *perf_evlist__alloc_mmap (struct evlist *evlist, +static struct mmap *perf_evlist__alloc_mmap (struct evlist *evlist,bool overwrite) {int i; -struct perf_mmap *map; +struct mmap *map; evlist->nr_mmaps =perf_cpu_map__nr (evlist->core.cpus);if (perf_cpu_map__empty (evlist->core.cpus)) evlist->nr_mmaps =perf_thread_map__nr (evlist->core.threads); - map =zalloc (evlist->nr_mmaps *sizeof (struct perf_mmap )); + map =zalloc (evlist->nr_mmaps *sizeof (struct mmap ));if (!map)return NULL ;
@@ -741,7 +741,7 @@ static int perf_evlist__mmap_per_evsel(struct evlist *evlist, int idx,int evlist_cpu =cpu_map__cpu (evlist->core.cpus, cpu_idx);evlist__for_each_entry (evlist, evsel) { -struct perf_mmap *maps = evlist->mmap; +struct mmap *maps = evlist->mmap;int *output = _output;int fd;int cpu;
@@ -1847,7 +1847,7 @@ static void *perf_evlist__poll_thread(void *arg)perf_evlist__poll (evlist,1000 );for (i =0 ; i < evlist->nr_mmaps; i++) { -struct perf_mmap *map = &evlist->mmap[i]; +struct mmap *map = &evlist->mmap[i];union perf_event *event;if (perf_mmap__read_init (map))
tools/perf/util/evlist.h +2 -2
@@ -39,8 +39,8 @@ struct evlist {pid_t pid; } workload;struct fdarray pollfd; -struct perf_mmap *mmap; -struct perf_mmap *overwrite_mmap; +struct mmap *mmap; +struct mmap *overwrite_mmap;struct evsel *selected;struct events_stats stats;struct perf_env *env;
tools/perf/util/mmap.c +27 -27
197 lines not on this page · raw message
tools/perf/util/mmap.h +16 -16
@@ -15,11 +15,11 @@struct aiocb ;/** -* struct perf_mmap - perf's ring buffer mmap details +* struct mmap - perf's ring buffer mmap details * * @refcnt - e.g. code using PERF_EVENT_IOC_SET_OUTPUT to share this */ -struct perf_mmap { +struct mmap {void *base;int mask;int fd;
@@ -78,33 +78,33 @@ struct mmap_params {struct auxtrace_mmap_params auxtrace_mp; }; -int perf_mmap__mmap (struct perf_mmap *map,struct mmap_params *mp,int fd,int cpu); -void perf_mmap__munmap (struct perf_mmap *map); +int perf_mmap__mmap (struct mmap *map,struct mmap_params *mp,int fd,int cpu); +void perf_mmap__munmap (struct mmap *map); -void perf_mmap__get (struct perf_mmap *map); -void perf_mmap__put (struct perf_mmap *map); +void perf_mmap__get (struct mmap *map); +void perf_mmap__put (struct mmap *map); -void perf_mmap__consume (struct perf_mmap *map); +void perf_mmap__consume (struct mmap *map); -static inline u64 perf_mmap__read_head (struct perf_mmap *mm) +static inline u64 perf_mmap__read_head (struct mmap *mm) {return ring_buffer_read_head (mm->base); } -static inline void perf_mmap__write_tail (struct perf_mmap *md,u64 tail) +static inline void perf_mmap__write_tail (struct mmap *md,u64 tail) {ring_buffer_write_tail (md->base, tail); } -union perf_event *perf_mmap__read_forward (struct perf_mmap *map); +union perf_event *perf_mmap__read_forward (struct mmap *map); -union perf_event *perf_mmap__read_event (struct perf_mmap *map); +union perf_event *perf_mmap__read_event (struct mmap *map); -int perf_mmap__push (struct perf_mmap *md,void *to, -int push (struct perf_mmap *map,void *to,void *buf,size_t size)); +int perf_mmap__push (struct mmap *md,void *to, +int push (struct mmap *map,void *to,void *buf,size_t size)); -size_t perf_mmap__mmap_len (struct perf_mmap *map); +size_t perf_mmap__mmap_len (struct mmap *map); -int perf_mmap__read_init (struct perf_mmap *md); -void perf_mmap__read_done (struct perf_mmap *map); +int perf_mmap__read_init (struct mmap *md); +void perf_mmap__read_done (struct mmap *map);#endif /*__PERF_MMAP_H */
tools/perf/util/python.c +3 -3
@@ -984,12 +984,12 @@ static PyObject *pyrf_evlist__add(struct pyrf_evlist *pevlist,return Py_BuildValue ("i" , evlist->core.nr_entries); } -static struct perf_mmap *get_md (struct evlist *evlist,int cpu) +static struct mmap *get_md (struct evlist *evlist,int cpu) {int i;for (i =0 ; i < evlist->nr_mmaps; i++) { -struct perf_mmap *md = &evlist->mmap[i]; +struct mmap *md = &evlist->mmap[i];if (md->cpu == cpu)return md;
@@ -1005,7 +1005,7 @@ static PyObject *pyrf_evlist__read_on_cpu(struct pyrf_evlist *pevlist,union perf_event *event;int sample_id_all =1 , cpu;static char *kwlist[] = {"cpu" ,"sample_id_all" ,NULL }; -struct perf_mmap *md; +struct mmap *md;int err; if (!PyArg_ParseTupleAndKeywords(args, kwargs, "i|i", kwlist,
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 16/66 perf tools: Rename perf_evlist__mmap() to evlist__mmap()
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Rename perf_evlist__mmap() to evlist__mmap(), so we don't have a name clash when we add perf_evlist__mmap() in libperf.
--- tools/perf/arch/x86/tests/perf-time-to-tsc.c | 2 +- tools/perf/builtin-kvm.c | 2 +- tools/perf/builtin-record.c | 6 ++-- tools/perf/builtin-top.c | 2 +- tools/perf/builtin-trace.c | 2 +- tools/perf/tests/backward-ring-buffer.c | 4 +-- tools/perf/tests/bpf.c | 4 +-- tools/perf/tests/code-reading.c | 4 +-- tools/perf/tests/keep-tracking.c | 2 +- tools/perf/tests/mmap-basic.c | 2 +- tools/perf/tests/openat-syscall-tp-fields.c | 4 +-- tools/perf/tests/perf-record.c | 4 +-- tools/perf/tests/sw-clock.c | 2 +- tools/perf/tests/switch-tracking.c | 4 +-- tools/perf/tests/task-exit.c | 2 +- tools/perf/util/evlist.c | 30 ++++++++++---------- tools/perf/util/evlist.h | 8 +++--- tools/perf/util/python.c | 2 +- 18 files changed, 43 insertions(+), 43 deletions(-)
tools/perf/arch/x86/tests/perf-time-to-tsc.c +1 -1
@@ -90,7 +90,7 @@ int test__perf_time_to_tsc(struct test *test __maybe_unused, int subtest __maybeCHECK__ (evlist__open (evlist)); -CHECK__ (perf_evlist__mmap (evlist,UINT_MAX )); +CHECK__ (evlist__mmap (evlist,UINT_MAX )); pc = evlist->mmap[0 ].base; ret =perf_read_tsc_conversion (pc, &tc);
tools/perf/builtin-kvm.c +1 -1
@@ -1060,7 +1060,7 @@ static int kvm_live_open_events(struct perf_kvm_stat *kvm) goto out; } -if (perf_evlist__mmap (evlist, kvm->opts.mmap_pages) <0 ) { +if (evlist__mmap (evlist, kvm->opts.mmap_pages) <0 ) {ui__error ("Failed to mmap the events: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));evlist__close (evlist);
tools/perf/builtin-record.c +3 -3
@@ -439,7 +439,7 @@ static int record__mmap_flush_parse(const struct option *opt,if (!opts->mmap_flush) opts->mmap_flush =MMAP_FLUSH_DEFAULT ; - flush_max =perf_evlist__mmap_size (opts->mmap_pages); + flush_max =evlist__mmap_size (opts->mmap_pages); flush_max /=4 ;if (opts->mmap_flush > flush_max) opts->mmap_flush = flush_max;
@@ -707,7 +707,7 @@ static int record__mmap_evlist(struct record *rec,if (opts->affinity !=PERF_AFFINITY_SYS )cpu__setup_cpunode_map (); - if (perf_evlist__mmap_ex(evlist, opts->mmap_pages, + if (evlist__mmap_ex(evlist, opts->mmap_pages, opts->auxtrace_mmap_pages, opts->auxtrace_snapshot_mode, opts->nr_cblocks, opts->affinity,
@@ -1980,7 +1980,7 @@ static int record__parse_mmap_pages(const struct option *opt,static void switch_output_size_warn (struct record *rec) { -u64 wakeup_size =perf_evlist__mmap_size (rec->opts.mmap_pages); +u64 wakeup_size =evlist__mmap_size (rec->opts.mmap_pages);struct switch_output *s = &rec->switch_output; wakeup_size /=2 ;
tools/perf/builtin-top.c +1 -1
@@ -1042,7 +1042,7 @@ static int perf_top__start_counters(struct perf_top *top) } } -if (perf_evlist__mmap (evlist, opts->mmap_pages) <0 ) { +if (evlist__mmap (evlist, opts->mmap_pages) <0 ) {ui__error ("Failed to mmap with %d (%s)\n" , errno,str_error_r (errno, msg,sizeof (msg))); goto out_err;
tools/perf/builtin-trace.c +1 -1
@@ -3409,7 +3409,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)if (trace->dump.map)bpf_map__fprintf (trace->dump.map, trace->output); - err =perf_evlist__mmap (evlist, trace->opts.mmap_pages); + err =evlist__mmap (evlist, trace->opts.mmap_pages);if (err <0 ) goto out_error_mmap;
tools/perf/tests/backward-ring-buffer.c +2 -2
@@ -63,9 +63,9 @@ static int do_test(struct evlist *evlist, int mmap_pages,int err;char sbuf[STRERR_BUFSIZE ]; - err =perf_evlist__mmap (evlist, mmap_pages); + err =evlist__mmap (evlist, mmap_pages);if (err <0 ) { -pr_debug ("perf_evlist__mmap: %s\n" , +pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf)));return TEST_FAIL ; }
tools/perf/tests/bpf.c +2 -2
@@ -167,9 +167,9 @@ static int do_test(struct bpf_object *obj, int (*func)(void), goto out_delete_evlist; } - err =perf_evlist__mmap (evlist, opts.mmap_pages); + err =evlist__mmap (evlist, opts.mmap_pages);if (err <0 ) { -pr_debug ("perf_evlist__mmap: %s\n" , +pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); goto out_delete_evlist; }
tools/perf/tests/code-reading.c +2 -2
@@ -685,9 +685,9 @@ static int do_test_code_reading(bool try_kcore)break ; } - ret =perf_evlist__mmap (evlist,UINT_MAX ); + ret =evlist__mmap (evlist,UINT_MAX );if (ret <0 ) { -pr_debug ("perf_evlist__mmap failed\n" ); +pr_debug ("evlist__mmap failed\n" ); goto out_put; }
tools/perf/tests/keep-tracking.c +1 -1
@@ -104,7 +104,7 @@ int test__keep_tracking(struct test *test __maybe_unused, int subtest __maybe_un goto out_err; } -CHECK__ (perf_evlist__mmap (evlist,UINT_MAX )); +CHECK__ (evlist__mmap (evlist,UINT_MAX ));/* * First, test that a 'comm' event can be found when the event is
tools/perf/tests/mmap-basic.c +1 -1
@@ -99,7 +99,7 @@ int test__basic_mmap(struct test *test __maybe_unused, int subtest __maybe_unuse expected_nr_events[i] =1 +rand () %127 ; } -if (perf_evlist__mmap (evlist,128 ) <0 ) { +if (evlist__mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); goto out_delete_evlist;
tools/perf/tests/openat-syscall-tp-fields.c +2 -2
@@ -69,9 +69,9 @@ int test__syscall_openat_tp_fields(struct test *test __maybe_unused, int subtest goto out_delete_evlist; } - err =perf_evlist__mmap (evlist,UINT_MAX ); + err =evlist__mmap (evlist,UINT_MAX );if (err <0 ) { -pr_debug ("perf_evlist__mmap: %s\n" , +pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); goto out_delete_evlist; }
tools/perf/tests/perf-record.c +2 -2
@@ -143,9 +143,9 @@ int test__PERF_RECORD(struct test *test __maybe_unused, int subtest __maybe_unus* fds in the same CPU to be injected in the same mmap ring buffer * (using ioctl(PERF_EVENT_IOC_SET_OUTPUT)). */ - err =perf_evlist__mmap (evlist, opts.mmap_pages); + err =evlist__mmap (evlist, opts.mmap_pages);if (err <0 ) { -pr_debug ("perf_evlist__mmap: %s\n" , +pr_debug ("evlist__mmap: %s\n" ,str_error_r (errno, sbuf,sizeof (sbuf))); goto out_delete_evlist; }
tools/perf/tests/sw-clock.c +1 -1
@@ -82,7 +82,7 @@ static int __test__sw_clock_freq(enum perf_sw_ids clock_id) goto out_delete_evlist; } - err =perf_evlist__mmap (evlist,128 ); + err =evlist__mmap (evlist,128 );if (err <0 ) {pr_debug ("failed to mmap event: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf)));
tools/perf/tests/switch-tracking.c +2 -2
@@ -460,9 +460,9 @@ int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_ goto out; } - err =perf_evlist__mmap (evlist,UINT_MAX ); + err =evlist__mmap (evlist,UINT_MAX );if (err) { -pr_debug ("perf_evlist__mmap failed!\n" ); +pr_debug ("evlist__mmap failed!\n" ); goto out_err; }
tools/perf/tests/task-exit.c +1 -1
@@ -106,7 +106,7 @@ int test__task_exit(struct test *test __maybe_unused, int subtest __maybe_unused goto out_delete_evlist; } -if (perf_evlist__mmap (evlist,128 ) <0 ) { +if (evlist__mmap (evlist,128 ) <0 ) {pr_debug ("failed to mmap events: %d (%s)\n" , errno,str_error_r (errno, sbuf,sizeof (sbuf))); goto out_delete_evlist;
tools/perf/util/evlist.c +15 -15
@@ -732,7 +732,7 @@ perf_evlist__should_poll(struct evlist *evlist __maybe_unused,return true; } -static int perf_evlist__mmap_per_evsel (struct evlist *evlist,int idx, +static int evlist__mmap_per_evsel (struct evlist *evlist,int idx,struct mmap_params *mp,int cpu_idx,int thread,int *_output,int *_output_overwrite) {
@@ -810,7 +810,7 @@ static int perf_evlist__mmap_per_evsel(struct evlist *evlist, int idx,return 0 ; } -static int perf_evlist__mmap_per_cpu (struct evlist *evlist, +static int evlist__mmap_per_cpu (struct evlist *evlist,struct mmap_params *mp) {int cpu, thread;
@@ -826,7 +826,7 @@ static int perf_evlist__mmap_per_cpu(struct evlist *evlist, true);for (thread =0 ; thread < nr_threads; thread++) { -if (perf_evlist__mmap_per_evsel (evlist, cpu, mp, cpu, +if (evlist__mmap_per_evsel (evlist, cpu, mp, cpu, thread, &output, &output_overwrite)) goto out_unmap; }
@@ -839,7 +839,7 @@ static int perf_evlist__mmap_per_cpu(struct evlist *evlist,return -1 ; } -static int perf_evlist__mmap_per_thread (struct evlist *evlist, +static int evlist__mmap_per_thread (struct evlist *evlist,struct mmap_params *mp) {int thread;
@@ -853,7 +853,7 @@ static int perf_evlist__mmap_per_thread(struct evlist *evlist,auxtrace_mmap_params__set_idx (&mp->auxtrace_mp, evlist, thread, false); -if (perf_evlist__mmap_per_evsel (evlist, thread, mp,0 , thread, +if (evlist__mmap_per_evsel (evlist, thread, mp,0 , thread, &output, &output_overwrite)) goto out_unmap; }
@@ -888,7 +888,7 @@ unsigned long perf_event_mlock_kb_in_pages(void)return pages; } -size_t perf_evlist__mmap_size (unsigned long pages) +size_t evlist__mmap_size (unsigned long pages) {if (pages ==UINT_MAX ) pages =perf_event_mlock_kb_in_pages ();
@@ -971,7 +971,7 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str, }/** -* perf_evlist__mmap_ex - Create mmaps to receive events. +* evlist__mmap_ex - Create mmaps to receive events. * @evlist: list of events * @pages: map length in pages * @overwrite: overwrite older events?
@@ -979,7 +979,7 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,* @auxtrace_overwrite - overwrite older auxtrace data? * * If @overwrite is %false the user needs to signal event consumption using -* perf_mmap__write_tail(). Using perf_evlist__mmap_read() does this +* perf_mmap__write_tail(). Using evlist__mmap_read() does this * automatically. * * Similarly, if @auxtrace_overwrite is %false the user needs to signal data
@@ -987,7 +987,7 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,* * Return: %0 on success, negative error code otherwise. */ -int perf_evlist__mmap_ex (struct evlist *evlist,unsigned int pages, +int evlist__mmap_ex (struct evlist *evlist,unsigned int pages,unsigned int auxtrace_pages,bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush,int comp_level)
@@ -1011,7 +1011,7 @@ int perf_evlist__mmap_ex(struct evlist *evlist, unsigned int pages,if (evlist->pollfd.entries ==NULL &&perf_evlist__alloc_pollfd (evlist) <0 )return -ENOMEM ; - evlist->mmap_len =perf_evlist__mmap_size (pages); + evlist->mmap_len =evlist__mmap_size (pages);pr_debug ("mmap size %zuB\n" , evlist->mmap_len); mp.mask = evlist->mmap_len - page_size -1 ;
@@ -1026,14 +1026,14 @@ int perf_evlist__mmap_ex(struct evlist *evlist, unsigned int pages, }if (perf_cpu_map__empty (cpus)) -return perf_evlist__mmap_per_thread (evlist, &mp); +return evlist__mmap_per_thread (evlist, &mp); -return perf_evlist__mmap_per_cpu (evlist, &mp); +return evlist__mmap_per_cpu (evlist, &mp); } -int perf_evlist__mmap (struct evlist *evlist,unsigned int pages) +int evlist__mmap (struct evlist *evlist,unsigned int pages) { -return perf_evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ,0 ); +return evlist__mmap_ex (evlist, pages,0 , false,0 ,PERF_AFFINITY_SYS ,1 ,0 ); }int perf_evlist__create_maps (struct evlist *evlist,struct target *target)
@@ -1889,7 +1889,7 @@ int perf_evlist__start_sb_thread(struct evlist *evlist, goto out_delete_evlist; } -if (perf_evlist__mmap (evlist,UINT_MAX )) +if (evlist__mmap (evlist,UINT_MAX )) goto out_delete_evlist;evlist__for_each_entry (evlist, counter) {
tools/perf/util/evlist.h +4 -4
@@ -139,7 +139,7 @@ struct perf_sample_id *perf_evlist__id2sid(struct evlist *evlist, u64 id);void perf_evlist__toggle_bkw_mmap (struct evlist *evlist,enum bkw_mmap_state state); -void perf_evlist__mmap_consume (struct evlist *evlist,int idx); +void evlist__mmap_consume (struct evlist *evlist,int idx);int evlist__open (struct evlist *evlist);void evlist__close (struct evlist *evlist);
@@ -170,14 +170,14 @@ int perf_evlist__parse_mmap_pages(const struct option *opt,unsigned long perf_event_mlock_kb_in_pages (void ); -int perf_evlist__mmap_ex (struct evlist *evlist,unsigned int pages, +int evlist__mmap_ex (struct evlist *evlist,unsigned int pages,unsigned int auxtrace_pages,bool auxtrace_overwrite,int nr_cblocks,int affinity,int flush,int comp_level); -int perf_evlist__mmap (struct evlist *evlist,unsigned int pages); +int evlist__mmap (struct evlist *evlist,unsigned int pages);void perf_evlist__munmap (struct evlist *evlist); -size_t perf_evlist__mmap_size (unsigned long pages); +size_t evlist__mmap_size (unsigned long pages);void evlist__disable (struct evlist *evlist);void evlist__enable (struct evlist *evlist);
tools/perf/util/python.c +1 -1
@@ -899,7 +899,7 @@ static PyObject *pyrf_evlist__mmap(struct pyrf_evlist *pevlist, &pages, &overwrite))return NULL ; -if (perf_evlist__mmap (evlist, pages) <0 ) { +if (evlist__mmap (evlist, pages) <0 ) {PyErr_SetFromErrno (PyExc_OSError);return NULL ; }
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 17/66 perf tools: Rename perf_evlist__munmap() to evlist__munmap()
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Rename perf_evlist__munmap() to evlist__munmap(), so we don't have a name clash when we add perf_evlist__munmap() in libperf.
--- tools/perf/tests/backward-ring-buffer.c | 2 +- tools/perf/util/evlist.c | 12 ++++++------ tools/perf/util/evlist.h | 2 +- 3 files changed, 8 insertions(+), 8 deletions(-)
tools/perf/tests/backward-ring-buffer.c +1 -1
@@ -75,7 +75,7 @@ static int do_test(struct evlist *evlist, int mmap_pages,evlist__disable (evlist); err =count_samples (evlist, sample_count, comm_count); -perf_evlist__munmap (evlist); +evlist__munmap (evlist);return err; }
tools/perf/util/evlist.c +6 -6
@@ -149,7 +149,7 @@ void evlist__delete(struct evlist *evlist)if (evlist ==NULL )return ; -perf_evlist__munmap (evlist); +evlist__munmap (evlist);evlist__close (evlist);perf_cpu_map__put (evlist->core.cpus);perf_thread_map__put (evlist->core.threads);
@@ -673,7 +673,7 @@ static int perf_evlist__resume(struct evlist *evlist)return perf_evlist__set_paused (evlist, false); } -static void perf_evlist__munmap_nofree (struct evlist *evlist) +static void evlist__munmap_nofree (struct evlist *evlist) {int i;
@@ -686,9 +686,9 @@ static void perf_evlist__munmap_nofree(struct evlist *evlist)perf_mmap__munmap (&evlist->overwrite_mmap[i]); } -void perf_evlist__munmap (struct evlist *evlist) +void evlist__munmap (struct evlist *evlist) { -perf_evlist__munmap_nofree (evlist); +evlist__munmap_nofree (evlist);zfree (&evlist->mmap);zfree (&evlist->overwrite_mmap); }
@@ -835,7 +835,7 @@ static int evlist__mmap_per_cpu(struct evlist *evlist,return 0 ; out_unmap: -perf_evlist__munmap_nofree (evlist); +evlist__munmap_nofree (evlist);return -1 ; }
@@ -861,7 +861,7 @@ static int evlist__mmap_per_thread(struct evlist *evlist,return 0 ; out_unmap: -perf_evlist__munmap_nofree (evlist); +evlist__munmap_nofree (evlist);return -1 ; }
tools/perf/util/evlist.h +1 -1
@@ -175,7 +175,7 @@ int evlist__mmap_ex(struct evlist *evlist, unsigned int pages, bool auxtrace_overwrite, int nr_cblocks, int affinity, int flush, int comp_level);int evlist__mmap (struct evlist *evlist,unsigned int pages); -void perf_evlist__munmap (struct evlist *evlist); +void evlist__munmap (struct evlist *evlist);size_t evlist__mmap_size (unsigned long pages);
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 18/66 perf tools: Rename perf_evlist__alloc_mmap() to evlist__alloc_mmap()
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Rename perf_evlist__alloc_mmap() to evlist__alloc_mmap(), so we don't have a name clash when we add perf_evlist__alloc_mmap() to libperf.
--- tools/perf/util/evlist.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
tools/perf/util/evlist.c +4 -4
@@ -693,8 +693,8 @@ void evlist__munmap(struct evlist *evlist)zfree (&evlist->overwrite_mmap); } -static struct mmap *perf_evlist__alloc_mmap (struct evlist *evlist, -bool overwrite) +static struct mmap *evlist__alloc_mmap (struct evlist *evlist, +bool overwrite) {int i;struct mmap *map;
@@ -752,7 +752,7 @@ static int evlist__mmap_per_evsel(struct evlist *evlist, int idx, maps = evlist->overwrite_mmap;if (!maps) { - maps =perf_evlist__alloc_mmap (evlist, true); + maps =evlist__alloc_mmap (evlist, true);if (!maps)return -1 ; evlist->overwrite_mmap = maps;
@@ -1004,7 +1004,7 @@ int evlist__mmap_ex(struct evlist *evlist, unsigned int pages, .comp_level = comp_level };if (!evlist->mmap) - evlist->mmap =perf_evlist__alloc_mmap (evlist, false); + evlist->mmap =evlist__alloc_mmap (evlist, false);if (!evlist->mmap)return -ENOMEM ;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 19/66 perf tools: Rename perf_evlist__exit() to evlist__exit()
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Rename perf_evlist__exit() to evlist__exit(), so we don't have a name clash when we add perf_evlist__exit() to libperf.
--- tools/perf/util/evlist.c | 4 ++-- tools/perf/util/evlist.h | 2 +- tools/perf/util/python.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-)
tools/perf/util/evlist.c +2 -2
@@ -137,7 +137,7 @@ static void perf_evlist__purge(struct evlist *evlist) evlist->core.nr_entries =0 ; } -void perf_evlist__exit (struct evlist *evlist) +void evlist__exit (struct evlist *evlist) {zfree (&evlist->mmap);zfree (&evlist->overwrite_mmap);
@@ -156,7 +156,7 @@ void evlist__delete(struct evlist *evlist) evlist->core.cpus =NULL ; evlist->core.threads =NULL ;perf_evlist__purge (evlist); -perf_evlist__exit (evlist); +evlist__exit (evlist);free (evlist); }
tools/perf/util/evlist.h +1 -1
@@ -65,7 +65,7 @@ struct evlist *perf_evlist__new_default(void);struct evlist *perf_evlist__new_dummy (void );void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus,struct perf_thread_map *threads); -void perf_evlist__exit (struct evlist *evlist); +void evlist__exit (struct evlist *evlist);void evlist__delete (struct evlist *evlist);void evlist__add (struct evlist *evlist,struct evsel *entry);
tools/perf/util/python.c +1 -1
@@ -884,7 +884,7 @@ static int pyrf_evlist__init(struct pyrf_evlist *pevlist,static void pyrf_evlist__delete (struct pyrf_evlist *pevlist) { -perf_evlist__exit (&pevlist->evlist); +evlist__exit (&pevlist->evlist);Py_TYPE (pevlist)->tp_free ((PyObject *)pevlist); }
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 20/66 perf tools: Rename perf_evlist__purge() to evlist__purge()
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Rename (perf_evlist__purge) to evlist__purge(), so we don't have a name clash when we add (perf_evlist__purge) in libperf.
--- tools/perf/util/evlist.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/util/evlist.c +2 -2
@@ -124,7 +124,7 @@ static void perf_evlist__update_id_pos(struct evlist *evlist)perf_evlist__set_id_pos (evlist); } -static void perf_evlist__purge (struct evlist *evlist) +static void evlist__purge (struct evlist *evlist) {struct evsel *pos, *n;
@@ -155,7 +155,7 @@ void evlist__delete(struct evlist *evlist)perf_thread_map__put (evlist->core.threads); evlist->core.cpus =NULL ; evlist->core.threads =NULL ; -perf_evlist__purge (evlist); +evlist__purge (evlist);evlist__exit (evlist);free (evlist); }
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 21/66 libperf: Link libapi.a in libperf.so
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Linking libapi.a in libperf.so, because we are about to use some of the API functions in it.
--- tools/perf/lib/Makefile | 33 ++++++++++++++++++++++++++++++--- 1 file changed, 30 insertions(+), 3 deletions(-)
tools/perf/lib/Makefile +30 -3
@@ -59,7 +59,13 @@ else CFLAGS := -g -Wall endif -INCLUDES = -I$(srctree)/tools/perf/lib/include -I$(srctree)/tools/include -I$(srctree)/tools/arch/$(SRCARCH)/include/ -I$(srctree)/tools/arch/$(SRCARCH)/include/uapi -I$(srctree)/tools/include/uapi +INCLUDES = \ +-I$(srctree)/tools/perf/lib/include \ +-I$(srctree)/tools/lib/ \ +-I$(srctree)/tools/include \ +-I$(srctree)/tools/arch/$(SRCARCH)/include/ \ +-I$(srctree)/tools/arch/$(SRCARCH)/include/uapi \ +-I$(srctree)/tools/include/uapi # Append required CFLAGS override CFLAGS += $(EXTRA_WARNINGS)
@@ -88,13 +94,34 @@ LIBPERF_PC := $(OUTPUT)libperf.pc LIBPERF_ALL := $(LIBPERF_A) $(OUTPUT)libperf.so* +LIB_DIR := $(srctree)/tools/lib/api/ + +ifneq ($(OUTPUT),) +ifneq ($(subdir),) + API_PATH=$(OUTPUT)/../lib/api/ +else + API_PATH=$(OUTPUT) +endif +else + API_PATH=$(LIB_DIR) +endif + +LIBAPI = $(API_PATH)libapi.a + +$(LIBAPI): FORCE + $(Q)$(MAKE) -C $(LIB_DIR) O=$(OUTPUT) $(OUTPUT)libapi.a + +$(LIBAPI)-clean: + $(call QUIET_CLEAN, libapi) + $(Q)$(MAKE) -C $(LIB_DIR) O=$(OUTPUT) clean >/dev/null + $(LIBPERF_IN): FORCE $(Q)$(MAKE) $(build)=libperf $(LIBPERF_A): $(LIBPERF_IN) $(QUIET_AR)$(RM) $@ && $(AR) rcs $@ $(LIBPERF_IN) -$(LIBPERF_SO): $(LIBPERF_IN) +$(LIBPERF_SO): $(LIBPERF_IN) $(LIBAPI) $(QUIET_LINK)$(CC) --shared -Wl,-soname,libperf.so \ -Wl,--version-script=$(VERSION_SCRIPT) $^ -o $@ @ln -sf $(@F) $(OUTPUT)libperf.so
@@ -106,7 +133,7 @@ libs: $(LIBPERF_A) $(LIBPERF_SO) $(LIBPERF_PC) all: fixdep $(Q)$(MAKE) libs -clean: +clean: $(LIBAPI)-clean $(call QUIET_CLEAN, libperf) $(RM) $(LIBPERF_A) \ *.o *~ *.a *.so *.so.$(VERSION) *.so.$(LIBPERF_VERSION) .*.d .*.cmd LIBPERF-CFLAGS $(LIBPERF_PC) $(Q)$(MAKE) -C tests clean
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 22/66 perf evlist: Adopt backwards ring buffer state enum
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
As this isn't used at all in mmap.h but in evlist.h, so to cut down the header dependency tree, move it to where it is used.
Also add mmap.h to the places using it but previously getting it indirectly via evlist.h.
Add missing pthread.h to evlist.h, as it has a pthread_t struct member and was getting the header via mmap.h.
Noticed while processing a Jiri's libperf batch touching mmap.h, where almost everything gets rebuilt because evlist.h is so popular, so cut down't this rebuild the world party.
--- tools/perf/arch/s390/util/auxtrace.c | 1 + tools/perf/arch/x86/tests/perf-time-to-tsc.c | 1 + tools/perf/arch/x86/util/intel-bts.c | 1 + tools/perf/arch/x86/util/intel-pt.c | 1 + tools/perf/builtin-kvm.c | 1 + tools/perf/builtin-record.c | 1 + tools/perf/builtin-top.c | 1 + tools/perf/builtin-trace.c | 1 + tools/perf/tests/backward-ring-buffer.c | 1 + tools/perf/tests/bpf.c | 1 + tools/perf/tests/code-reading.c | 1 + tools/perf/tests/hists_link.c | 1 + tools/perf/tests/keep-tracking.c | 1 + tools/perf/tests/mmap-basic.c | 1 + tools/perf/tests/openat-syscall-tp-fields.c | 1 + tools/perf/tests/perf-record.c | 1 + tools/perf/tests/sw-clock.c | 1 + tools/perf/tests/switch-tracking.c | 1 + tools/perf/tests/task-exit.c | 1 + tools/perf/ui/gtk/hists.c | 1 + tools/perf/util/annotate.c | 1 + tools/perf/util/auxtrace.c | 1 + tools/perf/util/evlist.c | 1 + tools/perf/util/evlist.h | 30 +++++++++++++++++++- tools/perf/util/mmap.h | 28 ------------------ tools/perf/util/parse-events.c | 1 + 26 files changed, 53 insertions(+), 29 deletions(-)
tools/perf/arch/s390/util/auxtrace.c +1 -0
@@ -1,4 +1,5 @@#include <stdbool.h> +#include <stdlib.h> #include <linux/kernel.h> #include <linux/types.h> #include <linux/bitops.h>
tools/perf/arch/x86/tests/perf-time-to-tsc.c +1 -0
@@ -17,6 +17,7 @@#include "thread_map.h" #include "record.h" #include "tsc.h" +#include "util/mmap.h" #include "tests/tests.h" #include "arch-tests.h"
tools/perf/arch/x86/util/intel-bts.c +1 -0
@@ -15,6 +15,7 @@#include "../../util/event.h" #include "../../util/evsel.h" #include "../../util/evlist.h" +#include "../../util/mmap.h" #include "../../util/session.h" #include "../../util/pmu.h" #include "../../util/debug.h"
tools/perf/arch/x86/util/intel-pt.c +1 -0
@@ -18,6 +18,7 @@#include "../../util/evlist.h" #include "../../util/evsel.h" #include "../../util/cpumap.h" +#include "../../util/mmap.h" #include <subcmd/parse-options.h> #include "../../util/parse-events.h" #include "../../util/pmu.h"
tools/perf/builtin-kvm.c +1 -0
@@ -5,6 +5,7 @@#include "util/build-id.h" #include "util/evsel.h" #include "util/evlist.h" +#include "util/mmap.h" #include "util/term.h" #include "util/symbol.h" #include "util/thread.h"
tools/perf/builtin-record.c +1 -0
@@ -20,6 +20,7 @@#include "util/evlist.h" #include "util/evsel.h" #include "util/debug.h" +#include "util/mmap.h" #include "util/target.h" #include "util/session.h" #include "util/tool.h"
tools/perf/builtin-top.c +1 -0
@@ -30,6 +30,7 @@#include "util/event.h" #include "util/machine.h" #include "util/map.h" +#include "util/mmap.h" #include "util/session.h" #include "util/symbol.h" #include "util/synthetic-events.h"
tools/perf/builtin-trace.c +1 -0
@@ -31,6 +31,7 @@#include "util/synthetic-events.h" #include "util/evlist.h" #include "util/evswitch.h" +#include "util/mmap.h" #include <subcmd/pager.h> #include <subcmd/exec-cmd.h> #include "util/machine.h"
tools/perf/tests/backward-ring-buffer.c +1 -0
@@ -10,6 +10,7 @@#include "tests.h" #include "debug.h" #include "parse-events.h" +#include "util/mmap.h" #include <errno.h> #include <linux/string.h>
tools/perf/tests/bpf.c +1 -0
@@ -19,6 +19,7 @@#include "llvm.h" #include "debug.h" #include "parse-events.h" +#include "util/mmap.h" #define NR_ITERS 111#define PERF_TEST_BPF_PATH "/sys/fs/bpf/perf_test"
tools/perf/tests/code-reading.c +1 -0
@@ -24,6 +24,7 @@#include "symbol.h" #include "event.h" #include "record.h" +#include "util/mmap.h" #include "util/synthetic-events.h" #include "thread.h"
tools/perf/tests/hists_link.c +1 -0
@@ -8,6 +8,7 @@#include "machine.h" #include "parse-events.h" #include "hists_common.h" +#include "util/mmap.h" #include <errno.h> #include <linux/kernel.h>
tools/perf/tests/keep-tracking.c +1 -0
@@ -13,6 +13,7 @@#include "record.h" #include "thread_map.h" #include "tests.h" +#include "util/mmap.h" #define CHECK__ (x) { \ while ((x) < 0) { \
tools/perf/tests/mmap-basic.c +1 -0
@@ -11,6 +11,7 @@#include "evsel.h" #include "thread_map.h" #include "tests.h" +#include "util/mmap.h" #include <linux/err.h> #include <linux/kernel.h> #include <linux/string.h>
tools/perf/tests/openat-syscall-tp-fields.c +1 -0
@@ -11,6 +11,7 @@#include "record.h" #include "tests.h" #include "debug.h" +#include "util/mmap.h" #include <errno.h> #ifndef O_DIRECTORY
tools/perf/tests/perf-record.c +1 -0
@@ -11,6 +11,7 @@#include "debug.h" #include "record.h" #include "tests.h" +#include "util/mmap.h" static int sched__get_first_possible_cpu (pid_t pid,cpu_set_t *maskp) {
tools/perf/tests/sw-clock.c +1 -0
@@ -12,6 +12,7 @@#include "util/evsel.h" #include "util/evlist.h" #include "util/cpumap.h" +#include "util/mmap.h" #include "util/thread_map.h" #include <perf/evlist.h>
tools/perf/tests/switch-tracking.c +1 -0
@@ -16,6 +16,7 @@#include "thread_map.h" #include "record.h" #include "tests.h" +#include "util/mmap.h" static int spin_sleep (void ) {
tools/perf/tests/task-exit.c +1 -0
@@ -5,6 +5,7 @@#include "target.h" #include "thread_map.h" #include "tests.h" +#include "util/mmap.h" #include <errno.h> #include <signal.h>
tools/perf/ui/gtk/hists.c +1 -0
@@ -8,6 +8,7 @@#include "../string2.h" #include "gtk.h" #include <signal.h> +#include <stdlib.h> #include <linux/string.h> #define MAX_COLUMNS 32
tools/perf/util/annotate.c +1 -0
@@ -34,6 +34,7 @@#include "bpf-event.h" #include "block-range.h" #include "string2.h" +#include "util/mmap.h" #include "arch/common.h" #include <regex.h> #include <pthread.h>
tools/perf/util/auxtrace.c +1 -0
@@ -51,6 +51,7 @@#include "arm-spe.h" #include "s390-cpumsf.h" #include "util.h" // page_size +#include "util/mmap.h" #include <linux/ctype.h> #include <linux/kernel.h>
tools/perf/util/evlist.c +1 -0
@@ -10,6 +10,7 @@#include <inttypes.h> #include <poll.h> #include "cpumap.h" +#include "util/mmap.h" #include "thread_map.h" #include "target.h" #include "evlist.h"
tools/perf/util/evlist.h +29 -1
@@ -11,7 +11,7 @@#include <internal/evlist.h> #include "events_stats.h" #include "evsel.h" -#include "mmap.h" +#include <pthread.h> #include <signal.h> #include <unistd.h>
@@ -20,6 +20,34 @@ struct thread_map;struct perf_cpu_map ;struct record_opts ; +/* +* State machine of bkw_mmap_state: +* +* .________________(forbid)_____________. +* | V +* NOTREADY --(0)--> RUNNING --(1)--> DATA_PENDING --(2)--> EMPTY +* ^ ^ | ^ | +* | |__(forbid)____/ |___(forbid)___/| +* | | +* \_________________(3)_______________/ +* +* NOTREADY : Backward ring buffers are not ready +* RUNNING : Backward ring buffers are recording +* DATA_PENDING : We are required to collect data from backward ring buffers +* EMPTY : We have collected data from backward ring buffers. +* +* (0): Setup backward ring buffer +* (1): Pause ring buffers for reading +* (2): Read from ring buffers +* (3): Resume ring buffers for recording +*/ +enum bkw_mmap_state { +BKW_MMAP_NOTREADY , +BKW_MMAP_RUNNING , +BKW_MMAP_DATA_PENDING , +BKW_MMAP_EMPTY , +}; +#define PERF_EVLIST__HLIST_BITS 8#define PERF_EVLIST__HLIST_SIZE (1 << PERF_EVLIST__HLIST_BITS)
tools/perf/util/mmap.h +0 -28
@@ -45,34 +45,6 @@ struct mmap {int comp_level; }; -/* -* State machine of bkw_mmap_state: -* -* .________________(forbid)_____________. -* | V -* NOTREADY --(0)--> RUNNING --(1)--> DATA_PENDING --(2)--> EMPTY -* ^ ^ | ^ | -* | |__(forbid)____/ |___(forbid)___/| -* | | -* \_________________(3)_______________/ -* -* NOTREADY : Backward ring buffers are not ready -* RUNNING : Backward ring buffers are recording -* DATA_PENDING : We are required to collect data from backward ring buffers -* EMPTY : We have collected data from backward ring buffers. -* -* (0): Setup backward ring buffer -* (1): Pause ring buffers for reading -* (2): Read from ring buffers -* (3): Resume ring buffers for recording -*/ -enum bkw_mmap_state { -BKW_MMAP_NOTREADY , -BKW_MMAP_RUNNING , -BKW_MMAP_DATA_PENDING , -BKW_MMAP_EMPTY , -}; -struct mmap_params {int prot, mask, nr_cblocks, affinity, flush, comp_level;struct auxtrace_mmap_params auxtrace_mp;
tools/perf/util/parse-events.c +1 -0
@@ -34,6 +34,7 @@#include "asm/bug.h" #include "util/parse-branch-options.h" #include "metricgroup.h" +#include "util/mmap.h" #define MAX_NAME_LEN 100
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 23/66 libperf: Add perf_mmap struct
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Add the perf_mmap struct to libperf.
The definition is added into:
include/internal/mmap.h
which is not to be included by users, but shared within perf and libperf.
Committer notes:
Remove unnecessary includes from tools/perf/lib/include/internal/mmap.h, those will be readded as they become necessary, later in the series.
--- tools/perf/arch/x86/tests/perf-time-to-tsc.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 2 +- tools/perf/arch/x86/util/intel-pt.c | 2 +- tools/perf/builtin-record.c | 14 ++++++------- tools/perf/lib/include/internal/mmap.h | 14 +++++++++++++ tools/perf/util/mmap.c | 22 ++++++++++---------- tools/perf/util/mmap.h | 7 ++++--- 7 files changed, 39 insertions(+), 24 deletions(-) create mode 100644 tools/perf/lib/include/internal/mmap.h
tools/perf/arch/x86/tests/perf-time-to-tsc.c +1 -1
@@ -93,7 +93,7 @@ int test__perf_time_to_tsc(struct test *test __maybe_unused, int subtest __maybeCHECK__ (evlist__mmap (evlist,UINT_MAX )); - pc = evlist->mmap[0 ].base; + pc = evlist->mmap[0 ].core.base; ret =perf_read_tsc_conversion (pc, &tc);if (ret) {if (ret == -EOPNOTSUPP ) {
tools/perf/arch/x86/util/intel-bts.c +1 -1
@@ -78,7 +78,7 @@ static int intel_bts_info_fill(struct auxtrace_record *itr,if (!session->evlist->nr_mmaps)return -EINVAL ; - pc = session->evlist->mmap[0 ].base; + pc = session->evlist->mmap[0 ].core.base;if (pc) { err =perf_read_tsc_conversion (pc, &tc);if (err) {
tools/perf/arch/x86/util/intel-pt.c +1 -1
@@ -355,7 +355,7 @@ static int intel_pt_info_fill(struct auxtrace_record *itr,if (!session->evlist->nr_mmaps)return -EINVAL ; - pc = session->evlist->mmap[0 ].base; + pc = session->evlist->mmap[0 ].core.base;if (pc) { err =perf_read_tsc_conversion (pc, &tc);if (err) {
tools/perf/builtin-record.c +7 -7
@@ -261,7 +261,7 @@ static int record__aio_pushfn(struct mmap *map, void *to, void *buf, size_t sizestruct record_aio *aio = to;/* -* map->base data pointed by buf is copied into free map->aio.data[] buffer +* map->core.base data pointed by buf is copied into free map->aio.data[] buffer * to release space in the kernel buffer as fast as possible, calling * perf_mmap__consume() from perf_mmap__push() function. *
@@ -360,7 +360,7 @@ static void record__aio_mmap_read_sync(struct record *rec)for (i =0 ; i < evlist->nr_mmaps; i++) {struct mmap *map = &maps[i]; -if (map->base) +if (map->core.base)record__aio_sync (map, true); } }
@@ -970,7 +970,7 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,u64 flush =0 ;struct mmap *map = &maps[i]; -if (map->base) { +if (map->core.base) {record__adjust_affinity (rec, map);if (synch) { flush = map->flush;
@@ -1198,10 +1198,10 @@ static const struct perf_event_mmap_page * perf_evlist__pick_pc(struct evlist *evlist) {if (evlist) { -if (evlist->mmap && evlist->mmap[0 ].base) -return evlist->mmap[0 ].base; -if (evlist->overwrite_mmap && evlist->overwrite_mmap[0 ].base) -return evlist->overwrite_mmap[0 ].base; +if (evlist->mmap && evlist->mmap[0 ].core.base) +return evlist->mmap[0 ].core.base; +if (evlist->overwrite_mmap && evlist->overwrite_mmap[0 ].core.base) +return evlist->overwrite_mmap[0 ].core.base; }return NULL ; }
tools/perf/lib/include/internal/mmap.h new file +14 -0
@@ -0,0 +1,14 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#ifndef __LIBPERF_INTERNAL_MMAP_H +#define __LIBPERF_INTERNAL_MMAP_H + +/** +* struct perf_mmap - perf's ring buffer mmap details +* +* @refcnt - e.g. code using PERF_EVENT_IOC_SET_OUTPUT to share this +*/ +struct perf_mmap { +void *base; +}; + +#endif /* __LIBPERF_INTERNAL_MMAP_H */
tools/perf/util/mmap.c +11 -11
@@ -31,7 +31,7 @@ size_t perf_mmap__mmap_len(struct mmap *map)static union perf_event *perf_mmap__read (struct mmap *map,u64 *startp,u64 end) { -unsigned char *data = map->base + page_size; +unsigned char *data = map->core.base + page_size;union perf_event *event =NULL ;int diff = end - *startp;
@@ -116,7 +116,7 @@ void perf_mmap__get(struct mmap *map)void perf_mmap__put (struct mmap *map) { -BUG_ON (map->base &&refcount_read (&map->refcnt) ==0 ); +BUG_ON (map->core.base &&refcount_read (&map->refcnt) ==0 );if (refcount_dec_and_test (&map->refcnt))perf_mmap__munmap (map);
@@ -317,9 +317,9 @@ void perf_mmap__munmap(struct mmap *map)munmap (map->data,perf_mmap__mmap_len (map)); map->data =NULL ; } -if (map->base !=NULL ) { -munmap (map->base,perf_mmap__mmap_len (map)); - map->base =NULL ; +if (map->core.base !=NULL ) { +munmap (map->core.base,perf_mmap__mmap_len (map)); + map->core.base =NULL ; map->fd =-1 ;refcount_set (&map->refcnt,0 ); }
@@ -370,12 +370,12 @@ int perf_mmap__mmap(struct mmap *map, struct mmap_params *mp, int fd, int cpu)refcount_set (&map->refcnt,2 ); map->prev =0 ; map->mask = mp->mask; - map->base =mmap (NULL ,perf_mmap__mmap_len (map), mp->prot, + map->core.base =mmap (NULL ,perf_mmap__mmap_len (map), mp->prot,MAP_SHARED , fd,0 ); -if (map->base ==MAP_FAILED ) { +if (map->core.base ==MAP_FAILED ) {pr_debug2 ("failed to mmap perf event ring buffer, error %d\n" , errno); - map->base =NULL ; + map->core.base =NULL ;return -1 ; } map->fd = fd;
@@ -399,7 +399,7 @@ int perf_mmap__mmap(struct mmap *map, struct mmap_params *mp, int fd, int cpu) }if (auxtrace_mmap__mmap (&map->auxtrace_mmap, - &mp->auxtrace_mp, map->base, fd)) + &mp->auxtrace_mp, map->core.base, fd))return -1 ;return perf_mmap__aio_mmap (map, mp);
@@ -444,7 +444,7 @@ static int __perf_mmap__read_init(struct mmap *md) {u64 head =perf_mmap__read_head (md);u64 old = md->prev; -unsigned char *data = md->base + page_size; +unsigned char *data = md->core.base + page_size;unsigned long size; md->start = md->overwrite ? head : old;
@@ -489,7 +489,7 @@ int perf_mmap__push(struct mmap *md, void *to, int push(struct mmap *map, void *to, void *buf, size_t size)) {u64 head =perf_mmap__read_head (md); -unsigned char *data = md->base + page_size; +unsigned char *data = md->core.base + page_size;unsigned long size;void *buf;int rc =0 ;
tools/perf/util/mmap.h +4 -3
@@ -1,6 +1,7 @@#ifndef __PERF_MMAP_H#define __PERF_MMAP_H 1 +#include <internal/mmap.h> #include <linux/compiler.h> #include <linux/refcount.h> #include <linux/types.h>
@@ -20,7 +21,7 @@ struct aiocb;* @refcnt - e.g. code using PERF_EVENT_IOC_SET_OUTPUT to share this */ struct mmap { -void *base; +struct perf_mmap core;int mask;int fd;int cpu;
@@ -60,12 +61,12 @@ void perf_mmap__consume(struct mmap *map);static inline u64 perf_mmap__read_head (struct mmap *mm) { -return ring_buffer_read_head (mm->base); +return ring_buffer_read_head (mm->core.base); }static inline void perf_mmap__write_tail (struct mmap *md,u64 tail) { -ring_buffer_write_tail (md->base, tail); +ring_buffer_write_tail (md->core.base, tail); }union perf_event *perf_mmap__read_forward (struct mmap *map);
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 24/66 libperf: Add 'mask' to struct perf_mmap
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'mask' from tools/perf's mmap to libperf's perf_mmap struct.
--- tools/perf/lib/include/internal/mmap.h | 1 + tools/perf/util/mmap.c | 24 ++++++++++++------------ tools/perf/util/mmap.h | 1 - 3 files changed, 13 insertions(+), 13 deletions(-)
tools/perf/lib/include/internal/mmap.h +1 -0
@@ -9,6 +9,7 @@*/ struct perf_mmap {void *base; +int mask; };#endif /* __LIBPERF_INTERNAL_MMAP_H */
tools/perf/util/mmap.c +12 -12
@@ -24,7 +24,7 @@size_t perf_mmap__mmap_len (struct mmap *map) { -return map->mask +1 + page_size; +return map->core.mask +1 + page_size; }/* When check_messup is true, 'end' must points to a good entry */
@@ -38,7 +38,7 @@ static union perf_event *perf_mmap__read(struct mmap *map,if (diff >= (int )sizeof (event->header)) {size_t size; - event = (union perf_event *)&data[*startp & map->mask]; + event = (union perf_event *)&data[*startp & map->core.mask]; size = event->header.size;if (size <sizeof (event->header) || diff < (int )size)
@@ -48,14 +48,14 @@ static union perf_event *perf_mmap__read(struct mmap *map,* Event straddles the mmap boundary -- header should always * be inside due to u64 alignment of output. */ -if ((*startp & map->mask) + size != ((*startp + size) & map->mask)) { +if ((*startp & map->core.mask) + size != ((*startp + size) & map->core.mask)) {unsigned int offset = *startp;unsigned int len =min (sizeof (*event), size), cpy;void *dst = map->event_copy; do { - cpy = min(map->mask + 1 - (offset & map->mask), len); - memcpy(dst, &data[offset & map->mask], cpy); + cpy = min(map->core.mask + 1 - (offset & map->core.mask), len); + memcpy(dst, &data[offset & map->core.mask], cpy); offset += cpy; dst += cpy; len -= cpy;
@@ -369,7 +369,7 @@ int perf_mmap__mmap(struct mmap *map, struct mmap_params *mp, int fd, int cpu)*/ refcount_set (&map->refcnt,2 ); map->prev =0 ; - map->mask = mp->mask; + map->core.mask = mp->mask; map->core.base =mmap (NULL ,perf_mmap__mmap_len (map), mp->prot,MAP_SHARED , fd,0 );if (map->core.base ==MAP_FAILED ) {
@@ -454,7 +454,7 @@ static int __perf_mmap__read_init(struct mmap *md)return -EAGAIN ; size = md->end - md->start; -if (size > (unsigned long )(md->mask) +1 ) { +if (size > (unsigned long )(md->core.mask) +1 ) {if (!md->overwrite) {WARN_ONCE (1 ,"failed to keep up with mmap data. (warn only once)\n" );
@@ -467,7 +467,7 @@ static int __perf_mmap__read_init(struct mmap *md)* Backward ring buffer is full. We still have a chance to read * most of data from it. */ -if (overwrite_rb_find_range (data, md->mask, &md->start, &md->end)) +if (overwrite_rb_find_range (data, md->core.mask, &md->start, &md->end))return -EINVAL ; }
@@ -500,9 +500,9 @@ int perf_mmap__push(struct mmap *md, void *to, size = md->end - md->start; -if ((md->start & md->mask) + size != (md->end & md->mask)) { - buf = &data[md->start & md->mask]; - size = md->mask +1 - (md->start & md->mask); +if ((md->start & md->core.mask) + size != (md->end & md->core.mask)) { + buf = &data[md->start & md->core.mask]; + size = md->core.mask +1 - (md->start & md->core.mask); md->start += size;if (push (md, to, buf, size) <0 ) {
@@ -511,7 +511,7 @@ int perf_mmap__push(struct mmap *md, void *to, } } - buf = &data[md->start & md->mask]; + buf = &data[md->start & md->core.mask]; size = md->end - md->start; md->start += size;
tools/perf/util/mmap.h +0 -1
@@ -22,7 +22,6 @@ struct aiocb;*/ struct mmap {struct perf_mmap core; -int mask;int fd;int cpu;refcount_t refcnt;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 25/66 libperf: Add 'fd' to struct perf_mmap
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'fd' from tools/perf's mmap to libperf's perf_mmap struct.
--- tools/perf/lib/include/internal/mmap.h | 1 + tools/perf/util/evlist.c | 4 ++-- tools/perf/util/mmap.c | 4 ++-- tools/perf/util/mmap.h | 1 - 4 files changed, 5 insertions(+), 5 deletions(-)
tools/perf/lib/include/internal/mmap.h +1 -0
@@ -10,6 +10,7 @@struct perf_mmap {void *base;int mask; +int fd; };#endif /* __LIBPERF_INTERNAL_MMAP_H */
tools/perf/util/evlist.c +2 -2
@@ -652,7 +652,7 @@ static int perf_evlist__set_paused(struct evlist *evlist, bool value)return 0 ;for (i =0 ; i < evlist->nr_mmaps; i++) { -int fd = evlist->overwrite_mmap[i].fd; +int fd = evlist->overwrite_mmap[i].core.fd;int err;if (fd <0 )
@@ -708,7 +708,7 @@ static struct mmap *evlist__alloc_mmap(struct evlist *evlist,return NULL ;for (i =0 ; i < evlist->nr_mmaps; i++) { - map[i].fd =-1 ; + map[i].core.fd =-1 ; map[i].overwrite = overwrite;/* * When the perf_mmap() call is made we grab one refcount, plus
tools/perf/util/mmap.c +2 -2
@@ -320,7 +320,7 @@ void perf_mmap__munmap(struct mmap *map)if (map->core.base !=NULL ) {munmap (map->core.base,perf_mmap__mmap_len (map)); map->core.base =NULL ; - map->fd =-1 ; + map->core.fd =-1 ;refcount_set (&map->refcnt,0 ); }auxtrace_mmap__munmap (&map->auxtrace_mmap);
@@ -378,7 +378,7 @@ int perf_mmap__mmap(struct mmap *map, struct mmap_params *mp, int fd, int cpu) map->core.base =NULL ;return -1 ; } - map->fd = fd; + map->core.fd = fd; map->cpu = cpu;perf_mmap__setup_affinity_mask (map, mp);
tools/perf/util/mmap.h +0 -1
@@ -22,7 +22,6 @@ struct aiocb;*/ struct mmap {struct perf_mmap core; -int fd;int cpu;refcount_t refcnt;u64 prev;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 26/66 libperf: Add 'cpu' to struct perf_mmap
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'cpu' from tools/perf's mmap to libperf's perf_mmap struct.
--- tools/perf/lib/include/internal/mmap.h | 1 + tools/perf/util/mmap.c | 8 ++++---- tools/perf/util/mmap.h | 1 - tools/perf/util/python.c | 2 +- 4 files changed, 6 insertions(+), 6 deletions(-)
tools/perf/lib/include/internal/mmap.h +1 -0
@@ -11,6 +11,7 @@ struct perf_mmap {void *base;int mask;int fd; + int cpu; };#endif /* __LIBPERF_INTERNAL_MMAP_H */
tools/perf/util/mmap.c +4 -4
@@ -256,7 +256,7 @@ static int perf_mmap__aio_mmap(struct mmap *map, struct mmap_params *mp)pr_debug2 ("failed to allocate data buffer area, error %m" );return -1 ; } - ret =perf_mmap__aio_bind (map, i, map->cpu, mp->affinity); + ret =perf_mmap__aio_bind (map, i, map->core.cpu, mp->affinity);if (ret ==-1 )return -1 ;/*
@@ -347,9 +347,9 @@ static void perf_mmap__setup_affinity_mask(struct mmap *map, struct mmap_params {CPU_ZERO (&map->affinity_mask);if (mp->affinity ==PERF_AFFINITY_NODE &&cpu__max_node () >1 ) -build_node_mask (cpu__get_node (map->cpu), &map->affinity_mask); +build_node_mask (cpu__get_node (map->core.cpu), &map->affinity_mask);else if (mp->affinity ==PERF_AFFINITY_CPU ) -CPU_SET (map->cpu, &map->affinity_mask); +CPU_SET (map->core.cpu, &map->affinity_mask); }int perf_mmap__mmap (struct mmap *map,struct mmap_params *mp,int fd,int cpu)
@@ -379,7 +379,7 @@ int perf_mmap__mmap(struct mmap *map, struct mmap_params *mp, int fd, int cpu)return -1 ; } map->core.fd = fd; - map->cpu = cpu; + map->core.cpu = cpu;perf_mmap__setup_affinity_mask (map, mp);
tools/perf/util/mmap.h +0 -1
@@ -22,7 +22,6 @@ struct aiocb;*/ struct mmap {struct perf_mmap core; -int cpu;refcount_t refcnt;u64 prev;u64 start;
tools/perf/util/python.c +1 -1
@@ -991,7 +991,7 @@ static struct mmap *get_md(struct evlist *evlist, int cpu)for (i =0 ; i < evlist->nr_mmaps; i++) {struct mmap *md = &evlist->mmap[i]; -if (md->cpu == cpu) +if (md->core.cpu == cpu)return md; }
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 27/66 libperf: Add 'refcnt' to struct perf_mmap
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'refcnt' from tools/perf's mmap to libperf's perf_mmap struct.
Committer notes:
Add the refcount.h include directive here, now it is needed.
--- tools/perf/lib/include/internal/mmap.h | 3 +++ tools/perf/util/evlist.c | 2 +- tools/perf/util/mmap.c | 18 +++++++++--------- tools/perf/util/mmap.h | 1 - 4 files changed, 13 insertions(+), 11 deletions(-)
tools/perf/lib/include/internal/mmap.h +3 -0
@@ -2,6 +2,8 @@#ifndef __LIBPERF_INTERNAL_MMAP_H#define __LIBPERF_INTERNAL_MMAP_H +#include <linux/refcount.h> +/** * struct perf_mmap - perf's ring buffer mmap details *
@@ -12,6 +14,7 @@ struct perf_mmap {int mask;int fd;int cpu; + refcount_t refcnt; };#endif /* __LIBPERF_INTERNAL_MMAP_H */
tools/perf/util/evlist.c +1 -1
@@ -719,7 +719,7 @@ static struct mmap *evlist__alloc_mmap(struct evlist *evlist,* Each PERF_EVENT_IOC_SET_OUTPUT points to this mmap and * thus does perf_mmap__get() on it. */ -refcount_set (&map[i].refcnt,0 ); +refcount_set (&map[i].core.refcnt,0 ); }return map; }
tools/perf/util/mmap.c +9 -9
@@ -89,7 +89,7 @@ union perf_event *perf_mmap__read_event(struct mmap *map)/* * Check if event was unmapped due to a POLLHUP/POLLERR. */ -if (!refcount_read (&map->refcnt)) +if (!refcount_read (&map->core.refcnt))return NULL ;/* non-overwirte doesn't pause the ringbuffer */
@@ -111,14 +111,14 @@ static bool perf_mmap__empty(struct mmap *map)void perf_mmap__get (struct mmap *map) { -refcount_inc (&map->refcnt); +refcount_inc (&map->core.refcnt); }void perf_mmap__put (struct mmap *map) { -BUG_ON (map->core.base &&refcount_read (&map->refcnt) ==0 ); +BUG_ON (map->core.base &&refcount_read (&map->core.refcnt) ==0 ); -if (refcount_dec_and_test (&map->refcnt)) +if (refcount_dec_and_test (&map->core.refcnt))perf_mmap__munmap (map); }
@@ -130,7 +130,7 @@ void perf_mmap__consume(struct mmap *map)perf_mmap__write_tail (map, old); } -if (refcount_read (&map->refcnt) ==1 &&perf_mmap__empty (map)) +if (refcount_read (&map->core.refcnt) ==1 &&perf_mmap__empty (map))perf_mmap__put (map); }
@@ -321,7 +321,7 @@ void perf_mmap__munmap(struct mmap *map)munmap (map->core.base,perf_mmap__mmap_len (map)); map->core.base =NULL ; map->core.fd =-1 ; -refcount_set (&map->refcnt,0 ); +refcount_set (&map->core.refcnt,0 ); }auxtrace_mmap__munmap (&map->auxtrace_mmap); }
@@ -367,7 +367,7 @@ int perf_mmap__mmap(struct mmap *map, struct mmap_params *mp, int fd, int cpu)* evlist layer can't just drop it when filtering events in * perf_evlist__filter_pollfd(). */ -refcount_set (&map->refcnt,2 ); +refcount_set (&map->core.refcnt,2 ); map->prev =0 ; map->core.mask = mp->mask; map->core.base = mmap(NULL, perf_mmap__mmap_len(map), mp->prot,
@@ -479,7 +479,7 @@ int perf_mmap__read_init(struct mmap *map)/* * Check if event was unmapped due to a POLLHUP/POLLERR. */ -if (!refcount_read (&map->refcnt)) +if (!refcount_read (&map->core.refcnt))return -ENOENT ;return __perf_mmap__read_init (map);
@@ -537,7 +537,7 @@ void perf_mmap__read_done(struct mmap *map)/* * Check if event was unmapped due to a POLLHUP/POLLERR. */ -if (!refcount_read (&map->refcnt)) +if (!refcount_read (&map->core.refcnt))return ; map->prev =perf_mmap__read_head (map);
tools/perf/util/mmap.h +0 -1
@@ -22,7 +22,6 @@ struct aiocb;*/ struct mmap {struct perf_mmap core; -refcount_t refcnt;u64 prev;u64 start;u64 end;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 28/66 libperf: Add prev/start/end to struct perf_mmap
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move prev/start/end from tools/perf's mmap to libperf's perf_mmap struct.
Committer notes:
Add linux/types.h as we use u64.
--- tools/perf/lib/include/internal/mmap.h | 4 +++ tools/perf/util/mmap.c | 50 +++++++++++++------------- tools/perf/util/mmap.h | 3 -- 3 files changed, 29 insertions(+), 28 deletions(-)
tools/perf/lib/include/internal/mmap.h +4 -0
@@ -3,6 +3,7 @@#define __LIBPERF_INTERNAL_MMAP_H#include <linux/refcount.h> +#include <linux/types.h> /** * struct perf_mmap - perf's ring buffer mmap details
@@ -15,6 +16,9 @@ struct perf_mmap {int fd;int cpu;refcount_t refcnt; +u64 prev; +u64 start; + u64 end; };#endif /* __LIBPERF_INTERNAL_MMAP_H */
tools/perf/util/mmap.c +25 -25
128 lines not on this page · raw message
tools/perf/util/mmap.h +0 -3
@@ -22,9 +22,6 @@ struct aiocb;*/ struct mmap {struct perf_mmap core; -u64 prev; -u64 start; -u64 end;bool overwrite;struct auxtrace_mmap auxtrace_mmap;char event_copy[PERF_SAMPLE_MAX_SIZE ]__aligned(8) ;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 29/66 libperf: Add 'overwrite' to 'struct perf_mmap'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'overwrite' from tools/perf's mmap to libperf's perf_mmap struct.
Committer notes:
Add stdbool.h as we start using 'bool'.
--- tools/perf/lib/include/internal/mmap.h | 2 ++ tools/perf/util/evlist.c | 2 +- tools/perf/util/mmap.c | 12 ++++++------ tools/perf/util/mmap.h | 1 - 4 files changed, 9 insertions(+), 8 deletions(-)
tools/perf/lib/include/internal/mmap.h +2 -0
@@ -4,6 +4,7 @@#include <linux/refcount.h> #include <linux/types.h> +#include <stdbool.h> /** * struct perf_mmap - perf's ring buffer mmap details
@@ -19,6 +20,7 @@ struct perf_mmap {u64 prev;u64 start;u64 end; + bool overwrite; };#endif /* __LIBPERF_INTERNAL_MMAP_H */
tools/perf/util/evlist.c +1 -1
@@ -709,7 +709,7 @@ static struct mmap *evlist__alloc_mmap(struct evlist *evlist,for (i =0 ; i < evlist->nr_mmaps; i++) { map[i].core.fd =-1 ; - map[i].overwrite = overwrite; + map[i].core.overwrite = overwrite;/* * When the perf_mmap() call is made we grab one refcount, plus * one extra to let perf_mmap__consume() get the last
tools/perf/util/mmap.c +6 -6
@@ -93,12 +93,12 @@ union perf_event *perf_mmap__read_event(struct mmap *map)return NULL ;/* non-overwirte doesn't pause the ringbuffer */ -if (!map->overwrite) +if (!map->core.overwrite) map->core.end =perf_mmap__read_head (map); event =perf_mmap__read (map, &map->core.start, map->core.end); -if (!map->overwrite) +if (!map->core.overwrite) map->core.prev = map->core.start;return event;
@@ -124,7 +124,7 @@ void perf_mmap__put(struct mmap *map)void perf_mmap__consume (struct mmap *map) { -if (!map->overwrite) { +if (!map->core.overwrite) {u64 old = map->core.prev;perf_mmap__write_tail (map, old);
@@ -447,15 +447,15 @@ static int __perf_mmap__read_init(struct mmap *md)unsigned char *data = md->core.base + page_size;unsigned long size; - md->core.start = md->overwrite ? head : old; - md->core.end = md->overwrite ? old : head; + md->core.start = md->core.overwrite ? head : old; + md->core.end = md->core.overwrite ? old : head;if ((md->core.end - md->core.start) < md->flush)return -EAGAIN ; size = md->core.end - md->core.start;if (size > (unsigned long )(md->core.mask) +1 ) { -if (!md->overwrite) { +if (!md->core.overwrite) {WARN_ONCE (1 ,"failed to keep up with mmap data. (warn only once)\n" ); md->core.prev = head;
tools/perf/util/mmap.h +0 -1
@@ -22,7 +22,6 @@ struct aiocb;*/ struct mmap {struct perf_mmap core; -bool overwrite;struct auxtrace_mmap auxtrace_mmap;char event_copy[PERF_SAMPLE_MAX_SIZE ]__aligned(8) ;#ifdef HAVE_AIO_SUPPORT
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 30/66 libperf: Add 'event_copy' to 'struct perf_mmap'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'event_copy' from tools/perf's mmap to libperf's perf_mmap struct.
Committer notes:
Add linux/compiler.h as we need it for '__aligned'.
--- tools/perf/lib/include/internal/mmap.h | 5 +++++ tools/perf/util/mmap.c | 4 ++-- tools/perf/util/mmap.h | 1 - 3 files changed, 7 insertions(+), 3 deletions(-)
tools/perf/lib/include/internal/mmap.h +5 -0
@@ -2,10 +2,14 @@#ifndef __LIBPERF_INTERNAL_MMAP_H#define __LIBPERF_INTERNAL_MMAP_H +#include <linux/compiler.h> #include <linux/refcount.h> #include <linux/types.h> #include <stdbool.h> +/* perf sample has 16 bits size limit */ +#define PERF_SAMPLE_MAX_SIZE (1 << 16) +/** * struct perf_mmap - perf's ring buffer mmap details *
@@ -21,6 +25,7 @@ struct perf_mmap {u64 start;u64 end;bool overwrite; + char event_copy[PERF_SAMPLE_MAX_SIZE] __aligned(8); };#endif /* __LIBPERF_INTERNAL_MMAP_H */
tools/perf/util/mmap.c +2 -2
@@ -51,7 +51,7 @@ static union perf_event *perf_mmap__read(struct mmap *map,if ((*startp & map->core.mask) + size != ((*startp + size) & map->core.mask)) {unsigned int offset = *startp;unsigned int len =min (sizeof (*event), size), cpy; -void *dst = map->event_copy; +void *dst = map->core.event_copy; do { cpy = min(map->core.mask + 1 - (offset & map->core.mask), len);
@@ -61,7 +61,7 @@ static union perf_event *perf_mmap__read(struct mmap *map, len -= cpy; }while (len); - event = (union perf_event *)map->event_copy; + event = (union perf_event *)map->core.event_copy; } *startp += size;
tools/perf/util/mmap.h +0 -1
@@ -23,7 +23,6 @@ struct aiocb;struct mmap {struct perf_mmap core;struct auxtrace_mmap auxtrace_mmap; -char event_copy[PERF_SAMPLE_MAX_SIZE ]__aligned(8) ; #ifdef HAVE_AIO_SUPPORT struct { void **data;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 31/66 libperf: Add 'flush' to 'struct perf_mmap'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'flush' from tools/perf's mmap to libperf's perf_mmap struct.
--- tools/perf/builtin-record.c | 10 +++++----- tools/perf/lib/include/internal/mmap.h | 1 + tools/perf/util/mmap.c | 4 ++-- tools/perf/util/mmap.h | 1 - 4 files changed, 8 insertions(+), 8 deletions(-)
tools/perf/builtin-record.c +5 -5
@@ -973,13 +973,13 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,if (map->core.base) {record__adjust_affinity (rec, map);if (synch) { - flush = map->flush; - map->flush =1 ; + flush = map->core.flush; + map->core.flush =1 ; }if (!record__aio_enabled (rec)) {if (perf_mmap__push (map, rec, record__pushfn) <0 ) {if (synch) - map->flush = flush; + map->core.flush = flush; rc =-1 ; goto out; }
@@ -987,13 +987,13 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,if (record__aio_push (rec, map, &off) <0 ) {record__aio_set_pos (trace_fd, off);if (synch) - map->flush = flush; + map->core.flush = flush; rc =-1 ; goto out; } }if (synch) - map->flush = flush; + map->core.flush = flush; } if (map->auxtrace_mmap.base && !rec->opts.auxtrace_snapshot_mode &&
tools/perf/lib/include/internal/mmap.h +1 -0
@@ -25,6 +25,7 @@ struct perf_mmap {u64 start;u64 end;bool overwrite; +u64 flush;char event_copy[PERF_SAMPLE_MAX_SIZE ]__aligned(8) ; };
tools/perf/util/mmap.c +2 -2
@@ -383,7 +383,7 @@ int perf_mmap__mmap(struct mmap *map, struct mmap_params *mp, int fd, int cpu)perf_mmap__setup_affinity_mask (map, mp); - map->flush = mp->flush; + map->core.flush = mp->flush; map->comp_level = mp->comp_level;
@@ -450,7 +450,7 @@ static int __perf_mmap__read_init(struct mmap *md) md->core.start = md->core.overwrite ? head : old; md->core.end = md->core.overwrite ? old : head; -if ((md->core.end - md->core.start) < md->flush) +if ((md->core.end - md->core.start) < md->core.flush)return -EAGAIN ; size = md->core.end - md->core.start;
tools/perf/util/mmap.h +0 -1
@@ -32,7 +32,6 @@ struct mmap { } aio;#endif cpu_set_t affinity_mask; -u64 flush;void *data;int comp_level; };
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 32/66 libperf: Move 'system_wide' from 'struct evsel' to 'struct perf_evsel'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move the 'system_wide 'member from perf's evsel to libperf's perf_evsel.
Committer notes:
Added stdbool.h as we now use bool here.
--- tools/perf/arch/x86/util/intel-pt.c | 4 ++-- tools/perf/builtin-script.c | 2 +- tools/perf/builtin-stat.c | 4 ++-- tools/perf/lib/include/internal/evsel.h | 2 ++ tools/perf/tests/switch-tracking.c | 6 +++--- tools/perf/util/evlist.c | 10 +++++----- tools/perf/util/evsel.c | 8 ++++---- tools/perf/util/evsel.h | 1 - tools/perf/util/parse-events.c | 2 +- tools/perf/util/stat.c | 2 +- 10 files changed, 21 insertions(+), 20 deletions(-)
tools/perf/arch/x86/util/intel-pt.c +2 -2
@@ -422,7 +422,7 @@ static int intel_pt_track_switches(struct evlist *evlist)perf_evsel__set_sample_bit (evsel,CPU );perf_evsel__set_sample_bit (evsel,TIME ); - evsel->system_wide = true; + evsel->core.system_wide = true; evsel->no_aux_samples = true; evsel->immediate = true;
@@ -723,7 +723,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr, switch_evsel->core.attr.sample_period =1 ; switch_evsel->core.attr.context_switch =1 ; - switch_evsel->system_wide = true; + switch_evsel->core.system_wide = true; switch_evsel->no_aux_samples = true; switch_evsel->immediate = true;
tools/perf/builtin-script.c +1 -1
@@ -1916,7 +1916,7 @@ static void __process_stat(struct evsel *counter, u64 tstamp)int cpu, thread;static int header_printed; -if (counter->system_wide) +if (counter->core.system_wide) nthreads =1 ;if (!header_printed) {
tools/perf/builtin-stat.c +2 -2
@@ -278,7 +278,7 @@ static int read_counter(struct evsel *counter, struct timespec *rs)if (!counter->supported)return -ENOENT ; -if (counter->system_wide) +if (counter->core.system_wide) nthreads =1 ;for (thread =0 ; thread < nthreads; thread++) {
@@ -1671,7 +1671,7 @@ static void setup_system_wide(int forks)struct evsel *counter;evlist__for_each_entry (evsel_list, counter) { -if (!counter->system_wide) +if (!counter->core.system_wide)return ; }
tools/perf/lib/include/internal/evsel.h +2 -0
@@ -4,6 +4,7 @@#include <linux/types.h> #include <linux/perf_event.h> +#include <stdbool.h> struct perf_cpu_map ;struct perf_thread_map ;
@@ -18,6 +19,7 @@ struct perf_evsel {/* parse modifier helper */ int nr_members; +bool system_wide; };int perf_evsel__alloc_fd (struct perf_evsel *evsel,int ncpus,int nthreads);
tools/perf/tests/switch-tracking.c +3 -3
@@ -144,7 +144,7 @@ static int process_sample_event(struct evlist *evlist,return err;/* * Check for no missing sched_switch events i.e. that the -* evsel->system_wide flag has worked. +* evsel->core.system_wide flag has worked. */ if (switch_tracking->tids[cpu] !=-1 && switch_tracking->tids[cpu] != prev_tid) {
@@ -316,7 +316,7 @@ static int process_events(struct evlist *evlist,* * This function implements a test that checks that sched_switch events and * tracking events can be recorded for a workload (current process) using the -* evsel->system_wide and evsel->tracking flags (respectively) with other events +* evsel->core.system_wide and evsel->tracking flags (respectively) with other events * sometimes enabled or disabled. */ int test__switch_tracking (struct test *test__maybe_unused ,int subtest__maybe_unused )
@@ -396,7 +396,7 @@ int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_perf_evsel__set_sample_bit (switch_evsel,CPU );perf_evsel__set_sample_bit (switch_evsel,TIME ); - switch_evsel->system_wide = true; + switch_evsel->core.system_wide = true; switch_evsel->no_aux_samples = true; switch_evsel->immediate = true;
tools/perf/util/evlist.c +5 -5
@@ -319,7 +319,7 @@ int perf_evlist__add_newtp(struct evlist *evlist,static int perf_evlist__nr_threads (struct evlist *evlist,struct evsel *evsel) { -if (evsel->system_wide) +if (evsel->core.system_wide)return 1 ;else return perf_thread_map__nr (evlist->core.threads);
@@ -410,7 +410,7 @@ int perf_evlist__alloc_pollfd(struct evlist *evlist)struct evsel *evsel;evlist__for_each_entry (evlist, evsel) { -if (evsel->system_wide) +if (evsel->core.system_wide) nfds += nr_cpus;else nfds += nr_cpus * nr_threads;
@@ -536,7 +536,7 @@ static void perf_evlist__set_sid_idx(struct evlist *evlist, sid->cpu = evlist->core.cpus->map[cpu]; else sid->cpu = -1; -if (!evsel->system_wide && evlist->core.threads && thread >=0 ) +if (!evsel->core.system_wide && evlist->core.threads && thread >=0 ) sid->tid =perf_thread_map__pid (evlist->core.threads, thread);else sid->tid =-1 ;
@@ -763,7 +763,7 @@ static int evlist__mmap_per_evsel(struct evlist *evlist, int idx, mp->prot &= ~PROT_WRITE ; } -if (evsel->system_wide && thread) +if (evsel->core.system_wide && thread)continue ; cpu =perf_cpu_map__idx (evsel->core.cpus, evlist_cpu);
@@ -793,7 +793,7 @@ static int evlist__mmap_per_evsel(struct evlist *evlist, int idx,* other events, so it should not need to be polled anyway. * Therefore don't add it for polling. */ -if (!evsel->system_wide && +if (!evsel->core.system_wide &&__perf_evlist__add_pollfd (evlist, fd, &maps[idx], revent) <0 ) {perf_mmap__put (&maps[idx]);return -1 ;
tools/perf/util/evsel.c +4 -4
@@ -1232,7 +1232,7 @@ int perf_evsel__alloc_id(struct evsel *evsel, int ncpus, int nthreads)if (ncpus ==0 || nthreads ==0 )return 0 ; -if (evsel->system_wide) +if (evsel->core.system_wide) nthreads =1 ; evsel->sample_id =xyarray__new (ncpus, nthreads,sizeof (struct perf_sample_id ));
@@ -1663,7 +1663,7 @@ static bool ignore_missing_thread(struct evsel *evsel,return false;/* The system wide setup does not work with threads. */ -if (evsel->system_wide) +if (evsel->core.system_wide)return false;/* The -ESRCH is perf event syscall errno for pid's not found. */
@@ -1772,7 +1772,7 @@ int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus, threads = empty_thread_map; } -if (evsel->system_wide) +if (evsel->core.system_wide) nthreads =1 ;else nthreads = threads->nr;
@@ -1819,7 +1819,7 @@ int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus,for (thread =0 ; thread < nthreads; thread++) {int fd, group_fd; -if (!evsel->cgrp && !evsel->system_wide) +if (!evsel->cgrp && !evsel->core.system_wide) pid =perf_thread_map__pid (threads, thread); group_fd =get_group_fd (evsel, cpu, thread);
tools/perf/util/evsel.h +0 -1
@@ -146,7 +146,6 @@ struct evsel {bool disabled;bool no_aux_samples;bool immediate; -bool system_wide;bool tracking;bool per_pkg;bool precise_max;
tools/perf/util/parse-events.c +1 -1
@@ -335,7 +335,7 @@ __add_event(struct list_head *list, int *idx, (*idx)++; evsel->core.cpus =perf_cpu_map__get (cpus); evsel->core.own_cpus =perf_cpu_map__get (cpus); - evsel->system_wide = pmu ? pmu->is_uncore : false; + evsel->core.system_wide = pmu ? pmu->is_uncore : false; evsel->auto_merge_stats = auto_merge_stats;if (name)
tools/perf/util/stat.c +1 -1
@@ -336,7 +336,7 @@ static int process_counter_maps(struct perf_stat_config *config,int ncpus =perf_evsel__nr_cpus (counter);int cpu, thread; -if (counter->system_wide) +if (counter->core.system_wide) nthreads =1 ;for (thread =0 ; thread < nthreads; thread++) {
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 33/66 libperf: Move 'nr_mmaps' from 'struct evlist' to 'struct perf_evlist'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Moving 'nr_mmaps' from 'struct evlist' to 'struct perf_evlist', it will be used in following patches.
--- tools/perf/arch/arm/util/cs-etm.c | 2 +- tools/perf/arch/arm64/util/arm-spe.c | 2 +- tools/perf/arch/x86/tests/perf-time-to-tsc.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 2 +- tools/perf/arch/x86/util/intel-pt.c | 2 +- tools/perf/builtin-kvm.c | 2 +- tools/perf/builtin-record.c | 6 +++--- tools/perf/builtin-top.c | 2 +- tools/perf/builtin-trace.c | 2 +- tools/perf/lib/include/internal/evlist.h | 1 + tools/perf/tests/backward-ring-buffer.c | 2 +- tools/perf/tests/bpf.c | 2 +- tools/perf/tests/code-reading.c | 2 +- tools/perf/tests/keep-tracking.c | 2 +- tools/perf/tests/openat-syscall-tp-fields.c | 2 +- tools/perf/tests/perf-record.c | 2 +- tools/perf/tests/switch-tracking.c | 2 +- tools/perf/util/evlist.c | 16 ++++++++-------- tools/perf/util/evlist.h | 1 - tools/perf/util/python.c | 2 +- 20 files changed, 28 insertions(+), 28 deletions(-)
tools/perf/arch/arm/util/cs-etm.c +1 -1
@@ -648,7 +648,7 @@ static int cs_etm_info_fill(struct auxtrace_record *itr,if (priv_size !=cs_etm_info_priv_size (itr, session->evlist))return -EINVAL ; -if (!session->evlist->nr_mmaps) +if (!session->evlist->core.nr_mmaps)return -EINVAL ;/* If the cpu_map is empty all online CPUs are involved */
tools/perf/arch/arm64/util/arm-spe.c +1 -1
@@ -51,7 +51,7 @@ static int arm_spe_info_fill(struct auxtrace_record *itr,if (priv_size !=ARM_SPE_AUXTRACE_PRIV_SIZE )return -EINVAL ; -if (!session->evlist->nr_mmaps) +if (!session->evlist->core.nr_mmaps)return -EINVAL ; auxtrace_info->type =PERF_AUXTRACE_ARM_SPE ;
tools/perf/arch/x86/tests/perf-time-to-tsc.c +1 -1
@@ -115,7 +115,7 @@ int test__perf_time_to_tsc(struct test *test __maybe_unused, int subtest __maybeevlist__disable (evlist); -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) { md = &evlist->mmap[i];if (perf_mmap__read_init (md) <0 )continue ;
tools/perf/arch/x86/util/intel-bts.c +1 -1
@@ -75,7 +75,7 @@ static int intel_bts_info_fill(struct auxtrace_record *itr,if (priv_size !=INTEL_BTS_AUXTRACE_PRIV_SIZE )return -EINVAL ; -if (!session->evlist->nr_mmaps) +if (!session->evlist->core.nr_mmaps)return -EINVAL ; pc = session->evlist->mmap[0 ].core.base;
tools/perf/arch/x86/util/intel-pt.c +1 -1
@@ -352,7 +352,7 @@ static int intel_pt_info_fill(struct auxtrace_record *itr, filter =intel_pt_find_filter (session->evlist, ptr->intel_pt_pmu); filter_str_len = filter ?strlen (filter) :0 ; -if (!session->evlist->nr_mmaps) +if (!session->evlist->core.nr_mmaps)return -EINVAL ; pc = session->evlist->mmap[0 ].core.base;
tools/perf/builtin-kvm.c +1 -1
@@ -802,7 +802,7 @@ static int perf_kvm__mmap_read(struct perf_kvm_stat *kvm)s64 n, ntotal =0 ;u64 flush_time =ULLONG_MAX , mmap_time; -for (i =0 ; i < kvm->evlist->nr_mmaps; i++) { +for (i =0 ; i < kvm->evlist->core.nr_mmaps; i++) { n =perf_kvm__mmap_read_idx (kvm, i, &mmap_time);if (n <0 )return -1 ;
tools/perf/builtin-record.c +3 -3
@@ -357,7 +357,7 @@ static void record__aio_mmap_read_sync(struct record *rec)if (!record__aio_enabled (rec))return ; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {struct mmap *map = &maps[i];if (map->core.base)
@@ -603,7 +603,7 @@ static int record__auxtrace_read_snapshot_all(struct record *rec)int i;int rc =0 ; -for (i =0 ; i < rec->evlist->nr_mmaps; i++) { +for (i =0 ; i < rec->evlist->core.nr_mmaps; i++) {struct mmap *map = &rec->evlist->mmap[i];if (!map->auxtrace_mmap.base)
@@ -966,7 +966,7 @@ static int record__mmap_read_evlist(struct record *rec, struct evlist *evlist,if (record__aio_enabled (rec)) off =record__aio_get_pos (trace_fd); -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {u64 flush =0 ;struct mmap *map = &maps[i];
tools/perf/builtin-top.c +1 -1
@@ -904,7 +904,7 @@ static void perf_top__mmap_read(struct perf_top *top)if (overwrite)perf_evlist__toggle_bkw_mmap (evlist,BKW_MMAP_DATA_PENDING ); -for (i =0 ; i < top->evlist->nr_mmaps; i++) +for (i =0 ; i < top->evlist->core.nr_mmaps; i++)perf_top__mmap_read_idx (top, i);if (overwrite) {
tools/perf/builtin-trace.c +1 -1
@@ -3443,7 +3443,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) again: before = trace->nr_events; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {union perf_event *event;struct mmap *md;
tools/perf/lib/include/internal/evlist.h +1 -0
@@ -13,6 +13,7 @@ struct perf_evlist {bool has_user_cpus;struct perf_cpu_map *cpus;struct perf_thread_map *threads; +int nr_mmaps; };/**
tools/perf/tests/backward-ring-buffer.c +1 -1
@@ -33,7 +33,7 @@ static int count_samples(struct evlist *evlist, int *sample_count, {int i; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {struct mmap *map = &evlist->overwrite_mmap[i];union perf_event *event;
tools/perf/tests/bpf.c +1 -1
@@ -179,7 +179,7 @@ static int do_test(struct bpf_object *obj, int (*func)(void), (*func)();evlist__disable (evlist); -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {union perf_event *event;struct mmap *md;
tools/perf/tests/code-reading.c +1 -1
@@ -423,7 +423,7 @@ static int process_events(struct machine *machine, struct evlist *evlist,struct mmap *md;int i, ret; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) { md = &evlist->mmap[i];if (perf_mmap__read_init (md) <0 )continue ;
tools/perf/tests/keep-tracking.c +1 -1
@@ -36,7 +36,7 @@ static int find_comm(struct evlist *evlist, const char *comm)int i, found; found =0 ; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) { md = &evlist->mmap[i];if (perf_mmap__read_init (md) <0 )continue ;
tools/perf/tests/openat-syscall-tp-fields.c +1 -1
@@ -87,7 +87,7 @@ int test__syscall_openat_tp_fields(struct test *test __maybe_unused, int subtestwhile (1 ) {int before = nr_events; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {union perf_event *event;struct mmap *md;
tools/perf/tests/perf-record.c +1 -1
@@ -165,7 +165,7 @@ int test__PERF_RECORD(struct test *test __maybe_unused, int subtest __maybe_unuswhile (1 ) {int before = total_events; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {union perf_event *event;struct mmap *md;
tools/perf/tests/switch-tracking.c +1 -1
@@ -267,7 +267,7 @@ static int process_events(struct evlist *evlist,struct mmap *md;int i, ret; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) { md = &evlist->mmap[i];if (perf_mmap__read_init (md) <0 )continue ;
tools/perf/util/evlist.c +8 -8
@@ -651,7 +651,7 @@ static int perf_evlist__set_paused(struct evlist *evlist, bool value)if (!evlist->overwrite_mmap)return 0 ; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {int fd = evlist->overwrite_mmap[i].core.fd;int err;
@@ -679,11 +679,11 @@ static void evlist__munmap_nofree(struct evlist *evlist)int i;if (evlist->mmap) -for (i =0 ; i < evlist->nr_mmaps; i++) +for (i =0 ; i < evlist->core.nr_mmaps; i++)perf_mmap__munmap (&evlist->mmap[i]);if (evlist->overwrite_mmap) -for (i =0 ; i < evlist->nr_mmaps; i++) +for (i =0 ; i < evlist->core.nr_mmaps; i++)perf_mmap__munmap (&evlist->overwrite_mmap[i]); }
@@ -700,14 +700,14 @@ static struct mmap *evlist__alloc_mmap(struct evlist *evlist,int i;struct mmap *map; - evlist->nr_mmaps =perf_cpu_map__nr (evlist->core.cpus); + evlist->core.nr_mmaps =perf_cpu_map__nr (evlist->core.cpus);if (perf_cpu_map__empty (evlist->core.cpus)) - evlist->nr_mmaps =perf_thread_map__nr (evlist->core.threads); - map =zalloc (evlist->nr_mmaps *sizeof (struct mmap )); + evlist->core.nr_mmaps =perf_thread_map__nr (evlist->core.threads); + map =zalloc (evlist->core.nr_mmaps *sizeof (struct mmap ));if (!map)return NULL ; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) { map[i].core.fd =-1 ; map[i].core.overwrite = overwrite;/*
@@ -1847,7 +1847,7 @@ static void *perf_evlist__poll_thread(void *arg)if (!draining)perf_evlist__poll (evlist,1000 ); -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {struct mmap *map = &evlist->mmap[i];union perf_event *event;
tools/perf/util/evlist.h +0 -1
@@ -55,7 +55,6 @@ struct evlist {struct perf_evlist core;struct hlist_head heads[PERF_EVLIST__HLIST_SIZE ];int nr_groups; -int nr_mmaps;bool enabled;size_t mmap_len;int id_pos;
tools/perf/util/python.c +1 -1
@@ -988,7 +988,7 @@ static struct mmap *get_md(struct evlist *evlist, int cpu) {int i; -for (i =0 ; i < evlist->nr_mmaps; i++) { +for (i =0 ; i < evlist->core.nr_mmaps; i++) {struct mmap *md = &evlist->mmap[i];if (md->core.cpu == cpu)
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 34/66 libperf: Move 'mmap_len' from 'struct evlist' to 'struct perf_evlist'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Moving 'mmap_len' from 'struct evlist' to 'struct perf_evlist' it will be used in following patches.
--- tools/perf/builtin-record.c | 2 +- tools/perf/lib/include/internal/evlist.h | 1 + tools/perf/util/evlist.c | 10 +++++----- tools/perf/util/evlist.h | 1 - 4 files changed, 7 insertions(+), 7 deletions(-)
tools/perf/builtin-record.c +1 -1
@@ -1412,7 +1412,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) err =-1 ; goto out_child; } - session->header.env.comp_mmap_len = session->evlist->mmap_len; + session->header.env.comp_mmap_len = session->evlist->core.mmap_len; err =bpf__apply_obj_config ();if (err) {
tools/perf/lib/include/internal/evlist.h +1 -0
@@ -14,6 +14,7 @@ struct perf_evlist {struct perf_cpu_map *cpus;struct perf_thread_map *threads;int nr_mmaps; +size_t mmap_len; };/**
tools/perf/util/evlist.c +5 -5
@@ -1012,11 +1012,11 @@ int evlist__mmap_ex(struct evlist *evlist, unsigned int pages,if (evlist->pollfd.entries ==NULL &&perf_evlist__alloc_pollfd (evlist) <0 )return -ENOMEM ; - evlist->mmap_len =evlist__mmap_size (pages); -pr_debug ("mmap size %zuB\n" , evlist->mmap_len); - mp.mask = evlist->mmap_len - page_size -1 ; + evlist->core.mmap_len =evlist__mmap_size (pages); +pr_debug ("mmap size %zuB\n" , evlist->core.mmap_len); + mp.mask = evlist->core.mmap_len - page_size -1 ; -auxtrace_mmap_params__init (&mp.auxtrace_mp, evlist->mmap_len, +auxtrace_mmap_params__init (&mp.auxtrace_mp, evlist->core.mmap_len, auxtrace_pages, auxtrace_overwrite);evlist__for_each_entry (evlist, evsel) {
@@ -1600,7 +1600,7 @@ int perf_evlist__strerror_open(struct evlist *evlist,int perf_evlist__strerror_mmap (struct evlist *evlist,int err,char *buf,size_t size) {char sbuf[STRERR_BUFSIZE ], *emsg =str_error_r (err, sbuf,sizeof (sbuf)); -int pages_attempted = evlist->mmap_len /1024 , pages_max_per_user, printed =0 ; +int pages_attempted = evlist->core.mmap_len /1024 , pages_max_per_user, printed =0 ;switch (err) {case EPERM :
tools/perf/util/evlist.h +0 -1
@@ -56,7 +56,6 @@ struct evlist {struct hlist_head heads[PERF_EVLIST__HLIST_SIZE ];int nr_groups;bool enabled; -size_t mmap_len;int id_pos;int is_pos;u64 combined_sample_type;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 35/66 libperf: Move 'pollfd' from 'struct evlist' to 'struct perf_evlist'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Moving 'pollfd' from 'struct evlist' to 'struct perf_evlist' it will be used in following patches.
--- tools/perf/builtin-kvm.c | 2 +- tools/perf/lib/include/internal/evlist.h | 2 ++ tools/perf/util/evlist.c | 18 +++++++++--------- tools/perf/util/evlist.h | 1 - tools/perf/util/python.c | 6 +++--- 5 files changed, 15 insertions(+), 14 deletions(-)
tools/perf/builtin-kvm.c +1 -1
@@ -981,7 +981,7 @@ static int kvm_events_live_report(struct perf_kvm_stat *kvm)evlist__enable (kvm->evlist);while (!done) { -struct fdarray *fda = &kvm->evlist->pollfd; +struct fdarray *fda = &kvm->evlist->core.pollfd;int rc; rc =perf_kvm__mmap_read (kvm);
tools/perf/lib/include/internal/evlist.h +2 -0
@@ -3,6 +3,7 @@#define __LIBPERF_INTERNAL_EVLIST_H#include <linux/list.h> +#include <api/fd/array.h> struct perf_cpu_map ;struct perf_thread_map ;
@@ -15,6 +16,7 @@ struct perf_evlist {struct perf_thread_map *threads;int nr_mmaps;size_t mmap_len; +struct fdarray pollfd; };/**
tools/perf/util/evlist.c +9 -9
@@ -61,7 +61,7 @@ void evlist__init(struct evlist *evlist, struct perf_cpu_map *cpus,INIT_HLIST_HEAD (&evlist->heads[i]);perf_evlist__init (&evlist->core);perf_evlist__set_maps (&evlist->core, cpus, threads); -fdarray__init (&evlist->pollfd,64 ); +fdarray__init (&evlist->core.pollfd,64 ); evlist->workload.pid =-1 ; evlist->bkw_mmap_state =BKW_MMAP_NOTREADY ; }
@@ -142,7 +142,7 @@ void evlist__exit(struct evlist *evlist) {zfree (&evlist->mmap);zfree (&evlist->overwrite_mmap); -fdarray__exit (&evlist->pollfd); +fdarray__exit (&evlist->core.pollfd); }void evlist__delete (struct evlist *evlist)
@@ -416,8 +416,8 @@ int perf_evlist__alloc_pollfd(struct evlist *evlist) nfds += nr_cpus * nr_threads; } -if (fdarray__available_entries (&evlist->pollfd) < nfds && -fdarray__grow (&evlist->pollfd, nfds) <0 ) +if (fdarray__available_entries (&evlist->core.pollfd) < nfds && +fdarray__grow (&evlist->core.pollfd, nfds) <0 )return -ENOMEM ;return 0 ;
@@ -426,13 +426,13 @@ int perf_evlist__alloc_pollfd(struct evlist *evlist)static int __perf_evlist__add_pollfd (struct evlist *evlist,int fd,struct mmap *map,short revent) { -int pos =fdarray__add (&evlist->pollfd, fd, revent |POLLERR |POLLHUP ); +int pos =fdarray__add (&evlist->core.pollfd, fd, revent |POLLERR |POLLHUP );/* * Save the idx so that when we filter out fds POLLHUP'ed we can * close the associated evlist->mmap[] entry. */ if (pos >=0 ) { - evlist->pollfd.priv[pos].ptr = map; + evlist->core.pollfd.priv[pos].ptr = map;fcntl (fd,F_SETFL ,O_NONBLOCK ); }
@@ -456,13 +456,13 @@ static void perf_evlist__munmap_filtered(struct fdarray *fda, int fd,int perf_evlist__filter_pollfd (struct evlist *evlist,short revents_and_mask) { -return fdarray__filter (&evlist->pollfd, revents_and_mask, +return fdarray__filter (&evlist->core.pollfd, revents_and_mask, perf_evlist__munmap_filtered,NULL ); }int perf_evlist__poll (struct evlist *evlist,int timeout) { -return fdarray__poll (&evlist->pollfd, timeout); +return fdarray__poll (&evlist->core.pollfd, timeout); } static void perf_evlist__id_hash(struct evlist *evlist,
@@ -1009,7 +1009,7 @@ int evlist__mmap_ex(struct evlist *evlist, unsigned int pages,if (!evlist->mmap)return -ENOMEM ; -if (evlist->pollfd.entries ==NULL &&perf_evlist__alloc_pollfd (evlist) <0 ) +if (evlist->core.pollfd.entries ==NULL &&perf_evlist__alloc_pollfd (evlist) <0 )return -ENOMEM ; evlist->core.mmap_len =evlist__mmap_size (pages);
tools/perf/util/evlist.h +0 -1
@@ -64,7 +64,6 @@ struct evlist {int cork_fd;pid_t pid; } workload; -struct fdarray pollfd;struct mmap *mmap;struct mmap *overwrite_mmap;struct evsel *selected;
tools/perf/util/python.c +3 -3
@@ -935,17 +935,17 @@ static PyObject *pyrf_evlist__get_pollfd(struct pyrf_evlist *pevlist,PyObject *list =PyList_New (0 );int i; -for (i =0 ; i < evlist->pollfd.nr; ++i) { +for (i =0 ; i < evlist->core.pollfd.nr; ++i) {PyObject *file;#if PY_MAJOR_VERSION <3 -FILE *fp =fdopen (evlist->pollfd.entries[i].fd,"r" ); +FILE *fp =fdopen (evlist->core.pollfd.entries[i].fd,"r" );if (fp ==NULL ) goto free_list; file =PyFile_FromFile (fp,"perf" ,"r" ,NULL );#else - file =PyFile_FromFd (evlist->pollfd.entries[i].fd,"perf" ,"r" ,-1 , + file =PyFile_FromFd (evlist->core.pollfd.entries[i].fd,"perf" ,"r" ,-1 ,NULL ,NULL ,NULL ,0 );#endif if (file ==NULL )
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 36/66 libperf: Add missing 'struct xyarray' forward declaration
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
We were getting it by luck, from files included before internal/evsel.h where it is being included.
--- tools/perf/lib/include/internal/evsel.h | 1 + 1 file changed, 1 insertion(+)
tools/perf/lib/include/internal/evsel.h +1 -0
@@ -8,6 +8,7 @@struct perf_cpu_map ;struct perf_thread_map ; +struct xyarray ;struct perf_evsel {struct list_head node;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 37/66 libperf: Move 'sample_id' from 'struct evsel' to 'struct perf_evsel'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'sample_id' array from 'struct evsel' to libperf's 'struct perf_evsel'.
Committer notes:
Removed the 'struct xyarray' from util/evsel.h, not needed anymore there.
--- tools/perf/builtin-stat.c | 2 +- tools/perf/lib/include/internal/evsel.h | 1 + tools/perf/util/evlist.c | 4 ++-- tools/perf/util/evsel.c | 12 ++++++------ tools/perf/util/evsel.h | 2 -- 5 files changed, 10 insertions(+), 11 deletions(-)
tools/perf/builtin-stat.c +1 -1
@@ -235,7 +235,7 @@ static int write_stat_round_event(u64 tm, u64 type)#define WRITE_STAT_ROUND_EVENT (time, interval) \ write_stat_round_event(time, PERF_STAT_ROUND_TYPE__ ## interval) -#define SID (e, x, y) xyarray__entry(e->sample_id, x, y) +#define SID (e, x, y) xyarray__entry(e->core.sample_id, x, y) static int perf_evsel__write_stat_event(struct evsel *counter, u32 cpu, u32 thread,
tools/perf/lib/include/internal/evsel.h +1 -0
@@ -17,6 +17,7 @@ struct perf_evsel {struct perf_cpu_map *own_cpus;struct perf_thread_map *threads;struct xyarray *fd; +struct xyarray *sample_id;/* parse modifier helper */ int nr_members;
tools/perf/util/evlist.c +2 -2
@@ -50,7 +50,7 @@ int sigqueue(pid_t pid, int sig, const union sigval value);#endif #define FD (e, x, y) (*(int *)xyarray__entry(e->core.fd, x, y)) -#define SID (e, x, y) xyarray__entry(e->sample_id, x, y) +#define SID (e, x, y) xyarray__entry(e->core.sample_id, x, y)void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus,struct perf_thread_map *threads)
@@ -1021,7 +1021,7 @@ int evlist__mmap_ex(struct evlist *evlist, unsigned int pages,evlist__for_each_entry (evlist, evsel) {if ((evsel->core.attr.read_format &PERF_FORMAT_ID ) && - evsel->sample_id ==NULL && + evsel->core.sample_id ==NULL &&perf_evsel__alloc_id (evsel,perf_cpu_map__nr (cpus), threads->nr) <0 )return -ENOMEM ; }
tools/perf/util/evsel.c +6 -6
@@ -1235,14 +1235,14 @@ int perf_evsel__alloc_id(struct evsel *evsel, int ncpus, int nthreads)if (evsel->core.system_wide) nthreads =1 ; - evsel->sample_id =xyarray__new (ncpus, nthreads,sizeof (struct perf_sample_id )); -if (evsel->sample_id ==NULL ) + evsel->core.sample_id =xyarray__new (ncpus, nthreads,sizeof (struct perf_sample_id )); +if (evsel->core.sample_id ==NULL )return -ENOMEM ; evsel->id =zalloc (ncpus * nthreads *sizeof (u64));if (evsel->id ==NULL ) { -xyarray__delete (evsel->sample_id); - evsel->sample_id =NULL ; +xyarray__delete (evsel->core.sample_id); + evsel->core.sample_id =NULL ;return -ENOMEM ; }
@@ -1251,8 +1251,8 @@ int perf_evsel__alloc_id(struct evsel *evsel, int ncpus, int nthreads)static void perf_evsel__free_id (struct evsel *evsel) { -xyarray__delete (evsel->sample_id); - evsel->sample_id =NULL ; +xyarray__delete (evsel->core.sample_id); + evsel->core.sample_id =NULL ;zfree (&evsel->id); evsel->ids =0 ; }
tools/perf/util/evsel.h +0 -2
@@ -96,7 +96,6 @@ enum perf_tool_event {struct bpf_object ;struct perf_counts ; -struct xyarray ;/** struct evsel - event selector *
@@ -117,7 +116,6 @@ struct evsel {struct perf_evsel core;struct evlist *evlist;char *filter; -struct xyarray *sample_id;u64 *id;struct perf_counts *counts;struct perf_counts *prev_raw_counts;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 38/66 libperf: Move 'id' from 'struct evsel' to 'struct perf_evsel'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move the 'id' array from 'struct evsel' to libperf's 'struct perf_evsel'.
Committer note:
Fix the tools/perf/util/cs-etm.c build, i.e. aarch64's CoreSight.
--- tools/perf/lib/include/internal/evsel.h | 1 + tools/perf/util/cs-etm.c | 2 +- tools/perf/util/evlist.c | 2 +- tools/perf/util/evsel.c | 6 +++--- tools/perf/util/evsel.h | 1 - tools/perf/util/header.c | 10 +++++----- tools/perf/util/intel-bts.c | 2 +- tools/perf/util/intel-pt.c | 8 ++++---- tools/perf/util/synthetic-events.c | 12 ++++++------ 9 files changed, 22 insertions(+), 22 deletions(-)
tools/perf/lib/include/internal/evsel.h +1 -0
@@ -18,6 +18,7 @@ struct perf_evsel {struct perf_thread_map *threads;struct xyarray *fd;struct xyarray *sample_id; +u64 *id;/* parse modifier helper */ int nr_members;
tools/perf/util/cs-etm.c +1 -1
@@ -1298,7 +1298,7 @@ static int cs_etm__synth_events(struct cs_etm_auxtrace *etm, attr.read_format = evsel->core.attr.read_format;/* create new id val to be a fixed offset from evsel id */ - id = evsel->id[0 ] +1000000000 ; + id = evsel->core.id[0 ] +1000000000 ;if (!id) id =1 ;
tools/perf/util/evlist.c +1 -1
@@ -482,7 +482,7 @@ void perf_evlist__id_add(struct evlist *evlist, struct evsel *evsel, int cpu, int thread, u64 id) {perf_evlist__id_hash (evlist ,evsel ,cpu ,thread ,id ); - evsel->id[evsel->ids++] = id; + evsel->core.id[evsel->ids++] = id; } int perf_evlist__id_add_fd(struct evlist *evlist,
tools/perf/util/evsel.c +3 -3
@@ -1239,8 +1239,8 @@ int perf_evsel__alloc_id(struct evsel *evsel, int ncpus, int nthreads)if (evsel->core.sample_id ==NULL )return -ENOMEM ; - evsel->id =zalloc (ncpus * nthreads *sizeof (u64)); -if (evsel->id ==NULL ) { + evsel->core.id =zalloc (ncpus * nthreads *sizeof (u64)); +if (evsel->core.id ==NULL ) {xyarray__delete (evsel->core.sample_id); evsel->core.sample_id =NULL ;return -ENOMEM ;
@@ -1253,7 +1253,7 @@ static void perf_evsel__free_id(struct evsel *evsel) {xyarray__delete (evsel->core.sample_id); evsel->core.sample_id =NULL ; -zfree (&evsel->id); +zfree (&evsel->core.id); evsel->ids =0 ; }
tools/perf/util/evsel.h +0 -1
@@ -116,7 +116,6 @@ struct evsel {struct perf_evsel core;struct evlist *evlist;char *filter; -u64 *id;struct perf_counts *counts;struct perf_counts *prev_raw_counts;int idx;
tools/perf/util/header.c +5 -5
@@ -530,7 +530,7 @@ static int write_event_desc(struct feat_fd *ff,/* * write unique ids for this event */ - ret =do_write (ff, evsel->id, evsel->ids *sizeof (u64)); + ret =do_write (ff, evsel->core.id, evsel->ids *sizeof (u64));if (ret <0 )return ret; }
@@ -1590,7 +1590,7 @@ static void free_event_desc(struct evsel *events)for (evsel = events; evsel->core.attr.size; evsel++) {zfree (&evsel->name); -zfree (&evsel->id); +zfree (&evsel->core.id); }free (events);
@@ -1657,7 +1657,7 @@ static struct evsel *read_event_desc(struct feat_fd *ff)if (!id) goto error; evsel->ids = nr; - evsel->id = id; + evsel->core.id = id;for (j =0 ; j < nr; j++) {if (do_read_u64 (ff, id))
@@ -1701,7 +1701,7 @@ static void print_event_desc(struct feat_fd *ff, FILE *fp)if (evsel->ids) {fprintf (fp,", id = {" ); -for (j =0 , id = evsel->id; j < evsel->ids; j++, id++) { +for (j =0 , id = evsel->core.id; j < evsel->ids; j++, id++) {if (j)fputc (',' , fp);fprintf (fp," %" PRIu64, *id);
@@ -3068,7 +3068,7 @@ int perf_session__write_header(struct perf_session *session,evlist__for_each_entry (session->evlist, evsel) { evsel->id_offset =lseek (fd,0 ,SEEK_CUR ); - err =do_write (&ff, evsel->id, evsel->ids *sizeof (u64)); + err =do_write (&ff, evsel->core.id, evsel->ids *sizeof (u64));if (err <0 ) {pr_debug ("failed to write perf header\n" );return err;
tools/perf/util/intel-bts.c +1 -1
@@ -795,7 +795,7 @@ static int intel_bts_synth_events(struct intel_bts *bts, attr.sample_id_all = evsel->core.attr.sample_id_all; attr.read_format = evsel->core.attr.read_format; - id = evsel->id[0 ] +1000000000 ; + id = evsel->core.id[0 ] +1000000000 ;if (!id) id =1 ;
tools/perf/util/intel-pt.c +4 -4
@@ -1705,7 +1705,7 @@ static int intel_pt_synth_pebs_sample(struct intel_pt_queue *ptq)struct intel_pt *pt = ptq->pt;struct evsel *evsel = pt->pebs_evsel;u64 sample_type = evsel->core.attr.sample_type; -u64 id = evsel->id[0 ]; +u64 id = evsel->core.id[0 ];u8 cpumode;if (intel_pt_skip_event (pt))
@@ -2720,7 +2720,7 @@ static void intel_pt_set_event_name(struct evlist *evlist, u64 id,struct evsel *evsel;evlist__for_each_entry (evlist, evsel) { -if (evsel->id && evsel->id[0 ] == id) { +if (evsel->core.id && evsel->core.id[0 ] == id) {if (evsel->name)zfree (&evsel->name); evsel->name =strdup (name);
@@ -2776,7 +2776,7 @@ static int intel_pt_synth_events(struct intel_pt *pt, attr.sample_id_all = evsel->core.attr.sample_id_all; attr.read_format = evsel->core.attr.read_format; - id = evsel->id[0 ] +1000000000 ; + id = evsel->core.id[0 ] +1000000000 ;if (!id) id =1 ;
@@ -2903,7 +2903,7 @@ static void intel_pt_setup_pebs_events(struct intel_pt *pt)return ;evlist__for_each_entry (pt->session->evlist, evsel) { -if (evsel->core.attr.aux_output && evsel->id) { +if (evsel->core.attr.aux_output && evsel->core.id) { pt->sample_pebs = true; pt->pebs_evsel = evsel;return ;
tools/perf/util/synthetic-events.c +6 -6
@@ -1442,7 +1442,7 @@ int perf_event__synthesize_id_index(struct perf_tool *tool, perf_event__handler_ e = &ev->id_index.entries[i++]; - e->id = evsel->id[j]; + e->id = evsel->core.id[j]; sid =perf_evlist__id2sid (evlist, e->id);if (!sid) {
@@ -1515,7 +1515,7 @@ int perf_event__synthesize_event_update_unit(struct perf_tool *tool, struct evsestruct perf_record_event_update *ev;int err; - ev =event_update_event__new (size +1 ,PERF_EVENT_UPDATE__UNIT , evsel->id[0 ]); + ev =event_update_event__new (size +1 ,PERF_EVENT_UPDATE__UNIT , evsel->core.id[0 ]);if (ev ==NULL )return -ENOMEM ;
@@ -1532,7 +1532,7 @@ int perf_event__synthesize_event_update_scale(struct perf_tool *tool, struct evsstruct perf_record_event_update_scale *ev_data;int err; - ev =event_update_event__new (sizeof (*ev_data),PERF_EVENT_UPDATE__SCALE , evsel->id[0 ]); + ev =event_update_event__new (sizeof (*ev_data),PERF_EVENT_UPDATE__SCALE , evsel->core.id[0 ]);if (ev ==NULL )return -ENOMEM ;
@@ -1550,7 +1550,7 @@ int perf_event__synthesize_event_update_name(struct perf_tool *tool, struct evsesize_t len =strlen (evsel->name);int err; - ev =event_update_event__new (len +1 ,PERF_EVENT_UPDATE__NAME , evsel->id[0 ]); + ev =event_update_event__new (len +1 ,PERF_EVENT_UPDATE__NAME , evsel->core.id[0 ]);if (ev ==NULL )return -ENOMEM ;
@@ -1578,7 +1578,7 @@ int perf_event__synthesize_event_update_cpus(struct perf_tool *tool, struct evse ev->header.type =PERF_RECORD_EVENT_UPDATE ; ev->header.size = (u16 )size; ev->type =PERF_EVENT_UPDATE__CPUS ; - ev->id = evsel->id[0 ]; + ev->id = evsel->core.id[0 ];cpu_map_data__synthesize ((struct perf_record_cpu_map_data *)ev->data, evsel->core.own_cpus, type, max);
@@ -1596,7 +1596,7 @@ int perf_event__synthesize_attrs(struct perf_tool *tool, struct evlist *evlist,evlist__for_each_entry (evlist, evsel) { err =perf_event__synthesize_attr (tool, &evsel->core.attr, evsel->ids, - evsel->id, process); + evsel->core.id, process);if (err) {pr_debug ("failed to create perf header attribute\n" );return err;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 39/66 libperf: Move 'ids' from 'struct evsel' to 'struct perf_evsel'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'ids' from 'struct evsel' to libperf's 'struct perf_evsel'.
--- tools/perf/lib/include/internal/evsel.h | 1 + tools/perf/util/evlist.c | 2 +- tools/perf/util/evsel.c | 2 +- tools/perf/util/evsel.h | 1 - tools/perf/util/header.c | 14 +++++++------- tools/perf/util/intel-bts.c | 2 +- tools/perf/util/intel-pt.c | 2 +- tools/perf/util/synthetic-events.c | 6 +++--- 8 files changed, 15 insertions(+), 15 deletions(-)
tools/perf/lib/include/internal/evsel.h +1 -0
@@ -19,6 +19,7 @@ struct perf_evsel {struct xyarray *fd;struct xyarray *sample_id;u64 *id; +u32 ids;/* parse modifier helper */ int nr_members;
tools/perf/util/evlist.c +1 -1
@@ -482,7 +482,7 @@ void perf_evlist__id_add(struct evlist *evlist, struct evsel *evsel, int cpu, int thread, u64 id) {perf_evlist__id_hash (evlist ,evsel ,cpu ,thread ,id ); - evsel->core.id[evsel->ids++] = id; + evsel->core.id[evsel->core.ids++] = id; } int perf_evlist__id_add_fd(struct evlist *evlist,
tools/perf/util/evsel.c +1 -1
@@ -1254,7 +1254,7 @@ static void perf_evsel__free_id(struct evsel *evsel)xyarray__delete (evsel->core.sample_id); evsel->core.sample_id =NULL ;zfree (&evsel->core.id); - evsel->ids =0 ; + evsel->core.ids =0 ; }static void perf_evsel__free_config_terms (struct evsel *evsel)
tools/perf/util/evsel.h +0 -1
@@ -119,7 +119,6 @@ struct evsel {struct perf_counts *counts;struct perf_counts *prev_raw_counts;int idx; -u32 ids;unsigned long max_events;unsigned long nr_events_printed;char *name;
tools/perf/util/header.c +7 -7
@@ -516,7 +516,7 @@ static int write_event_desc(struct feat_fd *ff,* copy into an nri to be independent of the * type of ids, */ - nri = evsel->ids; + nri = evsel->core.ids; ret =do_write (ff, &nri,sizeof (nri));if (ret <0 )return ret;
@@ -530,7 +530,7 @@ static int write_event_desc(struct feat_fd *ff,/* * write unique ids for this event */ - ret =do_write (ff, evsel->core.id, evsel->ids *sizeof (u64)); + ret =do_write (ff, evsel->core.id, evsel->core.ids *sizeof (u64));if (ret <0 )return ret; }
@@ -1656,7 +1656,7 @@ static struct evsel *read_event_desc(struct feat_fd *ff) id =calloc (nr,sizeof (*id));if (!id) goto error; - evsel->ids = nr; + evsel->core.ids = nr; evsel->core.id = id;for (j =0 ; j < nr; j++) {
@@ -1699,9 +1699,9 @@ static void print_event_desc(struct feat_fd *ff, FILE *fp)for (evsel = events; evsel->core.attr.size; evsel++) {fprintf (fp,"# event : name = %s, " , evsel->name); -if (evsel->ids) { +if (evsel->core.ids) {fprintf (fp,", id = {" ); -for (j =0 , id = evsel->core.id; j < evsel->ids; j++, id++) { +for (j =0 , id = evsel->core.id; j < evsel->core.ids; j++, id++) {if (j)fputc (',' , fp);fprintf (fp," %" PRIu64, *id);
@@ -3068,7 +3068,7 @@ int perf_session__write_header(struct perf_session *session,evlist__for_each_entry (session->evlist, evsel) { evsel->id_offset =lseek (fd,0 ,SEEK_CUR ); - err =do_write (&ff, evsel->core.id, evsel->ids *sizeof (u64)); + err =do_write (&ff, evsel->core.id, evsel->core.ids *sizeof (u64));if (err <0 ) {pr_debug ("failed to write perf header\n" );return err;
@@ -3082,7 +3082,7 @@ int perf_session__write_header(struct perf_session *session, .attr = evsel->core.attr, .ids = { .offset = evsel->id_offset, - .size = evsel->ids * sizeof(u64), + .size = evsel->core.ids * sizeof(u64), } }; err =do_write (&ff, &f_attr,sizeof (f_attr));
tools/perf/util/intel-bts.c +1 -1
@@ -768,7 +768,7 @@ static int intel_bts_synth_events(struct intel_bts *bts,int err;evlist__for_each_entry (evlist, evsel) { -if (evsel->core.attr.type == bts->pmu_type && evsel->ids) { +if (evsel->core.attr.type == bts->pmu_type && evsel->core.ids) { found = true;break ; }
tools/perf/util/intel-pt.c +1 -1
@@ -2735,7 +2735,7 @@ static struct evsel *intel_pt_evsel(struct intel_pt *pt,struct evsel *evsel;evlist__for_each_entry (evlist, evsel) { -if (evsel->core.attr.type == pt->pmu_type && evsel->ids) +if (evsel->core.attr.type == pt->pmu_type && evsel->core.ids)return evsel; }
tools/perf/util/synthetic-events.c +3 -3
@@ -1413,7 +1413,7 @@ int perf_event__synthesize_id_index(struct perf_tool *tool, perf_event__handler_sizeof (struct id_index_entry );evlist__for_each_entry (evlist, evsel) - nr += evsel->ids; + nr += evsel->core.ids; n = nr > max_nr ? max_nr : nr; sz =sizeof (struct perf_record_id_index ) + n *sizeof (struct id_index_entry );
@@ -1428,7 +1428,7 @@ int perf_event__synthesize_id_index(struct perf_tool *tool, perf_event__handler_evlist__for_each_entry (evlist, evsel) {u32 j; -for (j =0 ; j < evsel->ids; j++) { +for (j =0 ; j < evsel->core.ids; j++) {struct id_index_entry *e;struct perf_sample_id *sid;
@@ -1595,7 +1595,7 @@ int perf_event__synthesize_attrs(struct perf_tool *tool, struct evlist *evlist,int err =0 ;evlist__for_each_entry (evlist, evsel) { - err =perf_event__synthesize_attr (tool, &evsel->core.attr, evsel->ids, + err =perf_event__synthesize_attr (tool, &evsel->core.attr, evsel->core.ids, evsel->core.id, process);if (err) {pr_debug ("failed to create perf header attribute\n" );
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 40/66 libperf: Move 'heads' from 'struct evlist' to 'struct perf_evlist'
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move 'heads' hash table from 'struct evlist' to 'struct perf_evlist'.
--- tools/perf/lib/evlist.c | 4 ++++ tools/perf/lib/include/internal/evlist.h | 4 ++++ tools/perf/util/evlist.c | 10 +++------- tools/perf/util/evlist.h | 4 ---- 4 files changed, 11 insertions(+), 11 deletions(-)
tools/perf/lib/evlist.c +4 -0
@@ -11,6 +11,10 @@void perf_evlist__init (struct perf_evlist *evlist) { +int i; + +for (i =0 ; i <PERF_EVLIST__HLIST_SIZE ; ++i) +INIT_HLIST_HEAD (&evlist->heads[i]);INIT_LIST_HEAD (&evlist->entries); evlist->nr_entries =0 ; }
tools/perf/lib/include/internal/evlist.h +4 -0
@@ -5,6 +5,9 @@#include <linux/list.h> #include <api/fd/array.h> +#define PERF_EVLIST__HLIST_BITS 8 +#define PERF_EVLIST__HLIST_SIZE (1 << PERF_EVLIST__HLIST_BITS) +struct perf_cpu_map ;struct perf_thread_map ;
@@ -17,6 +20,7 @@ struct perf_evlist {int nr_mmaps;size_t mmap_len;struct fdarray pollfd; +struct hlist_head heads[PERF_EVLIST__HLIST_SIZE ]; };/**
tools/perf/util/evlist.c +3 -7
@@ -55,10 +55,6 @@ int sigqueue(pid_t pid, int sig, const union sigval value);void evlist__init (struct evlist *evlist,struct perf_cpu_map *cpus,struct perf_thread_map *threads) { -int i; - -for (i =0 ; i <PERF_EVLIST__HLIST_SIZE ; ++i) -INIT_HLIST_HEAD (&evlist->heads[i]);perf_evlist__init (&evlist->core);perf_evlist__set_maps (&evlist->core, cpus, threads);fdarray__init (&evlist->core.pollfd,64 );
@@ -475,7 +471,7 @@ static void perf_evlist__id_hash(struct evlist *evlist, sid->id = id; sid->evsel = evsel; hash =hash_64 (sid->id,PERF_EVLIST__HLIST_BITS ); -hlist_add_head (&sid->node, &evlist->heads[hash]); +hlist_add_head (&sid->node, &evlist->core.heads[hash]); } void perf_evlist__id_add(struct evlist *evlist, struct evsel *evsel,
@@ -549,7 +545,7 @@ struct perf_sample_id *perf_evlist__id2sid(struct evlist *evlist, u64 id)int hash; hash =hash_64 (id,PERF_EVLIST__HLIST_BITS ); - head = &evlist->heads[hash]; + head = &evlist->core.heads[hash];hlist_for_each_entry (sid, head, node)if (sid->id == id)
@@ -635,7 +631,7 @@ struct evsel *perf_evlist__event2evsel(struct evlist *evlist,return first; hash =hash_64 (id,PERF_EVLIST__HLIST_BITS ); - head = &evlist->heads[hash]; + head = &evlist->core.heads[hash];hlist_for_each_entry (sid, head, node) {if (sid->id == id)
tools/perf/util/evlist.h +0 -4
@@ -48,12 +48,8 @@ enum bkw_mmap_state { BKW_MMAP_EMPTY, }; -#define PERF_EVLIST__HLIST_BITS 8 -#define PERF_EVLIST__HLIST_SIZE (1 << PERF_EVLIST__HLIST_BITS) -struct evlist {struct perf_evlist core; -struct hlist_head heads[PERF_EVLIST__HLIST_SIZE ];int nr_groups;bool enabled;int id_pos;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 41/66 libperf: Add perf_evsel__alloc_id/perf_evsel__free_id functions
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Add perf_evsel__alloc_id()/perf_evsel__free_id() functions to libperf as internal functions.
Move 'struct perf_sample_id' to internal/evsel.h header and change 'struct perf_sample_id::evsel' to 'struct perf_evsel' and the related code that touches it.
--- tools/perf/lib/evsel.c | 30 +++++++++++++++++++++ tools/perf/lib/include/internal/evsel.h | 27 +++++++++++++++++++ tools/perf/tests/event_update.c | 2 +- tools/perf/util/evlist.c | 10 +++---- tools/perf/util/evsel.c | 36 +++---------------------- tools/perf/util/evsel.h | 25 ----------------- tools/perf/util/header.c | 4 +-- tools/perf/util/session.c | 4 ++- 8 files changed, 71 insertions(+), 67 deletions(-)
tools/perf/lib/evsel.c +30 -0
@@ -230,3 +230,33 @@ struct perf_event_attr *perf_evsel__attr(struct perf_evsel *evsel) {return &evsel->attr; } + +int perf_evsel__alloc_id (struct perf_evsel *evsel,int ncpus,int nthreads) +{ +if (ncpus ==0 || nthreads ==0 ) +return 0 ; + +if (evsel->system_wide) + nthreads =1 ; + + evsel->sample_id =xyarray__new (ncpus, nthreads,sizeof (struct perf_sample_id )); +if (evsel->sample_id ==NULL ) +return -ENOMEM ; + + evsel->id =zalloc (ncpus * nthreads *sizeof (u64)); +if (evsel->id ==NULL ) { +xyarray__delete (evsel->sample_id); + evsel->sample_id =NULL ; +return -ENOMEM ; + } + +return 0 ; +} + +void perf_evsel__free_id (struct perf_evsel *evsel) +{ +xyarray__delete (evsel->sample_id); + evsel->sample_id =NULL ; +zfree (&evsel->id); + evsel->ids =0 ; +}
tools/perf/lib/include/internal/evsel.h +27 -0
@@ -5,11 +5,35 @@#include <linux/types.h> #include <linux/perf_event.h> #include <stdbool.h> +#include <sys/types.h> struct perf_cpu_map ;struct perf_thread_map ;struct xyarray ; +/* +* Per fd, to map back from PERF_SAMPLE_ID to evsel, only used when there are +* more than one entry in the evlist. +*/ +struct perf_sample_id { +struct hlist_node node; +u64 id; +struct perf_evsel *evsel; +/* +* 'idx' will be used for AUX area sampling. A sample will have AUX area +* data that will be queued for decoding, where there are separate +* queues for each CPU (per-cpu tracing) or task (per-thread tracing). +* The sample ID can be used to lookup 'idx' which is effectively the +* queue number. +*/ +int idx; +int cpu; +pid_t tid; + +/* Holds total ID period value for PERF_SAMPLE_READ processing. */ +u64 period; +}; +struct perf_evsel {struct list_head node;struct perf_event_attr attr;
@@ -32,4 +56,7 @@ void perf_evsel__free_fd(struct perf_evsel *evsel);int perf_evsel__read_size (struct perf_evsel *evsel);int perf_evsel__apply_filter (struct perf_evsel *evsel,const char *filter); +int perf_evsel__alloc_id (struct perf_evsel *evsel,int ncpus,int nthreads); +void perf_evsel__free_id (struct perf_evsel *evsel); +#endif /* __LIBPERF_INTERNAL_EVSEL_H */
tools/perf/tests/event_update.c +1 -1
@@ -95,7 +95,7 @@ int test__event_update(struct test *test __maybe_unused, int subtest __maybe_unu evsel =perf_evlist__first (evlist);TEST_ASSERT_VAL ("failed to allocate ids" , - !perf_evsel__alloc_id (evsel,1 ,1 )); + !perf_evsel__alloc_id (&evsel->core,1 ,1 ));perf_evlist__id_add (evlist, evsel,0 ,0 ,123 );
tools/perf/util/evlist.c +5 -5
@@ -469,7 +469,7 @@ static void perf_evlist__id_hash(struct evlist *evlist,struct perf_sample_id *sid =SID (evsel, cpu, thread); sid->id = id; - sid->evsel = evsel; + sid->evsel = &evsel->core; hash =hash_64 (sid->id,PERF_EVLIST__HLIST_BITS );hlist_add_head (&sid->node, &evlist->core.heads[hash]); }
@@ -563,7 +563,7 @@ struct evsel *perf_evlist__id2evsel(struct evlist *evlist, u64 id) sid =perf_evlist__id2sid (evlist, id);if (sid) -return sid->evsel; + return container_of(sid->evsel, struct evsel, core);if (!perf_evlist__sample_id_all (evlist))return perf_evlist__first (evlist);
@@ -581,7 +581,7 @@ struct evsel *perf_evlist__id2evsel_strict(struct evlist *evlist, sid =perf_evlist__id2sid (evlist, id);if (sid) -return sid->evsel; + return container_of(sid->evsel, struct evsel, core);return NULL ; }
@@ -635,7 +635,7 @@ struct evsel *perf_evlist__event2evsel(struct evlist *evlist,hlist_for_each_entry (sid, head, node) {if (sid->id == id) -return sid->evsel; + return container_of(sid->evsel, struct evsel, core); }return NULL ; }
@@ -1018,7 +1018,7 @@ int evlist__mmap_ex(struct evlist *evlist, unsigned int pages,evlist__for_each_entry (evlist, evsel) {if ((evsel->core.attr.read_format &PERF_FORMAT_ID ) && evsel->core.sample_id ==NULL && -perf_evsel__alloc_id (evsel,perf_cpu_map__nr (cpus), threads->nr) <0 ) +perf_evsel__alloc_id (&evsel->core,perf_cpu_map__nr (cpus), threads->nr) <0 )return -ENOMEM ; }
tools/perf/util/evsel.c +3 -33
@@ -1227,36 +1227,6 @@ int evsel__disable(struct evsel *evsel)return err; } -int perf_evsel__alloc_id (struct evsel *evsel,int ncpus,int nthreads) -{ -if (ncpus ==0 || nthreads ==0 ) -return 0 ; - -if (evsel->core.system_wide) - nthreads =1 ; - - evsel->core.sample_id =xyarray__new (ncpus, nthreads,sizeof (struct perf_sample_id )); -if (evsel->core.sample_id ==NULL ) -return -ENOMEM ; - - evsel->core.id =zalloc (ncpus * nthreads *sizeof (u64)); -if (evsel->core.id ==NULL ) { -xyarray__delete (evsel->core.sample_id); - evsel->core.sample_id =NULL ; -return -ENOMEM ; - } - -return 0 ; -} - -static void perf_evsel__free_id (struct evsel *evsel) -{ -xyarray__delete (evsel->core.sample_id); - evsel->core.sample_id =NULL ; -zfree (&evsel->core.id); - evsel->core.ids =0 ; -} -static void perf_evsel__free_config_terms (struct evsel *evsel) {struct perf_evsel_config_term *term, *h;
@@ -1273,7 +1243,7 @@ void perf_evsel__exit(struct evsel *evsel)assert (evsel->evlist ==NULL );perf_evsel__free_counts (evsel);perf_evsel__free_fd (&evsel->core); -perf_evsel__free_id (evsel); +perf_evsel__free_id (&evsel->core);perf_evsel__free_config_terms (evsel);cgroup__put (evsel->cgrp);perf_cpu_map__put (evsel->core.cpus);
@@ -1992,7 +1962,7 @@ int evsel__open(struct evsel *evsel, struct perf_cpu_map *cpus,void evsel__close (struct evsel *evsel) {perf_evsel__close (&evsel->core); -perf_evsel__free_id (evsel); +perf_evsel__free_id (&evsel->core); } int perf_evsel__open_per_cpu(struct evsel *evsel,
@@ -2706,7 +2676,7 @@ int perf_evsel__store_ids(struct evsel *evsel, struct evlist *evlist)struct perf_cpu_map *cpus = evsel->core.cpus;struct perf_thread_map *threads = evsel->core.threads; -if (perf_evsel__alloc_id (evsel, cpus->nr, threads->nr)) +if (perf_evsel__alloc_id (&evsel->core, cpus->nr, threads->nr))return -ENOMEM ;return store_evsel_ids (evsel, evlist);
tools/perf/util/evsel.h +0 -25
@@ -17,29 +17,6 @@ struct addr_location;struct evsel ;union perf_event ; -/* -* Per fd, to map back from PERF_SAMPLE_ID to evsel, only used when there are -* more than one entry in the evlist. -*/ -struct perf_sample_id { -struct hlist_node node; -u64 id; -struct evsel *evsel; -/* -* 'idx' will be used for AUX area sampling. A sample will have AUX area -* data that will be queued for decoding, where there are separate -* queues for each CPU (per-cpu tracing) or task (per-thread tracing). -* The sample ID can be used to lookup 'idx' which is effectively the -* queue number. -*/ -int idx; -int cpu; -pid_t tid; - -/* Holds total ID period value for PERF_SAMPLE_READ processing. */ -u64 period; -}; -struct cgroup ;/*
@@ -272,8 +249,6 @@ const char *perf_evsel__name(struct evsel *evsel);const char *perf_evsel__group_name (struct evsel *evsel);int perf_evsel__group_desc (struct evsel *evsel,char *buf,size_t size); -int perf_evsel__alloc_id (struct evsel *evsel,int ncpus,int nthreads); -void __perf_evsel__set_sample_bit (struct evsel *evsel,enum perf_event_sample_format bit); void __perf_evsel__reset_sample_bit(struct evsel *evsel,
tools/perf/util/header.c +2 -2
@@ -3609,7 +3609,7 @@ int perf_session__read_header(struct perf_session *session)* for allocating the perf_sample_id table we fake 1 cpu and * hattr->ids threads. */ -if (perf_evsel__alloc_id (evsel,1 , nr_ids)) +if (perf_evsel__alloc_id (&evsel->core,1 , nr_ids)) goto out_delete_evlist;lseek (fd, f_attr.ids.offset,SEEK_SET );
@@ -3750,7 +3750,7 @@ int perf_event__process_attr(struct perf_tool *tool __maybe_unused,* for allocating the perf_sample_id table we fake 1 cpu and * hattr->ids threads. */ -if (perf_evsel__alloc_id (evsel,1 , n_ids)) +if (perf_evsel__alloc_id (&evsel->core,1 , n_ids))return -ENOMEM ;for (i =0 ; i < n_ids; i++) {
tools/perf/util/session.c +3 -1
@@ -1324,6 +1324,7 @@ static int deliver_sample_value(struct evlist *evlist, struct machine *machine) {struct perf_sample_id *sid =perf_evlist__id2sid (evlist, v->id); +struct evsel *evsel;if (sid) { sample->id = v->id;
@@ -1343,7 +1344,8 @@ static int deliver_sample_value(struct evlist *evlist,if (!sample->period)return 0 ; -return tool->sample (tool, event, sample, sid->evsel, machine); + evsel = container_of(sid->evsel, struct evsel, core); +return tool->sample (tool, event, sample, evsel, machine); } static int deliver_sample_group(struct evlist *evlist,
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 42/66 libperf: Add perf_evlist__first()/last() functions
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Add perf_evlist__first()/last() functions to libperf, as internal functions and rename perf's origins to evlist__first/last.
--- tools/perf/arch/arm/util/cs-etm.c | 2 +- tools/perf/arch/arm64/util/arm-spe.c | 2 +- tools/perf/arch/x86/tests/intel-cqm.c | 4 +- tools/perf/arch/x86/tests/perf-time-to-tsc.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 2 +- tools/perf/arch/x86/util/intel-pt.c | 6 +- tools/perf/builtin-record.c | 4 +- tools/perf/builtin-script.c | 2 +- tools/perf/builtin-top.c | 10 +- tools/perf/builtin-trace.c | 2 +- tools/perf/lib/include/internal/evlist.h | 11 ++ tools/perf/tests/code-reading.c | 2 +- tools/perf/tests/event-times.c | 14 +-- tools/perf/tests/event_update.c | 2 +- tools/perf/tests/evsel-roundtrip-name.c | 2 +- tools/perf/tests/hists_cumulate.c | 2 +- tools/perf/tests/hists_link.c | 4 +- tools/perf/tests/hists_output.c | 2 +- tools/perf/tests/keep-tracking.c | 4 +- tools/perf/tests/parse-events.c | 116 +++++++++---------- tools/perf/tests/perf-record.c | 2 +- tools/perf/tests/switch-tracking.c | 14 +-- tools/perf/tests/task-exit.c | 2 +- tools/perf/ui/browsers/hists.c | 6 +- tools/perf/util/bpf-loader.c | 2 +- tools/perf/util/evlist.c | 24 ++-- tools/perf/util/evlist.h | 13 ++- tools/perf/util/jitdump.c | 2 +- tools/perf/util/parse-events.c | 4 +- tools/perf/util/record.c | 6 +- tools/perf/util/sort.c | 2 +- tools/perf/util/top.c | 2 +- 32 files changed, 145 insertions(+), 129 deletions(-)
tools/perf/arch/arm/util/cs-etm.c +1 -1
@@ -416,7 +416,7 @@ static int cs_etm_recording_options(struct auxtrace_record *itr,if (err) goto out; - tracking_evsel =perf_evlist__last (evlist); + tracking_evsel =evlist__last (evlist);perf_evlist__set_tracking_event (evlist, tracking_evsel); tracking_evsel->core.attr.freq =0 ;
tools/perf/arch/arm64/util/arm-spe.c +1 -1
@@ -129,7 +129,7 @@ static int arm_spe_recording_options(struct auxtrace_record *itr,if (err)return err; - tracking_evsel =perf_evlist__last (evlist); + tracking_evsel =evlist__last (evlist);perf_evlist__set_tracking_event (evlist, tracking_evsel); tracking_evsel->core.attr.freq =0 ;
tools/perf/arch/x86/tests/intel-cqm.c +2 -2
@@ -62,9 +62,9 @@ int test__intel_cqm_count_nmi_context(struct test *test __maybe_unused, int subt goto out; } - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist);if (!evsel) { -pr_debug ("perf_evlist__first failed\n" ); +pr_debug ("evlist__first failed\n" ); goto out; }
tools/perf/arch/x86/tests/perf-time-to-tsc.c +1 -1
@@ -83,7 +83,7 @@ int test__perf_time_to_tsc(struct test *test __maybe_unused, int subtest __maybeperf_evlist__config (evlist, &opts,NULL ); - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist); evsel->core.attr.comm =1 ; evsel->core.attr.disabled =1 ;
tools/perf/arch/x86/util/intel-bts.c +1 -1
@@ -231,7 +231,7 @@ static int intel_bts_recording_options(struct auxtrace_record *itr,if (err)return err; - tracking_evsel =perf_evlist__last (evlist); + tracking_evsel =evlist__last (evlist);perf_evlist__set_tracking_event (evlist, tracking_evsel);
tools/perf/arch/x86/util/intel-pt.c +3 -3
@@ -417,7 +417,7 @@ static int intel_pt_track_switches(struct evlist *evlist)return err; } - evsel =perf_evlist__last (evlist); + evsel =evlist__last (evlist);perf_evsel__set_sample_bit (evsel,CPU );perf_evsel__set_sample_bit (evsel,TIME );
@@ -717,7 +717,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr,if (err)return err; - switch_evsel =perf_evlist__last (evlist); + switch_evsel =evlist__last (evlist); switch_evsel->core.attr.freq =0 ; switch_evsel->core.attr.sample_period =1 ;
@@ -775,7 +775,7 @@ static int intel_pt_recording_options(struct auxtrace_record *itr,if (err)return err; - tracking_evsel =perf_evlist__last (evlist); + tracking_evsel =evlist__last (evlist);perf_evlist__set_tracking_event (evlist, tracking_evsel);
tools/perf/builtin-record.c +2 -2
@@ -756,9 +756,9 @@ static int record__open(struct record *rec)if (perf_evlist__add_dummy (evlist))return -ENOMEM ; - pos =perf_evlist__first (evlist); + pos =evlist__first (evlist); pos->tracking =0 ; - pos =perf_evlist__last (evlist); + pos =evlist__last (evlist); pos->tracking =1 ; pos->core.attr.enable_on_exec =1 ; }
tools/perf/builtin-script.c +1 -1
@@ -2043,7 +2043,7 @@ static int process_attr(struct perf_tool *tool, union perf_event *event,return err; evlist = *pevlist; - evsel =perf_evlist__last (*pevlist); + evsel =evlist__last (*pevlist);if (!evsel->priv) {if (scr->per_event_dump) {
tools/perf/builtin-top.c +5 -5
@@ -531,7 +531,7 @@ static bool perf_top__handle_keypress(struct perf_top *top, int c)prompt_integer (&counter,"Enter details event counter" );if (counter >= top->evlist->core.nr_entries) { - top->sym_evsel =perf_evlist__first (top->evlist); + top->sym_evsel =evlist__first (top->evlist);fprintf (stderr,"Sorry, no such event, using %s.\n" ,perf_evsel__name (top->sym_evsel));sleep (1 );break ;
@@ -540,7 +540,7 @@ static bool perf_top__handle_keypress(struct perf_top *top, int c)if (top->sym_evsel->idx == counter)break ; }else - top->sym_evsel =perf_evlist__first (top->evlist); + top->sym_evsel =evlist__first (top->evlist);break ;case 'f' :prompt_integer (&top->count_filter,"Enter display event count filter" );
@@ -962,7 +962,7 @@ static int perf_top__overwrite_check(struct perf_top *top)/* has term for current event */ if ((overwrite <0 ) && (set >=0 )) {/* if it's first event, set overwrite */ -if (evsel ==perf_evlist__first (evlist)) +if (evsel ==evlist__first (evlist)) overwrite = set;else return -1 ;
@@ -986,7 +986,7 @@ static int perf_top_overwrite_fallback(struct perf_top *top,return 0 ;/* only fall back when first event fails */ -if (evsel !=perf_evlist__first (evlist)) +if (evsel !=evlist__first (evlist))return 0 ;evlist__for_each_entry (evlist, counter)
@@ -1644,7 +1644,7 @@ int cmd_top(int argc, const char **argv) goto out_delete_evlist; } - top.sym_evsel =perf_evlist__first (top.evlist); + top.sym_evsel =evlist__first (top.evlist);if (!callchain_param.enabled) { symbol_conf.cumulate_callchain = false;
tools/perf/builtin-trace.c +1 -1
@@ -3427,7 +3427,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv) trace->multiple_threads =perf_thread_map__pid (evlist->core.threads,0 ) ==-1 || evlist->core.threads->nr >1 || -perf_evlist__first (evlist)->core.attr.inherit; +evlist__first (evlist)->core.attr.inherit;/* * Now that we already used evsel->core.attr to ask the kernel to setup the
tools/perf/lib/include/internal/evlist.h +11 -0
@@ -4,6 +4,7 @@#include <linux/list.h> #include <api/fd/array.h> +#include <internal/evsel.h> #define PERF_EVLIST__HLIST_BITS 8#define PERF_EVLIST__HLIST_SIZE (1 << PERF_EVLIST__HLIST_BITS)
@@ -55,4 +56,14 @@ struct perf_evlist {#define perf_evlist__for_each_entry_reverse (evlist, evsel) \ __perf_evlist__for_each_entry_reverse(&(evlist)->entries, evsel) +static inline struct perf_evsel *perf_evlist__first (struct perf_evlist *evlist) +{ + return list_entry(evlist->entries.next, struct perf_evsel, node); +} + +static inline struct perf_evsel *perf_evlist__last (struct perf_evlist *evlist) +{ + return list_entry(evlist->entries.prev, struct perf_evsel, node); +} +#endif /* __LIBPERF_INTERNAL_EVLIST_H */
tools/perf/tests/code-reading.c +1 -1
@@ -652,7 +652,7 @@ static int do_test_code_reading(bool try_kcore)perf_evlist__config (evlist, &opts,NULL ); - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist); evsel->core.attr.comm =1 ; evsel->core.attr.disabled =1 ;
tools/perf/tests/event-times.c +7 -7
@@ -16,7 +16,7 @@static int attach__enable_on_exec (struct evlist *evlist) { -struct evsel *evsel =perf_evlist__last (evlist); +struct evsel *evsel =evlist__last (evlist);struct target target = { .uid =UINT_MAX , };
@@ -58,7 +58,7 @@ static int detach__enable_on_exec(struct evlist *evlist)static int attach__current_disabled (struct evlist *evlist) { -struct evsel *evsel =perf_evlist__last (evlist); +struct evsel *evsel =evlist__last (evlist);struct perf_thread_map *threads;int err;
@@ -84,7 +84,7 @@ static int attach__current_disabled(struct evlist *evlist)static int attach__current_enabled (struct evlist *evlist) { -struct evsel *evsel =perf_evlist__last (evlist); +struct evsel *evsel =evlist__last (evlist);struct perf_thread_map *threads;int err;
@@ -104,14 +104,14 @@ static int attach__current_enabled(struct evlist *evlist)static int detach__disable (struct evlist *evlist) { -struct evsel *evsel =perf_evlist__last (evlist); +struct evsel *evsel =evlist__last (evlist);return evsel__enable (evsel); }static int attach__cpu_disabled (struct evlist *evlist) { -struct evsel *evsel =perf_evlist__last (evlist); +struct evsel *evsel =evlist__last (evlist);struct perf_cpu_map *cpus;int err;
@@ -140,7 +140,7 @@ static int attach__cpu_disabled(struct evlist *evlist)static int attach__cpu_enabled (struct evlist *evlist) { -struct evsel *evsel =perf_evlist__last (evlist); +struct evsel *evsel =evlist__last (evlist);struct perf_cpu_map *cpus;int err;
@@ -180,7 +180,7 @@ static int test_times(int (attach)(struct evlist *), goto out_err; } - evsel =perf_evlist__last (evlist); + evsel =evlist__last (evlist); evsel->core.attr.read_format |=PERF_FORMAT_TOTAL_TIME_ENABLED |PERF_FORMAT_TOTAL_TIME_RUNNING ;
tools/perf/tests/event_update.c +1 -1
@@ -92,7 +92,7 @@ int test__event_update(struct test *test __maybe_unused, int subtest __maybe_unu evlist =perf_evlist__new_default ();TEST_ASSERT_VAL ("failed to get evlist" , evlist); - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist);TEST_ASSERT_VAL ("failed to allocate ids" , !perf_evsel__alloc_id (&evsel->core,1 ,1 ));
tools/perf/tests/evsel-roundtrip-name.c +1 -1
@@ -34,7 +34,7 @@ static int perf_evsel__roundtrip_cache_name_test(void) } idx =0 ; - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist);for (type =0 ; type <PERF_COUNT_HW_CACHE_MAX ; type++) {for (op =0 ; op <PERF_COUNT_HW_CACHE_OP_MAX ; op++) {
tools/perf/tests/hists_cumulate.c +1 -1
@@ -721,7 +721,7 @@ int test__hists_cumulate(struct test *test __maybe_unused, int subtest __maybe_uif (verbose >1 )machine__fprintf (machine, stderr); - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist);for (i =0 ; i <ARRAY_SIZE (testcases); i++) { err = testcases[i](evsel, machine);
tools/perf/tests/hists_link.c +2 -2
@@ -311,8 +311,8 @@ int test__hists_link(struct test *test __maybe_unused, int subtest __maybe_unuseprint_hists_in (hists); } - first =perf_evlist__first (evlist); - evsel =perf_evlist__last (evlist); + first =evlist__first (evlist); + evsel =evlist__last (evlist); first_hists =evsel__hists (first); hists =evsel__hists (evsel);
tools/perf/tests/hists_output.c +1 -1
@@ -608,7 +608,7 @@ int test__hists_output(struct test *test __maybe_unused, int subtest __maybe_unuif (verbose >1 )machine__fprintf (machine, stderr); - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist);for (i =0 ; i <ARRAY_SIZE (testcases); i++) { err = testcases[i](evsel, machine);
tools/perf/tests/keep-tracking.c +2 -2
@@ -93,7 +93,7 @@ int test__keep_tracking(struct test *test __maybe_unused, int subtest __maybe_unperf_evlist__config (evlist, &opts,NULL ); - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist); evsel->core.attr.comm =1 ; evsel->core.attr.disabled =1 ;
@@ -132,7 +132,7 @@ int test__keep_tracking(struct test *test __maybe_unused, int subtest __maybe_unevlist__enable (evlist); - evsel =perf_evlist__last (evlist); + evsel =evlist__last (evlist);CHECK__ (evsel__disable (evsel));
tools/perf/tests/parse-events.c +58 -58
464 lines not on this page · raw message
tools/perf/tests/perf-record.c +1 -1
@@ -104,7 +104,7 @@ int test__PERF_RECORD(struct test *test __maybe_unused, int subtest __maybe_unus/* * Config the evsels, setting attr->comm on the first one, etc. */ - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist);perf_evsel__set_sample_bit (evsel,CPU );perf_evsel__set_sample_bit (evsel,TID );perf_evsel__set_sample_bit (evsel,TIME );
tools/perf/tests/switch-tracking.c +7 -7
@@ -367,7 +367,7 @@ int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_ goto out_err; } - cpu_clocks_evsel =perf_evlist__last (evlist); + cpu_clocks_evsel =evlist__last (evlist);/* Second event */ err =parse_events (evlist,"cycles:u" ,NULL );
@@ -376,7 +376,7 @@ int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_ goto out_err; } - cycles_evsel =perf_evlist__last (evlist); + cycles_evsel =evlist__last (evlist);/* Third event */ if (!perf_evlist__can_select_event (evlist, sched_switch)) {
@@ -391,7 +391,7 @@ int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_ goto out_err; } - switch_evsel =perf_evlist__last (evlist); + switch_evsel =evlist__last (evlist);perf_evsel__set_sample_bit (switch_evsel,CPU );perf_evsel__set_sample_bit (switch_evsel,TIME );
@@ -401,12 +401,12 @@ int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_ switch_evsel->immediate = true;/* Test moving an event to the front */ -if (cycles_evsel ==perf_evlist__first (evlist)) { +if (cycles_evsel ==evlist__first (evlist)) {pr_debug ("cycles event already at front" ); goto out_err; }perf_evlist__to_front (evlist, cycles_evsel); -if (cycles_evsel !=perf_evlist__first (evlist)) { +if (cycles_evsel !=evlist__first (evlist)) {pr_debug ("Failed to move cycles event to front" ); goto out_err; }
@@ -421,7 +421,7 @@ int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_ goto out_err; } - tracking_evsel =perf_evlist__last (evlist); + tracking_evsel =evlist__last (evlist);perf_evlist__set_tracking_event (evlist, tracking_evsel);
@@ -434,7 +434,7 @@ int test__switch_tracking(struct test *test __maybe_unused, int subtest __maybe_perf_evlist__config (evlist, &opts,NULL );/* Check moved event is still at the front */ -if (cycles_evsel !=perf_evlist__first (evlist)) { +if (cycles_evsel !=evlist__first (evlist)) {pr_debug ("Front event no longer at front" ); goto out_err; }
tools/perf/tests/task-exit.c +1 -1
@@ -88,7 +88,7 @@ int test__task_exit(struct test *test __maybe_unused, int subtest __maybe_unused goto out_delete_evlist; } - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist); evsel->core.attr.task =1 ;#ifdef __s390x__ evsel->core.attr.sample_freq =1000000 ;
tools/perf/ui/browsers/hists.c +3 -3
@@ -3319,13 +3319,13 @@ static int perf_evsel_menu__run(struct evsel_menu *menu,switch (key) {case K_TAB :if (pos->core.node.next == &evlist->core.entries) - pos =perf_evlist__first (evlist); + pos =evlist__first (evlist);else pos =perf_evsel__next (pos); goto browse_hists;case K_UNTAB :if (pos->core.node.prev == &evlist->core.entries) - pos =perf_evlist__last (evlist); + pos =evlist__last (evlist);else pos =perf_evsel__prev (pos); goto browse_hists;
@@ -3417,7 +3417,7 @@ int perf_evlist__tui_browse_hists(struct evlist *evlist, const char *help, single_entry:if (nr_entries ==1 ) { -struct evsel *first =perf_evlist__first (evlist); +struct evsel *first =evlist__first (evlist); return perf_evsel__hists_browse(first, nr_entries, help, false, hbt, min_pcnt,
tools/perf/util/bpf-loader.c +1 -1
@@ -1568,7 +1568,7 @@ struct evsel *bpf__setup_output_event(struct evlist *evlist, const char *name)return ERR_PTR (-err); } - evsel =perf_evlist__last (evlist); + evsel =evlist__last (evlist); }bpf__for_each_map_named (map, obj, tmp, name) {
tools/perf/util/evlist.c +12 -12
@@ -105,7 +105,7 @@ struct evlist *perf_evlist__new_dummy(void)*/ void perf_evlist__set_id_pos (struct evlist *evlist) { -struct evsel *first =perf_evlist__first (evlist); +struct evsel *first =evlist__first (evlist); evlist->id_pos = first->id_pos; evlist->is_pos = first->is_pos;
@@ -559,14 +559,14 @@ struct evsel *perf_evlist__id2evsel(struct evlist *evlist, u64 id)struct perf_sample_id *sid;if (evlist->core.nr_entries ==1 || !id) -return perf_evlist__first (evlist); +return evlist__first (evlist); sid =perf_evlist__id2sid (evlist, id);if (sid) return container_of(sid->evsel, struct evsel, core);if (!perf_evlist__sample_id_all (evlist)) -return perf_evlist__first (evlist); +return evlist__first (evlist);return NULL ; }
@@ -610,7 +610,7 @@ static int perf_evlist__event2id(struct evlist *evlist,struct evsel *perf_evlist__event2evsel (struct evlist *evlist,union perf_event *event) { -struct evsel *first =perf_evlist__first (evlist); +struct evsel *first =evlist__first (evlist);struct hlist_head *head;struct perf_sample_id *sid;int hash;
@@ -1222,7 +1222,7 @@ u64 perf_evlist__combined_branch_type(struct evlist *evlist)bool perf_evlist__valid_read_format (struct evlist *evlist) { -struct evsel *first =perf_evlist__first (evlist), *pos = first; +struct evsel *first =evlist__first (evlist), *pos = first;u64 read_format = first->core.attr.read_format;u64 sample_type = first->core.attr.sample_type;
@@ -1242,13 +1242,13 @@ bool perf_evlist__valid_read_format(struct evlist *evlist)u64 perf_evlist__read_format (struct evlist *evlist) { -struct evsel *first =perf_evlist__first (evlist); +struct evsel *first =evlist__first (evlist);return first->core.attr.read_format; }u16 perf_evlist__id_hdr_size (struct evlist *evlist) { -struct evsel *first =perf_evlist__first (evlist); +struct evsel *first =evlist__first (evlist);struct perf_sample *data;u64 sample_type;u16 size =0 ;
@@ -1281,7 +1281,7 @@ u16 perf_evlist__id_hdr_size(struct evlist *evlist)bool perf_evlist__valid_sample_id_all (struct evlist *evlist) { -struct evsel *first =perf_evlist__first (evlist), *pos = first; +struct evsel *first =evlist__first (evlist), *pos = first;evlist__for_each_entry_continue (evlist, pos) {if (first->core.attr.sample_id_all != pos->core.attr.sample_id_all)
@@ -1293,7 +1293,7 @@ bool perf_evlist__valid_sample_id_all(struct evlist *evlist)bool perf_evlist__sample_id_all (struct evlist *evlist) { -struct evsel *first =perf_evlist__first (evlist); +struct evsel *first =evlist__first (evlist);return first->core.attr.sample_id_all; }
@@ -1568,7 +1568,7 @@ int perf_evlist__strerror_open(struct evlist *evlist, "Hint:\tThe current value is %d.", value);break ;case EINVAL : { -struct evsel *first =perf_evlist__first (evlist); +struct evsel *first =evlist__first (evlist);int max_freq;if (sysctl__read_int ("kernel/perf_event_max_sample_rate" , &max_freq) <0 )
@@ -1630,7 +1630,7 @@ void perf_evlist__to_front(struct evlist *evlist,struct evsel *evsel, *n;LIST_HEAD (move); -if (move_evsel ==perf_evlist__first (evlist)) +if (move_evsel ==evlist__first (evlist))return ;evlist__for_each_entry_safe (evlist, n, evsel) {
@@ -1751,7 +1751,7 @@ bool perf_evlist__exclude_kernel(struct evlist *evlist)void perf_evlist__force_leader (struct evlist *evlist) {if (!evlist->nr_groups) { -struct evsel *leader =perf_evlist__first (evlist); +struct evsel *leader =evlist__first (evlist);perf_evlist__set_leader (evlist); leader->forced_leader = true;
tools/perf/util/evlist.h +9 -4
@@ -9,6 +9,7 @@#include <api/fd/array.h> #include <stdio.h> #include <internal/evlist.h> +#include <internal/evsel.h> #include "events_stats.h" #include "evsel.h" #include <pthread.h>
@@ -242,14 +243,18 @@ static inline bool perf_evlist__empty(struct evlist *evlist)return list_empty (&evlist->core.entries); } -static inline struct evsel *perf_evlist__first (struct evlist *evlist) +static inline struct evsel *evlist__first (struct evlist *evlist) { - return list_entry(evlist->core.entries.next, struct evsel, core.node); +struct perf_evsel *evsel =perf_evlist__first (&evlist->core); + + return container_of(evsel, struct evsel, core); } -static inline struct evsel *perf_evlist__last (struct evlist *evlist) +static inline struct evsel *evlist__last (struct evlist *evlist) { - return list_entry(evlist->core.entries.prev, struct evsel, core.node); +struct perf_evsel *evsel =perf_evlist__last (&evlist->core); + + return container_of(evsel, struct evsel, core); }size_t perf_evlist__fprintf (struct evlist *evlist,FILE *fp);
tools/perf/util/jitdump.c +1 -1
@@ -777,7 +777,7 @@ jit_process(struct perf_session *session,* track sample_type to compute id_all layout * perf sets the same sample type to all events as of now */ - first =perf_evlist__first (session->evlist); + first =evlist__first (session->evlist); jd.sample_type = first->core.attr.sample_type; *nbytes =0 ;
tools/perf/util/parse-events.c +2 -2
@@ -1936,7 +1936,7 @@ int parse_events(struct evlist *evlist, const char *str,perf_evlist__splice_list_tail (evlist, &parse_state.list); evlist->nr_groups += parse_state.nr_groups; - last =perf_evlist__last (evlist); + last =evlist__last (evlist); last->cmdline_group_boundary = true;return 0 ;
@@ -2050,7 +2050,7 @@ foreach_evsel_in_last_glob(struct evlist *evlist,* So no need to WARN here, let *func do this. */ if (evlist->core.nr_entries >0 ) - last =perf_evlist__last (evlist); + last =evlist__last (evlist); do { err = (*func)(last, arg);
tools/perf/util/record.c +3 -3
@@ -30,7 +30,7 @@ static int perf_do_probe_api(setup_probe_fn_t fn, int cpu, const char *str)if (parse_events (evlist, str,NULL )) goto out_delete; - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist);while (1 ) { fd =sys_perf_event_open (&evsel->core.attr, pid, cpu,-1 , flags);
@@ -171,7 +171,7 @@ void perf_evlist__config(struct evlist *evlist, struct record_opts *opts, use_sample_identifier =perf_can_sample_identifier (); sample_id = true; }else if (evlist->core.nr_entries >1 ) { -struct evsel *first =perf_evlist__first (evlist); +struct evsel *first =evlist__first (evlist);evlist__for_each_entry (evlist, evsel) {if (evsel->core.attr.sample_type == first->core.attr.sample_type)
@@ -276,7 +276,7 @@ bool perf_evlist__can_select_event(struct evlist *evlist, const char *str)if (err) goto out_delete; - evsel =perf_evlist__last (temp_evlist); + evsel =evlist__last (temp_evlist);if (!evlist ||perf_cpu_map__empty (evlist->core.cpus)) {struct perf_cpu_map *cpus =perf_cpu_map__new (NULL );
tools/perf/util/sort.c +1 -1
@@ -2329,7 +2329,7 @@ static struct evsel *find_evsel(struct evlist *evlist, char *event_name)if (nr > evlist->core.nr_entries)return NULL ; - evsel =perf_evlist__first (evlist); + evsel =evlist__first (evlist);while (--nr >0 ) evsel =perf_evsel__next (evsel);
tools/perf/util/top.c +1 -1
@@ -71,7 +71,7 @@ size_t perf_top__header_snprintf(struct perf_top *top, char *bf, size_t size) }if (top->evlist->core.nr_entries ==1 ) { -struct evsel *first =perf_evlist__first (top->evlist); +struct evsel *first =evlist__first (top->evlist); ret +=SNPRINTF (bf + ret, size - ret,"%" PRIu64"%s " , (uint64_t )first->core.attr.sample_period, opts->freq ?"Hz" :"" );
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 43/66 libperf: Add perf_evlist__read_format() function
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Add the perf_evlist__read_format() function to libperf as internal function.
--- tools/perf/lib/evlist.c | 7 +++++++ tools/perf/lib/include/internal/evlist.h | 2 ++ tools/perf/util/evlist.c | 8 +------- tools/perf/util/evlist.h | 1 - 4 files changed, 10 insertions(+), 8 deletions(-)
tools/perf/lib/evlist.c +7 -0
@@ -161,3 +161,10 @@ void perf_evlist__disable(struct perf_evlist *evlist)perf_evlist__for_each_entry (evlist, evsel)perf_evsel__disable (evsel); } + +u64 perf_evlist__read_format (struct perf_evlist *evlist) +{ +struct perf_evsel *first =perf_evlist__first (evlist); + +return first->attr.read_format; +}
tools/perf/lib/include/internal/evlist.h +2 -0
@@ -66,4 +66,6 @@ static inline struct perf_evsel *perf_evlist__last(struct perf_evlist *evlist) return list_entry(evlist->entries.prev, struct perf_evsel, node); } +u64 perf_evlist__read_format (struct perf_evlist *evlist); +#endif /* __LIBPERF_INTERNAL_EVLIST_H */
tools/perf/util/evlist.c +1 -7
@@ -503,7 +503,7 @@ int perf_evlist__id_add_fd(struct evlist *evlist,* This way does not work with group format read, so bail * out in that case. */ -if (perf_evlist__read_format (evlist) &PERF_FORMAT_GROUP ) +if (perf_evlist__read_format (&evlist->core) &PERF_FORMAT_GROUP )return -1 ; if (!(evsel->core.attr.read_format & PERF_FORMAT_ID) ||
@@ -1240,12 +1240,6 @@ bool perf_evlist__valid_read_format(struct evlist *evlist)return true; } -u64 perf_evlist__read_format (struct evlist *evlist) -{ -struct evsel *first =evlist__first (evlist); -return first->core.attr.read_format; -} -u16 perf_evlist__id_hdr_size (struct evlist *evlist) {struct evsel *first =evlist__first (evlist);
tools/perf/util/evlist.h +0 -1
@@ -217,7 +217,6 @@ int perf_evlist__apply_filters(struct evlist *evlist, struct evsel **err_evsel);void __perf_evlist__set_leader (struct list_head *list);void perf_evlist__set_leader (struct evlist *evlist); -u64 perf_evlist__read_format (struct evlist *evlist);u64 __perf_evlist__combined_sample_type (struct evlist *evlist);u64 perf_evlist__combined_sample_type (struct evlist *evlist);u64 perf_evlist__combined_branch_type (struct evlist *evlist);
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 44/66 libperf: Add perf_evlist__id_add() function
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Add the perf_evlist__id_add() function to libperf as an internal function. We already have the 'heads' member in 'struct perf_evlist'.
--- tools/perf/lib/evlist.c | 26 ++++++++++++++++++++++++ tools/perf/lib/include/internal/evlist.h | 4 ++++ tools/perf/tests/event_update.c | 2 +- tools/perf/util/evlist.c | 22 +------------------- tools/perf/util/evlist.h | 2 -- tools/perf/util/header.c | 4 ++-- 6 files changed, 34 insertions(+), 26 deletions(-)
tools/perf/lib/evlist.c +26 -0
@@ -1,9 +1,12 @@// SPDX-License-Identifier: GPL-2.0 #include <perf/evlist.h> #include <perf/evsel.h> +#include <linux/bitops.h> #include <linux/list.h> +#include <linux/hash.h> #include <internal/evlist.h> #include <internal/evsel.h> +#include <internal/xyarray.h> #include <linux/zalloc.h> #include <stdlib.h> #include <perf/cpumap.h>
@@ -168,3 +171,26 @@ u64 perf_evlist__read_format(struct perf_evlist *evlist)return first->attr.read_format; } + +#define SID (e, x, y) xyarray__entry(e->sample_id, x, y) + +static void perf_evlist__id_hash (struct perf_evlist *evlist, +struct perf_evsel *evsel, +int cpu,int thread,u64 id) +{ +int hash; +struct perf_sample_id *sid =SID (evsel, cpu, thread); + + sid->id = id; + sid->evsel = evsel; + hash =hash_64 (sid->id,PERF_EVLIST__HLIST_BITS ); +hlist_add_head (&sid->node, &evlist->heads[hash]); +} + +void perf_evlist__id_add (struct perf_evlist *evlist, +struct perf_evsel *evsel, +int cpu,int thread,u64 id) +{ +perf_evlist__id_hash (evlist, evsel, cpu, thread, id); + evsel->id[evsel->ids++] = id; +}
tools/perf/lib/include/internal/evlist.h +4 -0
@@ -68,4 +68,8 @@ static inline struct perf_evsel *perf_evlist__last(struct perf_evlist *evlist)u64 perf_evlist__read_format (struct perf_evlist *evlist); +void perf_evlist__id_add (struct perf_evlist *evlist, +struct perf_evsel *evsel, +int cpu,int thread,u64 id); +#endif /* __LIBPERF_INTERNAL_EVLIST_H */
tools/perf/tests/event_update.c +1 -1
@@ -97,7 +97,7 @@ int test__event_update(struct test *test __maybe_unused, int subtest __maybe_unuTEST_ASSERT_VAL ("failed to allocate ids" , !perf_evsel__alloc_id (&evsel->core,1 ,1 )); -perf_evlist__id_add (evlist, evsel,0 ,0 ,123 ); +perf_evlist__id_add (&evlist->core, &evsel->core,0 ,0 ,123 ); evsel->unit =strdup ("KRAVA" );
tools/perf/util/evlist.c +1 -21
@@ -461,26 +461,6 @@ int perf_evlist__poll(struct evlist *evlist, int timeout)return fdarray__poll (&evlist->core.pollfd, timeout); } -static void perf_evlist__id_hash (struct evlist *evlist, -struct evsel *evsel, -int cpu,int thread,u64 id) -{ -int hash; -struct perf_sample_id *sid =SID (evsel, cpu, thread); - - sid->id = id; - sid->evsel = &evsel->core; - hash =hash_64 (sid->id,PERF_EVLIST__HLIST_BITS ); -hlist_add_head (&sid->node, &evlist->core.heads[hash]); -} - -void perf_evlist__id_add (struct evlist *evlist,struct evsel *evsel, -int cpu,int thread,u64 id) -{ -perf_evlist__id_hash (evlist, evsel, cpu, thread, id); - evsel->core.id[evsel->core.ids++] = id; -} -int perf_evlist__id_add_fd (struct evlist *evlist,struct evsel *evsel,int cpu,int thread,int fd)
@@ -518,7 +498,7 @@ int perf_evlist__id_add_fd(struct evlist *evlist, id = read_data[id_idx]; add: -perf_evlist__id_add (evlist, evsel, cpu, thread, id); +perf_evlist__id_add (&evlist->core, &evsel->core, cpu, thread, id);return 0 ; }
tools/perf/util/evlist.h +0 -2
@@ -141,8 +141,6 @@ struct evsel * perf_evlist__find_tracepoint_by_name(struct evlist *evlist, const char *name); -void perf_evlist__id_add (struct evlist *evlist,struct evsel *evsel, -int cpu,int thread,u64 id);int perf_evlist__id_add_fd (struct evlist *evlist,struct evsel *evsel,int cpu,int thread,int fd);
tools/perf/util/header.c +2 -2
@@ -3618,7 +3618,7 @@ int perf_session__read_header(struct perf_session *session)if (perf_header__getbuffer64 (header, fd, &f_id,sizeof (f_id))) goto out_errno; -perf_evlist__id_add (session->evlist, evsel,0 , j, f_id); +perf_evlist__id_add (&session->evlist->core, &evsel->core,0 , j, f_id); }lseek (fd, tmp,SEEK_SET );
@@ -3754,7 +3754,7 @@ int perf_event__process_attr(struct perf_tool *tool __maybe_unused,return -ENOMEM ;for (i =0 ; i < n_ids; i++) { -perf_evlist__id_add (evlist, evsel,0 , i, event->attr.id[i]); +perf_evlist__id_add (&evlist->core, &evsel->core,0 , i, event->attr.id[i]); }return 0 ;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 45/66 libperf: Add perf_evlist__id_add_fd() function
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Add the perf_evlist__id_add_fd() function to libperf as an internal function.
--- tools/perf/lib/evlist.c | 44 ++++++++++++++++++++++++ tools/perf/lib/include/internal/evlist.h | 4 +++ tools/perf/util/evlist.c | 43 +---------------------- tools/perf/util/evlist.h | 4 --- tools/perf/util/evsel.c | 2 +- 5 files changed, 50 insertions(+), 47 deletions(-)
tools/perf/lib/evlist.c +44 -0
@@ -4,11 +4,14 @@#include <linux/bitops.h> #include <linux/list.h> #include <linux/hash.h> +#include <sys/ioctl.h> #include <internal/evlist.h> #include <internal/evsel.h> #include <internal/xyarray.h> #include <linux/zalloc.h> #include <stdlib.h> +#include <errno.h> +#include <unistd.h> #include <perf/cpumap.h> #include <perf/threadmap.h>
@@ -194,3 +197,44 @@ void perf_evlist__id_add(struct perf_evlist *evlist,perf_evlist__id_hash (evlist, evsel, cpu, thread, id); evsel->id[evsel->ids++] = id; } + +int perf_evlist__id_add_fd (struct perf_evlist *evlist, +struct perf_evsel *evsel, +int cpu,int thread,int fd) +{ +u64 read_data[4 ] = {0 , }; +int id_idx =1 ;/* The first entry is the counter value */ +u64 id; +int ret; + + ret =ioctl (fd,PERF_EVENT_IOC_ID , &id); +if (!ret) + goto add; + +if (errno !=ENOTTY ) +return -1 ; + +/* Legacy way to get event id.. All hail to old kernels! */ + +/* +* This way does not work with group format read, so bail +* out in that case. +*/ +if (perf_evlist__read_format (evlist) &PERF_FORMAT_GROUP ) +return -1 ; + +if (!(evsel->attr.read_format &PERF_FORMAT_ID ) || +read (fd, &read_data,sizeof (read_data)) ==-1 ) +return -1 ; + +if (evsel->attr.read_format &PERF_FORMAT_TOTAL_TIME_ENABLED ) + ++id_idx; +if (evsel->attr.read_format &PERF_FORMAT_TOTAL_TIME_RUNNING ) + ++id_idx; + + id = read_data[id_idx]; + +add: +perf_evlist__id_add (evlist, evsel, cpu, thread, id); +return 0 ; +}
tools/perf/lib/include/internal/evlist.h +4 -0
@@ -72,4 +72,8 @@ void perf_evlist__id_add(struct perf_evlist *evlist,struct perf_evsel *evsel, int cpu, int thread, u64 id); +int perf_evlist__id_add_fd (struct perf_evlist *evlist, +struct perf_evsel *evsel, +int cpu,int thread,int fd); +#endif /* __LIBPERF_INTERNAL_EVLIST_H */
tools/perf/util/evlist.c +1 -42
55 lines not on this page · raw message
tools/perf/util/evlist.h +0 -4
@@ -141,10 +141,6 @@ struct evsel * perf_evlist__find_tracepoint_by_name(struct evlist *evlist, const char *name); -int perf_evlist__id_add_fd (struct evlist *evlist, -struct evsel *evsel, -int cpu,int thread,int fd); -int perf_evlist__add_pollfd (struct evlist *evlist,int fd);int perf_evlist__alloc_pollfd (struct evlist *evlist);int perf_evlist__filter_pollfd (struct evlist *evlist,short revents_and_mask);
tools/perf/util/evsel.c +1 -1
@@ -2662,7 +2662,7 @@ static int store_evsel_ids(struct evsel *evsel, struct evlist *evlist) thread++) {int fd =FD (evsel, cpu, thread); -if (perf_evlist__id_add_fd (evlist, evsel, +if (perf_evlist__id_add_fd (&evlist->core, &evsel->core, cpu, thread, fd) <0 )return -1 ; }
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 46/66 libperf: Move 'page_size' global variable to libperf
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
We need the 'page_size' variable in libperf, so move it there.
Add a libperf_init() as a global libperf init function to obtain this value via sysconf() at tool start.
Committer notes:
Add internal/lib.h to tools/perf/ files using 'page_size', sometimes replacing util.h with it if that was the only reason for having util.h included.
--- tools/perf/arch/arm/util/cs-etm.c | 2 +- tools/perf/arch/arm64/util/arm-spe.c | 2 +- tools/perf/arch/s390/util/machine.c | 2 +- tools/perf/arch/x86/tests/intel-cqm.c | 1 + tools/perf/arch/x86/tests/rdpmc.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 2 +- tools/perf/arch/x86/util/intel-pt.c | 2 +- tools/perf/arch/x86/util/machine.c | 2 +- tools/perf/lib/core.c | 7 +++++++ tools/perf/lib/include/internal/lib.h | 2 ++ tools/perf/lib/include/perf/core.h | 1 + tools/perf/lib/lib.c | 2 ++ tools/perf/lib/libperf.map | 1 + tools/perf/perf.c | 7 ++++--- tools/perf/tests/mmap-thread-lookup.c | 2 +- tools/perf/tests/vmlinux-kallsyms.c | 2 +- tools/perf/util/auxtrace.c | 1 - tools/perf/util/evlist.c | 2 +- tools/perf/util/header.c | 2 +- tools/perf/util/machine.c | 1 + tools/perf/util/mmap.c | 2 +- tools/perf/util/python.c | 2 +- tools/perf/util/session.c | 1 - tools/perf/util/srccode.c | 2 +- tools/perf/util/synthetic-events.c | 2 +- tools/perf/util/trace-event-info.c | 2 +- tools/perf/util/util.c | 3 +-- tools/perf/util/util.h | 2 -- 28 files changed, 36 insertions(+), 25 deletions(-)
tools/perf/arch/arm/util/cs-etm.c +1 -1
@@ -25,7 +25,7 @@#include "../../util/evsel.h" #include "../../util/pmu.h" #include "../../util/cs-etm.h" -#include "../../util/util.h" // page_size +#include <internal/lib.h> // page_size #include "../../util/session.h" #include <errno.h>
tools/perf/arch/arm64/util/arm-spe.c +1 -1
@@ -16,7 +16,7 @@#include "../../util/evsel.h" #include "../../util/evlist.h" #include "../../util/session.h" -#include "../../util/util.h" // page_size +#include <internal/lib.h> // page_size #include "../../util/pmu.h" #include "../../util/debug.h" #include "../../util/auxtrace.h"
tools/perf/arch/s390/util/machine.c +1 -1
@@ -2,7 +2,7 @@#include <unistd.h> #include <stdio.h> #include <string.h> -#include "util.h" // page_size +#include <internal/lib.h> // page_size #include "machine.h" #include "api/fs/fs.h" #include "debug.h"
tools/perf/arch/x86/tests/intel-cqm.c +1 -0
@@ -5,6 +5,7 @@#include "evlist.h" #include "evsel.h" #include "arch-tests.h" +#include <internal/lib.h> // page_size #include <signal.h> #include <sys/mman.h>
tools/perf/arch/x86/tests/rdpmc.c +1 -1
@@ -13,7 +13,7 @@#include "tests/tests.h" #include "cloexec.h" #include "event.h" -#include "util.h" // page_size +#include <internal/lib.h> // page_size #include "arch-tests.h" static u64 rdpmc (unsigned int counter)
tools/perf/arch/x86/util/intel-bts.c +1 -1
@@ -23,7 +23,7 @@#include "../../util/tsc.h" #include "../../util/auxtrace.h" #include "../../util/intel-bts.h" -#include "../../util/util.h" // page_size +#include <internal/lib.h> // page_size #define KiB (x) ((x) * 1024)#define MiB (x) ((x) * 1024 * 1024)
tools/perf/arch/x86/util/intel-pt.c +1 -1
@@ -27,7 +27,7 @@#include "../../util/record.h" #include "../../util/target.h" #include "../../util/tsc.h" -#include "../../util/util.h" // page_size +#include <internal/lib.h> // page_size #include "../../util/intel-pt.h" #define KiB (x) ((x) * 1024)
tools/perf/arch/x86/util/machine.c +1 -1
@@ -4,7 +4,7 @@#include <limits.h> #include <stdlib.h> -#include "../../util/util.h" // page_size +#include <internal/lib.h> // page_size #include "../../util/machine.h" #include "../../util/map.h" #include "../../util/symbol.h"
tools/perf/lib/core.c +7 -0
@@ -4,7 +4,9 @@#include <stdio.h> #include <stdarg.h> +#include <unistd.h> #include <perf/core.h> +#include <internal/lib.h> #include "internal.h" static int __base_pr(enum libperf_print_level level, const char *format,
@@ -32,3 +34,8 @@ void libperf_print(enum libperf_print_level level, const char *format, ...)__libperf_pr (level, format, args);va_end (args); } + +void libperf_init (void ) +{ + page_size =sysconf (_SC_PAGE_SIZE); +}
tools/perf/lib/include/internal/lib.h +2 -0
@@ -4,6 +4,8 @@#include <unistd.h> +extern unsigned int page_size; +ssize_t readn (int fd,void *buf,size_t n);ssize_t writen (int fd,const void *buf,size_t n);
tools/perf/lib/include/perf/core.h +1 -0
@@ -18,5 +18,6 @@ typedef int (*libperf_print_fn_t)(enum libperf_print_level level, const char *, va_list ap); LIBPERF_API void libperf_set_print(libperf_print_fn_t fn); +LIBPERF_API void libperf_init(void);#endif /* __LIBPERF_CORE_H */
tools/perf/lib/lib.c +2 -0
@@ -5,6 +5,8 @@#include <linux/kernel.h> #include <internal/lib.h> +unsigned int page_size; +static ssize_t ion (bool is_read,int fd,void *buf,size_t n) {void *buf_start = buf;
tools/perf/lib/libperf.map +1 -0
@@ -1,5 +1,6 @@ LIBPERF_0.0.1 { global: + libperf_init; libperf_set_print; perf_cpu_map__dummy_new; perf_cpu_map__get;
tools/perf/perf.c +4 -3
@@ -12,6 +12,7 @@#include "util/build-id.h" #include "util/cache.h" #include "util/env.h" +#include <internal/lib.h> // page_size #include <subcmd/exec-cmd.h> #include "util/config.h" #include <subcmd/run-command.h>
@@ -20,11 +21,12 @@#include "util/bpf-loader.h" #include "util/debug.h" #include "util/event.h" -#include "util/util.h" // page_size, usage() +#include "util/util.h" // usage() #include "ui/ui.h" #include "perf-sys.h" #include <api/fs/fs.h> #include <api/fs/tracing_path.h> +#include <perf/core.h> #include <errno.h> #include <pthread.h> #include <signal.h>
@@ -438,8 +440,7 @@ int main(int argc, const char **argv)exec_cmd_init ("perf" ,PREFIX ,PERF_EXEC_PATH ,EXEC_PATH_ENVIRONMENT );pager_init (PERF_PAGER_ENVIRONMENT ); -/* The page_size is placed in util object. */ - page_size =sysconf (_SC_PAGE_SIZE); +libperf_init (); cmd =extract_argv0_path (argv[0 ]);if (!cmd)
tools/perf/tests/mmap-thread-lookup.c +1 -1
@@ -16,7 +16,7 @@#include "symbol.h" #include "util/synthetic-events.h" #include "thread.h" -#include "util.h" // page_size +#include <internal/lib.h> // page_size #define THREADS 4
tools/perf/tests/vmlinux-kallsyms.c +1 -1
@@ -7,7 +7,7 @@#include "dso.h" #include "map.h" #include "symbol.h" -#include "util.h" // page_size +#include <internal/lib.h> // page_size #include "tests.h" #include "debug.h" #include "machine.h"
tools/perf/util/auxtrace.c +0 -1
@@ -50,7 +50,6 @@#include "intel-bts.h" #include "arm-spe.h" #include "s390-cpumsf.h" -#include "util.h" // page_size #include "util/mmap.h" #include <linux/ctype.h>
tools/perf/util/evlist.c +1 -1
@@ -17,7 +17,7 @@#include "evsel.h" #include "debug.h" #include "units.h" -#include "util.h" // page_size +#include <internal/lib.h> // page_size #include "../perf.h" #include "asm/bug.h" #include "bpf-event.h"
tools/perf/util/header.c +1 -1
@@ -42,7 +42,7 @@#include "tool.h" #include "time-utils.h" #include "units.h" -#include "util.h" // page_size, perf_exe() +#include "util/util.h" // perf_exe() #include "cputopo.h" #include "bpf-event.h"
tools/perf/util/machine.c +1 -0
@@ -32,6 +32,7 @@#include "linux/hash.h" #include "asm/bug.h" #include "bpf-event.h" +#include <internal/lib.h> // page_size #include <linux/ctype.h> #include <symbol/kallsyms.h>
tools/perf/util/mmap.c +1 -1
@@ -20,7 +20,7 @@#include "event.h" #include "mmap.h" #include "../perf.h" -#include "util.h" /* page_size */ +#include <internal/lib.h> /* page_size */ size_t perf_mmap__mmap_len (struct mmap *map) {
tools/perf/util/python.c +1 -1
@@ -14,7 +14,7 @@#include "thread_map.h" #include "trace-event.h" #include "mmap.h" -#include "util.h" +#include <internal/lib.h> #include "../perf-sys.h" #if PY_MAJOR_VERSION <3
tools/perf/util/session.c +0 -1
@@ -29,7 +29,6 @@#include "thread-stack.h" #include "sample-raw.h" #include "stat.h" -#include "util.h" #include "ui/progress.h" #include "../perf.h" #include "arch/common.h"
tools/perf/util/srccode.c +1 -1
@@ -15,7 +15,7 @@#include <string.h> #include "srccode.h" #include "debug.h" -#include "util.h" // page_size +#include <internal/lib.h> // page_size #define MAXSRCCACHE (32*1024*1024)#define MAXSRCFILES 64
tools/perf/util/synthetic-events.c +1 -1
@@ -16,7 +16,6 @@#include "util/synthetic-events.h" #include "util/target.h" #include "util/time-utils.h" -#include "util/util.h" #include <linux/bitops.h> #include <linux/kernel.h> #include <linux/string.h>
@@ -26,6 +25,7 @@#include <perf/evsel.h> #include <internal/cpumap.h> #include <perf/cpumap.h> +#include <internal/lib.h> // page_size #include <internal/threadmap.h> #include <perf/threadmap.h> #include <symbol/kallsyms.h>
tools/perf/util/trace-event-info.c +1 -1
@@ -2,7 +2,6 @@/* * Copyright (C) 2008,2009, Steven Rostedt <[email protected]> */ -#include "util.h" // page_size #include <dirent.h> #include <mntent.h> #include <stdio.h>
@@ -19,6 +18,7 @@#include <linux/list.h> #include <linux/kernel.h> #include <linux/zalloc.h> +#include <internal/lib.h> // page_size #include "trace-event.h" #include <api/fs/tracing_path.h>
tools/perf/util/util.c +1 -2
@@ -3,6 +3,7 @@#include "debug.h" #include "event.h" #include "namespaces.h" +#include <internal/lib.h> #include <api/fs/fs.h> #include <sys/mman.h> #include <sys/stat.h>
@@ -41,8 +42,6 @@ void perf_set_multithreaded(void) perf_singlethreaded = false; } -unsigned int page_size; -int sysctl_perf_event_max_stack =PERF_MAX_STACK_DEPTH ;int sysctl_perf_event_max_contexts_per_stack =PERF_MAX_CONTEXTS_PER_STACK ;
tools/perf/util/util.h +0 -2
@@ -33,8 +33,6 @@ int copyfile_offset(int ifd, loff_t off_in, int ofd, loff_t off_out, u64 size);size_t hex_width (u64 v); -extern unsigned int page_size; -int sysctl__max_stack (void ); int fetch_kernel_version(unsigned int *puint,
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 47/66 perf tools: No need to include internal/lib.h from util/util.h
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
That was done just to have users of writen() and readn(), that before had their prototypes in util/util.h to get it without having to add an include for internal/lib.h, but the right way is to add it and by now all places already do it.
Fix a fallout were readlink() was used but unistd.h was being obtained by luck thru util.h -> internal/lib.h, now to check why unistd.h is being included there...
--- tools/perf/tests/sdt.c | 1 + tools/perf/util/util.h | 1 - 2 files changed, 1 insertion(+), 1 deletion(-)
tools/perf/tests/sdt.c +1 -0
@@ -3,6 +3,7 @@#include <limits.h> #include <stdio.h> #include <stdlib.h> +#include <unistd.h> #include <sys/epoll.h> #include <util/symbol.h> #include <linux/filter.h>
tools/perf/util/util.h +0 -1
@@ -11,7 +11,6 @@#include <stddef.h> #include <linux/compiler.h> #include <sys/types.h> -#include <internal/lib.h> /* General helper functions */ void usage (const char *err)__noreturn ;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 48/66 libperf: Use sys/types.h to get ssize_t, not unistd.h
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
The sys/types.h header looks more sensible, from its name we can gather it should be there because of some needed typedef, and it is much smaller than unistd.h, so use it and fix up the fallout in places where it was being used for something else entirely but being obtained by sheer luck, indirectly.
--- tools/perf/lib/include/internal/lib.h | 2 +- tools/perf/util/mmap.c | 1 + 2 files changed, 2 insertions(+), 1 deletion(-)
tools/perf/lib/include/internal/lib.h +1 -1
@@ -2,7 +2,7 @@#ifndef __LIBPERF_INTERNAL_LIB_H#define __LIBPERF_INTERNAL_LIB_H -#include <unistd.h> +#include <sys/types.h> extern unsigned int page_size;
tools/perf/util/mmap.c +1 -0
@@ -12,6 +12,7 @@#include <linux/zalloc.h> #include <stdlib.h> #include <string.h> +#include <unistd.h> // sysconf() #ifdef HAVE_LIBNUMA_SUPPORT #include <numaif.h> #endif
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 49/66 libperf: Add libperf dependency for tests targets
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Add libperf dependency for tests targets.
--- tools/perf/lib/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/lib/Makefile +1 -1
@@ -138,7 +138,7 @@ clean: $(LIBAPI)-clean *.o *~ *.a *.so *.so.$(VERSION) *.so.$(LIBPERF_VERSION) .*.d .*.cmd LIBPERF-CFLAGS $(LIBPERF_PC) $(Q)$(MAKE) -C tests clean -tests: +tests: libs $(Q)$(MAKE) -C tests $(Q)$(MAKE) -C tests run
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 50/66 libperf: Merge libperf_set_print() into libperf_init()
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
The libperf_set_print() function needs to be called in any case so let's merge it with libperf_init(), so we have just one init function.
--- tools/perf/lib/core.c | 8 ++------ tools/perf/lib/include/perf/core.h | 3 +-- tools/perf/lib/libperf.map | 1 - tools/perf/perf.c | 8 +++++++- 4 files changed, 10 insertions(+), 10 deletions(-)
tools/perf/lib/core.c +2 -6
@@ -17,11 +17,6 @@ static int __base_pr(enum libperf_print_level level, const char *format,static libperf_print_fn_t __libperf_pr = __base_pr; -void libperf_set_print (libperf_print_fn_t fn) -{ - __libperf_pr = fn; -} - __printf(2, 3)void libperf_print (enum libperf_print_level level,const char *format, ...) {
@@ -35,7 +30,8 @@ void libperf_print(enum libperf_print_level level, const char *format, ...)va_end (args); } -void libperf_init (void ) +void libperf_init (libperf_print_fn_t fn) { page_size =sysconf (_SC_PAGE_SIZE); + __libperf_pr = fn; }
tools/perf/lib/include/perf/core.h +1 -2
@@ -17,7 +17,6 @@ enum libperf_print_level {typedef int (*libperf_print_fn_t )(enum libperf_print_level level,const char *,va_list ap); -LIBPERF_API void libperf_set_print(libperf_print_fn_t fn); -LIBPERF_API void libperf_init(void); +LIBPERF_API void libperf_init(libperf_print_fn_t fn);#endif /* __LIBPERF_CORE_H */
tools/perf/lib/libperf.map +0 -1
@@ -1,7 +1,6 @@ LIBPERF_0.0.1 { global: libperf_init; - libperf_set_print; perf_cpu_map__dummy_new; perf_cpu_map__get; perf_cpu_map__put;
tools/perf/perf.c +7 -1
@@ -430,6 +430,12 @@ void pthread__unblock_sigwinch(void)pthread_sigmask (SIG_UNBLOCK , &set,NULL ); } +static int libperf_print (enum libperf_print_level level, +const char *fmt,va_list ap) +{ +return eprintf (level, verbose, fmt, ap); +} +int main (int argc,const char **argv) {int err;
@@ -440,7 +446,7 @@ int main(int argc, const char **argv)exec_cmd_init ("perf" ,PREFIX ,PERF_EXEC_PATH ,EXEC_PATH_ENVIRONMENT );pager_init (PERF_PAGER_ENVIRONMENT ); -libperf_init (); +libperf_init (libperf_print); cmd =extract_argv0_path (argv[0 ]);if (!cmd)
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 51/66 libperf: Add libperf_init() call to the tests
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Add libperf_init() call to the automated tests.
Committer notes:
Added missing stdarg.h and/or stdio.h to places using vfprintf.
--- tools/perf/lib/tests/test-cpumap.c | 10 ++++++++++ tools/perf/lib/tests/test-evlist.c | 10 ++++++++++ tools/perf/lib/tests/test-evsel.c | 10 ++++++++++ tools/perf/lib/tests/test-threadmap.c | 10 ++++++++++ 4 files changed, 40 insertions(+)
tools/perf/lib/tests/test-cpumap.c +10 -0
@@ -1,13 +1,23 @@// SPDX-License-Identifier: GPL-2.0 +#include <stdarg.h> +#include <stdio.h> #include <perf/cpumap.h> #include <internal/tests.h> +static int libperf_print (enum libperf_print_level level, +const char *fmt,va_list ap) +{ +return vfprintf (stderr, fmt, ap); +} +int main (int argc,char **argv) {struct perf_cpu_map *cpus; __T_START; +libperf_init (libperf_print); + cpus =perf_cpu_map__dummy_new ();if (!cpus)return -1 ;
tools/perf/lib/tests/test-evlist.c +10 -0
@@ -1,4 +1,6 @@// SPDX-License-Identifier: GPL-2.0 +#include <stdio.h> +#include <stdarg.h> #include <linux/perf_event.h> #include <perf/cpumap.h> #include <perf/threadmap.h>
@@ -6,6 +8,12 @@#include <perf/evsel.h> #include <internal/tests.h> +static int libperf_print (enum libperf_print_level level, +const char *fmt,va_list ap) +{ +return vfprintf (stderr, fmt, ap); +} +static int test_stat_cpu (void ) {struct perf_cpu_map *cpus;
@@ -177,6 +185,8 @@ int main(int argc, char **argv) { __T_START; +libperf_init (libperf_print); +test_stat_cpu ();test_stat_thread ();test_stat_thread_enable ();
tools/perf/lib/tests/test-evsel.c +10 -0
@@ -1,10 +1,18 @@// SPDX-License-Identifier: GPL-2.0 +#include <stdarg.h> +#include <stdio.h> #include <linux/perf_event.h> #include <perf/cpumap.h> #include <perf/threadmap.h> #include <perf/evsel.h> #include <internal/tests.h> +static int libperf_print (enum libperf_print_level level, +const char *fmt,va_list ap) +{ +return vfprintf (stderr, fmt, ap); +} +static int test_stat_cpu (void ) {struct perf_cpu_map *cpus;
@@ -116,6 +124,8 @@ int main(int argc, char **argv) { __T_START; +libperf_init (libperf_print); +test_stat_cpu ();test_stat_thread ();test_stat_thread_enable ();
tools/perf/lib/tests/test-threadmap.c +10 -0
@@ -1,13 +1,23 @@// SPDX-License-Identifier: GPL-2.0 +#include <stdarg.h> +#include <stdio.h> #include <perf/threadmap.h> #include <internal/tests.h> +static int libperf_print (enum libperf_print_level level, +const char *fmt,va_list ap) +{ +return vfprintf (stderr, fmt, ap); +} +int main (int argc,char **argv) {struct perf_thread_map *threads; __T_START; +libperf_init (libperf_print); + threads =perf_thread_map__new_dummy ();if (!threads)return -1 ;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 52/66 libperf: Add perf_evlist__alloc_pollfd() function
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move perf_evlist__alloc_pollfd() from tools/perf to libperf, it will be used in the following patches.
--- tools/perf/lib/evlist.c | 22 ++++++++++++++++++++++ tools/perf/lib/include/internal/evlist.h | 2 ++ tools/perf/util/evlist.c | 23 +---------------------- tools/perf/util/evlist.h | 1 - 4 files changed, 25 insertions(+), 23 deletions(-)
tools/perf/lib/evlist.c +22 -0
@@ -14,6 +14,7 @@#include <unistd.h> #include <perf/cpumap.h> #include <perf/threadmap.h> +#include <api/fd/array.h> void perf_evlist__init (struct perf_evlist *evlist) {
@@ -238,3 +239,24 @@ int perf_evlist__id_add_fd(struct perf_evlist *evlist,perf_evlist__id_add (evlist, evsel, cpu, thread, id);return 0 ; } + +int perf_evlist__alloc_pollfd (struct perf_evlist *evlist) +{ +int nr_cpus =perf_cpu_map__nr (evlist->cpus); +int nr_threads =perf_thread_map__nr (evlist->threads); +int nfds =0 ; +struct perf_evsel *evsel; + +perf_evlist__for_each_entry (evlist, evsel) { +if (evsel->system_wide) + nfds += nr_cpus; +else + nfds += nr_cpus * nr_threads; + } + +if (fdarray__available_entries (&evlist->pollfd) < nfds && +fdarray__grow (&evlist->pollfd, nfds) <0 ) +return -ENOMEM ; + +return 0 ; +}
tools/perf/lib/include/internal/evlist.h +2 -0
@@ -24,6 +24,8 @@ struct perf_evlist {struct hlist_head heads[PERF_EVLIST__HLIST_SIZE ]; }; +int perf_evlist__alloc_pollfd (struct perf_evlist *evlist); +/** * __perf_evlist__for_each_entry - iterate thru all the evsels * @list: list_head instance to iterate
tools/perf/util/evlist.c +1 -22
@@ -398,27 +398,6 @@ int perf_evlist__enable_event_idx(struct evlist *evlist,return perf_evlist__enable_event_thread (evlist, evsel, idx); } -int perf_evlist__alloc_pollfd (struct evlist *evlist) -{ -int nr_cpus =perf_cpu_map__nr (evlist->core.cpus); -int nr_threads =perf_thread_map__nr (evlist->core.threads); -int nfds =0 ; -struct evsel *evsel; - -evlist__for_each_entry (evlist, evsel) { -if (evsel->core.system_wide) - nfds += nr_cpus; -else - nfds += nr_cpus * nr_threads; - } - -if (fdarray__available_entries (&evlist->core.pollfd) < nfds && -fdarray__grow (&evlist->core.pollfd, nfds) <0 ) -return -ENOMEM ; - -return 0 ; -} -static int __perf_evlist__add_pollfd (struct evlist *evlist,int fd,struct mmap *map,short revent) {
@@ -944,7 +923,7 @@ int evlist__mmap_ex(struct evlist *evlist, unsigned int pages,if (!evlist->mmap)return -ENOMEM ; -if (evlist->core.pollfd.entries ==NULL &&perf_evlist__alloc_pollfd (evlist) <0 ) +if (evlist->core.pollfd.entries ==NULL &&perf_evlist__alloc_pollfd (&evlist->core) <0 )return -ENOMEM ; evlist->core.mmap_len =evlist__mmap_size (pages);
tools/perf/util/evlist.h +0 -1
@@ -142,7 +142,6 @@ perf_evlist__find_tracepoint_by_name(struct evlist *evlist, const char *name);int perf_evlist__add_pollfd (struct evlist *evlist,int fd); -int perf_evlist__alloc_pollfd (struct evlist *evlist);int perf_evlist__filter_pollfd (struct evlist *evlist,short revents_and_mask);int perf_evlist__poll (struct evlist *evlist,int timeout);
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 53/66 libperf: Add perf_evlist__add_pollfd() function
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move perf_evlist__add_pollfd() from tools/perf to libperf, it will be used in the following patches.
Also rename perf's perf_evlist__add_pollfd()/perf_evlist__filter_pollfd() to evlist__add_pollfd()/evlist__filter_pollfd().
--- tools/perf/builtin-kvm.c | 4 ++-- tools/perf/builtin-record.c | 2 +- tools/perf/builtin-trace.c | 2 +- tools/perf/lib/evlist.c | 16 +++++++++++++++ tools/perf/lib/include/internal/evlist.h | 2 ++ tools/perf/util/evlist.c | 25 ++++-------------------- tools/perf/util/evlist.h | 4 ++-- 7 files changed, 28 insertions(+), 27 deletions(-)
tools/perf/builtin-kvm.c +2 -2
@@ -967,10 +967,10 @@ static int kvm_events_live_report(struct perf_kvm_stat *kvm) goto out; } -if (perf_evlist__add_pollfd (kvm->evlist, kvm->timerfd) <0 ) +if (evlist__add_pollfd (kvm->evlist, kvm->timerfd) <0 ) goto out; - nr_stdin =perf_evlist__add_pollfd (kvm->evlist,fileno (stdin)); + nr_stdin =evlist__add_pollfd (kvm->evlist,fileno (stdin));if (nr_stdin <0 ) goto out;
tools/perf/builtin-record.c +1 -1
@@ -1624,7 +1624,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv) err =0 ; waking++; -if (perf_evlist__filter_pollfd (rec->evlist,POLLERR |POLLHUP ) ==0 ) +if (evlist__filter_pollfd (rec->evlist,POLLERR |POLLHUP ) ==0 ) draining = true; }
tools/perf/builtin-trace.c +1 -1
@@ -3475,7 +3475,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)int timeout = done ?100 :-1 ;if (!draining &&perf_evlist__poll (evlist, timeout) >0 ) { -if (perf_evlist__filter_pollfd (evlist,POLLERR |POLLHUP |POLLNVAL ) ==0 ) +if (evlist__filter_pollfd (evlist,POLLERR |POLLHUP |POLLNVAL ) ==0 ) draining = true; goto again;
tools/perf/lib/evlist.c +16 -0
@@ -12,6 +12,9 @@#include <stdlib.h> #include <errno.h> #include <unistd.h> +#include <fcntl.h> +#include <signal.h> +#include <poll.h> #include <perf/cpumap.h> #include <perf/threadmap.h> #include <api/fd/array.h>
@@ -260,3 +263,16 @@ int perf_evlist__alloc_pollfd(struct perf_evlist *evlist)return 0 ; } + +int perf_evlist__add_pollfd (struct perf_evlist *evlist,int fd, +void *ptr,short revent) +{ +int pos =fdarray__add (&evlist->pollfd, fd, revent |POLLERR |POLLHUP ); + +if (pos >=0 ) { + evlist->pollfd.priv[pos].ptr = ptr; +fcntl (fd,F_SETFL ,O_NONBLOCK ); + } + +return pos; +}
tools/perf/lib/include/internal/evlist.h +2 -0
@@ -25,6 +25,8 @@ struct perf_evlist { };int perf_evlist__alloc_pollfd (struct perf_evlist *evlist); +int perf_evlist__add_pollfd (struct perf_evlist *evlist,int fd, +void *ptr,short revent);/** * __perf_evlist__for_each_entry - iterate thru all the evsels
tools/perf/util/evlist.c +4 -21
@@ -398,26 +398,9 @@ int perf_evlist__enable_event_idx(struct evlist *evlist,return perf_evlist__enable_event_thread (evlist, evsel, idx); } -static int __perf_evlist__add_pollfd (struct evlist *evlist,int fd, -struct mmap *map,short revent) +int evlist__add_pollfd (struct evlist *evlist,int fd) { -int pos =fdarray__add (&evlist->core.pollfd, fd, revent |POLLERR |POLLHUP ); -/* -* Save the idx so that when we filter out fds POLLHUP'ed we can -* close the associated evlist->mmap[] entry. -*/ -if (pos >=0 ) { - evlist->core.pollfd.priv[pos].ptr = map; - -fcntl (fd,F_SETFL ,O_NONBLOCK ); - } - -return pos; -} - -int perf_evlist__add_pollfd (struct evlist *evlist,int fd) -{ -return __perf_evlist__add_pollfd (evlist, fd,NULL ,POLLIN ); +return perf_evlist__add_pollfd (&evlist->core, fd,NULL ,POLLIN ); } static void perf_evlist__munmap_filtered(struct fdarray *fda, int fd,
@@ -429,7 +412,7 @@ static void perf_evlist__munmap_filtered(struct fdarray *fda, int fd,perf_mmap__put (map); } -int perf_evlist__filter_pollfd (struct evlist *evlist,short revents_and_mask) +int evlist__filter_pollfd (struct evlist *evlist,short revents_and_mask) {return fdarray__filter (&evlist->core.pollfd, revents_and_mask, perf_evlist__munmap_filtered,NULL );
@@ -708,7 +691,7 @@ static int evlist__mmap_per_evsel(struct evlist *evlist, int idx,* Therefore don't add it for polling. */ if (!evsel->core.system_wide && -__perf_evlist__add_pollfd (evlist, fd, &maps[idx], revent) <0 ) { +perf_evlist__add_pollfd (&evlist->core, fd, &maps[idx], revent) <0 ) {perf_mmap__put (&maps[idx]);return -1 ; }
tools/perf/util/evlist.h +2 -2
@@ -141,8 +141,8 @@ struct evsel * perf_evlist__find_tracepoint_by_name(struct evlist *evlist, const char *name); -int perf_evlist__add_pollfd (struct evlist *evlist,int fd); -int perf_evlist__filter_pollfd (struct evlist *evlist,short revents_and_mask); +int evlist__add_pollfd (struct evlist *evlist,int fd); +int evlist__filter_pollfd (struct evlist *evlist,short revents_and_mask);int perf_evlist__poll (struct evlist *evlist,int timeout);
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 54/66 libperf: Add perf_evlist__poll() function
In reply to Arnaldo Carvalho de MeloFrom: Jiri Olsa <[email protected]>
Move perf_evlist__poll() from tools/perf to libperf, it will be used in the following patches.
And rename the existing perf's function to evlist__poll().
--- tools/perf/builtin-record.c | 2 +- tools/perf/builtin-top.c | 4 ++-- tools/perf/builtin-trace.c | 2 +- tools/perf/lib/evlist.c | 5 +++++ tools/perf/lib/include/perf/evlist.h | 1 + tools/perf/lib/libperf.map | 1 + tools/perf/tests/openat-syscall-tp-fields.c | 2 +- tools/perf/tests/perf-record.c | 2 +- tools/perf/tests/task-exit.c | 2 +- tools/perf/util/evlist.c | 6 +++--- tools/perf/util/evlist.h | 2 +- tools/perf/util/python.c | 2 +- 12 files changed, 19 insertions(+), 12 deletions(-)
tools/perf/builtin-record.c +1 -1
@@ -1615,7 +1615,7 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)if (hits == rec->samples) {if (done || draining)break ; - err =perf_evlist__poll (rec->evlist,-1 ); + err =evlist__poll (rec->evlist,-1 );/* * Propagate error, only if there's any. Ignore positive * number of returned events and interrupt error.
tools/perf/builtin-top.c +2 -2
@@ -1307,7 +1307,7 @@ static int __cmd_top(struct perf_top *top) }/* Wait for a minimal set of events before starting the snapshot */ -perf_evlist__poll (top->evlist,100 ); +evlist__poll (top->evlist,100 );perf_top__mmap_read (top);
@@ -1317,7 +1317,7 @@ static int __cmd_top(struct perf_top *top)perf_top__mmap_read (top);if (opts->overwrite || (hits == top->samples)) - ret =perf_evlist__poll (top->evlist,100 ); + ret =evlist__poll (top->evlist,100 );if (resize) {perf_top__resize (top);
tools/perf/builtin-trace.c +1 -1
@@ -3474,7 +3474,7 @@ static int trace__run(struct trace *trace, int argc, const char **argv)if (trace->nr_events == before) {int timeout = done ?100 :-1 ; -if (!draining &&perf_evlist__poll (evlist, timeout) >0 ) { +if (!draining &&evlist__poll (evlist, timeout) >0 ) {if (evlist__filter_pollfd (evlist,POLLERR |POLLHUP |POLLNVAL ) ==0 ) draining = true;
tools/perf/lib/evlist.c +5 -0
@@ -276,3 +276,8 @@ int perf_evlist__add_pollfd(struct perf_evlist *evlist, int fd,return pos; } + +int perf_evlist__poll (struct perf_evlist *evlist,int timeout) +{ +return fdarray__poll (&evlist->pollfd, timeout); +}
tools/perf/lib/include/perf/evlist.h +1 -0
@@ -31,5 +31,6 @@ LIBPERF_API void perf_evlist__disable(struct perf_evlist *evlist); LIBPERF_API void perf_evlist__set_maps(struct perf_evlist *evlist,struct perf_cpu_map *cpus,struct perf_thread_map *threads); +LIBPERF_API int perf_evlist__poll(struct perf_evlist *evlist, int timeout);#endif /* __LIBPERF_EVLIST_H */
tools/perf/lib/libperf.map +1 -0
@@ -39,6 +39,7 @@ LIBPERF_0.0.1 { perf_evlist__remove; perf_evlist__next; perf_evlist__set_maps; + perf_evlist__poll; local: *; };
tools/perf/tests/openat-syscall-tp-fields.c +1 -1
@@ -127,7 +127,7 @@ int test__syscall_openat_tp_fields(struct test *test __maybe_unused, int subtest }if (nr_events == before) -perf_evlist__poll (evlist,10 ); +evlist__poll (evlist,10 );if (++nr_polls >5 ) {pr_debug ("%s: no events!\n" , __func__);
tools/perf/tests/perf-record.c +1 -1
@@ -287,7 +287,7 @@ int test__PERF_RECORD(struct test *test __maybe_unused, int subtest __maybe_unus* perf_event_attr.wakeup_events, just PERF_EVENT_SAMPLE does. */ if (total_events == before && false) -perf_evlist__poll (evlist,-1 ); +evlist__poll (evlist,-1 );sleep (1 );if (++wakeups >5 ) {
tools/perf/tests/task-exit.c +1 -1
@@ -130,7 +130,7 @@ int test__task_exit(struct test *test __maybe_unused, int subtest __maybe_unused out_init:if (!exited || !nr_exit) { -perf_evlist__poll (evlist,-1 ); +evlist__poll (evlist,-1 ); goto retry; }
tools/perf/util/evlist.c +3 -3
@@ -418,9 +418,9 @@ int evlist__filter_pollfd(struct evlist *evlist, short revents_and_mask) perf_evlist__munmap_filtered, NULL); } -int perf_evlist__poll (struct evlist *evlist,int timeout) +int evlist__poll (struct evlist *evlist,int timeout) { -return fdarray__poll (&evlist->core.pollfd, timeout); +return perf_evlist__poll (&evlist->core, timeout); } static void perf_evlist__set_sid_idx(struct evlist *evlist,
@@ -1736,7 +1736,7 @@ static void *perf_evlist__poll_thread(void *arg) draining = true;if (!draining) -perf_evlist__poll (evlist,1000 ); +evlist__poll (evlist,1000 );for (i =0 ; i < evlist->core.nr_mmaps; i++) {struct mmap *map = &evlist->mmap[i];
tools/perf/util/evlist.h +1 -1
@@ -144,7 +144,7 @@ perf_evlist__find_tracepoint_by_name(struct evlist *evlist,int evlist__add_pollfd (struct evlist *evlist,int fd);int evlist__filter_pollfd (struct evlist *evlist,short revents_and_mask); -int perf_evlist__poll (struct evlist *evlist,int timeout); +int evlist__poll (struct evlist *evlist,int timeout);struct evsel *perf_evlist__id2evsel (struct evlist *evlist,u64 id); struct evsel *perf_evlist__id2evsel_strict(struct evlist *evlist,
tools/perf/util/python.c +1 -1
@@ -918,7 +918,7 @@ static PyObject *pyrf_evlist__poll(struct pyrf_evlist *pevlist,if (!PyArg_ParseTupleAndKeywords (args, kwargs,"|i" , kwlist, &timeout))return NULL ; - n =perf_evlist__poll (evlist, timeout); + n =evlist__poll (evlist, timeout);if (n <0 ) {PyErr_SetFromErrno (PyExc_OSError);return NULL ;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 55/66 perf copyfile: Move copyfile routines to separate files
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
Further reducing the util.c hodgepodge files.
--- tools/perf/util/Build | 1 + tools/perf/util/build-id.c | 3 +- tools/perf/util/copyfile.c | 144 +++++++++++++++++++++++++++++++++++ tools/perf/util/copyfile.h | 16 ++++ tools/perf/util/symbol-elf.c | 2 +- tools/perf/util/util.c | 135 -------------------------------- tools/perf/util/util.h | 5 -- 7 files changed, 164 insertions(+), 142 deletions(-) create mode 100644 tools/perf/util/copyfile.c create mode 100644 tools/perf/util/copyfile.h
tools/perf/util/Build +1 -0
@@ -3,6 +3,7 @@ perf-y += block-range.o perf-y += build-id.o perf-y += cacheline.o perf-y += config.o +perf-y += copyfile.o perf-y += ctype.o perf-y += db-export.o perf-y += env.o
tools/perf/util/build-id.c +2 -1
@@ -7,12 +7,13 @@* Copyright (C) 2009, 2010 Red Hat Inc. * Copyright (C) 2009, 2010 Arnaldo Carvalho de Melo <[email protected]> */ -#include "util.h" // copyfile_ns(), lsdir(), mkdir_p(), rm_rf() +#include "util.h" // lsdir(), mkdir_p(), rm_rf() #include <dirent.h> #include <errno.h> #include <stdio.h> #include <sys/stat.h> #include <sys/types.h> +#include "util/copyfile.h" #include "dso.h" #include "build-id.h" #include "event.h"
tools/perf/util/copyfile.c new file +144 -0
144 lines not on this page · raw message
tools/perf/util/copyfile.h new file +16 -0
@@ -0,0 +1,16 @@ +// SPDX-License-Identifier: GPL-2.0 +#ifndef PERF_COPYFILE_H_ +#define PERF_COPYFILE_H_ + +#include <linux/types.h> +#include <sys/types.h> +#include <fcntl.h> + +struct nsinfo ; + +int copyfile (const char *from,const char *to); +int copyfile_mode (const char *from,const char *to,mode_t mode); +int copyfile_ns (const char *from,const char *to,struct nsinfo *nsi); +int copyfile_offset (int ifd,loff_t off_in,int ofd,loff_t off_out,u64 size); + +#endif // PERF_COPYFILE_H_
tools/perf/util/symbol-elf.c +1 -1
@@ -17,7 +17,7 @@#include "machine.h" #include "vdso.h" #include "debug.h" -#include "util.h" +#include "util/copyfile.h" #include <linux/ctype.h> #include <linux/kernel.h> #include <linux/zalloc.h>
tools/perf/util/util.c +0 -135
148 lines not on this page · raw message
tools/perf/util/util.h +0 -5
@@ -17,7 +17,6 @@ void usage(const char *err) __noreturn; void die(const char *err, ...) __noreturn __printf(1, 2);struct dirent ; -struct nsinfo ;struct strlist ;int mkdir_p (char *path,mode_t mode);
@@ -25,10 +24,6 @@ int rm_rf(const char *path);int rm_rf_perf_data (const char *path);struct strlist *lsdir (const char *name,bool (*filter)(const char *,struct dirent *));bool lsdir_no_dot_filter (const char *name,struct dirent *d); -int copyfile (const char *from,const char *to); -int copyfile_mode (const char *from,const char *to,mode_t mode); -int copyfile_ns (const char *from,const char *to,struct nsinfo *nsi); -int copyfile_offset (int ifd,loff_t off_in,int ofd,loff_t off_out,u64 size);size_t hex_width (u64 v);
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 56/66 perf evsel: Remove need for symbol_conf in evsel_fprintf.c
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
So that we an later link it to the python binding without having to drag the symbol object files.
--- tools/perf/builtin-sched.c | 2 +- tools/perf/builtin-script.c | 6 ++++-- tools/perf/builtin-trace.c | 2 +- tools/perf/util/evsel.h | 7 ++++--- tools/perf/util/evsel_fprintf.c | 14 ++++++-------- 5 files changed, 16 insertions(+), 15 deletions(-)
tools/perf/builtin-sched.c +1 -1
@@ -2055,7 +2055,7 @@ static void timehist_print_sample(struct perf_sched *sched,EVSEL__PRINT_SYM |EVSEL__PRINT_ONELINE |EVSEL__PRINT_CALLCHAIN_ARROW |EVSEL__PRINT_SKIP_IGNORED , - &callchain_cursor, stdout); + &callchain_cursor, symbol_conf.bt_stop_list, stdout); out:printf ("\n" );
tools/perf/builtin-script.c +4 -2
@@ -1325,7 +1325,8 @@ static int perf_sample__fprintf_bts(struct perf_sample *sample, }else printed +=fprintf (fp,"\n" ); - printed +=sample__fprintf_sym (sample, al,0 , print_opts, cursor, fp); + printed +=sample__fprintf_sym (sample, al,0 , print_opts, cursor, + symbol_conf.bt_stop_list, fp); }/* print branch_to information */
@@ -1867,7 +1868,8 @@ static void process_event(struct perf_script *script, cursor = &callchain_cursor;fputc (cursor ?'\n' :' ' , fp); -sample__fprintf_sym (sample, al,0 , output[type].print_ip_opts, cursor, fp); +sample__fprintf_sym (sample, al,0 , output[type].print_ip_opts, cursor, + symbol_conf.bt_stop_list, fp); }if (PRINT_FIELD (IREGS ))
tools/perf/builtin-trace.c +1 -1
@@ -2076,7 +2076,7 @@ static int trace__fprintf_callchain(struct trace *trace, struct perf_sample *samEVSEL__PRINT_DSO |EVSEL__PRINT_UNKNOWN_AS_ADDR ; -return sample__fprintf_callchain (sample,38 , print_opts, &callchain_cursor, trace->output); +return sample__fprintf_callchain (sample,38 , print_opts, &callchain_cursor, symbol_conf.bt_stop_list, trace->output); }static const char *errno_to_name (struct evsel *evsel,int err)
tools/perf/util/evsel.h +4 -3
@@ -428,12 +428,13 @@ int perf_evsel__fprintf(struct evsel *evsel,struct callchain_cursor ;int sample__fprintf_callchain (struct perf_sample *sample,int left_alignment, -unsigned int print_opts, -struct callchain_cursor *cursor,FILE *fp); +unsigned int print_opts,struct callchain_cursor *cursor, +struct strlist *bt_stop_list,FILE *fp);int sample__fprintf_sym (struct perf_sample *sample,struct addr_location *al,int left_alignment,unsigned int print_opts, -struct callchain_cursor *cursor,FILE *fp); +struct callchain_cursor *cursor, +struct strlist *bt_stop_list,FILE *fp);bool perf_evsel__fallback (struct evsel *evsel,int err,char *msg,size_t msgsize);
tools/perf/util/evsel_fprintf.c +6 -8
@@ -102,7 +102,7 @@ int perf_evsel__fprintf(struct evsel *evsel,int sample__fprintf_callchain (struct perf_sample *sample,int left_alignment,unsigned int print_opts,struct callchain_cursor *cursor, -FILE *fp) +struct strlist *bt_stop_list,FILE *fp) {int printed =0 ;struct callchain_cursor_node *node;
@@ -175,10 +175,8 @@ int sample__fprintf_callchain(struct perf_sample *sample, int left_alignment, printed +=fprintf (fp,"\n" );/* Add srccode here too? */ -if (symbol_conf.bt_stop_list && - node->sym && -strlist__has_entry (symbol_conf.bt_stop_list, - node->sym->name)) { +if (bt_stop_list && node->sym && +strlist__has_entry (bt_stop_list, node->sym->name)) {break ; }
@@ -193,7 +191,7 @@ int sample__fprintf_callchain(struct perf_sample *sample, int left_alignment,int sample__fprintf_sym (struct perf_sample *sample,struct addr_location *al,int left_alignment,unsigned int print_opts, -struct callchain_cursor *cursor,FILE *fp) +struct callchain_cursor *cursor,struct strlist *bt_stop_list,FILE *fp) {int printed =0 ;int print_ip = print_opts &EVSEL__PRINT_IP ;
@@ -204,8 +202,8 @@ int sample__fprintf_sym(struct perf_sample *sample, struct addr_location *al,int print_unknown_as_addr = print_opts &EVSEL__PRINT_UNKNOWN_AS_ADDR ;if (cursor !=NULL ) { - printed +=sample__fprintf_callchain (sample, left_alignment, - print_opts, cursor, fp); + printed +=sample__fprintf_callchain (sample, left_alignment, print_opts, + cursor, bt_stop_list, fp); }else { printed +=fprintf (fp,"%-*.*s" , left_alignment, left_alignment," " );
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 57/66 perf evsel: Introduce evsel_fprintf.h
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
We already had evsel_fprintf.c, add its counterpart, so that we can reduce evsel.h a bit more.
We needed a new perf_event_attr_fprintf.c file so as to have a separate object to link with the python binding in tools/perf/util/python-ext-sources and not drag symbol_conf, etc into the python binding.
--- tools/perf/builtin-evlist.c | 1 + tools/perf/builtin-sched.c | 1 + tools/perf/builtin-script.c | 1 + tools/perf/builtin-trace.c | 2 + tools/perf/util/Build | 1 + tools/perf/util/evsel.c | 153 +--------------------- tools/perf/util/evsel.h | 51 +------- tools/perf/util/evsel_fprintf.c | 1 + tools/perf/util/evsel_fprintf.h | 50 +++++++ tools/perf/util/header.c | 1 + tools/perf/util/perf_event_attr_fprintf.c | 148 +++++++++++++++++++++ tools/perf/util/python-ext-sources | 1 + 12 files changed, 218 insertions(+), 193 deletions(-) create mode 100644 tools/perf/util/evsel_fprintf.h create mode 100644 tools/perf/util/perf_event_attr_fprintf.c
tools/perf/builtin-evlist.c +1 -0
@@ -10,6 +10,7 @@#include "perf.h" #include "util/evlist.h" #include "util/evsel.h" +#include "util/evsel_fprintf.h" #include "util/parse-events.h" #include <subcmd/parse-options.h> #include "util/session.h"
tools/perf/builtin-sched.c +1 -0
@@ -6,6 +6,7 @@#include "util/cpumap.h" #include "util/evlist.h" #include "util/evsel.h" +#include "util/evsel_fprintf.h" #include "util/symbol.h" #include "util/thread.h" #include "util/header.h"
tools/perf/builtin-script.c +1 -0
@@ -17,6 +17,7 @@#include "util/trace-event.h" #include "util/evlist.h" #include "util/evsel.h" +#include "util/evsel_fprintf.h" #include "util/evswitch.h" #include "util/sort.h" #include "util/data.h"
tools/perf/builtin-trace.c +2 -0
@@ -28,6 +28,8 @@#include "util/dso.h" #include "util/env.h" #include "util/event.h" +#include "util/evsel.h" +#include "util/evsel_fprintf.h" #include "util/synthetic-events.h" #include "util/evlist.h" #include "util/evswitch.h"
tools/perf/util/Build +1 -0
@@ -11,6 +11,7 @@ perf-y += event.o perf-y += evlist.o perf-y += evsel.o perf-y += evsel_fprintf.o +perf-y += perf_event_attr_fprintf.o perf-y += evswitch.o perf-y += find_bit.o perf-y += get_current_dir_name.o
tools/perf/util/evsel.c +7 -146
171 lines not on this page · raw message
tools/perf/util/evsel.h +4 -47
88 lines not on this page · raw message
tools/perf/util/evsel_fprintf.c +1 -0
@@ -4,6 +4,7 @@#include <stdbool.h> #include <traceevent/event-parse.h> #include "evsel.h" +#include "util/evsel_fprintf.h" #include "util/event.h" #include "callchain.h" #include "map.h"
tools/perf/util/evsel_fprintf.h new file +50 -0
50 lines not on this page · raw message
tools/perf/util/header.c +1 -0
@@ -25,6 +25,7 @@#include "dso.h" #include "evlist.h" #include "evsel.h" +#include "util/evsel_fprintf.h" #include "header.h" #include "memswap.h" #include "trace-event.h"
tools/perf/util/perf_event_attr_fprintf.c new file +148 -0
148 lines not on this page · raw message
tools/perf/util/python-ext-sources +1 -0
@@ -10,6 +10,7 @@ util/python.c util/cap.c util/evlist.c util/evsel.c +util/perf_event_attr_fprintf.c util/cpumap.c util/memswap.c util/mmap.c
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 58/66 perf evlist: Remove unused perf_evlist__fprintf() method
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
Ditch it, noone is using it, one more stdio.h include in a hot header.
Fix the fallout in parse-events.y, where we end up using a FILE pointer, I think due to YYDEBUG being set and in some places, like Amazon Linux 1 we don't get stdio.h included by luck, like in most other places, add a explicit stdio.h include directive.
--- tools/perf/util/evlist.c | 13 ------------- tools/perf/util/evlist.h | 3 --- tools/perf/util/parse-events.y | 1 + 3 files changed, 1 insertion(+), 16 deletions(-)
tools/perf/util/evlist.c +0 -13
@@ -1421,19 +1421,6 @@ int perf_evlist__parse_sample_timestamp(struct evlist *evlist,return perf_evsel__parse_sample_timestamp (evsel, event, timestamp); } -size_t perf_evlist__fprintf (struct evlist *evlist,FILE *fp) -{ -struct evsel *evsel; -size_t printed =0 ; - -evlist__for_each_entry (evlist, evsel) { - printed +=fprintf (fp,"%s%s" , evsel->idx ?", " :"" , -perf_evsel__name (evsel)); - } - -return printed +fprintf (fp,"\n" ); -} -int perf_evlist__strerror_open (struct evlist *evlist,int err,char *buf,size_t size) {
tools/perf/util/evlist.h +0 -3
@@ -7,7 +7,6 @@#include <linux/refcount.h> #include <linux/list.h> #include <api/fd/array.h> -#include <stdio.h> #include <internal/evlist.h> #include <internal/evsel.h> #include "events_stats.h"
@@ -249,8 +248,6 @@ static inline struct evsel *evlist__last(struct evlist *evlist) return container_of(evsel, struct evsel, core); } -size_t perf_evlist__fprintf (struct evlist *evlist,FILE *fp); -int perf_evlist__strerror_open (struct evlist *evlist,int err,char *buf,size_t size);int perf_evlist__strerror_mmap (struct evlist *evlist,int err,char *buf,size_t size);
tools/perf/util/parse-events.y +1 -0
@@ -9,6 +9,7 @@ #define YYDEBUG 1 #include <fnmatch.h> +#include <stdio.h> #include <linux/compiler.h> #include <linux/list.h> #include <linux/types.h>
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 59/66 perf evsel: Move config terms to a separate header
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
Further reducing the size of util/evsel.h.
--- tools/perf/arch/arm/util/cs-etm.c | 1 + tools/perf/builtin-top.c | 1 + tools/perf/util/evsel.c | 1 + tools/perf/util/evsel.h | 43 -------------------------- tools/perf/util/evsel_config.h | 50 +++++++++++++++++++++++++++++++ tools/perf/util/parse-events.c | 1 + 6 files changed, 54 insertions(+), 43 deletions(-) create mode 100644 tools/perf/util/evsel_config.h
tools/perf/arch/arm/util/cs-etm.c +1 -0
@@ -23,6 +23,7 @@#include "../../util/event.h" #include "../../util/evlist.h" #include "../../util/evsel.h" +#include "../../util/evsel_config.h" #include "../../util/pmu.h" #include "../../util/cs-etm.h" #include <internal/lib.h> // page_size
tools/perf/builtin-top.c +1 -0
@@ -27,6 +27,7 @@#include "util/dso.h" #include "util/evlist.h" #include "util/evsel.h" +#include "util/evsel_config.h" #include "util/event.h" #include "util/machine.h" #include "util/map.h"
tools/perf/util/evsel.c +1 -0
@@ -30,6 +30,7 @@#include "counts.h" #include "event.h" #include "evsel.h" +#include "util/evsel_config.h" #include "util/evsel_fprintf.h" #include "evlist.h" #include <perf/cpumap.h>
tools/perf/util/evsel.h +0 -43
49 lines not on this page · raw message
tools/perf/util/evsel_config.h new file +50 -0
50 lines not on this page · raw message
tools/perf/util/parse-events.c +1 -0
@@ -34,6 +34,7 @@#include "asm/bug.h" #include "util/parse-branch-options.h" #include "metricgroup.h" +#include "util/evsel_config.h" #include "util/mmap.h" #define MAX_NAME_LEN 100
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 60/66 perf tools: Replace needless mmap.h with what is needed, event.h
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
The perf_sample struct definition and the event_attr_init() are in util/event.h, but some places were getting it thru an otherwise needless util/mmap.h header, fix it by including util/event.h directly.
--- tools/perf/util/annotate.c | 2 +- tools/perf/util/parse-events.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
tools/perf/util/annotate.c +1 -1
@@ -34,7 +34,7 @@#include "bpf-event.h" #include "block-range.h" #include "string2.h" -#include "util/mmap.h" +#include "util/event.h" #include "arch/common.h" #include <regex.h> #include <pthread.h>
tools/perf/util/parse-events.c +1 -1
@@ -35,7 +35,7 @@#include "util/parse-branch-options.h" #include "metricgroup.h" #include "util/evsel_config.h" -#include "util/mmap.h" +#include "util/event.h" #define MAX_NAME_LEN 100
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 61/66 perf stat: Fix free memory access / memory leaks in metrics
In reply to Arnaldo Carvalho de MeloFrom: Andi Kleen <[email protected]>
Make sure to not free the name passed in by the caller, but free all the allocated ids when parsing expressions.
The loop at the end knows that the first entry shouldn't be freed, so make sure the caller name is the first entry.
Fixes
% perf stat -M IpB,IpCall,IpTB,IPC,Retiring_SMT,Frontend_Bound_SMT,Kernel_Utilization,CPU_Utilization --metric-only -a -I 1000 sleep 2
valgrind:
1.009943231 ==21527== Invalid read of size 1 ==21527== at 0x483CB74: strcmp (vg_replace_strmem.c:849) ==21527== by 0x582CF8: collect_all_aliases (stat-display.c:554) ==21527== by 0x582EB3: collect_data (stat-display.c:577) ==21527== by 0x583A32: print_counter_aggr (stat-display.c:806) ==21527== by 0x584FAD: perf_evlist__print_counters (stat-display.c:1200) ==21527== by 0x45133A: print_counters (builtin-stat.c:655) ==21527== by 0x450629: process_interval (builtin-stat.c:353) ==21527== by 0x450FBD: __run_perf_stat (builtin-stat.c:564) ==21527== by 0x451285: run_perf_stat (builtin-stat.c:636) ==21527== by 0x454619: cmd_stat (builtin-stat.c:1966) ==21527== by 0x4D557D: run_builtin (perf.c:310) ==21527== by 0x4D57EA: handle_internal_command (perf.c:362) ==21527== Address 0x12826cd0 is 0 bytes inside a block of size 25 free'd ==21527== at 0x4839A0C: free (vg_replace_malloc.c:540) ==21527== by 0x627041: __zfree (zalloc.c:13) ==21527== by 0x57F66A: generic_metric (stat-shadow.c:814) ==21527== by 0x580B21: perf_stat__print_shadow_stats (stat-shadow.c:1057) ==21527== by 0x58418E: print_metric_headers (stat-display.c:943) ==21527== by 0x5844BC: print_interval (stat-display.c:1004) ==21527== by 0x584DEB: perf_evlist__print_counters (stat-display.c:1172) ==21527== by 0x45133A: print_counters (builtin-stat.c:655) ==21527== by 0x450629: process_interval (builtin-stat.c:353) ==21527== by 0x450FBD: __run_perf_stat (builtin-stat.c:564) ==21527== by 0x451285: run_perf_stat (builtin-stat.c:636) ==21527== by 0x454619: cmd_stat (builtin-stat.c:1966) ==21527== Block was alloc'd at ==21527== at 0x483880B: malloc (vg_replace_malloc.c:309) ==21527== by 0x51677DE: strdup (in /usr/lib64/libc-2.29.so) ==21527== by 0x506457: parse_events_name (parse-events.c:1754) ==21527== by 0x5550BB: parse_events_parse (parse-events.y:214) ==21527== by 0x50694D: parse_events__scanner (parse-events.c:1887) ==21527== by 0x506AEF: parse_events (parse-events.c:1927) ==21527== by 0x521D8B: metricgroup__parse_groups (metricgroup.c:527) ==21527== by 0x45156F: parse_metric_groups (builtin-stat.c:721) ==21527== by 0x6228A9: get_value (parse-options.c:243) ==21527== by 0x62363F: parse_short_opt (parse-options.c:348) ==21527== by 0x62363F: parse_options_step (parse-options.c:536) ==21527== by 0x62363F: parse_options_subcommand (parse-options.c:651) ==21527== by 0x453C1D: cmd_stat (builtin-stat.c:1718) ==21527== by 0x4D557D: run_builtin (perf.c:310)
and also a leak report.
Committer testing:
Before:
# perf stat -M IpB,IpCall,IpTB,IPC,Retiring_SMT,Frontend_Bound_SMT,Kernel_Utilization,CPU_Utilization --metric-only -a -I 1000 sleep 2
# time CPU_Utilization
1.000470810 free(): double free detected in tcache 2
Aborted (core dumped)
#After:
# perf stat -M IpB,IpCall,IpTB,IPC,Retiring_SMT,Frontend_Bound_SMT,Kernel_Utilization,CPU_Utilization --metric-only -a -I 1000 sleep 2
# time CPU_Utilization
1.000494752 0.1
2.001105112 0.1
#--- tools/perf/util/stat-shadow.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tools/perf/util/stat-shadow.c +2 -2
@@ -738,6 +738,8 @@ static void generic_metric(struct perf_stat_config *config,char *n, *pn;expr__ctx_init (&pctx); +/* Must be first id entry */ +expr__add_id (&pctx, name, avg);for (i =0 ; metric_events[i]; i++) {struct saved_value *v;struct stats *stats;
@@ -776,8 +778,6 @@ static void generic_metric(struct perf_stat_config *config,expr__add_id (&pctx, n,avg_stats (stats)*scale); } -expr__add_id (&pctx, name, avg); -if (!metric_events[i]) {const char *p = metric_expr;
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 62/66 perf evlist: Fix access of freed id arrays
In reply to Arnaldo Carvalho de MeloFrom: Andi Kleen <[email protected]>
I'm not fully sure if this is the correct fix, but without this I get crashes on more complex perf stat metric usages. The problem is that part of the state gets freed when a weak group fails, but then is later still used. Just don't free the ids, we're going to reuse them anyways on the weak group retry.
For example:
% perf stat -M IpB,IpCall,IpTB,IPC,Retiring_SMT,Frontend_Bound_SMT,Kernel_Utilization,CPU_Utilization --metric-only -a -I 1000 sleep 2
crashes and gives in valgrind:
=21527== Invalid write of size 8
==21527== at 0x4EE582: hlist_add_head (list.h:644) ==21527== by 0x4EFD3C: perf_evlist__id_hash (evlist.c:477) ==21527== by 0x4EFD99: perf_evlist__id_add (evlist.c:483) ==21527== by 0x4EFF15: perf_evlist__id_add_fd (evlist.c:524) ==21527== by 0x4FC693: store_evsel_ids (evsel.c:2969) ==21527== by 0x4FC76C: perf_evsel__store_ids (evsel.c:2986) ==21527== by 0x450DA7: __run_perf_stat (builtin-stat.c:519) ==21527== by 0x451285: run_perf_stat (builtin-stat.c:636) ==21527== by 0x454619: cmd_stat (builtin-stat.c:1966) ==21527== by 0x4D557D: run_builtin (perf.c:310) ==21527== by 0x4D57EA: handle_internal_command (perf.c:362) ==21527== by 0x4D5931: run_argv (perf.c:406) ==21527== Address 0x12e3f008 is 104 bytes inside a block of size 2,056 free'd ==21527== at 0x4839A0C: free (vg_replace_malloc.c:540) ==21527== by 0x627139: xyarray__delete (xyarray.c:32) ==21527== by 0x4F6BE4: perf_evsel__free_id (evsel.c:1253) ==21527== by 0x4FA11F: evsel__close (evsel.c:1994) ==21527== by 0x4F30A3: perf_evlist__reset_weak_group (evlist.c:1783) ==21527== by 0x450B47: __run_perf_stat (builtin-stat.c:466) ==21527== by 0x451285: run_perf_stat (builtin-stat.c:636) ==21527== by 0x454619: cmd_stat (builtin-stat.c:1966) ==21527== by 0x4D557D: run_builtin (perf.c:310) ==21527== by 0x4D57EA: handle_internal_command (perf.c:362) ==21527== by 0x4D5931: run_argv (perf.c:406) ==21527== by 0x4D5CAE: main (perf.c:531) ==21527== Block was alloc'd at ==21527== at 0x483AB1A: calloc (vg_replace_malloc.c:762) ==21527== by 0x627024: zalloc (zalloc.c:8) ==21527== by 0x627088: xyarray__new (xyarray.c:10) ==21527== by 0x4F6B20: perf_evsel__alloc_id (evsel.c:1237) ==21527== by 0x4FC74E: perf_evsel__store_ids (evsel.c:2983) ==21527== by 0x450DA7: __run_perf_stat (builtin-stat.c:519) ==21527== by 0x451285: run_perf_stat (builtin-stat.c:636) ==21527== by 0x454619: cmd_stat (builtin-stat.c:1966) ==21527== by 0x4D557D: run_builtin (perf.c:310) ==21527== by 0x4D57EA: handle_internal_command (perf.c:362) ==21527== by 0x4D5931: run_argv (perf.c:406) ==21527== by 0x4D5CAE: main (perf.c:531)
--- tools/perf/util/evlist.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/util/evlist.c +1 -1
@@ -1659,7 +1659,7 @@ struct evsel *perf_evlist__reset_weak_group(struct evlist *evsel_list, is_open = false;if (c2->leader == leader) {if (is_open) -evsel__close (c2); +perf_evsel__close (&evsel->core); c2->leader = c2; c2->core.nr_members =0 ; }
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 63/66 perf vendor events: Remove P8 HW events which are not supported
In reply to Arnaldo Carvalho de MeloFrom: Mamatha Inamdar <[email protected]>
This patch is to remove following hardware events
from JSON file which are not supported on POWER8.
pm_l3_p0_grp_pumppm_l3_p0_lco_datapm_l3_p0_lco_no_datapm_l3_p0_lco_rty
Note: Unfortunately power8 event list is not publicly available.
--- .../pmu-events/arch/powerpc/power8/other.json | 24 ------------------- 1 file changed, 24 deletions(-)
tools/perf/pmu-events/arch/powerpc/power8/other.json +0 -24
@@ -1775,30 +1775,6 @@ "BriefDescription": "L3 Load Prefetches", "PublicDescription": "" }, - {, - "EventCode": "0xa29084", - "EventName": "PM_L3_P0_GRP_PUMP", - "BriefDescription": "L3 pf sent with grp scope port 0", - "PublicDescription": "" - }, - {, - "EventCode": "0x528084", - "EventName": "PM_L3_P0_LCO_DATA", - "BriefDescription": "lco sent with data port 0", - "PublicDescription": "" - }, - {, - "EventCode": "0x518080", - "EventName": "PM_L3_P0_LCO_NO_DATA", - "BriefDescription": "dataless l3 lco sent port 0", - "PublicDescription": "" - }, - {, - "EventCode": "0xa4908c", - "EventName": "PM_L3_P0_LCO_RTY", - "BriefDescription": "L3 LCO received retry port 0", - "PublicDescription": "" - }, {, "EventCode": "0x84908d", "EventName": "PM_L3_PF0_ALLOC",
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 64/66 perf jvmti: Include JVMTI support for s390
In reply to Arnaldo Carvalho de MeloFrom: Thomas Richter <[email protected]>
Enable JVMTI support for s390 perf tool chain.
--- tools/perf/arch/s390/Makefile | 1 + tools/perf/util/genelf.h | 3 +++ 2 files changed, 4 insertions(+)
tools/perf/arch/s390/Makefile +1 -0
@@ -4,6 +4,7 @@ PERF_HAVE_DWARF_REGS := 1 endif HAVE_KVM_STAT_SUPPORT := 1 PERF_HAVE_ARCH_REGS_QUERY_REGISTER_OFFSET := 1 +PERF_HAVE_JITDUMP := 1 # # Syscall table generation for perf
tools/perf/util/genelf.h +3 -0
@@ -35,6 +35,9 @@ int jit_add_debug_info(Elf *e, uint64_t code_addr, void *debug, int nr_debug_ent#elif defined(__sparc__)#define GEN_ELF_ARCH EM_SPARC#define GEN_ELF_CLASS ELFCLASS32 +#elif defined(__s390x__) +#define GEN_ELF_ARCH EM_S390 +#define GEN_ELF_CLASS ELFCLASS64#else #error "unsupported architecture"#endif
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 65/66 perf build: Add detection of java-11-openjdk-devel package
In reply to Arnaldo Carvalho de MeloFrom: Thomas Richter <[email protected]>
With Java 11 there is no seperate JRE anymore.
Details:
https://coderanch.com/t/701603/java/JRE-JDK
Therefore the detection of the JRE needs to be adapted.
This change works for s390 and x86. I have not tested other platforms.
Committer testing:
Continues to work with the OpenJDK 8:
$ rm -f ~acme/lib64/libperf-jvmti.so $ rpm -qa | grep jdk-devel java-1.8.0-openjdk-devel-1.8.0.222.b10-0.fc30.x86_64 $ git log --oneline -1 a51937170f33 (HEAD -> perf/core) perf build: Add detection of java-11-openjdk-devel package $ rm -rf /tmp/build/perf ; mkdir -p /tmp/build/perf ; make -C tools/perf O=/tmp/build/perf install > /dev/null 2>1 $ ls -la ~acme/lib64/libperf-jvmti.so -rwxr-xr-x. 1 acme acme 230744 Sep 24 16:46 /home/acme/lib64/libperf-jvmti.so $
--- tools/perf/Makefile.config | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
tools/perf/Makefile.config +1 -1
@@ -924,7 +924,7 @@ ifndef NO_JVMTI JDIR=$(shell /usr/sbin/update-java-alternatives -l | head -1 | awk '{print $$3}') else ifneq (,$(wildcard /usr/sbin/alternatives)) - JDIR=$(shell /usr/sbin/alternatives --display java | tail -1 | cut -d' ' -f 5 | sed 's%/jre/bin/java.%%g') + JDIR=$(shell /usr/sbin/alternatives --display java | tail -1 | cut -d' ' -f 5 | sed -e 's%/jre/bin/java.%%g' -e 's%/bin/java.%%g') endif endif ifndef JDIR
Signature · 1 line
2.21.0
Arnaldo Carvalho de Melo
maintainerPATCH 66/66 perf parser: Remove needless include directives
In reply to Arnaldo Carvalho de MeloFrom: Arnaldo Carvalho de Melo <[email protected]>
They go on accumulating there like the debug.h one, that was introduced here:
f23610245c1a ("perf list: Add debug support for outputing alias string")
But then, when that need is removed via:
2073ad3326b7 ("perf tools: Factor out PMU matching in parser")
The thing stays there, so continue the house cleaning spree...
list.h not needed, no macros from there are used, and 'struct list_head' is in linux/types.h, ditto for util.h, no need for that as well.
--- tools/perf/util/parse-events.y | 3 --- 1 file changed, 3 deletions(-)
tools/perf/util/parse-events.y +0 -3
@@ -11,12 +11,9 @@ #include <fnmatch.h> #include <stdio.h> #include <linux/compiler.h> -#include <linux/list.h> #include <linux/types.h> -#include "util.h" #include "pmu.h" #include "evsel.h" -#include "debug.h" #include "parse-events.h" #include "parse-events-bison.h"
Signature · 1 line
2.21.0
Quoted 348 lines · Arnaldo Carvalho de Melo
Hi Ingo/Thomas,
Please consider pulling,
Best regards,
- Arnaldo
Test results at the end of this message, as usual.
The following changes since commit
2b32769700f857a8e608a8ee24080833889965b9:Merge tag 'perf-urgent-for-mingo-5.4-20190921' of
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linuxinto perf/urgent (2019-09-22 12:45:11 +0200)are available in the Git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.gittags/perf-core-for-mingo-5.5-20190925for you to fetch changes up to
d6840d87b2d148e19e244ad2b44d28ba07f437a0:perf parser: Remove needless include directives (2019-09-25 16:26:41 -0300)
----------------------------------------------------------------
perf/core improvements and fixes:perf record:
Stephane Eranian:
- Fix priv level with branch sampling for paranoid=2, i.e. the kernel checks if perf_event_attr_attr.exclude_hv is set in addition to .exclude_kernel, so reset both to zero.
Arnaldo Carvalho de Melo:
- Don't warn about not being able to read kernel maps (kallsyms, etc) when kernel samples aren't being collected.
perf list:
Kim Phillips:
- Allow plurals for metric, metricgroup., i.e.:
$ perf list metricswas showing nothing, which is very confusing, make it work like:
$ perf stat metric
perf stat:
Andi Kleen:
- Free memory access/leaks detected via valgrind, related to metrics.
Libraries:
libperf:
Jiri Olsa:
- Move more stuff from
tools/perf, this time a first stab at movingperf_mmapmethods.libtracevent:
Steven Rostedt (VMware):
- Round up in
tep_print_event()time precision.Tzvetomir Stoyanov (VMware):
- Man pages for event print and related and plugins APIs.
- Move traceevent plugins in its own subdirectory.
Feature detection:
Thomas Richter:
- Add detection of java-11-openjdk-devel package, in addition to the older versions supported.
Architecture specific:
S/390:
Thomas Richter (2):
- Include JVMTI support for s390
Vendor events:
AMD:
Kim Phillips:
- Add L3 cache events for Family 17h.
- Remove redundant '['.
PowerPC:
Mamatha Inamdar:
- Remove P8 HW events which are not supported.
Cleanups:
Arnaldo Carvalho de Melo:
- Remove needless headers, add needed ones, move things around to reduce the headers dependency tree, speeding up builds by not doing needless compiles when unrelated stuff gets changed.
- Ditch unused code that was dragging headers.
Signed-off-by Arnaldo Carvalho de Melo <[email protected]>----------------------------------------------------------------
Andi Kleen (2):
perf stat: Fix free memory access / memory leaks in metrics perf evlist: Fix access of freed id arraysArnaldo Carvalho de Melo (12):
perf record: Move restricted maps check to after a possible fallback to not collect kernel samples perf evlist: Adopt backwards ring buffer state enum
libperf: Add missing 'struct xyarray' forward declaration
perf tools: No need to includeinternal/lib.hfromutil/util.h
libperf: Usesys/types.hto getssize_t, notunistd.h
perf copyfile: Move copyfile routines to separate files
perf evsel: Remove need forsymbol_confinevsel_fprintf.c
perf evsel: Introduceevsel_fprintf.h
perf evlist: Remove unusedperf_evlist__fprintf()method
perf evsel: Move config terms to a separate header
perf tools: Replace needlessmmap.hwith what is needed,event.h
perf parser: Remove needless include directivesJiri Olsa (37):
tools: Add missingstdio.hinclude toasm/bug.hheader perf tools: Rename 'structperf_mmap' to 'struct mmap' perf tools: Renameperf_evlist__mmap()toevlist__mmap()perf tools: Renameperf_evlist__munmap()toevlist__munmap()perf tools: Renameperf_evlist__alloc_mmap()toevlist__alloc_mmap()perf tools: Renameperf_evlist__exit()toevlist__exit()perf tools: Renameperf_evlist__purge()toevlist__purge()
libperf: Link libapi.a in libperf.so
libperf: Addperf_mmapstruct
libperf: Add 'mask' to structperf_mmap
libperf: Add 'fd' to structperf_mmap
libperf: Add 'cpu' to structperf_mmap
libperf: Add 'refcnt' to structperf_mmap
libperf: Add prev/start/end to structperf_mmap
libperf: Add 'overwrite' to 'structperf_mmap'
libperf: Add 'event_copy' to 'structperf_mmap'
libperf: Add 'flush' to 'structperf_mmap'
libperf: Move 'system_wide' from 'struct evsel' to 'structperf_evsel'
libperf: Move 'nr_mmaps' from 'struct evlist' to 'structperf_evlist'
libperf: Move 'mmap_len' from 'struct evlist' to 'structperf_evlist'
libperf: Move 'pollfd' from 'struct evlist' to 'structperf_evlist'
libperf: Move 'sample_id' from 'struct evsel' to 'structperf_evsel'
libperf: Move 'id' from 'struct evsel' to 'structperf_evsel'
libperf: Move 'ids' from 'struct evsel' to 'structperf_evsel'
libperf: Move 'heads' from 'struct evlist' to 'structperf_evlist'
libperf: Add perf_evsel__alloc_id/perf_evsel__free_id functions
libperf: Add perf_evlist__first()/last() functions
libperf: Addperf_evlist__read_format()function
libperf: Addperf_evlist__id_add()function
libperf: Addperf_evlist__id_add_fd()function
libperf: Move 'page_size' global variable to libperf
libperf: Add libperf dependency for tests targets
libperf: Mergelibperf_set_print()intolibperf_init()
libperf: Addlibperf_init()call to the tests
libperf: Addperf_evlist__alloc_pollfd()function
libperf: Addperf_evlist__add_pollfd()function
libperf: Addperf_evlist__poll()functionKim Phillips (4):
perf vendor events amd: Add L3 cache events for Family 17h perf vendor events amd: Remove redundant '[' perf vendor events: Minor fixes to the README perf list: Allow plurals for metric, metricgroupMamatha Inamdar (1):
perf vendor events: Remove P8 HW events which are not supportedStephane Eranian (1):
perf record: Fix priv level with branch sampling for paranoid=2Steven Rostedt (VMware) (1):
libtraceevent: Round up intep_print_event()time precisionThomas Richter (2):
perf jvmti: Include JVMTI support for s390
perf build: Add detection of java-11-openjdk-devel packageTzvetomir Stoyanov (2):
libtraceevent: Man pages for libtraceevent event print related API
libtraceevent: Man pages for tep plugins APIsTzvetomir Stoyanov (VMware) (4):
libtraceevent: Man pages fix, renametep_ref_get()totep_get_ref()
libtraceevent: Man pages fix, changes in event printing APIs
libtraceevent: Addtep_get_event()inevent-parse.h
libtraceevent: Move traceevent plugins in its own subdirectorytools/include/asm/bug.h | 1 + tools/lib/traceevent/Build | 11 - .../Documentation/libtraceevent-event_print.txt | 130 +++++++++ .../Documentation/libtraceevent-handle.txt | 8 +- .../Documentation/libtraceevent-plugins.txt | 99 +++++++ .../lib/traceevent/Documentation/libtraceevent.txt | 15 +- tools/lib/traceevent/Makefile | 94 ++----- tools/lib/traceevent/event-parse.c | 4 +- tools/lib/traceevent/event-parse.h | 2 + tools/lib/traceevent/plugins/Build | 10 + tools/lib/traceevent/plugins/Makefile | 222 ++++++++++++++++ .../lib/traceevent/{ => plugins}/plugin_cfg80211.c | 0 .../lib/traceevent/{ => plugins}/plugin_function.c | 0 .../lib/traceevent/{ => plugins}/plugin_hrtimer.c | 0 tools/lib/traceevent/{ => plugins}/plugin_jbd2.c | 0 tools/lib/traceevent/{ => plugins}/plugin_kmem.c | 0 tools/lib/traceevent/{ => plugins}/plugin_kvm.c | 0 .../lib/traceevent/{ => plugins}/plugin_mac80211.c | 0 .../traceevent/{ => plugins}/plugin_sched_switch.c | 0 tools/lib/traceevent/{ => plugins}/plugin_scsi.c | 0 tools/lib/traceevent/{ => plugins}/plugin_xen.c | 0 tools/perf/Makefile.config | 2 +- tools/perf/Makefile.perf | 4 +- tools/perf/arch/arm/util/cs-etm.c | 7 +- tools/perf/arch/arm64/util/arm-spe.c | 6 +- tools/perf/arch/s390/Makefile | 1 + tools/perf/arch/s390/util/auxtrace.c | 1 + tools/perf/arch/s390/util/machine.c | 2 +- tools/perf/arch/x86/tests/intel-cqm.c | 5 +- tools/perf/arch/x86/tests/perf-time-to-tsc.c | 11 +- tools/perf/arch/x86/tests/rdpmc.c | 2 +- tools/perf/arch/x86/util/intel-bts.c | 9 +- tools/perf/arch/x86/util/intel-pt.c | 17 +- tools/perf/arch/x86/util/machine.c | 2 +- tools/perf/builtin-evlist.c | 1 + tools/perf/builtin-kvm.c | 13 +- tools/perf/builtin-list.c | 4 +- tools/perf/builtin-record.c | 102 +++---- tools/perf/builtin-sched.c | 3 +- tools/perf/builtin-script.c | 11 +- tools/perf/builtin-stat.c | 6 +- tools/perf/builtin-top.c | 22 +- tools/perf/builtin-trace.c | 17 +- tools/perf/lib/Makefile | 35 ++- tools/perf/lib/core.c | 13 +- tools/perf/lib/evlist.c | 124 +++++++++ tools/perf/lib/evsel.c | 30 +++ tools/perf/lib/include/internal/evlist.h | 33 +++ tools/perf/lib/include/internal/evsel.h | 33 +++ tools/perf/lib/include/internal/lib.h | 4 +- tools/perf/lib/include/internal/mmap.h | 32 +++ tools/perf/lib/include/perf/core.h | 2 +- tools/perf/lib/include/perf/evlist.h | 1 + tools/perf/lib/lib.c | 2 + tools/perf/lib/libperf.map | 3 +- tools/perf/lib/tests/test-cpumap.c | 10 + tools/perf/lib/tests/test-evlist.c | 10 + tools/perf/lib/tests/test-evsel.c | 10 + tools/perf/lib/tests/test-threadmap.c | 10 + tools/perf/perf.c | 13 +- tools/perf/pmu-events/README | 22 +- .../perf/pmu-events/arch/powerpc/power8/other.json | 24 -- .../perf/pmu-events/arch/x86/amdfam17h/cache.json | 42 +++ tools/perf/pmu-events/arch/x86/amdfam17h/core.json | 2 +- tools/perf/pmu-events/jevents.c | 1 + tools/perf/tests/backward-ring-buffer.c | 11 +- tools/perf/tests/bpf.c | 9 +- tools/perf/tests/code-reading.c | 11 +- tools/perf/tests/event-times.c | 14 +- tools/perf/tests/event_update.c | 6 +- tools/perf/tests/evsel-roundtrip-name.c | 2 +- tools/perf/tests/hists_cumulate.c | 2 +- tools/perf/tests/hists_link.c | 5 +- tools/perf/tests/hists_output.c | 2 +- tools/perf/tests/keep-tracking.c | 11 +- tools/perf/tests/mmap-basic.c | 5 +- tools/perf/tests/mmap-thread-lookup.c | 2 +- tools/perf/tests/openat-syscall-tp-fields.c | 11 +- tools/perf/tests/parse-events.c | 116 ++++---- tools/perf/tests/perf-record.c | 13 +- tools/perf/tests/sdt.c | 1 + tools/perf/tests/sw-clock.c | 5 +- tools/perf/tests/switch-tracking.c | 29 +- tools/perf/tests/task-exit.c | 9 +- tools/perf/tests/vmlinux-kallsyms.c | 2 +- tools/perf/ui/browsers/hists.c | 6 +- tools/perf/ui/gtk/hists.c | 1 + tools/perf/util/Build | 2 + tools/perf/util/annotate.c | 1 + tools/perf/util/auxtrace.c | 8 +- tools/perf/util/auxtrace.h | 8 +- tools/perf/util/bpf-loader.c | 2 +- tools/perf/util/build-id.c | 3 +- tools/perf/util/copyfile.c | 144 ++++++++++ tools/perf/util/copyfile.h | 16 ++ tools/perf/util/cs-etm.c | 2 +- tools/perf/util/evlist.c | 295 ++++++--------------- tools/perf/util/evlist.h | 81 +++--- tools/perf/util/evsel.c | 204 ++------------ tools/perf/util/evsel.h | 121 +-------- tools/perf/util/evsel_config.h | 50 ++++ tools/perf/util/evsel_fprintf.c | 15 +- tools/perf/util/evsel_fprintf.h | 50 ++++ tools/perf/util/genelf.h | 3 + tools/perf/util/header.c | 29 +- tools/perf/util/intel-bts.c | 4 +- tools/perf/util/intel-pt.c | 10 +- tools/perf/util/jitdump.c | 2 +- tools/perf/util/machine.c | 1 + tools/perf/util/mmap.c | 185 ++++++------- tools/perf/util/mmap.h | 77 ++---- tools/perf/util/parse-events.c | 8 +- tools/perf/util/parse-events.y | 4 +- tools/perf/util/perf_event_attr_fprintf.c | 148 +++++++++++ tools/perf/util/python-ext-sources | 1 + tools/perf/util/python.c | 24 +- tools/perf/util/record.c | 6 +- tools/perf/util/session.c | 5 +- tools/perf/util/sort.c | 2 +- tools/perf/util/srccode.c | 2 +- tools/perf/util/stat-shadow.c | 4 +- tools/perf/util/stat.c | 2 +- tools/perf/util/symbol-elf.c | 2 +- tools/perf/util/synthetic-events.c | 20 +- tools/perf/util/top.c | 2 +- tools/perf/util/trace-event-info.c | 2 +- tools/perf/util/util.c | 136 ---------- tools/perf/util/util.h | 8 - 128 files changed, 1941 insertions(+), 1321 deletions(-) create mode 100644 tools/lib/traceevent/Documentation/libtraceevent-event_print.txt create mode 100644 tools/lib/traceevent/Documentation/libtraceevent-plugins.txt create mode 100644 tools/lib/traceevent/plugins/Build create mode 100644 tools/lib/traceevent/plugins/Makefile rename tools/lib/traceevent/{ => plugins}/plugin_cfg80211.c (100%) rename tools/lib/traceevent/{ => plugins}/plugin_function.c (100%) rename tools/lib/traceevent/{ => plugins}/plugin_hrtimer.c (100%) rename tools/lib/traceevent/{ => plugins}/plugin_jbd2.c (100%) rename tools/lib/traceevent/{ => plugins}/plugin_kmem.c (100%) rename tools/lib/traceevent/{ => plugins}/plugin_kvm.c (100%) rename tools/lib/traceevent/{ => plugins}/plugin_mac80211.c (100%) rename tools/lib/traceevent/{ => plugins}/plugin_sched_switch.c (100%)
Quoted 8 lines
rename tools/lib/traceevent/{ => plugins}/plugin_scsi.c (100%)
rename tools/lib/traceevent/{ => plugins}/plugin_xen.c (100%)
create mode 100644 tools/perf/lib/include/internal/mmap.h
create mode 100644 tools/perf/util/copyfile.c
create mode 100644 tools/perf/util/copyfile.h
create mode 100644 tools/perf/util/evsel_config.h
create mode 100644 tools/perf/util/evsel_fprintf.h
create mode 100644 tools/perf/util/perf_event_attr_fprintf.cPulled, thanks a lot Arnaldo!
Ingo
Re: [PATCH 57/66] perf evsel: Introduce evsel_fprintf.h
In reply to Arnaldo Carvalho de MeloQuoted 14 lines · Arnaldo Carvalho de Melo
From: Arnaldo Carvalho de Melo <[email protected]>
We already had
evsel_fprintf.c, add its counterpart, so that we can reduceevsel.ha bit more.We needed a new
perf_event_attr_fprintf.cfile so as to have a separate object to link with the python binding intools/perf/util/python-ext-sourcesand not dragsymbol_conf, etc into the python binding.Cc Adrian Hunter <[email protected]>Cc Jiri Olsa <[email protected]>Cc Namhyung Kim <[email protected]>Signed-off-by Arnaldo Carvalho de Melo <[email protected]>
Quoted 16 lines
--- tools/perf/builtin-evlist.c | 1 + tools/perf/builtin-sched.c | 1 + tools/perf/builtin-script.c | 1 + tools/perf/builtin-trace.c | 2 + tools/perf/util/Build | 1 + tools/perf/util/evsel.c | 153 +--------------------- tools/perf/util/evsel.h | 51 +------- tools/perf/util/evsel_fprintf.c | 1 + tools/perf/util/evsel_fprintf.h | 50 +++++++ tools/perf/util/header.c | 1 + tools/perf/util/perf_event_attr_fprintf.c | 148 +++++++++++++++++++++ tools/perf/util/python-ext-sources | 1 + 12 files changed, 218 insertions(+), 193 deletions(-) create mode 100644 tools/perf/util/evsel_fprintf.h create mode 100644 tools/perf/util/perf_event_attr_fprintf.c
hum, I see this file in this patch, but not when I checkout yours perf/core
jirka
Re: [PATCH 57/66] perf evsel: Introduce evsel_fprintf.h
Em Thu, Sep 26, 2019 at 01:15:18PM +0200, Jiri Olsa escreveu:
Quoted 32 lines · Arnaldo Carvalho de Melo
Quoted 14 lines
From: Arnaldo Carvalho de Melo <[email protected]>
We already had
evsel_fprintf.c, add its counterpart, so that we can reduceevsel.ha bit more.We needed a new
perf_event_attr_fprintf.cfile so as to have a separate object to link with the python binding intools/perf/util/python-ext-sourcesand not dragsymbol_conf, etc into the python binding.Cc Adrian Hunter <[email protected]>Cc Jiri Olsa <[email protected]>Cc Namhyung Kim <[email protected]>Signed-off-by Arnaldo Carvalho de Melo <[email protected]>Quoted 16 lines
--- tools/perf/builtin-evlist.c | 1 + tools/perf/builtin-sched.c | 1 + tools/perf/builtin-script.c | 1 + tools/perf/builtin-trace.c | 2 + tools/perf/util/Build | 1 + tools/perf/util/evsel.c | 153 +--------------------- tools/perf/util/evsel.h | 51 +------- tools/perf/util/evsel_fprintf.c | 1 + tools/perf/util/evsel_fprintf.h | 50 +++++++ tools/perf/util/header.c | 1 + tools/perf/util/perf_event_attr_fprintf.c | 148 +++++++++++++++++++++ tools/perf/util/python-ext-sources | 1 + 12 files changed, 218 insertions(+), 193 deletions(-) create mode 100644 tools/perf/util/evsel_fprintf.h create mode 100644 tools/perf/util/perf_event_attr_fprintf.c
Quoted 1 line
hum, I see this file in this patch, but not when I checkout yours perf/core
I made a mistake and had to force push it, it went on the signed tag to Ingo, so what made upstream is sane, now what is in my perf/core branch should be as well:
https://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git/tree/tools/perf/util/evsel_fprintf.h?h=perf/core
https://git.kernel.org/pub/scm/linux/kernel/git/acme/linux.git/tree/tools/perf/util/perf_event_attr_fprintf.c?h=perf/core
- Arnaldo