Add a new branch filter, "counter", for the branch counter option. It is
used to mark the events which should be logged in the branch. If it is
applied with the -j option, the counters of all the events should be
logged in the branch. If the legacy kernel doesn't support the new
branch sample type, switching off the branch counter filter.
The stored counter values in each branch are displayed right after the
regular branch stack information via perf report -D.
Usage examples:
# perf record -e "{branch-instructions,branch-misses}:S" -j any,counter
Only the first event, branch-instructions, collect the LBR. Both
branch-instructions and branch-misses are marked as logged events. The
occurrences information of them can be found in the branch stack
extension space of each branch.
# perf record -e "{cpu/branch-instructions,branch_type=any/,cpu/branch-misses,branch_type=counter/}"
Only the first event, branch-instructions, collect the LBR. Only the
branch-misses event is marked as a logged event.
Committer notes:
I noticed 'perf test "Sample parsing"' failing, reported to the list and
Kan provided a patch that checks if the evsel has a leader and that
evsel->evlist is set, the comment in the source code further explains
it.
Reviewed-by: Ian Rogers <irogers@google.com>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Alexey Bayduraev <alexey.v.bayduraev@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Cc: Tinghao Zhang <tinghao.zhang@intel.com>
Link: https://lore.kernel.org/r/20231025201626.3000228-8-kan.liang@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
To get the latest fixes in the perf tools including perf stat output,
dlfilter and LLVM feature detection.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
While building on a wide range of distros and clang versions it was
noticed that at least version 12.0.1 (noticed on Alpine 3.15 with
"Alpine clang version 12.0.1") is needed to not fail with BTF generation
errors such as:
Debian:10
Debian clang version 11.0.1-2~deb10u1:
CLANG /tmp/build/perf/util/bpf_skel/.tmp/sample_filter.bpf.o
<SNIP>
GENSKEL /tmp/build/perf/util/bpf_skel/sample_filter.skel.h
libbpf: failed to find BTF for extern 'bpf_cast_to_kern_ctx' [21] section: -2
Error: failed to open BPF object file: No such file or directory
make[2]: *** [Makefile.perf:1121: /tmp/build/perf/util/bpf_skel/sample_filter.skel.h] Error 254
make[2]: *** Deleting file '/tmp/build/perf/util/bpf_skel/sample_filter.skel.h'
Amazon Linux 2:
clang version 11.1.0 (Amazon Linux 2 11.1.0-1.amzn2.0.2)
GENSKEL /tmp/build/perf/util/bpf_skel/sample_filter.skel.h
libbpf: elf: skipping unrecognized data section(18) .eh_frame
libbpf: elf: skipping relo section(19) .rel.eh_frame for section(18) .eh_frame
libbpf: failed to find BTF for extern 'bpf_cast_to_kern_ctx' [21] section: -2
Error: failed to open BPF object file: No such file or directory
make[2]: *** [/tmp/build/perf/util/bpf_skel/sample_filter.skel.h] Error 254
make[2]: *** Deleting file `/tmp/build/perf/util/bpf_skel/sample_filter.skel.h'
Ubuntu 20.04:
clang version 10.0.0-4ubuntu1
CLANG /tmp/build/perf/util/bpf_skel/.tmp/augmented_raw_syscalls.bpf.o
GENSKEL /tmp/build/perf/util/bpf_skel/bench_uprobe.skel.h
GENSKEL /tmp/build/perf/util/bpf_skel/bperf_leader.skel.h
libbpf: sec '.reluprobe': corrupted symbol #27 pointing to invalid section #65522 for relo #0
GENSKEL /tmp/build/perf/util/bpf_skel/bperf_follower.skel.h
Error: failed to open BPF object file: BPF object format invalid
make[2]: *** [Makefile.perf:1121: /tmp/build/perf/util/bpf_skel/bench_uprobe.skel.h] Error 95
make[2]: *** Deleting file '/tmp/build/perf/util/bpf_skel/bench_uprobe.skel.h'
So check if the version is at least 12.0.1 otherwise disable building
BPF skels and provide a message about it, continuing the build.
The message, when running on amazonlinux:2:
Makefile.config:698: Warning: Disabled BPF skeletons as reliable BTF generation needs at least clang version 12.0.1
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Reviewed-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/ZTvGx/Ou6BVnYBqi@kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The changes in ("perf evsel: Rename evsel__increase_rlimit to
rlimit__increase_nofile") ended up breaking the python binding that now
references the rlimit__increase_nofile function, add the util/rlimit.o
to the tools/perf/util/python-ext-sources to cure that.
This was detected by the 'perf test python' regression test:
$ perf test python
14: 'import perf' in python : FAILED!
$ perf test -v python
Couldn't bump rlimit(MEMLOCK), failures may take place when creating BPF maps, etc
14: 'import perf' in python :
--- start ---
test child forked, pid 2912462
python usage test: "echo "import sys ; sys.path.insert(0, '/tmp/build/perf-tools-next/python'); import perf" | '/usr/bin/python3' "
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ImportError: /tmp/build/perf-tools-next/python/perf.cpython-311-x86_64-linux-gnu.so: undefined symbol: rlimit__increase_nofile
test child finished with -1
---- end ----
'import perf' in python: FAILED!
$
Fixes: e093a222d7 ("perf evsel: Rename evsel__increase_rlimit to rlimit__increase_nofile")
Acked-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Yang Jihong <yangjihong1@huawei.com>
Link: https://lore.kernel.org/lkml/ZTrCS5Z3PZAmfPdV@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
To address this grep 3.8 warning:
grep: warning: stray \ before #
We needed to remove the '' around the grep expression and keep the \
before # so that it is escaped by the $(shell grep ...) and thus doesn't
get to grep.
We need that \ before the #, otherwise we get this:
Makefile.perf:364: *** unterminated call to function 'shell': missing ')'. Stop.
As everything after the # will be considered a comment.
Removing the single quotes needs some more escaping so that _some_ of
the escaped chars gets to grep, like the '\|' that becomes '\\\|´.
Running on debian:10, where there is no libtraceevent-devel available,
we get:
Makefile.perf:367: *** PYTHON_EXT_SRCS= util/python.c ../lib/ctype.c util/cap.c util/evlist.c util/evsel.c util/evsel_fprintf.c util/perf_event_attr_fprintf.c util/cpumap.c util/memswap.c util/mmap.c util/namespaces.c ../lib/bitmap.c ../lib/find_bit.c ../lib/list_sort.c ../lib/hweight.c ../lib/string.c ../lib/vsprintf.c util/thread_map.c util/util.c util/cgroup.c util/parse-branch-options.c util/rblist.c util/counts.c util/print_binary.c util/strlist.c ../lib/rbtree.c util/string.c util/symbol_fprintf.c util/units.c util/affinity.c util/rwsem.c util/hashmap.c util/perf_regs.c util/fncache.c util/perf-regs-arch/perf_regs_aarch64.c util/perf-regs-arch/perf_regs_arm.c util/perf-regs-arch/perf_regs_csky.c util/perf-regs-arch/perf_regs_loongarch.c util/perf-regs-arch/perf_regs_mips.c util/perf-regs-arch/perf_regs_powerpc.c util/perf-regs-arch/perf_regs_riscv.c util/perf-regs-arch/perf_regs_s390.c util/perf-regs-arch/perf_regs_x86.c. Stop.
make[1]: *** [Makefile.perf:242: sub-make] Error 2
I.e. both the comments and the util/trace-event.c were removed.
When using:
msg := $(error PYTHON_EXT_SRCS=$(PYTHON_EXT_SRCS))
While on the more recent fedora:38, with the new grep and make packages
and libtraceevent-devel installed:
Makefile.perf:367: *** PYTHON_EXT_SRCS= util/python.c ../lib/ctype.c util/cap.c util/evlist.c util/evsel.c util/evsel_fprintf.c util/perf_event_attr_fprintf.c util/cpumap.c util/memswap.c util/mmap.c util/namespaces.c ../lib/bitmap.c ../lib/find_bit.c ../lib/list_sort.c ../lib/hweight.c ../lib/string.c ../lib/vsprintf.c util/thread_map.c util/util.c util/cgroup.c util/parse-branch-options.c util/rblist.c util/counts.c util/print_binary.c util/strlist.c util/trace-event.c ../lib/rbtree.c util/string.c util/symbol_fprintf.c util/units.c util/affinity.c util/rwsem.c util/hashmap.c util/perf_regs.c util/fncache.c util/perf-regs-arch/perf_regs_aarch64.c util/perf-regs-arch/perf_regs_arm.c util/perf-regs-arch/perf_regs_csky.c util/perf-regs-arch/perf_regs_loongarch.c util/perf-regs-arch/perf_regs_mips.c util/perf-regs-arch/perf_regs_powerpc.c util/perf-regs-arch/perf_regs_riscv.c util/perf-regs-arch/perf_regs_s390.c util/perf-regs-arch/perf_regs_x86.c. Stop.
make[1]: *** [Makefile.perf:242: sub-make] Error 2
make: *** [Makefile:113: install-bin] Error 2
make: Leaving directory '/home/acme/git/perf-tools-next/tools/perf'
$
I.e. only the comments were removed.
If we build it on the same fedora:38 system, but using NO_LIBTRACEEVENT=1
$ make NO_LIBTRACEEVENT=1 CORESIGHT=1 O=/tmp/build/$(basename $PWD) -C tools/perf install-bin
Makefile.perf:367: *** PYTHON_EXT_SRCS= util/python.c ../lib/ctype.c util/cap.c util/evlist.c util/evsel.c util/evsel_fprintf.c util/perf_event_attr_fprintf.c util/cpumap.c util/memswap.c util/mmap.c util/namespaces.c ../lib/bitmap.c ../lib/find_bit.c ../lib/list_sort.c ../lib/hweight.c ../lib/string.c ../lib/vsprintf.c util/thread_map.c util/util.c util/cgroup.c util/parse-branch-options.c util/rblist.c util/counts.c util/print_binary.c util/strlist.c ../lib/rbtree.c util/string.c util/symbol_fprintf.c util/units.c util/affinity.c util/rwsem.c util/hashmap.c util/perf_regs.c util/fncache.c util/perf-regs-arch/perf_regs_aarch64.c util/perf-regs-arch/perf_regs_arm.c util/perf-regs-arch/perf_regs_csky.c util/perf-regs-arch/perf_regs_loongarch.c util/perf-regs-arch/perf_regs_mips.c util/perf-regs-arch/perf_regs_powerpc.c util/perf-regs-arch/perf_regs_riscv.c util/perf-regs-arch/perf_regs_s390.c util/perf-regs-arch/perf_regs_x86.c. Stop.
make[1]: *** [Makefile.perf:242: sub-make] Error 2
make: *** [Makefile:113: install-bin] Error 2
make: Leaving directory '/home/acme/git/perf-tools-next/tools/perf'
$
Both comments and the util/trace-event.c file removed.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/ZTj6mfM9UqY2DggC@kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The hierarchy mode needs to setup output formats for each evsel.
Normally setup_sorting() handles this at the beginning, but it cannot
do that if data comes from a pipe since there's no evsel info before
reading the data. And then perf report cannot process the samples
in hierarchy mode and think as if there's no sample.
Let's check the condition and setup the output formats after reading
data so that it can find evsels.
Before:
$ ./perf record -o- true | ./perf report -i- --hierarchy -q
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.000 MB - ]
Error:
The - data has no samples!
After:
$ ./perf record -o- true | ./perf report -i- --hierarchy -q
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.000 MB - ]
94.76% true
94.76% [kernel.kallsyms]
94.76% [k] filemap_fault
5.24% perf-ex
5.24% [kernel.kallsyms]
5.06% [k] __memset
0.18% [k] native_write_msr
Acked-by: Ian Rogers <irogers@google.com>
Link: https://lore.kernel.org/r/20231025003121.2811738-1-namhyung@kernel.org
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Currently lock contention timestamp is maintained in a hash map keyed by
pid. That means it needs to get and release a map element (which is
proctected by spinlock!) on each contention begin and end pair. This
can impact on performance if there are a lot of contention (usually from
spinlocks).
It used to go with task local storage but it had an issue on memory
allocation in some critical paths. Although it's addressed in recent
kernels IIUC, the tool should support old kernels too. So it cannot
simply switch to the task local storage at least for now.
As spinlocks create lots of contention and they disabled preemption
during the spinning, it can use per-cpu array to keep the timestamp to
avoid overhead in hashmap update and delete.
In contention_begin, it's easy to check the lock types since it can see
the flags. But contention_end cannot see it. So let's try to per-cpu
array first (unconditionally) if it has an active element (lock != 0).
Then it should be used and per-task tstamp map should not be used until
the per-cpu array element is cleared which means nested spinlock
contention (if any) was finished and it nows see (the outer) lock.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Hao Luo <haoluo@google.com>
Cc: Song Liu <song@kernel.org>
Cc: bpf@vger.kernel.org
Link: https://lore.kernel.org/r/20231020204741.1869520-3-namhyung@kernel.org
It checks the current lock to calculated the delta of contention time.
The address is saved in the tstamp map which is allocated at begining of
contention and released at end of contention.
But it's possible for bpf_map_delete_elem() to fail. In that case, the
element in the tstamp map kept for the current lock and it makes the
next contention for the same lock tracked incorrectly. Specificially
the next contention begin will see the existing element for the task and
it'd just return. Then the next contention end will see the element and
calculate the time using the timestamp for the previous begin.
This can result in a large value for two small contentions happened from
time to time. Let's clear the lock address so that it can be updated
next time even if the bpf_map_delete_elem() failed.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Ian Rogers <irogers@google.com>
Cc: Hao Luo <haoluo@google.com>
Cc: Song Liu <song@kernel.org>
Cc: bpf@vger.kernel.org
Link: https://lore.kernel.org/r/20231020204741.1869520-1-namhyung@kernel.org
evsel__increase_rlimit() helper does nothing with evsel, and description
of the functionality is inaccurate, rename it and move to util/rlimit.c.
By the way, fix a checkppatch warning about misplaced license tag:
WARNING: Misplaced SPDX-License-Identifier tag - use line 1 instead
#160: FILE: tools/perf/util/rlimit.h:3:
/* SPDX-License-Identifier: LGPL-2.1 */
No functional change.
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Link: https://lore.kernel.org/r/20231023033144.1011896-1-yangjihong1@huawei.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The -G/--cgroups option is to put sender and receiver in different
cgroups in order to measure cgroup context switch overheads.
Users need to make sure the cgroups exist and accessible. The following
example should the effect of this change. Please don't forget taskset
before the perf bench to measure cgroup switches properly. Otherwise
each task would run on a different CPU and generate cgroup switches
regardless of this change.
# perf stat -e context-switches,cgroup-switches \
> taskset -c 0 perf bench sched pipe -l 10000 > /dev/null
Performance counter stats for 'taskset -c 0 perf bench sched pipe -l 10000':
20,001 context-switches
2 cgroup-switches
0.053449651 seconds time elapsed
0.011286000 seconds user
0.041869000 seconds sys
# perf stat -e context-switches,cgroup-switches \
> taskset -c 0 perf bench sched pipe -l 10000 -G AAA,BBB > /dev/null
Performance counter stats for 'taskset -c 0 perf bench sched pipe -l 10000 -G AAA,BBB':
20,001 context-switches
20,001 cgroup-switches
0.052768627 seconds time elapsed
0.006284000 seconds user
0.046266000 seconds sys
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Link: https://lore.kernel.org/r/20231017202342.1353124-1-namhyung@kernel.org
If using parallel threads to collect data, perf record needs at least 6 fds
per CPU. (one for sys_perf_event_open, four for pipe msg and ack of the
pipe, see record__thread_data_open_pipes(), and one for open perf.data.XXX)
For an environment with more than 100 cores, if perf record uses both
`-a` and `--threads` options, it is easy to exceed the upper limit of the
file descriptor number, when we run out of them try to increase the limits.
Before:
$ ulimit -n
1024
$ lscpu | grep 'On-line CPU(s)'
On-line CPU(s) list: 0-159
$ perf record --threads -a sleep 1
Failed to create data directory: Too many open files
After:
$ ulimit -n
1024
$ lscpu | grep 'On-line CPU(s)'
On-line CPU(s) list: 0-159
$ perf record --threads -a sleep 1
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.394 MB perf.data (1576 samples) ]
Signed-off-by: Yang Jihong <yangjihong1@huawei.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Link: https://lore.kernel.org/r/20231013075945.698874-1-yangjihong1@huawei.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The CPI_STALL_RATIO metric group can be used to present the high
level CPI stall breakdown metrics in powerpc, which will show:
- DISPATCH_STALL_CPI ( Dispatch stall cycles per insn )
- ISSUE_STALL_CPI ( Issue stall cycles per insn )
- EXECUTION_STALL_CPI ( Execution stall cycles per insn )
- COMPLETION_STALL_CPI ( Completion stall cycles per insn )
Commit cf26e043c2 ("perf vendor events power10: Add JSON
metric events to present CPI stall cycles in powerpc)" which added
the CPI_STALL_RATIO metric group, also modified
the PMC value used in PM_RUN_INST_CMPL event from PMC4 to PMC5,
to avoid multiplexing of events.
But that got revert in recent changes. Fix this issue by changing
back the PMC value used in PM_RUN_INST_CMPL to PMC5.
Result with the fix:
./perf stat --metric-no-group -M CPI_STALL_RATIO <workload>
Performance counter stats for 'workload':
68,745,426 PM_CMPL_STALL # 0.21 COMPLETION_STALL_CPI
7,692,827 PM_ISSUE_STALL # 0.02 ISSUE_STALL_CPI
322,638,223 PM_RUN_INST_CMPL # 0.05 DISPATCH_STALL_CPI
# 0.48 EXECUTION_STALL_CPI
16,858,553 PM_DISP_STALL_CYC
153,880,133 PM_EXEC_STALL
0.089774592 seconds time elapsed
"--metric-no-group" is used for forcing PM_RUN_INST_CMPL to be scheduled
in all group for more accuracy.
Fixes: 7d473f475b ("perf vendor events: Move JSON/events to appropriate files for power10 platform")
Reported-by: Disha Goel <disgoel@linux.vnet.ibm.com>
Signed-off-by: Kajol Jain <kjain@linux.ibm.com>
Reviewed-by: Athira Rajeev <atrajeev@linux.vnet.ibm.com>
Tested-by: Disha Goel<disgoel@linux.ibm.com>
Cc: maddy@linux.ibm.com
Link: https://lore.kernel.org/r/20231016143110.244255-1-kjain@linux.ibm.com
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
