Add two new columns to the annotate display and display the average
cycles and the compute IPC if available.
When the LBR was not in any branch mode the IPC computation is
automatically disabled. We still display the cycle information.
Example output (with made up numbers):
The second column is the IPC and third average cycles.
│ __attribute__((noinline)) f2()
│ {
5.15 0.07 │ push %rbp
0.01 0.07 │ mov %rsp,%rbp
│ c = a / b;
9.87 0.07 │ mov a,%eax
0.07 │ mov b,%ecx
0.07 │ cltd
4.92 0.07 123│ idiv %ecx
70.79 0.07 │ mov %eax,__TMC_END__
│ }
9.25 0.07 │ pop %rbp
0.01 0.07 123│ ← retq
v2: Fix display problems.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1437233094-12844-7-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Compute the IPC and the basic block cycles for the annotate display.
IPC is computed by counting the instructions, and then dividing the
accounted cycles by that count.
The actual IPC computation can only be done at annotate time, because we
need to parse the objdump output first to know the number of
instructions in the basic block.
The cycles/IPC are also put into the perf function annotation so that
the display code can show them.
Again basic block overlaps are not handled, with the longest winning,
but there are some heuristics to hide the IPC when the longest is not
the most common.
v2: Compute IPC correctly.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1437233094-12844-6-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
At some point:
commit 2c86c7ca76
Author: Namhyung Kim <namhyung@kernel.org>
Date: Mon Mar 17 18:18:54 2014 -0300
perf report: Merge al->filtered with hist_entry->filtered
We stopped dropping samples for things filtered via the --comms, --dsos,
--symbols, etc, i.e. things marked as filtered in the symbol resolution
routines (thread__find_addr_map(), perf_event__preprocess_sample(),
etc).
But then, in:
commit 268397cb2a
Author: Namhyung Kim <namhyung@kernel.org>
Date: Tue Apr 22 14:49:31 2014 +0900
perf top/tui: Update nr_entries properly after a filter is applied
We don't take into account entries that were filtered in
perf_event__preprocess_sample() and friends, which leads to
inconsistency in the browser seek routines, that expects the number of
hist_entry->filtered entries to match what it thinks is the number of
unfiltered, browsable entries.
So, for instance, when we do:
perf top --symbols ___non_existent_symbol___
the hist_browser__nr_entries() routine thinks there are no filters in
place, uses the hists->nr_entries but all entries are filtered, leading
to a segfault.
Tested with:
perf top --symbols malloc,free --percentage=relative
Freezing, by pressing 'f', at any time and doing the math on the
percentages ends up with 100%, ditto for:
perf top --dsos libpthread-2.20.so,libxul.so --percentage=relative
Both were segfaulting, all fixed now.
More work needed to do away with checking if filters are in place, we
should just use the nr_non_filtered_samples counter, no need to
conditionally use it or hists.nr_filter, as what the browser does is
just show unfiltered stuff. An audit of how it is being accounted is
needed, this is the minimal fix.
Reported-by: Michael Petlan <mpetlan@redhat.com>
Fixes: 268397cb2a ("perf top/tui: Update nr_entries properly after a filter is applied")
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-6w01d5q97qk0d64kuojme5in@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Now it is possible to press CTRL+z at anytime and that will disable the
events being monitored, essentially turning 'top' into 'report', with
pressing CTRL+z again making it enable the events again, returning to
the 'top' behaviour, i.e. dynamic + decaying of older samples.
One may want, for instance, play with:
-d, --delay <n> number of seconds to delay between refreshes
and:
-z, --zero zero history across updates
Plus CTRL+z to see only the events since last zeroing, etc.
Suggested-by: Ingo Molnar <mingo@kernel.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Borislav Petkov <bp@suse.de>
Cc: David Ahern <dsahern@gmail.com>
Cc: Don Zickus <dzickus@redhat.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/n/tip-zq7tnh5462blt2yda0bcxh5b@git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
It was inconvenient that perf cannot be quit with SIGINT during
processing samples on TUI especially for large data files.
This was because the first argument of SLang_init_tty(), abort_char,
being 0. The manual says it's the ascii value of the control character
that will be used to generate the interrupt signal [1]. Passing -1
means to use the default value (Ctrl-C).
However, after processing samples, Ctrl-C was used to in other cases as
well - like stepping back from annotate. So recover the original
behavior after processing.
[1] http://jedsoft.org/slang/doc/html/cslang-6.html#ss6.1
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Tested-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: David Ahern <dsahern@gmail.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1432904024-13170-1-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently perf_evsel__hists_browse() function spins on a huge loop and
handles many key actions. Since it's hard to read and modify, let's
split it out into small helper functions.
The add_XXX_opt() functions are to register popup menu item on the
selected entry. When it adds an item, it also saves related data into
struct popup_action and returns 1 so that it can increase the number of
items (nr_options).
With this change, we can simplify the code just to call selected
callback function without considering various conditions. A callback
function named do_XXX is called with saved data when the item is
selected by user.
No functional change intended.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1429687101-4360-9-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently perf_evsel__hists_browse() function spins on a huge loop and
handles many key actions. Since it's hard to read and modify, let's
split it out into small helper functions.
This patch introduces do_XXX() functions which corresponds to each goto
label. This way we can call such functions both from key press actions
and popup menu actions.
No functional change intended.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: David Ahern <dsahern@gmail.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1429687101-4360-8-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
In perf hists browser, the fold/unfold stat of each hist entry is
recorded but hb->nr_callchain_rows loses its value after zoom out and
zoom in back. This causes a wrong row cursor range that restrict user to
move down anymore.
This bug can be reproduced as follows:
$ perf record -g -e syscalls:* ls
$ perf report
Available samples
================================================================
2 syscalls:sys_enter_mprotect <= [enter one of the entries]
2 syscalls:sys_exit_mprotect
13 syscalls:sys_enter_brk
...
In the hists brower, unfold some of the items, now the cursor can reach
to any rows:
Children Self Command Shared Object Symbol
================================================================
- 100.00% 100.00% ls libuClibc-0.9.33.2.so [.] lstat64
- lstat64
16.67% 0x6469702e64
8.33% 0x646970
8.33% 0x617461
8.33% 0x65
- 16.67% 0.00% ls [unknown] [.]0x6469702e64
0x6469702e64 <= [cursor can reach to bottom line, everything is ok]
Now, zoom back to "Available samples" and enter again:
Children Self Command Shared Object Symbol
================================================================
- 100.00% 100.00% ls libuClibc-0.9.33.2.so [.] lstat64
- lstat64
16.67% 0x6469702e64
8.33% 0x646970
8.33% 0x617461 <= [cursor may stop here, can't move down anymore]
8.33% 0x65
- 16.67% 0.00% ls [unknown] [.]0x6469702e64
0x6469702e64
This patch recalculates hb->nr_callchain_rows to fix the bug.
Signed-off-by: He Kuang <hekuang@huawei.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Wang Nan <wangnan0@huawei.com>
Link: http://lkml.kernel.org/r/1426144909-18951-1-git-send-email-hekuang@huawei.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When perf report on TUI shows callchain it checks first node has
siblings to determine whether it needs to print percentage value.
But it missed a case that first node is NULL. So sometimes it segfaults
like below:
$ perf top -g
perf: Segmentation fault
-------- backtrace --------
perf[0x4fcefb]
/usr/lib/libc.so.6(+0x33b20)[0x7f2a35839b20]
perf(rb_next+0x8)[0x47d3d8]
perf[0x4f6058]
perf[0x4f833b]
perf[0x4f8610]
perf[0x4f209e]
perf(ui_browser__run+0x3a)[0x4f2e6a]
perf[0x4f94ee]
perf(perf_evlist__tui_browse_hists+0x94)[0x4fbbf4]
perf[0x444d10]
/usr/lib/libpthread.so.0(+0x7314)[0x7f2a37070314]
/usr/lib/libc.so.6(clone+0x6d)[0x7f2a358ee5bd]
$ addr2line -e `which perf` 0x4f6058
/home/namhyung/project/linux/tools/perf/ui/browsers/hists.c:553
I don't know why the backtrace didn't print some symbols..
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Fixes: 4087d11cd9 ("perf hists browser: Print overhead percent value for first-level callchain")
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1419401076-21700-1-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Currently perf report on TUI doesn't print percent for first-level
callchain entry.
I guess it (wrongly) assumes that there's only a single callchain in the
first level.
This patch fixes it by handling the first level callchains same as
others - if it's not 100% it should print the percent value.
Also it'll affect other callchains in the other way around - if it's
100% (single callchain) it should not print the percentage.
Before:
- 30.95% 6.84% abc2 abc2 [.] a
- a
- 70.00% c
- 100.00% apic_timer_interrupt
smp_apic_timer_interrupt
local_apic_timer_interrupt
hrtimer_interrupt
...
+ 30.00% b
+ __libc_start_main
After:
- 30.95% 6.84% abc2 abc2 [.] a
- 77.90% a
- 70.00% c
- apic_timer_interrupt
smp_apic_timer_interrupt
local_apic_timer_interrupt
hrtimer_interrupt
...
+ 30.00% b
+ 22.10% __libc_start_main
Reported-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Cc: Andi Kleen <andi@firstfloor.org>
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: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1416816807-6495-1-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
With srcline key/sort'ing it's useful to have line numbers in the
annotate window. This patch implements this.
Use objdump -l to request the line numbers and save them in the line
structure. Then the browser displays them for source lines.
The line numbers are not displayed by default, but can be toggled on
with 'k'
There is one unfortunate problem with this setup. For lines not
containing source and which are outside functions objdump -l reports
line numbers off by a few: it always reports the first line number in
the next function even for lines that are outside the function.
I haven't found a nice way to detect/correct this. Probably objdump has
to be fixed.
See https://sourceware.org/bugzilla/show_bug.cgi?id=16433
The line numbers are still useful even with these problems, as most are
correct and the ones which are not are nearby.
v2: Fix help text. Handle (discriminator...) output in objdump.
Left align the line numbers.
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Link: http://lkml.kernel.org/r/1415844328-4884-9-git-send-email-andi@firstfloor.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
When 'perf top' is run, one can't easily find a difference
between -z option and normal output.
So I added a visual cue to know whether it is the zeroing or not.
Output is as below.
Before:
$ perf top
Samples: 61K of event 'cycles', Event count (approx.): 3908136933
Overhead Shared Object Symbol
1.42% firefox [.] 0x0000000000011e76
1.32% libpthread-2.17.so [.] pthread_mutex_lock
If you press key 'z' or run with zero option like '$ perf top --zero', it is as below.
After:
Samples: 61K of event 'cycles', Event count (approx.): 3908136933 [z]
Overhead Shared Object Symbol
1.42% firefox [.] 0x0000000000011e76
1.32% libpthread-2.17.so [.] pthread_mutex_lock
Signed-off-by: Taeung Song <treeze.taeung@gmail.com>
Acked-by: Namhyung Kim <namhyung@kernel.org>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Stephane Eranian <eranian@google.com>
Link: http://lkml.kernel.org/r/1412665995-26359-1-git-send-email-treeze.taeung@gmail.com
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
The hist_browser__show_callchain() and friends don't need to be that
complex. They're splitted in 3 pieces - one for traversing top-level
tree, other one for special casing first chains in the top-level
entries, and last one for recursive traversing inner trees. It led to
code duplication and unnecessary complexity IMHO.
Simplify the function and consolidate the logic into a single function
- it can recursively call itself. A little difference in printing
callchains in top-level tree can be handled with a small change.
It should have no functional change.
Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Acked-by: Jiri Olsa <jolsa@kernel.org>
Cc: Andi Kleen <andi@firstfloor.org>
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: Namhyung Kim <namhyung.kim@lge.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/1408583746-5540-2-git-send-email-namhyung@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
