Add support to perf stat to print the basic transactional execution statistics:
Total cycles, Cycles in Transaction, Cycles in aborted transsactions
using the in_tx and in_tx_checkpoint qualifiers.
Transaction Starts and Elision Starts, to compute the average transaction
length.
This is a reasonable overview over the success of the transactions.
Also support architectures that have a transaction aborted cycles
counter like POWER8. Since that is awkward to handle in the kernel
abstract handle both cases here.
Enable with a new --transaction / -T option.
This requires measuring these events in a group, since they depend on each
other.
This is implemented by using TM sysfs events exported by the kernel
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Arnaldo Carvalho de Melo <acme@infradead.org>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1377128846-977-5-git-send-email-andi@firstfloor.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
On ARM the debug info is not present in the .eh_frame sections but
in .debug_frame instead, in dwarf format.
Use libunwind to load and parse the debug info.
Dependencies:
. if present, libunwind >= 1.1 is needed to prevent a segfault when
parsing the dwarf info,
. libunwind needs to be configured with --enable-debug-frame. Note:
--enable-debug-frame is automatically selected on ARM.
Acked-by: Jiri Olsa <jolsa@redhat.com>
Signed-off-by: Jean Pihet <jean.pihet@linaro.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
This reverts commit de95ab5364.
Markus Trippelsdorf reported that this commit broke 'perf top':
> I just see a gray screen with no text at all. Sometimes the
> following error messages are printed:
>
> *** Error in `perf': invalid fastbin entry (free): 0x00000000029b18c0
> ***
> *** Error in `perf': malloc(): memory corruption (fast): 0x0000000000ee0b10 ***
While this code is fixable, the commit itself fails on several levels:
- it should have been a separate helper function
- why the heck does it do strchr() twice
- it casts a const char * over into char *
- sloppy style
- it's not even a regression fix!
So lets revert it and re-try the patch in v3.13.
Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Fix perf probe to probe on some symbols which have some optimzation
suffixes, e.g. ".part", ".isra", and ".constprop".
To fix this issue, instead of using the DIE name, perf probe uses the
symbol name found by dwfl_module_addrsym().
This also involves a perf probe --vars operation update which now shows
the symbol name instead of the DIE name.
Without this patch, putting a probe on an inlined function which was
compiled with a suffixed symbol will fail like this:
$ perf probe -v getname_flags
probe-definition(0): getname_flags
symbol:getname_flags file:(null) line:0 offset:0 return:0 lazy:(null)
0 arguments
Looking at the vmlinux_path (6 entries long)
Using /lib/modules/3.11.0+/build/vmlinux for symbols
found inline addr: 0xffffffff8119bb70
Probe point found: getname_flags+0
found inline addr: 0xffffffff8119bcb6
Probe point found: getname+6
found inline addr: 0xffffffff811a06a6
Probe point found: user_path_at_empty+6
find 3 probe_trace_events.
Opening /sys/kernel/debug//tracing/kprobe_events write=1
Added new events:
Writing event: p:probe/getname_flags getname_flags+0
Failed to write event: No such file or directory
Error: Failed to add events. (-1)
Because the debuginfo knows only the original (non suffix) symbol name,
it uses the original symbol for probe address but the kernel (kallsyms)
knows only suffixed symbol. Then, the kernel rejects that original
symbol.
This patch uses dwfl_module_addrsym() to get the correct (suffixed)
symbol from symtab when a probe point is found.
Reported-by: Arnaldo Carvalho de Melo <acme@ghostprotocols.net>
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20130925131616.31632.46658.stgit@udc4-manage.rcp.hitachi.co.jp
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When parsing lines from objdump a line containing source code starting
with a numeric label is mistaken for a line of disassembly starting with
a memory address.
Current validation fails to recognise that the "memory address" is out
of range and calculates an invalid offset which later causes this
segfault:
Program received signal SIGSEGV, Segmentation fault.
0x0000000000457315 in disasm__calc_percent (notes=0xc98970, evidx=0, offset=143705, end=2127526177, path=0x7fffffffbf50)
at util/annotate.c:631
631 hits += h->addr[offset++];
(gdb) bt
#0 0x0000000000457315 in disasm__calc_percent (notes=0xc98970, evidx=0, offset=143705, end=2127526177, path=0x7fffffffbf50)
at util/annotate.c:631
#1 0x00000000004d65e3 in annotate_browser__calc_percent (browser=0x7fffffffd130, evsel=0xa01da0) at ui/browsers/annotate.c:364
#2 0x00000000004d7433 in annotate_browser__run (browser=0x7fffffffd130, evsel=0xa01da0, hbt=0x0) at ui/browsers/annotate.c:672
#3 0x00000000004d80c9 in symbol__tui_annotate (sym=0xc989a0, map=0xa02660, evsel=0xa01da0, hbt=0x0) at ui/browsers/annotate.c:962
#4 0x00000000004d7aa0 in hist_entry__tui_annotate (he=0xdf73f0, evsel=0xa01da0, hbt=0x0) at ui/browsers/annotate.c:823
#5 0x00000000004dd648 in perf_evsel__hists_browse (evsel=0xa01da0, nr_events=1, helpline=
0x58b768 "For a higher level overview, try: perf report --sort comm,dso", ev_name=0xa02cd0 "cycles", left_exits=false, hbt=
0x0, min_pcnt=0, env=0xa011e0) at ui/browsers/hists.c:1659
#6 0x00000000004de372 in perf_evlist__tui_browse_hists (evlist=0xa01520, help=
0x58b768 "For a higher level overview, try: perf report --sort comm,dso", hbt=0x0, min_pcnt=0, env=0xa011e0)
at ui/browsers/hists.c:1950
#7 0x000000000042cf6b in __cmd_report (rep=0x7fffffffd6c0) at builtin-report.c:581
#8 0x000000000042e25d in cmd_report (argc=0, argv=0x7fffffffe4b0, prefix=0x0) at builtin-report.c:965
#9 0x000000000041a0e1 in run_builtin (p=0x801548, argc=1, argv=0x7fffffffe4b0) at perf.c:319
#10 0x000000000041a319 in handle_internal_command (argc=1, argv=0x7fffffffe4b0) at perf.c:376
#11 0x000000000041a465 in run_argv (argcp=0x7fffffffe38c, argv=0x7fffffffe380) at perf.c:420
#12 0x000000000041a707 in main (argc=1, argv=0x7fffffffe4b0) at perf.c:521
After the fix is applied the symbol can be annotated showing the
problematic line "1: rep"
copy_user_generic_string /usr/lib/debug/lib/modules/3.9.10-100.fc17.x86_64/vmlinux
*/
ENTRY(copy_user_generic_string)
CFI_STARTPROC
ASM_STAC
andl %edx,%edx
and %edx,%edx
jz 4f
je 37
cmpl $8,%edx
cmp $0x8,%edx
jb 2f /* less than 8 bytes, go to byte copy loop */
jb 33
ALIGN_DESTINATION
mov %edi,%ecx
and $0x7,%ecx
je 28
sub $0x8,%ecx
neg %ecx
sub %ecx,%edx
1a: mov (%rsi),%al
mov %al,(%rdi)
inc %rsi
inc %rdi
dec %ecx
jne 1a
movl %edx,%ecx
28: mov %edx,%ecx
shrl $3,%ecx
shr $0x3,%ecx
andl $7,%edx
and $0x7,%edx
1: rep
100.00 rep movsq %ds:(%rsi),%es:(%rdi)
movsq
2: movl %edx,%ecx
33: mov %edx,%ecx
3: rep
rep movsb %ds:(%rsi),%es:(%rdi)
movsb
4: xorl %eax,%eax
37: xor %eax,%eax
data32 xchg %ax,%ax
ASM_CLAC
ret
retq
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1379009721-27667-1-git-send-email-adrian.hunter@intel.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This patch adds support for the new PERF_RECORD_MMAP2 record type
exposed by the kernel. This is an extended PERF_RECORD_MMAP record.
It adds for each file-backed mapping the device major, minor number and
the inode number and generation.
This triplet uniquely identifies the source of a file-backed mapping. It
can be used to detect identical virtual mappings between processes, for
instance.
The patch will prefer MMAP2 over MMAP.
Signed-off-by: Stephane Eranian <eranian@google.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1377079825-19057-3-git-send-email-eranian@google.com
[ Cope with 314add6 "Change machine__findnew_thread() to set thread pid",
fix 'perf test' regression test entry affected,
use perf_missing_features.mmap2 to fallback to not using .mmap2 in older kernels,
so that new tools can work with kernels where this feature is not present ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Before:
perf trace -i perf.data
Segmentation fault (core dumped)
#
After:
# perf trace -i perf.data
Data file does not have raw_syscalls:sys_enter events
#
When there are no tracepoints in a perf.data file the struct pevent
that contains the list of tracepoints that will be used to lookup the
tracepoint id by name will not be populated, causing a NULL deref.
And we don't need to do all that dance to look at pevents for an entry
with a slighly different name to then lookup the tracepoint by its id on
the evlist, just use the perf_evlist__find_tracepoint_by_name() routine,
that will find the tracepoint, if present.
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-egcm21k1e6gcyxpcgjxtmsq3@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently when processing events in the __perf_session__process_events
function we update a progress bar based on the file_size. During the
same processing we update the progress bar from within
flush_sample_queue which is based on number of samples count.
Having 2 different based updates is causing the progress bar to jump
heavily back and forth giving not much usefull info.
Fixing this by keeping only __perf_session__process_events based
progress bar update. And turning on flush_sample_queue progress bar
update only for final flushing.
This reduces the number of time the progress bar update function is
called and it significantly reduces the loading time for TUI, where the
progress bar update takes quite a lot of time.
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Cc: Corey Ashford <cjashfor@linux.vnet.ibm.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20130905091449.GC1100@krava.brq.redhat.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Occassionally events (e.g., context-switch, sched tracepoints) are losing
the conversion of sample data associated with a thread. For example:
$ perf record -e sched:sched_switch -c 1 -a -- sleep 5
$ perf script
<selected events shown>
ls 30482 [000] 1379727.583037: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
ls 30482 [000] 1379727.586339: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
:30482 30482 [000] 1379727.589462: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
The last line lost the conversion from tid to comm. If you look at the events
(perf script -D) you see why - a SAMPLE event is generated after the EXIT:
0 1379727589449774 0x1540b0 [0x38]: PERF_RECORD_EXIT(30482:30482):(30482:30482)
0 1379727589462497 0x1540e8 [0x80]: PERF_RECORD_SAMPLE(IP, 1): 30482/30482: 0xffffffff816416f1 period: 1 addr: 0
... thread: :30482:30482
When perf processes the EXIT event the thread is moved to the dead_threads
list. When the SAMPLE event is processed no thread exists for the pid so a new
one is created by machine__findnew_thread.
This patch address the problem by delaying the move to the dead_threads list
until the tid is re-used (per Adrian's suggestion).
With this patch we get the previous example shows:
ls 30482 [000] 1379727.583037: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
ls 30482 [000] 1379727.586339: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
ls 30482 [000] 1379727.589462: sched:sched_switch: prev_comm=ls prev_pid=30482 ...
and
0 1379727589449774 0x1540b0 [0x38]: PERF_RECORD_EXIT(30482:30482):(30482:30482)
0 1379727589462497 0x1540e8 [0x80]: PERF_RECORD_SAMPLE(IP, 1): 30482/30482: 0xffffffff816416f1 period: 1 addr: 0
... thread: ls:30482
v4: per Arnaldo's request add dead flag to thread struct and set when task exits
v3: re-do from a time based check to a delayed move to dead_threads list
v2: Rebased to latest perf/core branch. Changed time comparison to use
a macro which explicitly shows the time basis
Signed-off-by: David Ahern <dsahern@gmail.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1376491767-84171-1-git-send-email-dsahern@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
perf stat -a needs 10 open file descriptors per logical CPU
perf stat -a -dddd needs 20 open fds for each.
This implies that stat -a doesn't work on any system with the default
ulimit -n 1024 which has more than ~100 CPUs and stat -a -dddd doesn't
work on anything with more than 46 CPUs.
Longer term there needs to be probably some way to lower the file
descriptor requirements. This would need some changes in the kernel/user
interface.
But short term this patch just tries to increase the file descriptor
limit in perf itself, when it runs into a EMFILE.
It first sets it to the hard limit, and then tries to increase the hard
limit.
On Fedora systems the default seems to be soft limit 1024 and hard limit
4*1024. So even non root can support 409 or 186 CPUs respectively. root
can go far higher.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Link: http://lkml.kernel.org/r/1375670486-15480-1-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
For some types of work loads and special guest environments, you might
have a kernel that has no kernel modules. The perf kvm record tool
fails instantiate vmlinux maps when the kernel modules directory cannot
be opened, even though the kallsyms has been properly processed. This
leads to a perf kvm report that has no guest symbols resolved.
This patch changes the failure to locate kernel modules to be non-fatal.
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Acked-by: David Ahern <dsahern@gmail.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1373920073-4874-1-git-send-email-jason.wessel@windriver.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This commit adds support for a new modifier "D", which requests that the
event, or group of events, be pinned to the PMU.
The "p" modifier is already taken for precise, and "P" may be used in
future to mean "fully precise".
So we use "D", which stands for pinneD - and looks like a padlock, or if
you're using the ":D" syntax perf smiles at you.
This is an oft-requested feature from our HW folks, who want to be able
to run a large number of events, but also want 100% accurate results for
instructions per cycle.
Comparison of results with and without pinning:
$ perf stat -e '{cycles,instructions}:D' -e cycles,instructions,...
79,590,480,683 cycles # 0.000 GHz
166,123,716,524 instructions # 2.09 insns per cycle
# 0.11 stalled cycles per insn
79,352,134,463 cycles # 0.000 GHz [11.11%]
165,178,301,818 instructions # 2.08 insns per cycle
# 0.11 stalled cycles per insn [11.13%]
As you can see although perf does a very good job of scaling the values
in the non-pinned case, there is some small discrepancy.
The patch is fairly straight forward, the one detail is that we need to
make sure we only request pinning for the group leader when we have a
group.
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Jiri Olsa <jolsa@redhat.com>
Tested-by: Jiri Olsa <jolsa@redhat.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1375795686-4226-1-git-send-email-michael@ellerman.id.au
[ Use perf_evsel__is_group_leader instead of open coded equivalent, as
suggested by Jiri Olsa ]
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
