Adding test to validate perf_event_attr data for command:
'record'
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
---
tools/perf/tests/attr/base-record
Adding test to validate perf_event_attr data for command:
'record -n'
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
---
tools/perf/tests/attr/test-record-no-samples | 6
Adding test to validate perf_event_attr data for command:
'record -F 100'
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
---
tools/perf/tests/attr/test-record-freq | 6
Adding test to validate perf_event_attr data for command:
'record --group -e cycles,instructions'
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
---
tools/perf/tests/attr/t
ython ./tests/attr.py -d ./tests/attr/ -p ./perf -v
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
---
tools/perf/Makefile | 1 +
tools/perf/perf.c| 2 +
tools/perf/p
On Tue, Oct 30, 2012 at 11:01:44PM +0100, Jiri Olsa wrote:
> Adding automated test to check event's perf_event_attr values.
>
> The idea is run perf session with kidnapping sys_perf_event_open
> function. For each sys_perf_event_open call we store the
> perf_event_attr da
On Tue, Aug 27, 2013 at 10:12:08AM -0600, David Ahern wrote:
> [Added Jiri who added the feature to perf]
>
> On 8/27/13 5:58 AM, Jean Pihet wrote:
> >Hi!
> >
> >I am looking for info and help on the libunwind API.
> >
> >What I am trying to achieve is to allow the callchain feature on perf
> >for
On Wed, Oct 02, 2013 at 10:18:32AM +0900, Namhyung Kim wrote:
> Hi Jiri,
>
> On Sun, 29 Sep 2013 15:52:06 +0900, Namhyung Kim wrote:
> > Hi Jiri,
> >
> > 2013-09-27 (금), 16:32 +0200, Jiri Olsa:
> >> We fail build with NO_DEMANGLE with missing -lbfd externals e
On Mon, Sep 30, 2013 at 06:42:10PM +0200, Ingo Molnar wrote:
>
> This series (with combo patch attached) implements (much) faster
> perf-tools feature-auto-detection.
>
> I used 3 tricks to implement feature auto-dependencies and to speed up
> feature detection:
>
> - standalone Makefile in
On Wed, Oct 02, 2013 at 04:36:45PM +0530, Ramkumar Ramachandra wrote:
> Hi,
>
> I tried using the `perf timechart` command after a `sudo perf
> timechart record -- git status`, and it segfaulted. Backtrace follows:
>
> Program received signal SIGSEGV, Segmentation fault.
> 0x in ?
On Wed, Oct 02, 2013 at 05:03:48PM +0530, Ramkumar Ramachandra wrote:
> Hi Jiri,
>
> Jiri Olsa wrote:
> > what perf version are you running?
>
> I'm running off the one in torvalds/linux.git.
>
> $ perf --version
> perf version 3.12.rc3.g34b22d5
>
> The
On Fri, Sep 06, 2013 at 01:37:01PM -0600, David Ahern wrote:
> When processing events the session code has an ordered samples queue which is
> used to time-sort events coming in across multiple mmaps. At a later point in
> time samples on the queue are flushed up to some timestamp at which point th
On Wed, Oct 02, 2013 at 02:18:21PM +0200, Jiri Olsa wrote:
> On Fri, Sep 06, 2013 at 01:37:01PM -0600, David Ahern wrote:
> > When processing events the session code has an ordered samples queue which
> > is
> > used to time-sort events coming in across multiple mmaps. At
hi,
got a segfault in the tsc test on latest acme's tree.
I'm dealing with some other issues right now, so just reporting ;-)
thanks,
jirka
(gdb) r test 20
Starting program: /home/jolsa/linux-perf/tools/perf/perf test 20
[Thread debugging using libthread_db enabled]
Using host libthread_db libra
;struct perf_mmap' is quite big now, maybe we
could use some lazy allocation for event_copy size.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: David Ahern
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Frederic Weisbecker
---
On Wed, Oct 02, 2013 at 04:46:59PM +0300, Adrian Hunter wrote:
> On 02/10/13 16:23, Jiri Olsa wrote:
> > hi,
> > got a segfault in the tsc test on latest acme's tree.
> >
> > I'm dealing with some other issues right now, so just reporting ;-)
>
> The c
On Mon, Oct 07, 2013 at 11:47:13AM +0530, Ramkumar Ramachandra wrote:
> Hi,
>
> `perf test` doesn't give me a clean run; it seems to be out-of-date. I
> spent many hours on tracing "parse events tests" with gdb, but
> preprocessor macros and other complexity make my job very difficult. I
> haven't
Removing 'fd, fd_pipe, filename, size' from struct perf_session
and replacing them with struct perf_data_file object.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho d
Adding perf_data_file__open interface to data object
to open the perf.data file for both read and write.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: David Ahern
aces using
perf_session are changed now.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: David Ahern
Cc: Adrian Hunter
Cc: Andi Kleen
---
tools/perf/builtin-annotate.c
threads.
This patchset is extracted and rebased as generic one
from my earlier 'multiple file support' patchset:
http://marc.info/?t=13780318941&r=1&w=2
It's reachable here:
git://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf.git
perf/file38
thanks,
jirka
Signed-off-b
On Mon, Jul 08, 2013 at 08:11:50AM -0400, Greg Price wrote:
> We use these macros in a few files in ui/gtk/ and they work fine
> because GTK includes GLib which defines them, but here in util/
> they break the build, at least on my machine. So stick to 0 and 1.
and here's another one, pick at you
On Wed, Jul 03, 2013 at 09:49:29AM +0200, Peter Zijlstra wrote:
> On Tue, Jul 02, 2013 at 11:15:59PM -0400, Vince Weaver wrote:
> >
> > OK, I have bisected this problem to the following change:
> >
> > commit 8dc85d547285668e509f86c177bcd4ea055bcaaf
> > Author: Peter Zijlstra
> > Date: Thu Sep
On Mon, Jul 08, 2013 at 07:57:46AM -0400, Greg Price wrote:
> On Sun, Jul 07, 2013 at 03:13:40PM +0200, Jiri Olsa wrote:
> > On Mon, Jul 01, 2013 at 10:28:45AM -0400, Greg Price wrote:
> > > For example, in an application with an expensive function
> > > implemented w
On Tue, Jul 09, 2013 at 05:44:11PM +0200, Jiri Olsa wrote:
> The '!ctx->is_active' check has a valid scenario, so
> there's no need for the WARN macro.
>
> The reason is that there's a time window between
> 'ctx->is_active' check in perf_event
984420] [] retint_signal+0x46/0x84
[ 324.984420] ---[ end trace 442ec2f04db3771a ]---
Reported-by: Vince Weaver
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Vince Weav
trace 827178d8a5966c3d ]---
Reported-by: Vince Weaver
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Vince Weaver
---
kernel/events/core.c | 2 +-
1 file changed, 1 inse
Removing event types data storing/reading to/from perf
data file as it's no longer needed. The only user of
this data 'perf timechart' is switched to use traceevent
library.
The event_types offset and size stays in the perf data
file header but are ignored from now on.
Signed-of
Removing event types data pushing from record command. It's
no longer needed, because this data is ignored.
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc:
ed
in perf_session__process_user_event function.
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc: David Ahern
Cc: Thomas Renninger
---
tools/perf/builtin-inject.c | 7
thanks for comments,
jirka
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc: David Ahern
Cc: Thomas Renninger
---
tools/perf/builtin-inject.c| 7 --
tools/pe
no longer compare strings but numbers.
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc: David Ahern
Cc: Thomas Renninger
---
tools/perf/builtin-timech
Removing match callback for auxiliary events processing
with following benefits:
- only one function pointer is passed along the way
- the match function is called within output function
and could be inlined by compiler
Suggested-by: Peter Zijlstra
Signed-off-by: Jiri Olsa
Cc: Corey
On Tue, Jul 09, 2013 at 06:48:55PM +0200, Jiri Olsa wrote:
> hi,
> following up on the 'perf timechart' FIXME note and changing
> its tracepoint match not to use event types data.
>
> In my tests the old and new timechart outputs look the same,
> so.. any other
On Wed, Jul 10, 2013 at 08:11:48AM +0200, Thomas Renninger wrote:
> On Tuesday, July 09, 2013 06:48:55 PM Jiri Olsa wrote:
> > hi,
> > following up on the 'perf timechart' FIXME note and changing
> > its tracepoint match not to use event types data.
>
> L
On Wed, Jun 26, 2013 at 04:14:09PM +0900, Namhyung Kim wrote:
> From: Namhyung Kim
>
> The -p (--pid) option enables to trace existing process by its pid.
hi,
I can't get any output from -p for live subcommand:
[jolsa@krava perf]$ pgrep yes
6443
[jolsa@krava perf]$ sudo ./perf ftrace live -p `p
On Wed, Jun 26, 2013 at 04:14:18PM +0900, Namhyung Kim wrote:
> From: Namhyung Kim
>
> Cc: Steven Rostedt
> Cc: Frederic Weisbecker
> Signed-off-by: Namhyung Kim
> ---
SNIP
> +OPTIONS
> +---
> +...::
> + Any command you can specify in a shell.
> +
> +-t::
> +--tracer=::
> +Th
On Wed, Jun 26, 2013 at 04:14:12PM +0900, Namhyung Kim wrote:
> From: Namhyung Kim
SNIP
> +
> +static int create_perf_directory(struct perf_ftrace *ftrace)
> +{
> + int err;
> + char buf[PATH_MAX];
> + struct stat statbuf;
> +
> + scnprintf(buf, sizeof(buf), "%s.dir", ftrace->dir
On Wed, Jun 26, 2013 at 04:14:13PM +0900, Namhyung Kim wrote:
> From: Namhyung Kim
>
> The ftrace show subcommand is for viewing recorded ftrace files. It
> enters perf.data.dir directory and open perf.header file to find out
> necessary information. And then read out per-cpu trace records usin
On Wed, Jun 26, 2013 at 04:14:14PM +0900, Namhyung Kim wrote:
SNIP
>
> +static int
> +__cmd_ftrace_report(struct perf_ftrace *ftrace, int argc, const char **argv)
> +{
> + int ret = -1;
> + const char * const report_usage[] = {
> + "perf ftrace report []",
> + NU
On Thu, Jul 11, 2013 at 05:18:37PM +0900, Namhyung Kim wrote:
> On Wed, 10 Jul 2013 16:22:19 +0200, Jiri Olsa wrote:
> > On Wed, Jun 26, 2013 at 04:14:18PM +0900, Namhyung Kim wrote:
> >> From: Namhyung Kim
> >>
> >> Cc: Steven Rostedt
> >> Cc: Fred
On Thu, Jul 11, 2013 at 03:16:45PM +0900, Namhyung Kim wrote:
> Hi Jiri,
>
> On Tue, 9 Jul 2013 18:48:56 +0200, Jiri Olsa wrote:
> > The only user of the event types data is 'perf timechart'
> > command and uses this info to identify proper tracepoints
> >
ed
in perf_session__process_user_event function.
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc: David Ahern
---
tools/perf/builtin-inject.c | 7
tools/perf/builtin-rec
using the perf_session__set_tracepoints_handlers function.
This way we get rid of the only event types user and can
remove them completely in next patches.
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Adding traceevent lib event-parse.h include to timechart
command and removing duplicated local 'enum trace_flag_type'
definition.
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weis
t command (patch 2)
- separating enum removal of 'enum trace_flag_type'
(patch 1)
Again, timechart (and perf.data) tests look ok, but would
appreciate another tester ;-).
It's reachable here:
git://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf.git
perf/et2
thanks,
jir
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc: David Ahern
---
tools/perf/util/header.c | 28 +---
tools/perf/util/header.h | 3 +--
2 f
Removing event types data pushing from record command. It's
no longer needed, because this data is ignored.
Signed-off-by: Jiri Olsa
Cc: Arnaldo Carvalho de Melo
Cc: Peter Zijlstra
Cc: Ingo Molnar
Cc: Paul Mackerras
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc:
The prev_state is defined as long which is 4 bytes long
on 32-bit x86. Changing the check against sizeof(long).
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
---
tools/perf/tests/evsel
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
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
On Wed, Sep 04, 2013 at 10:47:22AM -0600, David Ahern wrote:
> On 9/4/13 10:27 AM, Jiri Olsa wrote:
SNIP
> >index 1fc0c62..4df449c 100644
> >--- a/tools/perf/util/session.c
> >+++ b/tools/perf/util/session.c
> >@@ -495,7 +495,7 @@ static int perf_session_deliver_ev
On Wed, Sep 04, 2013 at 08:04:14PM +0200, Jean Pihet wrote:
> On ARM the debug info is not present in the .eh_frame sections but
> instead in .debug_frame.
> Use libunwind to load and parse the debug info.
hum, cannot make final link:
$ make LIBUNWIND_DIR=/opt/libunwind/
CHK -fstack-protector
On Thu, Sep 05, 2013 at 06:49:13PM +0200, Jean Pihet wrote:
> Hi Jiri,
>
> On 5 September 2013 18:30, Jiri Olsa wrote:
> > On Wed, Sep 04, 2013 at 08:04:14PM +0200, Jean Pihet wrote:
> >> On ARM the debug info is not present in the .eh_frame sections but
> >>
On Fri, Sep 06, 2013 at 11:31:17AM +0200, Jean Pihet wrote:
> Hi Jiri,
>
> On 5 September 2013 18:30, Jiri Olsa wrote:
> > On Wed, Sep 04, 2013 at 08:04:14PM +0200, Jean Pihet wrote:
> >> On ARM the debug info is not present in the .eh_frame sections but
> >>
On Mon, Sep 09, 2013 at 01:17:49PM +0200, Peter Zijlstra wrote:
> On Sun, Sep 01, 2013 at 12:36:11PM +0200, Jiri Olsa wrote:
> > hi,
> > sending the support for multiple file storage. Initial
> > RFC is here:
> > http://marc.info/?l=linux-kernel&m=13740838
On Mon, Sep 09, 2013 at 01:55:18PM +0200, Peter Zijlstra wrote:
> On Mon, Sep 09, 2013 at 01:36:41PM +0200, Jiri Olsa wrote:
> > On Mon, Sep 09, 2013 at 01:17:49PM +0200, Peter Zijlstra wrote:
> > > On Sun, Sep 01, 2013 at 12:36:11PM +0200, Jiri Olsa wrote:
> > This patc
On Mon, Sep 09, 2013 at 07:11:11AM -0700, David Ahern wrote:
> On 9/9/13 7:03 AM, Jiri Olsa wrote:
> >my usage currently is to having this running during the day:
> >(well whenever I remember to run it.. ;-) )
> >
> > [jolsa@krava perf]$ sudo ./perf record -g -M 1m -a
On Tue, Sep 10, 2013 at 05:24:16PM +0900, Namhyung Kim wrote:
> From: Namhyung Kim
>
> Current collapse stage has a scalability problem which can be
> reproduced easily with parallel kernel build. This is because it
> needs to traverse every children of callchain linearly during the
> collapse/m
On Thu, Sep 12, 2013 at 03:34:19PM +0900, Namhyung Kim wrote:
> Separate out GTK codes to a shared object called libperf-gtk.so. This
SNIP
> pthread_mutex_t ui__lock = PTHREAD_MUTEX_INITIALIZER;
> +void *perf_gtk_handle;
> +
> +#ifdef GTK2_SUPPORT
> +static int setup_gtk_browser(void)
> +{
> +
.eh_frame code is still functional, but
haven't tested the .debug_frame functionality itself
otherwise it looks ok:
Rewieved-by: Jiri Olsa
thanks,
jirka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
Mo
Removing unused trace-event-* code.
Signed-off-by: Jiri Olsa
Acked-by: Namhyung Kim
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: David Ahern
---
tools/perf/util/trace-event-parse.c | 36
Making page_size global from the util object.
Removing the not needed one.
Signed-off-by: Jiri Olsa
Acked-by: Namhyung Kim
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: David Ahern
pt/perf/libexec/perf-core': No such file or directory
make: *** [install-bin] Error 1
Fixing this by properly creating '/opt/perf/libexec/perf-core'
directory.
Signed-off-by: Jiri Olsa
Acked-by: Namhyung Kim
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyun
Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: David Ahern
---
Jiri Olsa (4):
perf tools: Fix 'make install prefix=...' build rule
perf tools: Remove unused trace-event-* code
perf tools: Unify page_size usage
perf tools: Make kmem work for non num
execution.
Signed-off-by: Jiri Olsa
Cc: Namhyung Kim
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
---
tools/perf/builtin-kmem.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/bu
On Mon, Oct 07, 2013 at 01:56:12PM +0200, Ingo Molnar wrote:
> Cc: Arnaldo Carvalho de Melo
> Cc: Peter Zijlstra
> Cc: Namhyung Kim
> Cc: David Ahern
> Cc: Jiri Olsa
> Link: http://lkml.kernel.org/n/tip-orhejqtjao3vf4wxwbudz...@git.kernel.org
> Signed-off-by: Ingo Molnar
On Mon, Oct 07, 2013 at 09:11:15AM -0600, David Ahern wrote:
> On 10/7/13 5:56 AM, Ingo Molnar wrote:
> >diff --git a/tools/perf/config/Makefile b/tools/perf/config/Makefile
> >index 77a2e6b..4f60ecc 100644
> >--- a/tools/perf/config/Makefile
> >+++ b/tools/perf/config/Makefile
> >@@ -121,6 +121,7
On Mon, Oct 07, 2013 at 01:56:24PM +0200, Ingo Molnar wrote:
SNIP
> +
> +#
> +# Print the result of the feature test:
> +#
> +define feature_print_code
> + ifeq ($(feature-$(1)), 1)
> +MSG := $(shell printf '...%30s: [ \033[32mon\033[m ]' $(1))
> + else
> +MSG := $(shell printf '...%30
On Mon, Oct 07, 2013 at 01:56:29PM +0200, Ingo Molnar wrote:
SNIP
> -FEATURE_TESTS = \
> - hello \
> - stackprotector-all \
> - stackprotector \
> - volatile-register-var \
> +#
> +# Not
On Mon, Oct 07, 2013 at 01:56:24PM +0200, Ingo Molnar wrote:
SNIP
> diff --git a/tools/perf/config/feature-checks/Makefile
> b/tools/perf/config/feature-checks/Makefile
> index c65bdac..b031669 100644
> --- a/tools/perf/config/feature-checks/Makefile
> +++ b/tools/perf/config/feature-checks/Make
On Mon, Oct 07, 2013 at 01:56:30PM +0200, Ingo Molnar wrote:
SNIP
> +#
> +# Only print out our features if we rebuilt the testcases or if a test
> failed:
> +#
> +ifeq ($(test-all-failed), 1)
> + $(foreach feat,$(CORE_FEATURE_TESTS) DUMMY,$(call feature_print,$(feat)))
> + $(info )
> +endif
f
On Mon, Oct 07, 2013 at 01:56:37PM +0200, Ingo Molnar wrote:
SNIP
> diff --git a/tools/perf/config/feature-checks/test-backtrace.c
> b/tools/perf/config/feature-checks/test-backtrace.c
> index 6a5fd96..5b33bcf 100644
> --- a/tools/perf/config/feature-checks/test-backtrace.c
> +++ b/tools/perf/co
On Mon, Oct 07, 2013 at 01:56:42PM +0200, Ingo Molnar wrote:
SNIP
> diff --git a/tools/perf/Documentation/Makefile
> b/tools/perf/Documentation/Makefile
> index eaa477f..86a4ba8 100644
> --- a/tools/perf/Documentation/Makefile
> +++ b/tools/perf/Documentation/Makefile
> @@ -1,3 +1,5 @@
> +$(info
On Tue, Oct 08, 2013 at 11:45:50AM +0300, Adrian Hunter wrote:
> objdump fails to annotate module symbols when looking
> at kcore. Workaround this by extracting object code
> from kcore and putting it in a temporary file for
> objdump to use instead. The temporary file is created
> to look like k
On Tue, Oct 08, 2013 at 10:46:10AM +0200, Ingo Molnar wrote:
SNIP
> So, this is a bit of a GNU Make mystery to me. If I have a failure for at
> least one of the features, and if I leave out that DUMMY then I get this
> printout:
>
> ... libperl: [ OFF ]
> ...
On Tue, Oct 08, 2013 at 12:11:02PM +0200, Ingo Molnar wrote:
SNIP
> diff --git a/tools/perf/config/feature-checks/Makefile
> b/tools/perf/config/feature-checks/Makefile
> index c65bdac..b031669 100644
> --- a/tools/perf/config/feature-checks/Makefile
> +++ b/tools/perf/config/feature-checks/Make
On Wed, Oct 09, 2013 at 08:42:05AM +0200, Ingo Molnar wrote:
SNIP
> >
> > I think the issue might be in the eval handling the $(info $(MSG)).
> > With following change on top of your v2 patchset it works for me.
>
> Indeed - and that works here too. Initially I was trying to solve it by
> forc
On Wed, Oct 09, 2013 at 10:33:25AM +0300, Adrian Hunter wrote:
> On 08/10/13 17:02, Jiri Olsa wrote:
> > On Tue, Oct 08, 2013 at 11:45:50AM +0300, Adrian Hunter wrote:
> >> objdump fails to annotate module symbols when looking
> >> at kcore. Workaround this by extra
On Wed, Oct 09, 2013 at 01:38:04PM +0300, Adrian Hunter wrote:
> On 09/10/13 13:12, Jiri Olsa wrote:
> > On Wed, Oct 09, 2013 at 10:33:25AM +0300, Adrian Hunter wrote:
> >> On 08/10/13 17:02, Jiri Olsa wrote:
> >>> On Tue, Oct 08, 2013 at 11:45:50AM +0300, Adrian Hu
On Tue, Oct 08, 2013 at 09:26:53PM -0600, David Ahern wrote:
> When enabled dumps a summary of all syscalls by task with the usual
> statistics -- min, max, average and relative stddev. For example,
>
> make - 26341 : 3344 [ 17.4% ] 0.000 ms
>
> read : 520.000
On Sun, Sep 22, 2013 at 08:05:59PM -0600, David Ahern wrote:
SNIP
> +
> + if (size > remaining) {
> + memcpy(rec->mmap_addr + rec->mmap_offset, buf, remaining);
> + rec->bytes_written += remaining;
> +
> + size -= remaining;
> + buf += remainin
On Wed, Oct 09, 2013 at 11:00:38PM -0300, Felipe Pena wrote:
> Fix for a memory leak on test_file() function in dso-data.c
oops, thanks ;-)
Acked-by: Jiri Olsa
>
> Signed-off-by: Felipe Pena
> ---
> tools/perf/tests/dso-data.c |1 +
> 1 file changed, 1 insertion(+)
Arnaldo, David,
wrt to our IRC discussion..
attached patch fixies the static build David is using:
make O=/tmp/perf LDFLAGS=-static DEBUG=1 -j 4
I still got some warning though:
LINK /tmp/perf/perf
/tmp/perf/libperf.a(target.o): In function `perf_target__parse_uid':
/home/jolsa/kernel.org/l
On Thu, Aug 29, 2013 at 07:15:00PM +0200, Jean Pihet wrote:
SNIP
>
> The problem is that it is slow as hell (couple of minutes to report
> the callchain from a simple test app ;-|). I will look at the sampling
> rate and the caching next.
FYI there's elfutils alternative grewing now via libdw p
On Fri, Aug 30, 2013 at 01:49:03PM +0200, Jiri Olsa wrote:
> On Thu, Aug 29, 2013 at 07:15:00PM +0200, Jean Pihet wrote:
>
> SNIP
>
> >
> > The problem is that it is slow as hell (couple of minutes to report
> > the callchain from a simple test app ;-|). I will loo
for both endianity
- added record '-M time' support
It's reachable here:
git://git.kernel.org/pub/scm/linux/kernel/git/jolsa/perf.git
perf/core_file
thanks,
jirka
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Introducing swap_features function to make the swapping
code more clear and extensible.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Cc: David Ahern
as
of now, like:
$ perf record -m 8 ls
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Cc: David Ahern
---
tools/perf/Documentation/perf-kvm.txt| 4
Adding perf data version 3 header swap same way
as it's done for v2.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Cc: David Ahern
---
tools/perf
Adding session reading test for little endian perf data
version 3 included as binary data in the test object.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi
No need to use post_processing_offset variable now when
we know where the data starts.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Cc: David Ahern
Moving synthesizing into single function, so it
could be reused.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Cc: David Ahern
---
tools/perf/builtin
Adding perf data version 3 header write code and
switch perf tool storing to version 3.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Cc: David Ahern
Change the target so the sythesizing routine would think
it needs to synthesized whole process, because there's no
exec COMM event for subsequent data files.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc:
Introducing swap_header function to make the swapping
code more clear and extensible.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Cc: David Ahern
read_lock
0.05% +0.01% [kernel.kallsyms] [k]
__srcu_read_unlock
...
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi
Adding session reading test for big endian perf data
version 3 included as binary data in the test object.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Adding perf_data_file__open interface to data object
to open the perf.data file for both read and write.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Removing 'fd, fd_pipe, filename, size' from struct perf_session
and replacing it with struct perf_data_file.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Adding perf data version 3 header read code to load
data for v3 format.
Signed-off-by: Jiri Olsa
Cc: Corey Ashford
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Andi Kleen
Cc: David Ahern
---
tools/perf
201 - 300 of 11474 matches
Mail list logo