14 Nov, 2016
2 commits
-
Adding support to detect jvmti support. It is not plugged into the
FEATURE_TESTS machinery, because it's quite rare and will be used
separately from perf via feature_check call.Signed-off-by: Jiri Olsa
Tested-by: Stephane Eranian
Cc: Andi Kleen
Cc: David Ahern
Cc: Namhyung Kim
Cc: Peter Zijlstra
Cc: William Cohen
Link: http://lkml.kernel.org/r/1478093749-5602-3-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Adding support to remove options from final CFLAGS for both object file
and build target. It's now possible to remove CFLAGS options like:CFLAGS_REMOVE_krava.o += -Wstrict-prototypes
Committer notes:
This comes from the kernel's kbuild infrastructure, the subset that is
supported in tools/ is being documented at tools/build/Documentation/Build.txt.Signed-off-by: Jiri Olsa
Tested-by: Stephane Eranian
Cc: Andi Kleen
Cc: David Ahern
Cc: Namhyung Kim
Cc: Peter Zijlstra
Cc: William Cohen
Link: http://lkml.kernel.org/r/1478093749-5602-2-git-send-email-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
12 Nov, 2016
1 commit
-
Since the unprivileged sched switch event was added in perf, PT doesn't
need need perf_event_paranoid=-1 anymore for per cpu decoding.Add a note stating that that is only needed for kernels < 4.2.
Reported-by: Andi Kleen
Report-Link: http://lkml.kernel.org/r/http://lkml.kernel.org/n/tip-x2ybghpqxxn3zu0m8o7qi42r@git.kernel.org
Acked-by: Adrian Hunter
Fixes: 45ac1403f564 ("perf: Add PERF_RECORD_SWITCH to indicate context switches")
Link: http://lkml.kernel.org/n/tip-x2ybghpqxxn3zu0m8o7qi42r@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
08 Nov, 2016
1 commit
-
Since 841e3558b2d ("perf callchain: Recording 'dwarf' callchains do not
need DWARF unwinding support"), --call-graph dwarf is allowed in 'perf
record' even without unwind support. A couple of other places don't
reflect this yet though: the help text should list dwarf as a valid
record mode and the dump_size config should be respected too.Signed-off-by: Rabin Vincent
Cc: He Kuang
Fixes: 841e3558b2de ("perf callchain: Recording 'dwarf' callchains do not need DWARF unwinding support")
Link: http://lkml.kernel.org/r/1470837148-7642-1-git-send-email-rabin.vincent@axis.com
Signed-off-by: Arnaldo Carvalho de Melo
29 Oct, 2016
1 commit
-
…ernel/git/acme/linux into perf/core
perf/core improvements and fixes from Arnaldo Carvalho de Melo:
New features:
- Support matching by topic in 'perf list' (Andi Kleen)
User visible:
- Apply cpu color only when there was activity in 'perf sched map' (Namhyung Kim)
- Always show the task's COMM in 'perf sched map -v' (Namhyung Kim)
- Fix hierarchy column counts in the perf hist browser (top, report), avoiding
showing nothing after pressing the RIGHT key a number of times (Namhyung Kim)Infrastructure:
- Support cascading options in libsubcmd and use it to share common options in
'perf sched' subcommands (Namhyung Kim)- Avoid worker cacheline bouncing in 'perf bench futex' (Davidlohr Bueso)
- Sanitize numeric parameters in 'perf bench futex' (Davidlohr Bueso)
- Update copies of kernel files (Arnaldo Carvalho de Melo)
- Fix scripting (perl, python) setup to avoid leaks (Arnaldo Carvalho de Melo)
- Add missing object file to the python binding linkage list (Arnaldo Carvalho de Melo)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
28 Oct, 2016
8 commits
-
In ac12f6764c50 ("perf tools: Implement branch_type event parameter") we
started using the parse_branch_str() function from one of the files used
in the python binding, which caused this entry in 'perf test' to fail:# perf test -v python
16: Try 'import perf' in python, checking link problems :
--- start ---
test child forked, pid 16667
Traceback (most recent call last):
File "", line 1, in
ImportError: /tmp/build/perf/python/perf.so: undefined symbol:
parse_branch_str
test child finished with -1
---- end ----
Try 'import perf' in python, checking link problems: FAILED!
#I must've commited some mistake when running 'perf test' to send the
pull request for the perf-core-for-mingo-20161024 tag, to have let this
regression to pass, sigh.Just add tools/perf/util/parse-branch-options.c and switch from using
ui__warning(), that is not available in the python binding, use
pr_warning() instead, which is good enough for this case.Now:
# perf test python
16: Try 'import perf' in python, checking link problems : Ok
#Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Cc: Andi Kleen
Fixes: ac12f6764c50 ("perf tools: Implement branch_type event parameter")
Link: http://lkml.kernel.org/n/tip-9kn1ct1cx9ppwqlmzl6z0xhs@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Removing one more set of die() calls.
Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-6pyil685m5i2tugg56gcy0tg@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Both register_perl_scripting() and register_python_scripting() allocate
this variable, fix it by checking if it already was.Cc: Adrian Hunter
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Tom Zanussi
Cc: Wang Nan
Fixes: 7e4b21b84c43 ("perf/scripts: Add Python scripting engine")
Signed-off-by: Arnaldo Carvalho de Melo -
Introduced in commit f9afc6197e9b ("x86: Wire up protection keys system
calls")This will make 'perf trace' aware of them on x86_64.
Cc: Adrian Hunter
Cc: Dave Hansen
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Thomas Gleixner
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-s1ta2ttv2xacecqogmd3a9p1@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
To get the defines introduced in the commit e8c24d3a23a4 ("x86/pkeys:
Allocation/free syscalls")Silencing this perf build warning:
Warning: tools/include/uapi/asm-generic/mman-common.h differs from kernel
Need to change 'perf trace' to beautify those syscalls, as soon as
booting with a kernel with it.Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-yev9rexu02cl7cjeozzmrl9t@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Ignore export.h and EXPORT_SYMBOL in:
784d5699eddc ("x86: move exports to actual definitions")
We're not dragging this stuff, not useful in tools/
This silences the following warnings while building perf:
Warning: tools/arch/x86/lib/memcpy_64.S differs from kernel
Warning: tools/arch/x86/lib/memset_64.S differs from kernelCc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-h9vw3pe0fq79zmyqsfr0s0mo@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Add support in perf list topic to only show events belonging to a
specific vendor events topic. For example the following works now:% perf list frontend
List of pre-defined events (to be used in -e):stalled-cycles-frontend OR idle-cycles-frontend [Hardware event]
stalled-cycles-frontend OR cpu/stalled-cycles-frontend/ [Kernel PMU event]
frontend:
dsb2mite_switches.count
[Decode Stream Buffer (DSB)-to-MITE switches]
dsb2mite_switches.penalty_cycles
[Decode Stream Buffer (DSB)-to-MITE switch true penalty cycles]
dsb_fill.exceed_dsb_lines
[Cycles when Decode Stream Buffer (DSB) fill encounter more than 3 Decode Stream Buffer (DSB)
lines]
icache.hit
[Number of Instruction Cache, Streaming Buffer and Victim Cache Reads. both cacheable and
noncacheable, including UC fetches]
...Signed-off-by: Andi Kleen
Tested-by: Arnaldo Carvalho de Melo
Cc: Jiri Olsa
Link: http://lkml.kernel.org/r/1476902724-9586-2-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo -
Joonwoo reported that there's a mismatch between timestamps in script
and sched commands. This was because of difference in printing the
timestamp. Factor out the code and share it so that they can be in
sync. Also I found that sched map has similar problem, fix it too.Committer notes:
Fixed the max_lat_at bug introduced by Namhyung's original patch, as
pointed out by Joonwoo, and made it a function following the scnprintf()
model, i.e. returning the number of bytes formatted, and receiving as
the first parameter the object from where the data to the formatting is
obtained, renaming it from:char *timestamp_in_usec(char *bf, size_t size, u64 timestamp)
to
int timestamp__scnprintf_usec(u64 timestamp, char *bf, size_t size)
Reported-by: Joonwoo Park
Signed-off-by: Namhyung Kim
Cc: David Ahern
Cc: Jiri Olsa
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/20161024020246.14928-3-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
26 Oct, 2016
2 commits
-
I'd like to see the name of tasks with perf sched map, but it only shows
name of new tasks and then use short names after all. This is not good
for long running tasks since it's hard for users to track the short
names. This patch makes it show the names (except the idle task) when
-v option is used. Probably we may make it as default behavior.Signed-off-by: Namhyung Kim
Acked-by: Jiri Olsa
Cc: David Ahern
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/20161024020246.14928-2-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Applying cpu color always doesn't help readability IMHO. Instead it
might be better to applying the color when there's an activity on those
CPUs.Signed-off-by: Namhyung Kim
Acked-by: Jiri Olsa
Cc: David Ahern
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/20161024020246.14928-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo
25 Oct, 2016
6 commits
-
The -i and -v options can be used in subcommands so enable cascading the
sched_options. This fixes the following inconvenience in 'perf sched':$ perf sched -i perf.data.sched map
... (it works well) ...$ perf sched map -i perf.data.sched
Error: unknown switch `i'Usage: perf sched map []
--color-cpus
highlight given CPUs in map
--color-pids
highlight given pids in map
--compact map output in compact mode
--cpus display given CPUs in mapWith this patch, the second command line works with the perf.data.sched
data file.Signed-off-by: Namhyung Kim
Tested-by: Arnaldo Carvalho de Melo
Acked-by: Jiri Olsa
Cc: Andi Kleen
Cc: David Ahern
Cc: Josh Poimboeuf
Cc: Peter Zijlstra
Cc: Wang Nan
Link: http://lkml.kernel.org/r/20161024030003.28534-2-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Sometimes subcommand have common options and it can only handled in the
upper level command unless it duplicates the options.This patch adds a parent field and fallback to the parent if the given
argument was not found in the current options.Signed-off-by: Namhyung Kim
Acked-by: Jiri Olsa
Cc: Andi Kleen
Cc: David Ahern
Cc: Josh Poimboeuf
Cc: Peter Zijlstra
Cc: Wang Nan
Link: http://lkml.kernel.org/r/20161024030003.28534-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
The perf report/top on TUI supports horizontal scrolling using LEFT and
RIGHT keys.But it calculate the number of columns incorrectly when hierarchy mode
is enabled so that keep pressing RIGHT key can make the output
disappeared.In the hierarchy mode, all sort keys are collapsed into a single column,
so it needs to be applied when calculating column numbers.Reported-and-Tested-by: Markus Trippelsdorf
Signed-off-by: Namhyung Kim
Tested-by: Arnaldo Carvalho de Melo
Cc: Jiri Olsa
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/20161024162110.17918-1-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
This gets rid of oddities such as:
perf bench futex hash -t -4
perf: calloc: Cannot allocate memoryRuntime (and many more) are equally busted, i.e. run for bogus amounts of
time. Just use the abs, instead of, for example errorring out.Committer note:
After the patch:
$ perf bench futex hash -t -4
# Running 'futex/hash' benchmark:
Run summary [PID 10178]: 4 threads, each operating on 1024 [private] futexes for 10 secs.[thread 0] futexes: 0x34f9fa0 ... 0x34faf9c [ 4702208 ops/sec ]
[thread 1] futexes: 0x34fb140 ... 0x34fc13c [ 4707020 ops/sec ]
[thread 2] futexes: 0x34fc2e0 ... 0x34fd2dc [ 4711526 ops/sec ]
[thread 3] futexes: 0x34fd480 ... 0x34fe47c [ 4709683 ops/sec ]Averaged 4707609 operations/sec (+- 0.04%), total secs = 10
$Signed-off-by: Davidlohr Bueso
Tested-by: Arnaldo Carvalho de Melo
Link: http://lkml.kernel.org/r/1477342613-9938-3-git-send-email-dave@stgolabs.net
Signed-off-by: Arnaldo Carvalho de Melo -
Sebastian noted that overhead for worker thread ops (throughput)
accounting was producing 'perf' to appear in the profiles, consuming a
non-trivial (i.e. 13%) amount of CPU.This is due to cacheline bouncing due to the increment of w->ops.
We can easily fix this by just working on a local copy and updating the
actual worker once done running, and ready to show the program summary.
There is no danger of the worker being concurrent, so we can trust that
no stale value is being seen by another thread.This also gets rid of the unnecessary cache alignment hack; its not
worth it.Reported-by: Sebastian Andrzej Siewior
Signed-off-by: Davidlohr Bueso
Acked-by: Sebastian Andrzej Siewior
Link: http://lkml.kernel.org/r/1477342613-9938-2-git-send-email-dave@stgolabs.net
Signed-off-by: Arnaldo Carvalho de Melo -
…ernel/git/acme/linux into perf/core
Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo:
New features:
- Dynamicly change verbosity level by pressing 'V' in the 'perf top/report'
hists TUI browser (Alexis Berlemont)- Implement 'perf trace --delay' in the same fashion as in 'perf record --delay',
to skip sampling workload initialization events (Alexis Berlemont)- Make vendor named events case insensitive in 'perf list', i.e.
'perf list LONGEST_LAT' works just the same as 'perf list longest_lat' (Andi Kleen)- Show instruction bytes and lenght in 'perf script' for Intel PT and BTS (Andi Kleen, Adrian Hunter)
E.g:
% perf record -e intel_pt// foo
% perf script --itrace=i0ns -F ip,insn,insnlen
ffffffff8101232f ilen: 5 insn: 0f 1f 44 00 00
ffffffff81012334 ilen: 1 insn: 5b
ffffffff81012335 ilen: 1 insn: 5d
ffffffff81012336 ilen: 1 insn: c3
ffffffff810123e3 ilen: 1 insn: 5b
ffffffff810123e4 ilen: 2 insn: 41 5c
ffffffff810123e6 ilen: 1 insn: 5d
ffffffff810123e7 ilen: 1 insn: c3
ffffffff810124a6 ilen: 2 insn: 31 c0
ffffffff810124a8 ilen: 9 insn: 41 83 bc 24 a8 01 00 00 01
ffffffff810124b1 ilen: 2 insn: 75 87- Allow enabling the perf_event_attr.branch_type attribute member: (Andi Kleen)
perf record -e sched:sched_switch,cpu/cpu-cycles,branch_type=any/ ...
- Add unwinding support for jitdump (Stefano Sanfilippo)
Fixes:
- Use raw_syscall:sys_enter timestamp in 'perf trace' (Arnaldo Carvalho de Melo)
Infrastructure:
- Allow jitdump to be built without libdwarf (Maciej Debski)
- Sync x86's syscall table tools/ copy (Arnaldo Carvalho de Melo)
- Fixes to avoid calling die() in library fuctions already propagating other
errors (Arnaldo Carvalho de Melo)- Improvements to allow libtraceevent to be properly installed in distro
packages (Jiri Olsa)- Removing coresight miscellaneous debug output (Mathieu Poirier)
- Cache align the 'perf bench futex' worker struct (Sebastian Andrzej Siewior)
Documentation:
- Minor improvements on the documentation of event parameters (Andi Kleen)
- Add jitdump format specification document (Stephane Eranian)
Spelling fixes:
- Fix typo "No enough" to "Not enough" (Alexander Alemayhu)
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
24 Oct, 2016
19 commits
-
Printing the full path of the selected link is obviously not needed,
hence removing.Signed-off-by: Mathieu Poirier
Cc: Peter Zijlstra
Cc: linux-arm-kernel@lists.infradead.org
Link: http://lkml.kernel.org/r/1476913323-6836-1-git-send-email-mathieu.poirier@linaro.org
Signed-off-by: Arnaldo Carvalho de Melo -
Make the 'perf list' glob matching for vendor events case insensitive.
This allows to use the upper case vendor events with perf list too.Now the following works:
% perf list LONGEST_LAT
...
cache:
longest_lat_cache.miss
[Core-originated cacheable demand requests missed LLC]
longest_lat_cache.reference
[Core-originated cacheable demand requests that refer to LLC]Signed-off-by: Andi Kleen
Suggested-by: Ingo Molnar
Tested-by: Arnaldo Carvalho de Melo
Cc: Jiri Olsa
Link: http://lkml.kernel.org/r/1476899402-31460-1-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo -
Instead of the one when another syscall takes place while another is being
processed (in another CPU, but we show it serialized, so need to "interrupt"
the other), and also when finally showing the sys_enter + sys_exit + duration,
where we were showing the sample->time for the sys_exit, duh.Before:
# perf trace sleep 1
0.373 ( 0.001 ms): close(fd: 3 ) = 0
1000.626 (1000.211 ms): nanosleep(rqtp: 0x7ffd6ddddfb0) = 0
1000.653 ( 0.003 ms): close(fd: 1 ) = 0
1000.657 ( 0.002 ms): close(fd: 2 ) = 0
1000.667 ( 0.000 ms): exit_group( )
#After:
# perf trace sleep 1
0.336 ( 0.001 ms): close(fd: 3 ) = 0
0.373 (1000.086 ms): nanosleep(rqtp: 0x7ffe303e9550) = 0
1000.481 ( 0.002 ms): close(fd: 1 ) = 0
1000.485 ( 0.001 ms): close(fd: 2 ) = 0
1000.494 ( 0.000 ms): exit_group( )
[root@jouet linux]#Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-ecbzgmu2ni6glc6zkw8p1zmx@git.kernel.org
Fixes: 752fde44fd1c ("perf trace: Support interrupted syscalls")
Signed-off-by: Arnaldo Carvalho de Melo -
Not used at all, we need just the entry_time to calculate the syscall
duration.Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-js6r09zdwlzecvaei7t4l3vd@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
It popped up in perf testing that the worker consumes some amount of
CPU. It boils down to the increment of `ops` which causes cache line
bouncing between the individual threads.This patch aligns the struct by 256 bytes to ensure that not a cache
line is shared among CPUs. 128 byte is the x86 worst case and grep says
that L1_CACHE_SHIFT is set to 8 on s390.Signed-off-by: Sebastian Andrzej Siewior
Cc: Davidlohr Bueso
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/20161016190803.3392-1-bigeasy@linutronix.de
Signed-off-by: Arnaldo Carvalho de Melo -
We already have handling for errors when processing PERF_RECORD_ events,
so instead of calling die() when not being able to alloc, propagate the
error, so that the normal UI exit sequence can take place, the user be
warned and possibly the terminal be properly reset to a sane mode.Cc: Adrian Hunter
Cc: Brice Goglin
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-r90je3c009a125dvs3525yge@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
It already returns whatever strbuf_(grow|addch)() returns in case of
failure, so just return -ENOSPC in the only case where it was die()ing.
When it returns, its only caller will call die() anyway, so no need to
be so eager, die later.Cc: Adrian Hunter
Cc: David Ahern
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-as05b7mbogprlwi8iarwns8e@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
Instead of having all tests perform alloc/free, do it in the code that
calls the do_cycles() and do_gettimeofday() functions.Cc: Adrian Hunter
Cc: David Ahern
Cc: Hitoshi Mitake
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Wang Nan
Link: http://lkml.kernel.org/n/tip-lywj4mbdb1m9x1z9asivwuuy@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo -
In the perf wiki todo-list[1], there is an entry regarding initial-delay
and 'perf trace'; the following small patch tries to fulfill this point.
It has been generated against the branch tip/perf/core.It has only been implemented in the "trace__run" case.
Ex.:
$ sudo strace -- ./perf trace --delay 5 sleep 1 2>&1
...
fcntl(7, F_SETFL, O_RDONLY|O_NONBLOCK) = 0
ioctl(7, PERF_EVENT_IOC_ID, 0x7ffc8fd35718) = 0
ioctl(11, PERF_EVENT_IOC_SET_OUTPUT, 0x7) = 0
fcntl(11, F_SETFL, O_RDONLY|O_NONBLOCK) = 0
ioctl(11, PERF_EVENT_IOC_ID, 0x7ffc8fd35718) = 0
write(6, "\0", 1) = 1
close(6) = 0
nanosleep({0, 5000000}, NULL) = 0 # DELAY OF 5 MS BEFORE ENABLING THE EVENTS
ioctl(3, PERF_EVENT_IOC_ENABLE, 0) = 0
ioctl(4, PERF_EVENT_IOC_ENABLE, 0) = 0
ioctl(5, PERF_EVENT_IOC_ENABLE, 0) = 0
ioctl(7, PERF_EVENT_IOC_ENABLE, 0) = 0
...[1]: https://perf.wiki.kernel.org/index.php/Todo
Signed-off-by: Alexis Berlemont
Suggested-and-Tested-by: Arnaldo Carvalho de Melo
Cc: Alexander Shishkin
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/20161010054328.4028-2-alexis.berlemont@gmail.com
[ Add entry to the manpage, cut'n'pasted from stat's and record's ]
Signed-off-by: Arnaldo Carvalho de Melo -
Here is a small patch which tries to fulfill a point in the perf todo
list:* Make pressing 'V' multiple times to go on cycling thru various
verbosity levels in 'perf top', so that info that is present in
'perf top -v' can be obtained without having to restart the tool
(acme).After a small grep in the code, the max verbosity level seems 3; so,
we cycle at 4; I did not dare define a MAX_VERBOSE_LEVEL constant.Signed-off-by: Alexis Berlemont
Suggested-and-Tested-by: Arnaldo Carvalho de Melo
Cc: Alexander Shishkin
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/20161012214823.14324-2-alexis.berlemont@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo -
The latter version occurs much more when running git grep.
Signed-off-by: Alexander Alemayhu
Cc: Alexander Shishkin
Cc: Peter Zijlstra
Cc: Wang Nan
Link: http://lkml.kernel.org/r/20161013161811.4939-1-alexander@alemayhu.com
Signed-off-by: Arnaldo Carvalho de Melo -
With uncore event aliases which are duplicated over multiple PMUs the
"Using CPUID" message with -v could be printed many times. Only print
it once.Signed-off-by: Andi Kleen
Cc: Jiri Olsa
Cc: Stephane Eranian
Cc: Sukadev Bhattiprolu
Link: http://lkml.kernel.org/r/1476393332-20732-3-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo -
This patch adds a formal specification of the jitdump format. The goal
is to help jit runtime developers implement the jitdump support without
having to read the jvmti code.Signed-off-by: Stephane Eranian
Cc: Anton Blanchard
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1476356383-30100-10-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo -
Check the version number when opening a jitdump file. Accept older
versions, but not newer ones.Signed-off-by: Stefano Sanfilippo
Signed-off-by: Ross McIlroy
Reviewed-by: Stephane Eranian
Cc: Anton Blanchard
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1476356383-30100-9-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo -
When the jit_buf_desc contains unwinding information, it is emitted as
eh_frame unwinding sections in the DSOs generated by perf inject.The unwinding information is required to unwind of JITed code which do
not maintain the frame pointer register during function calls. It can
be emitted by V8 / Chromium when the --perf_prof_unwinding_info is
passed to V8.The eh_frame and eh_frame_hdr sections are emitted immediately after the
.text.The .eh_frame is aligned at a 8-byte boundary, and .eh_frame_hdr at a
4-byte one. Since size of the .eh_frame is required to be a multiple of
the word size, which means there will never be additional padding
between it and the .eh_frame_hdr on machines where the word size is 4 or
8 bytes.However, additional padding might be inserted between .text and
.eh_frame to reach the correct alignment, which will always be 8 bytes,
also on 32bit machines. The reasoning behind this choice is that 4 extra
bytes of padding worst case are not a large cost for the advantage of
removing word-size dependent offset calculations when emitting the
jitdump.Signed-off-by: Stefano Sanfilippo
Signed-off-by: Ross McIlroy
Reviewed-by: Stephane Eranian
Cc: Anton Blanchard
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1476356383-30100-8-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo -
This record is intended to provide unwinding information in the
eh_frame format. This is required to unwind JITed code which
does not maintain the frame pointer register during function calls.The eh_frame unwinding information can be emitted by V8 / Chromium
when the --perf_prof_unwinding_info is passed.A record of type jr_code_unwinding_info comes before the jr_code_load
it referred to and contains both the .eh_frame and .eh_frame_hdr.The fields in the header have the following meaning:
* unwinding_size: size of the eh_frame and eh_frame_hdr, necessary
for distinguishing the content from the padding.* eh_frame_hdr_size: as the name says.
* mapped_size: size of the payload that was in memory at runtime.
typically unwinding_size if the .eh_frame_hdr and .eh_frame were
mapped, or 0 if they weren't. It should always be the former case,
since the .eh_frame is guaranteed to be mapped in memory. However,
certain JITs might want to inject an .eh_frame_hdr with an empty LUT
to trigger fp-based unwinding fallback in libunwind. The only part
of the .eh_frame_hdr that libunwind reads from remote memory is the
LUT, and since there is none, mapping the unwinding info in memory
is not necessary, and 0 in this field signifies that it wasn't.
This practical hack allows to save bytes in code memory for those
JIT compilers that might or might not maintain a valid frame pointer.The payload that follows is assumed to contain first the .eh_frame and
then the .eh_header_hdr, with no padding between the two.Signed-off-by: Stefano Sanfilippo
Signed-off-by: Ross McIlroy
Reviewed-by: Stephane Eranian
Cc: Anton Blanchard
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1476356383-30100-7-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo -
When calculating .eh_frame_hdr base and LUT offsets do not always assume
that pgoff is zero.The assumption is false for DSOs built from the jitdump by perf inject,
because the ELF header did not exist in memory at sampling time.Signed-off-by: Stefano Sanfilippo
Signed-off-by: Ross McIlroy
Reviewed-by: Stephane Eranian
Cc: Anton Blanchard
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1476356383-30100-6-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo -
The behavior before this commit was to skip the remaining portion of the
jitdump in case an unknown record was found, including those records
that perf could handle.With this change, parsing a record with an unknown id will cause a
warning to be emitted, the record will be skipped and parsing will
resume from the next (valid) one.The patch aims at making perf more future proof, by extracting as much
information as possible from jitdumps.Signed-off-by: Stefano Sanfilippo
Signed-off-by: Ross McIlroy
Reviewed-by: Stephane Eranian
Cc: Anton Blanchard
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1476356383-30100-5-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo -
This patch removes all the string padding generated in the jitdump file.
They are not necessary and were adding unnecessary complexity. Modern
processors can handle unaligned accesses quite well. The perf.data/
jitdump file are always post-processed, no need to add extra complexity
for no real gain.Signed-off-by: Stephane Eranian
Cc: Anton Blanchard
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Peter Zijlstra
Link: http://lkml.kernel.org/r/1476356383-30100-4-git-send-email-eranian@google.com
Signed-off-by: Arnaldo Carvalho de Melo
