perf report/script/... have a --time option to limit the time range of
output. That's very useful to slice large traces, e.g. when processing
the output of perf script for some analysis.
But right now --time only supports absolute time. Also there is no fast
way to get the start/end times of a given trace except for looking at
it. This makes it hard to e.g. only decode the first half of the trace,
which is useful for parallelization of scripts
Another problem is that perf records are variable size and there is no
synchronization mechanism. So the only way to find the last sample
reliably would be to walk all samples. But we want to avoid that in perf
report/... because it is already quite expensive. That is why storing
the first sample time and last sample time in perf record is better.
This patch creates a new header feature type HEADER_SAMPLE_TIME and
related ops. Save the first sample time and the last sample time to the
feature section in perf file header. That will be done when, for
instance, processing build-ids, where we already have to process all
samples to create the build-id table, take advantage of that to further
amortize that processing by storing HEADER_SAMPLE_TIME to make 'perf
report/script' faster when using --time.
Committer testing:
After this patch is applied the header is written with zeroes, we need
the next patch, for "perf record" to actually write the timestamps:
# perf report -D | grep PERF_RECORD_SAMPLE\(
22501155244406 0x44f0 [0x28]: PERF_RECORD_SAMPLE(IP, 0x4001): 25016/25016: 0xffffffffa21be8c5 period: 1 addr: 0
<SNIP>
22501155793625 0x4a30 [0x28]: PERF_RECORD_SAMPLE(IP, 0x4001): 25016/25016: 0xffffffffa21ffd50 period: 2828043 addr: 0
# perf report --header | grep "time of "
# time of first sample : 0.000000
# time of last sample : 0.000000
#
Changelog:
v7: 1. Rebase to latest perf/core branch.
2. Add following clarification in patch description according to
Arnaldo's suggestion.
"That will be done when, for instance, processing build-ids,
where we already have to process all samples to create the
build-id table, take advantage of that to further amortize
that processing by storing HEADER_SAMPLE_TIME to make
'perf report/script' faster when using --time."
v4: Use perf script time style for timestamp printing. Also add with
the printing of sample duration.
v3: Remove the definitions of first_sample_time/last_sample_time from
perf_session. Just define them in perf_evlist
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1512738826-2628-2-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When enabling '-b' option in perf record, for example,
perf record -b ...
perf report
and then browsing the annotate browser from perf report (press 'A'), it
would fail (annotate browser can't be displayed).
It's because the '.add_entry_cb' op of struct report is overwritten by
hist_iter__branch_callback() in builtin-report.c. But this function doesn't do
something like mapping symbols and sources. So next, do_annotate() will return
directly.
notes = symbol__annotation(act->ms.sym);
if (!notes->src)
return 0;
This patch adds the lost code to hist_iter__branch_callback (refer to
hist_iter__report_callback).
v2:
Fix a crash bug when perform 'perf report --stdio'.
The reason is that we init the symbol annotation only in browser mode, it
doesn't allocate/init resources for stdio mode.
So now in hist_iter__branch_callback(), it will return directly if it's not in
browser mode.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1514284963-18587-1-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When a valid vmlinux is not found, 'perf report' falls back to look at
/proc/kcore. In this case, it will report the impossible large offset.
For example:
# perf record -b -e cycles:k find /etc/ > /dev/null
# perf report --stdio --branch-history
22.77% _vm_normal_page+18446603336221188162
|
---page_remove_rmap +18446603336221188324
page_remove_rmap +18446603336221188487 (cycles:5)
unlock_page_memcg +18446603336221188096
page_remove_rmap +18446603336221188327 (cycles:1)
The issue is the value which is passed to parameter 'addr' in
__get_srcline() is the objdump address. It's not correct if we calculate
the offset by using 'addr - sym->start'.
This patch creates a new parameter 'ip' in __get_srcline(). It is not
converted to objdump address.
With this patch, the perf report output is:
22.77% _vm_normal_page+66
|
---page_remove_rmap +228
page_remove_rmap +391 (cycles:5)
unlock_page_memcg +0
page_remove_rmap +231 (cycles:1)
page_remove_rmap +236
Committer testing:
Make sure you get any valid vmlinux out of the way, using '-v' on the
'perf report' case and deleting it from places where perf searches them,
like your kernel build dir and the build-id cache, in ~/.debug/.
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1514564812-17344-1-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Fix a compile error:
...
CC util/libunwind/x86_32.o
In file included from util/libunwind/x86_32.c:33:0:
util/libunwind/../../arch/x86/util/unwind-libunwind.c: In function 'libunwind__x86_reg_id':
util/libunwind/../../arch/x86/util/unwind-libunwind.c:110:11: error: 'EINVAL' undeclared (first use in this function)
return -EINVAL;
^
util/libunwind/../../arch/x86/util/unwind-libunwind.c:110:11: note: each undeclared identifier is reported only once for each function it appears in
mv: cannot stat 'util/libunwind/.x86_32.o.tmp': No such file or directory
make[4]: *** [util/libunwind/x86_32.o] Error 1
make[3]: *** [util] Error 2
make[2]: *** [libperf-in.o] Error 2
make[1]: *** [sub-make] Error 2
make: *** [all] Error 2
It happens when libunwind-x86 feature is detected.
Signed-off-by: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/20171206015040.114574-1-wangnan0@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The 'perf test bpf' was hooking a eBPF program on the SyS_epoll_wait()
kernel function, that was what the epoll_wait() glibc function ended up
calling, but since at least glibc 2.26, the one that comes with, for
instance, Fedora 27, glibc ends up calling SyS_epoll_pwait() when
epoll_wait() is used, causing this 'perf test' entry to fail.
So switch to using epoll_pwait() and hook the eBPF program to the
SyS_epoll_pwait() kernel function to make it work on a wider range of
glibc and kernel versions.
Tested-by: Wang Nan <wangnan0@huawei.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: https://lkml.kernel.org/n/tip-zynvquy63er8s5mrgsz65pto@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When failing on one of the BPF tests we were just stating:
BPF filter result incorrect
Add some more info to help figuring out the problem:
BPF filter result incorrect, expected 56, got 0 samples
This came out while investigating this failure, first seen after
updating the kernel to the 4.15.0-rc6 tag:
[root@jouet ~]# perf test bpf
39: BPF filter :
39.1: Basic BPF filtering : FAILED!
39.2: BPF pinning : Skip
39.3: BPF prologue generation: Skip
39.4: BPF relocation checker : Skip
[root@jouet ~]#
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Link: https://lkml.kernel.org/n/tip-403npu7daupv6b2bmxliv5pk@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
It's a follow up patch for a previous patch "perf tool: Return all
events as auto-completions after comma".
With this patch, auto-completion can work well for events with a ':'.
For example:
root@skl:/tmp# perf stat -e block:block_<TAB>
block:block_bio_backmerge block:block_rq_complete
block:block_bio_bounce block:block_rq_insert
block:block_bio_complete block:block_rq_issue
block:block_bio_frontmerge block:block_rq_remap
block:block_bio_queue block:block_rq_requeue
block:block_bio_remap block:block_sleeprq
block:block_dirty_buffer block:block_split
block:block_getrq block:block_touch_buffer
block:block_plug block:block_unplug
root@skl:/tmp# perf stat -e block:block_rq_<TAB>
block:block_rq_complete block:block_rq_issue block:block_rq_requeue
block:block_rq_insert block:block_rq_remap
root@skl:/tmp# perf stat -e block:block_rq_complete<TAB>
block:block_rq_complete
root@skl:/tmp# perf stat -e block:block_rq_complete
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1513973758-19109-1-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
It's a follow up for one previous patch "perf tool: Improve bash command
line auto-complete for multiple events with comma."
It fixes an issue that no events are displayed when <TAB> is directly
typed after comma.
With this patch, now the result is:
root@skl:/tmp# perf stat -e cpu-cycles,<TAB>
Display all 2389 possibilities? (y or n)
alarmtimer:alarmtimer_cancel
alarmtimer:alarmtimer_fired
alarmtimer:alarmtimer_start
alarmtimer:alarmtimer_suspend
alignment-faults
arith.divider_active
BAClear_Cost
baclears.any
block:block_bio_backmerge
block:block_bio_bounce
block:block_bio_complete
block:block_bio_frontmerge
block:block_bio_queue
block:block_bio_remap
block:block_dirty_buffer
block:block_getrq
block:block_plug
block:block_rq_complete
block:block_rq_insert
block:block_rq_issue
block:block_rq_remap
block:block_rq_requeue
block:block_sleeprq
--More--
One remaining issue is that the auto-completions doesn't work well
for the event with ':'. For example, clk:clk_enable.
Because ':' is set as WORDBREAK by default in bash. Need more work
for this case.
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1513940255-16528-1-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf has perf-completion.sh to define command line auto-completion in
bash/zsh.
For record/stat -e it works for single events, but isn't working when
specifying multiple events with comma.
It would be very useful if it could be fixed to make it easier by
supporting multiple events, comma separated.
With this patch, the result can be like this:
1. Support the events returned from 'perf list --raw-dump'
root@skl:/tmp# perf stat -e cpu/cache<TAB>
cpu/cache-misses/ cpu/cache-references/
root@skl:/tmp# perf stat -e cpu/cache-misses/,cpu/branch-<TAB>
cpu/branch-instructions/ cpu/branch-misses/
root@skl:/tmp# perf stat -e cpu/cache-misses/,cpu/branch-i<TAB>
root@skl:/tmp# perf stat -e cpu/cache-misses/,cpu/branch-instructions/
2. Support the events listed in /sys/bus/event_source/devices/cpu/events
root@skl:/tmp# perf stat -e cycle<TAB>
cycle_activity.cycles_l1d_miss cycle_activity.stalls_l3_miss
cycle_activity.cycles_l2_miss cycle_activity.stalls_mem_any
cycle_activity.cycles_l3_miss cycle_activity.stalls_total
cycle_activity.cycles_mem_any cycles-ct
cycle_activity.stalls_l1d_miss cycles-t
cycle_activity.stalls_l2_miss
root@skl:/tmp# perf stat -e cycles-<TAB>
cycles-ct cycles-t
root@skl:/tmp# perf stat -e cycles-t,cpu/c<TAB>
cpu/cache-misses/ cpu/cpu-cycles/ cpu/cycles-t/
cpu/cache-references/ cpu/cycles-ct/
root@skl:/tmp# perf stat -e cycles-t,cpu/cache-<TAB>
cpu/cache-misses/ cpu/cache-references/
root@skl:/tmp# perf stat -e cycles-t,cpu/cache-misses/
3. Support the uppercase event which is with prefix "cpu/"
root@skl:/tmp# perf stat -e cpu/c<TAB>
cpu/cache-misses/ cpu/cpu-cycles/ cpu/cycles-t/
cpu/cache-references/ cpu/cycles-ct/
root@skl:/tmp# perf stat -e cpu/cache-misses/,cpu/C<TAB>
cpu/CACHE-MISSES/ cpu/CPU-CYCLES/ cpu/CYCLES-T/
cpu/CACHE-REFERENCES/ cpu/CYCLES-CT/
root@skl:/tmp# perf stat -e cpu/cache-misses/,cpu/CACHE-REFERENCES/
Note that:
a) This patch only supports bash.
b) It doesn't support the cases like {},{} or {...,...}.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1513848370-8098-1-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
On an arm64 machine running a CONFIG_RANDOMIZE_BASE=y kernel, perf
kernel symbol resolution fails. Debugging saw symsrc_init calling the
default elf__needs_adjust_symbols() where checks for an ET_DYN (3)
ehdr.e_type failed when they should have succeeded.
Fix by adopting powerpc version of the weak elf__needs_adjust_symbols()
function, as done in commit d233209833 ("perf probe ppc: Fix symbol
fixup issues due to ELF type").
Prior to this patch, perf test 1 would fail:
$ sudo oldperf test -v 1 |& head
1: vmlinux symtab matches kallsyms :
test child forked, pid 33374
Looking at the vmlinux_path (8 entries long)
Using /usr/lib/debug/boot/vmlinux for symbols
ERR : 0xfffe0000100f1000: do_undefinstr not on kallsyms
ERR : 0xfffe0000100f1320: do_sysinstr not on kallsyms
ERR : 0xfffe0000100f13b0: do_debug_exception not on kallsyms
ERR : 0xfffe0000100f1498: do_mem_abort not on kallsyms
ERR : 0xfffe0000100f1580: do_sp_pc_abort not on kallsyms
...
After applying this patch, perf test 1 now succeeds:
$ sudo ./newperf test -v 1 |& head
1: vmlinux symtab matches kallsyms :
test child forked, pid 33378
Looking at the vmlinux_path (8 entries long)
Using /usr/lib/debug/boot/vmlinux for symbols
WARN: 0xffff000008081000: diff name v: do_undefinstr k: __exception_text_start
WARN: 0xffff0000080819e8: diff name v: __irqentry_text_end k: __softirqentry_text_start
WARN: 0xffff000008081d08: diff name v: __entry_text_start k: __softirqentry_text_end
WARN: 0xffff00000809db5c: diff name v: flush_icache_range k: __flush_cache_user_range
WARN: 0xffff000008101908: diff name v: sys_ni_syscall k: sys_vm86old
...
Signed-off-by: Kim Phillips <kim.phillips@arm.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ganapatrao Kulkarni <ganapatrao.kulkarni@cavium.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Naveen N. Rao <naveen.n.rao@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lkml.kernel.org/r/20171214175242.e30450f17f93ad675d968fa3@arm.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
While monitoring a multithread process with pid option, perf sometimes
may return sys_perf_event_open failure with 3(No such process) if any of
the process's threads die before we open the event. However, we want
perf continue monitoring the remaining threads and do not exit with
error.
Here, the patch enables perf_evsel::ignore_missing_thread for -p option
to ignore complete failure if any of threads die before we open the event.
But it may still return sys_perf_event_open failure with 22(Invalid) if we
monitors several event groups.
sys_perf_event_open: pid 28960 cpu 40 group_fd 118202 flags 0x8
sys_perf_event_open: pid 28961 cpu 40 group_fd 118203 flags 0x8
WARNING: Ignored open failure for pid 28962
sys_perf_event_open: pid 28962 cpu 40 group_fd [118203] flags 0x8
sys_perf_event_open failed, error -22
That is because when we ignore a missing thread, we change the thread_idx
without dealing with its fds, FD(evsel, cpu, thread). Then get_group_fd()
may return a wrong group_fd for the next thread and sys_perf_event_open()
return with 22.
sys_perf_event_open(){
...
if (group_fd != -1)
perf_fget_light()//to get corresponding group_leader by group_fd
...
if (group_leader)
if (group_leader->ctx->task != ctx->task)//should on the same task
goto err_context
...
}
This patch also fixes this bug by introducing perf_evsel__remove_fd() and
update_fds to allow removing fds for the missing thread.
Changes since v1:
- Change group_fd__remove() into a more genetic way without changing code logic
- Remove redundant condition
Changes since v2:
- Use a proper function name and add some comment.
- Multiline comment style fixes.
Committer testing:
Before this patch the recently added 'perf stat --per-thread' for system
wide counting would race while enumerating all threads using /proc:
[root@jouet ~]# perf stat --per-thread
failed to parse CPUs map: No such file or directory
Usage: perf stat [<options>] [<command>]
-C, --cpu <cpu> list of cpus to monitor in system-wide
-a, --all-cpus system-wide collection from all CPUs
[root@jouet ~]# perf stat --per-thread
failed to parse CPUs map: No such file or directory
Usage: perf stat [<options>] [<command>]
-C, --cpu <cpu> list of cpus to monitor in system-wide
-a, --all-cpus system-wide collection from all CPUs
[root@jouet ~]#
When, say, the kernel was being built, so lots of shortlived threads,
after this patch this doesn't happen.
Signed-off-by: Mengting Zhang <zhangmengting@huawei.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Cc: Cheng Jian <cj.chengjian@huawei.com>
Cc: Li Bin <huawei.libin@huawei.com>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1513148513-6974-1-git-send-email-zhangmengting@huawei.com
[ Remove one use 'evlist' alias variable ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This one made x86 always build with -fPIC, when the intention was for
s390 to be built that way, due to a rebase mistake.
Reported-by: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
This reverts commit 1dc4ddf112.
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When we detect a different endianity we swap event before processing.
It's tricky for samples because we have no idea what's inside. We treat
it as an array of u64s, swap them and later on we swap back parts which
are different.
We mangle this way also the tracepoint raw data, which ends up in report
showing wrong data:
1.95% comm=Q^B pid=29285 prio=16777216 target_cpu=000
1.67% comm=l^B pid=0 prio=16777216 target_cpu=000
Luckily the traceevent library handles the endianity by itself (thank
you Steven!), so we can pass the RAW data directly in the other
endianity.
2.51% comm=beah-rhts-task pid=1175 prio=120 target_cpu=002
2.23% comm=kworker/0:0 pid=11566 prio=120 target_cpu=000
The fix is basically to swap back the raw data if different endianity is
detected.
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Link: http://lkml.kernel.org/r/20171129184346.3656-1-jolsa@kernel.org
[ Add util/memswap.c to python-ext-sources to link missing mem_bswap_64() ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Support the special characters escaped by '\' in parser. This allows
user to specify versions directly like below.
=====
# ./perf probe -x /lib64/libc-2.25.so malloc_get_state\\@GLIBC_2.2.5
Added new event:
probe_libc:malloc_get_state (on malloc_get_state@GLIBC_2.2.5 in /usr/lib64/libc-2.25.so)
You can now use it in all perf tools, such as:
perf record -e probe_libc:malloc_get_state -aR sleep 1
=====
Or, you can use separators in source filename, e.g.
=====
# ./perf probe -x /opt/test/a.out foo+bar.c:3
Semantic error :There is non-digit character in offset.
Error: Command Parse Error.
=====
Usually "+" in source file cause parser error, but
=====
# ./perf probe -x /opt/test/a.out foo\\+bar.c:4
Added new event:
probe_a:main (on @foo+bar.c:4 in /opt/test/a.out)
You can now use it in all perf tools, such as:
perf record -e probe_a:main -aR sleep 1
=====
escaped "\+" allows you to specify that.
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Reviewed-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Acked-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: bhargavb <bhargavaramudu@gmail.com>
Cc: linux-rt-users@vger.kernel.org
Link: http://lkml.kernel.org/r/151309111236.18107.5634753157435343410.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Commit d80406453a ("perf symbols: Allow user probes on versioned
symbols") allows user to find default versioned symbols (with "@@") in
map. However, it did not enable normal versioned symbol (with "@") for
perf-probe. E.g.
=====
# ./perf probe -x /lib64/libc-2.25.so malloc_get_state
Failed to find symbol malloc_get_state in /usr/lib64/libc-2.25.so
Error: Failed to add events.
=====
This solves above issue by improving perf-probe symbol search function,
as below.
=====
# ./perf probe -x /lib64/libc-2.25.so malloc_get_state
Added new event:
probe_libc:malloc_get_state (on malloc_get_state in /usr/lib64/libc-2.25.so)
You can now use it in all perf tools, such as:
perf record -e probe_libc:malloc_get_state -aR sleep 1
# ./perf probe -l
probe_libc:malloc_get_state (on malloc_get_state@GLIBC_2.2.5 in /usr/lib64/libc-2.25.so)
=====
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Reviewed-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Acked-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: bhargavb <bhargavaramudu@gmail.com>
Cc: linux-rt-users@vger.kernel.org
Link: http://lkml.kernel.org/r/151275049269.24652.1639103455496216255.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Add __return suffix for function return events automatically. Without
this, user have to give --force option and will see the number suffix
for each event like "function_1", which is not easy to recognize.
Instead, this adds __return suffix to it automatically. E.g.
=====
# ./perf probe -x /lib64/libc-2.25.so 'malloc*%return'
Added new events:
probe_libc:malloc_printerr__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_consolidate__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_check__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_hook_ini__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_trim__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_usable_size__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_stats__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_info__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:mallochook__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_get_state__return (on malloc*%return in /usr/lib64/libc-2.25.so)
probe_libc:malloc_set_state__return (on malloc*%return in /usr/lib64/libc-2.25.so)
You can now use it in all perf tools, such as:
perf record -e probe_libc:malloc_set_state__return -aR sleep 1
=====
Reported-by: Arnaldo Carvalho de Melo <acme@kernel.org>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Reviewed-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: bhargavb <bhargavaramudu@gmail.com>
Cc: linux-rt-users@vger.kernel.org
Link: http://lkml.kernel.org/r/151275046418.24652.6696011972866498489.stgit@devbox
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cut off the version suffix (e.g. @GLIBC_2.2.5 etc.) from automatic
generated event name. This fixes wildcard event adding like below case;
=====
# perf probe -x /lib64/libc-2.25.so malloc*
Internal error: "malloc_get_state@GLIBC_2" is wrong event name.
Error: Failed to add events.
=====
This failure was caused by a versioned suffix symbol.
With this fix, perf probe automatically cuts the suffix after @ as
below.
=====
# ./perf probe -x /lib64/libc-2.25.so malloc*
Added new events:
probe_libc:malloc_printerr (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_consolidate (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_check (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_hook_ini (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_trim (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_usable_size (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_stats (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_info (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:mallochook (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_get_state (on malloc* in /usr/lib64/libc-2.25.so)
probe_libc:malloc_set_state (on malloc* in /usr/lib64/libc-2.25.so)
You can now use it in all perf tools, such as:
perf record -e probe_libc:malloc_set_state -aR sleep 1
=====
Reported-by: Arnaldo Carvalho de Melo <acme@kernel.org>
Reported-by: bhargavb <bhargavaramudu@gmail.com>
Signed-off-by: Masami Hiramatsu <mhiramat@kernel.org>
Acked-by: Ravi Bangoria <ravi.bangoria@linux.vnet.ibm.com>
Reviewed-by: Thomas Richter <tmricht@linux.vnet.ibm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Paul Clarke <pc@us.ibm.com>
Cc: linux-rt-users@vger.kernel.org
Link: http://lkml.kernel.org/r/None
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently, if we execute 'perf stat --per-thread' without specifying
pid/tid, perf will return error.
root@skl:/tmp# perf stat --per-thread
The --per-thread option is only available when monitoring via -p -t options.
-p, --pid <pid> stat events on existing process id
-t, --tid <tid> stat events on existing thread id
This patch removes this limitation. If no pid/tid specified, it returns
all threads (get threads from /proc).
Note that it doesn't support cpu_list yet so if it's a cpu_list case,
then skip.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1512482591-4646-11-git-send-email-yao.jin@linux.intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The function perf_stat__print_shadow_stats() is called to print the
shadow stats on a set of static variables.
But the static variables are the limitations to support
per-thread shadow stats.
This patch lets the perf_stat__print_shadow_stats() support
to print the shadow stats from a input parameter 'st'.
It will not directly get value from static variable. Instead,
it now uses runtime_stat_avg() and runtime_stat_n() to get and
compute the values.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1512482591-4646-6-git-send-email-yao.jin@linux.intel.com
[ Rename 'stat' variables to 'st' to build on centos:{5,6} and others where it shadows a global declaration ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The functions perf_stat__update_shadow_stats() is called to update the
shadow stats on a set of static variables.
But the static variables are the limitations to be extended to support
per-thread shadow stats.
This patch lets the perf_stat__update_shadow_stats() support to update
the shadow stats on a input parameter 'st' and uses
update_runtime_stat() to update the stats. It will not directly update
the static variables as before.
Signed-off-by: Jin Yao <yao.jin@linux.intel.com>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Kan Liang <kan.liang@intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1512482591-4646-5-git-send-email-yao.jin@linux.intel.com
[ Rename 'stat' variables to 'st' to build on centos:{5,6} and others where it shadows a global declaration ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Long ago we decided to be verbotten including files in the kernel git
sources from tools/ living source code, to avoid disturbing kernel
development (and perf's and other tools/) when, say, a kernel hacker
adds something, tests everything but tools/ and have tools/ build
broken.
This got broken recently by s/390, fix it by copying
arch/s390/include/uapi/asm/perf_regs.h to tools/arch/s390/include/uapi/asm/,
making this one be used by means of <asm/perf_regs.h> and updating
tools/perf/check_headers.sh to make sure we are notified when the
original changes, so that we can check if anything is needed on the
tooling side.
This would have been caught by the 'tarkpg' test entry in:
$ make -C tools/perf build-test
When run on a s/390 build system or container.
Acked-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Hendrik Brueckner <brueckner@linux.vnet.ibm.com>
Cc: Thomas Richter <tmricht@linux.vnet.ibm.com>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Wang Nan <wangnan0@huawei.com>
Fixes: f704ef4460 ("s390/perf: add support for perf_regs and libdw")
Link: https://lkml.kernel.org/n/tip-n57139ic0v9uffx8wdqi3d8a@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
tools/perf/jvmti is broken in so far as it generates incorrect debug
information. Specifically it attributes all debug lines to the original
method being output even in the case that some code is being inlined
from elsewhere. This patch fixes the issue.
To test (from within linux/tools/perf):
export JDIR=/usr/lib/jvm/java-8-openjdk-amd64/
make
cat << __EOF > Test.java
public class Test
{
private StringBuilder b = new StringBuilder();
private void loop(int i, String... args)
{
for (String a : args)
b.append(a);
long hc = b.hashCode() * System.nanoTime();
b = new StringBuilder();
b.append(hc);
System.out.printf("Iteration %d = %d\n", i, hc);
}
public void run(String... args)
{
for (int i = 0; i < 10000; ++i)
{
loop(i, args);
}
}
public static void main(String... args)
{
Test t = new Test();
t.run(args);
}
}
__EOF
$JDIR/bin/javac Test.java
./perf record -F 10000 -g -k mono $JDIR/bin/java -agentpath:`pwd`/libperf-jvmti.so Test
./perf inject --jit -i perf.data -o perf.data.jitted
./perf annotate -i perf.data.jitted --stdio | grep Test\.java: | sort -u
Before this patch, Test.java line numbers get reported that are greater
than the number of lines in the Test.java file. They come from the
source file of the inlined function, e.g. java/lang/String.java:1085.
For further validation one can examine those lines in the JDK source
distribution and confirm that they map to inlined functions called by
Test.java.
After this patch, the filename of the inlined function is output
rather than the incorrect original source filename.
Signed-off-by: Ben Gainey <ben.gainey@arm.com>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Tested-by: Stephane Eranian <eranian@google.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ben Gainey <ben.gainey@arm.com>
Cc: Colin King <colin.king@canonical.com>
Cc: Darren Hart <dvhart@infradead.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: 598b7c6919 ("perf jit: add source line info support")
Link: http://lkml.kernel.org/r/20171122182541.d25599a3eb1ada3480d142fa@arm.com
Signed-off-by: Kim Phillips <kim.phillips@arm.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
On Fedora systems the perl and python CFLAGS/LDFLAGS include the
hardened specs from redhat-rpm-config package. We apply them only for
perl/python objects, which makes them not compatible with the rest of
the objects and the build fails with:
/usr/bin/ld: perf-in.o: relocation R_X86_64_32 against `.rodata.str1.1' can not be used when making a shared object; recompile with -f
+PIC
/usr/bin/ld: libperf.a(libperf-in.o): relocation R_X86_64_32S against `.text' can not be used when making a shared object; recompile w
+ith -fPIC
/usr/bin/ld: final link failed: Nonrepresentable section on output
collect2: error: ld returned 1 exit status
make[2]: *** [Makefile.perf:507: perf] Error 1
make[1]: *** [Makefile.perf:210: sub-make] Error 2
make: *** [Makefile:69: all] Error 2
Mainly it's caused by perl/python objects being compiled with:
-specs=/usr/lib/rpm/redhat/redhat-hardened-cc1
which prevent the final link impossible, because it will check
for 'proper' objects with following option:
-specs=/usr/lib/rpm/redhat/redhat-hardened-ld
Signed-off-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20171204082437.GC30564@krava
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Pull x86 fixes from Ingo Molnar:
"Misc fixes:
- fix the s2ram regression related to confusion around segment
register restoration, plus related cleanups that make the code more
robust
- a guess-unwinder Kconfig dependency fix
- an isoimage build target fix for certain tool chain combinations
- instruction decoder opcode map fixes+updates, and the syncing of
the kernel decoder headers to the objtool headers
- a kmmio tracing fix
- two 5-level paging related fixes
- a topology enumeration fix on certain SMP systems"
* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
objtool: Resync objtool's instruction decoder source code copy with the kernel's latest version
x86/decoder: Fix and update the opcodes map
x86/power: Make restore_processor_context() sane
x86/power/32: Move SYSENTER MSR restoration to fix_processor_context()
x86/power/64: Use struct desc_ptr for the IDT in struct saved_context
x86/unwinder/guess: Prevent using CONFIG_UNWINDER_GUESS=y with CONFIG_STACKDEPOT=y
x86/build: Don't verify mtools configuration file for isoimage
x86/mm/kmmio: Fix mmiotrace for page unaligned addresses
x86/boot/compressed/64: Print error if 5-level paging is not supported
x86/boot/compressed/64: Detect and handle 5-level paging at boot-time
x86/smpboot: Do not use smp_num_siblings in __max_logical_packages calculation
