On 6/3/14, 4:05 PM, Alexis Berlemont wrote:
From: Jiri Olsa
Fixing several sources config dependencies to allow
separate config builds.
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
C
On 6/3/14, 9:35 AM, Jiri Olsa wrote:
+ static char buf_templ[sizeof(TEMPL)];
why static? and ...
I need to use it outside the test_file function
Oh, it's a return variable.
David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
On 6/3/14, 8:02 AM, Prankul Garg wrote:
Qualcomm MSM 8974 is not able to detect [kernel.kallsyms] dso either with
Hardware or Tracepoint events while Exynos do.
Add -v to perf-record and perf-report.
And try:
perf report -D | grep kallsyms
Do you see something like:
0 0 0xd8 [0x58]: PERF_RECO
On 6/2/14, 3:18 PM, Jiri Olsa wrote:
Testing that perf properly closes opened dso objects
and tries to reopen in case we run out of allowed file
descriptors for dso data.
Reviewed by: David Ahern
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the
On 6/2/14, 3:18 PM, Jiri Olsa wrote:
+static int dsos__create(int cnt, int size)
+{
+ int i;
+
+ dsos = malloc(sizeof(dsos) * cnt);
+ TEST_ASSERT_VAL("failed to alloc dsos array", dsos);
+
+ for (i = 0; i < cnt; i++) {
+ char *file;
+
+ file = t
On 6/2/14, 3:18 PM, Jiri Olsa wrote:
Making the test_file function to be reusable for
new tests coming in following patches.
Cc: Arnaldo Carvalho de Melo
Cc: Corey Ashford
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Jean Pihet
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc
On 6/2/14, 3:18 PM, Jiri Olsa wrote:
In upcoming tests we will setup process limits, which
might affect other tests. Spawning child for each test
to prevent this.
Reviewed-by: David Ahern
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a
On 6/2/14, 3:18 PM, Jiri Olsa wrote:
Adding descriptions/explanations for dso__data_* interface
functions.
Shouldn't the descriptions be added when the functions are?
David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.ke
On 6/2/14, 3:18 PM, Jiri Olsa wrote:
+static rlim_t get_fd_limit(void)
+{
+ struct rlimit l;
+ rlim_t limit = 0;
+
+ /* Allow half of the current open fd limit. */
+ if (getrlimit(RLIMIT_NOFILE, &l) == 0) {
+ if (l.rlim_cur == RLIM_INFINITY)
+
On 6/2/14, 3:18 PM, Jiri Olsa wrote:
We keep dsos data file descriptors open until their count reaches
the half of the current fd open limit (RLIMIT_NOFILE). In this case
we close file descriptor of the first opened dso object.
We've got overall speedup (24% for my workload) of report:
'perf r
On 5/16/14, 5:14 PM, Stephane Eranian wrote:
From: Michael Lentine
This patch adds a fallback to cat for the pager. This is useful
on environmnents, such as Android, where less does not exist.
It is better to default to cat than to abort.
Reviewed-by: Stephane Eranian
Signed-off-by: Michael L
On 5/13/14, 4:02 PM, Alexis Berlemont wrote:
Hello,
A few months ago, I tried to make a proposal to introduce Kconfig in
perf's generation procedure.
(cf. https://lkml.org/lkml/2013/12/20/511)
I started from David Ahern's work; I was not aware that Jiri Olsa
pushed further the idea and proposed
bug when each thread has different name, it only reports a
main thread for samples in other threads.
Cc: Adrian Hunter
Cc: David Ahern
Cc: Stephane Eranian
Signed-off-by: Namhyung Kim
Acked-by: David Ahern
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel&qu
On 5/5/14, 12:24 PM, Jiri Olsa wrote:
On Mon, May 05, 2014 at 04:05:55PM +0900, Dongsheng Yang wrote:
From: Dongsheng
In output of perf sched map, any shortname of thread will be explained
at the first time when it appear.
Example:
*A0 228836.978985 secs A0 => perf:23032
On 5/5/14, 4:27 AM, Christian Borntraeger wrote:
diff --git a/tools/perf/builtin-kvm.c b/tools/perf/builtin-kvm.c
index 922706c..806c0e4 100644
--- a/tools/perf/builtin-kvm.c
+++ b/tools/perf/builtin-kvm.c
@@ -75,7 +75,7 @@ struct kvm_events_ops {
bool (*is_end_event)(struct perf_evsel *e
perf tools changes look good to me
Review feedback is welcome.
CC-ing David Ahern
I don't have the original emails, but looking at
https://lkml.org/lkml/2014/4/25/331
[PATCH 01/11] s390: add sie exit reasons tables
[PATCH 02/11] KVM: s390: Use trace tables from sie.h
[PATCH 03/11] KVM: s390:
sts/make make_static_O
- make_static_O: cd . && make -f Makefile O=/tmp/tmp.Ay6r3wEmtX
DESTDIR=/tmp/tmp.vK0KQwO0Vi LDFLAGS=-static
Cc: Arnaldo Carvalho de Melo
Cc: Corey Ashford
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter
On 4/28/14, 9:17 AM, Jiri Olsa wrote:
On Mon, Apr 28, 2014 at 09:01:57AM -0600, David Ahern wrote:
On 4/28/14, 3:22 AM, Jiri Olsa wrote:
From: Jiri Olsa
There's false assumption in the library detection code
assuming -liberty and -lz are always present once bfd
is detected. The fai
hford
Cc: David Ahern
Cc: Frederic Weisbecker
Cc: Ingo Molnar
Cc: Namhyung Kim
Cc: Paul Mackerras
Cc: Peter Zijlstra
Signed-off-by: Jiri Olsa
---
tools/perf/config/Makefile | 34 +++---
1 file changed, 23 insertions(+), 11 deletions(-)
diff --git a/tools
On 4/28/14, 7:16 AM, Namhyung Kim wrote:
2014-04-28 (월), 12:01 +0200, Jiri Olsa:
On Sun, Apr 27, 2014 at 11:36:35PM +0900, Namhyung Kim wrote:
2014-04-17 (목), 19:39 +0200, Jiri Olsa:
Keeping the data file description open for the whole life
of the dso object.
I suspect there might be an issu
On 4/23/14, 10:31 AM, Jiri Olsa wrote:
thoughts?;-) thanks,
Only that libperf is still a pipe dream. Don't hold up progress for
this, but we should be thinking about making it thread safe.
David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a mess
On 4/22/14, 1:05 PM, Stephane Eranian wrote:
I intend to contribute that code for perf once it is finalized.
Note that it uses a module developed by Sonny Rao to
export the perf timestamp time source via a posix-clock.
This clock discussion has been going on for a while and
never reached a conclu
ree_terms(&list);
return ret;
}
- list_add_tail(&clone->list, &list);
+ list_add_tail(&cloned->list, &list);
}
list_splice(&list, terms);
return 0;
Acked-by: David Ahern
--
On 4/15/14, 3:56 AM, Jiri Olsa wrote:
hum.. I think using atomic type is not enough, we'd need
to make map_groups__put/get atomic as well
not sure what's the support in user space for that.. will check
also not to be negative, but libperf is not thread safe anyway, right? ;-)
should be. pret
On 4/14/14, 9:52 AM, Jiri Olsa wrote:
From: Arnaldo Carvalho de Melo
We will share it among threads in the same process.
Signed-off-by: Arnaldo Carvalho de Melo
Cc: Adrian Hunter
Cc: David Ahern
Cc: Don Zickus
Cc: Frederic Weisbecker
Cc: Mike Galbraith
Cc: Namhyung Kim
Cc: Paul
et_modules_path(struct machine
*machine)
if (!version)
return -1;
- snprintf(modules_path, sizeof(modules_path), "%s/lib/modules/%s/kernel",
+ snprintf(modules_path, sizeof(modules_path), "%s/lib/modules/%s",
machine->root_
y;
+ init_stats(&event->total.stats);
return event;
}
Reviewed-by: David Ahern
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
On 3/26/14, 12:57 PM, Josef Bacik wrote:
We are trying to debug a problem internally and noticed that if we use perf
record -g -e and then try to use perf script to get the events and
their backtraces that we weren't getting the backtraces, even though they were
collected and would show up with
On 3/18/14, 7:30 PM, Namhyung Kim wrote:
Hi David,
On Tue, Mar 18, 2014 at 3:07 PM, David Ahern wrote:
On 3/18/14, 12:32 AM, Namhyung Kim wrote:
This is a patchset implementing -F/--field option to setup output
field/column as Ingo requested. It depends on my perf/percentage
patchset (with
On 3/18/14, 12:32 AM, Namhyung Kim wrote:
This is a patchset implementing -F/--field option to setup output
field/column as Ingo requested. It depends on my perf/percentage
patchset (with minor updates) [1].
perf-report should be consistent with perf-script which uses --fields.
David
--
To un
-list-cmds for use by scripts
perf mem: introduce --list-cmds for use by scripts
perf lock: introduce --list-cmds for use by scripts
perf sched: introduce --list-cmds for use by scripts
All 4 look good to me.
Acked-by: David Ahern
--
To unsubscribe from this list: send the line "un
On 3/14/14, 8:27 PM, Andi Kleen wrote:
On Fri, Mar 14, 2014 at 04:41:31PM -0600, David Ahern wrote:
On 3/14/14, 3:31 PM, Andi Kleen wrote:
@@ -374,6 +376,8 @@ LIB_OBJS += $(OUTPUT)util/stat.o
LIB_OBJS += $(OUTPUT)util/record.o
LIB_OBJS += $(OUTPUT)util/srcline.o
LIB_OBJS += $(OUTPUT)util
On 3/14/14, 3:31 PM, Andi Kleen wrote:
@@ -374,6 +376,8 @@ LIB_OBJS += $(OUTPUT)util/stat.o
LIB_OBJS += $(OUTPUT)util/record.o
LIB_OBJS += $(OUTPUT)util/srcline.o
LIB_OBJS += $(OUTPUT)util/data.o
+LIB_OBJS += $(OUTPUT)util/jsmn.o
+LIB_OBJS += $(OUTPUT)util/json.o
LIB_OBJS += $(OUTPUT)ui/
On 3/14/14, 10:17 AM, Andi Kleen wrote:
The Intel ISR section for RDMSR seems to say: "Specifying a reserved
or unimplemented
MSR address in ECX will also cause a general protection exception".
From a guest's perspective, MSR_RAPL_POWER_UNIT is unimplemented; kvm matches
this behavior.
MSRs a
On 3/3/14, 6:26 PM, Ramkumar Ramachandra wrote:
Introduce
$ perf kvm --list-cmds
to dump a raw list of commands for use by the completion script. In
order to do this, introduce parse_options_subcommand() for handling
subcommands as a special case in the parse-options machinery.
Cc: David
On 2/28/14, 2:29 PM, Peter Zijlstra wrote:
Hurm; didn't I suggest using -fno-strict-aliasing just like the kernel
does? Because the C aliasing rules are bonghits heavy?
you, and Ingo in 2009 -- 65014ab3
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of
On 2/26/14, 1:53 PM, Andi Kleen wrote:
Is there some reason not to enable frame pointers?
It makes code slower.
Sure there is some overhead because of the push, mov, pop instructions
per function. But, take for example the simple program below. Compile
with and without frame pointers
gcc
On 2/26/14, 12:25 PM, Andy Lutomirski wrote:
On Wed, Feb 26, 2014 at 11:19 AM, David Ahern wrote:
On 2/26/14, 11:59 AM, Andy Lutomirski wrote:
I wonder if anyone who uses perf for userspace profiling *ever* uses
FP and gets away with it. There's precious little userspace software
com
On 2/26/14, 11:59 AM, Andy Lutomirski wrote:
I wonder if anyone who uses perf for userspace profiling *ever* uses
FP and gets away with it. There's precious little userspace software
compiled with frame pointers these days on most architectures.
yes and yes. With control over the entire stack
ME_CHAR(tty) ? MIN_CHAR(tty) : 1;
+ int amt = poll && !TIME_CHAR(tty) && MIN_CHAR(tty) ? MIN_CHAR(tty) : 1;
if (ldata->icanon && !L_EXTPROC(tty)) {
if (ldata->canon_head != ldata->read_tail)
Worked for me.
Tested-by: David Ahern
On 2/6/14, 8:24 AM, Stephane Eranian wrote:
On Thu, Feb 6, 2014 at 3:09 PM, Arnaldo Carvalho de Melo
wrote:
Em Thu, Feb 06, 2014 at 12:07:05PM -0200, Arnaldo Carvalho de Melo escreveu:
Em Thu, Feb 06, 2014 at 02:58:01PM +0100, Stephane Eranian escreveu:
On Thu, Feb 6, 2014 at 2:56 PM, Arnaldo
On 1/9/14, 8:19 AM, Frederic Weisbecker wrote:
On Wed, Jan 08, 2014 at 02:48:37PM -0700, David Ahern wrote:
The existing code does not work. Your unstable tsc patch did not
work. I have not tried Joseph's patch. Are you proposing that one or
do you have something else in mind?
I thi
On 1/11/14, 12:42 PM, Andi Kleen wrote:
From: Andi Kleen
opensuse libbfd requires -lz -liberty to build. Add those
to the BFD feature detection.
Fixes static builds on Fedora as well.
David
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message t
On 1/9/14, 7:36 AM, Dongsheng Yang wrote:
On 01/08/2014 04:44 PM, David Ahern wrote:
On 1/8/14, 8:20 AM, Arnaldo Carvalho de Melo wrote:
Em Mon, Dec 30, 2013 at 10:26:41AM -0500, Dongsheng Yang escreveu:
>Ramkumar reported that perf list --raw-dump was broken by 44d742e.
>Fix by maki
On 1/4/14, 8:05 AM, Frederic Weisbecker wrote:
On Fri, Jan 03, 2014 at 03:45:36PM -0700, David Ahern wrote:
On 1/3/14, 3:07 PM, Frederic Weisbecker wrote:
I'm not sure I understand why we need that. Why doesn't it work by simply
flushing
events prior to the earliest timestamp among
On 1/8/14, 8:20 AM, Arnaldo Carvalho de Melo wrote:
Em Mon, Dec 30, 2013 at 10:26:41AM -0500, Dongsheng Yang escreveu:
>Ramkumar reported that perf list --raw-dump was broken by 44d742e.
>Fix by making raw-dump a proper argument.
>
>Signed-off-by: David Ahern
Are you sure David S
On 1/3/14, 3:07 PM, Frederic Weisbecker wrote:
On Wed, Jan 01, 2014 at 11:37:55AM -0700, David Ahern wrote:
On 12/26/13, 8:30 AM, Frederic Weisbecker wrote:
On Thu, Dec 26, 2013 at 10:24:03AM -0500, David Ahern wrote:
On 12/26/13, 10:14 AM, Frederic Weisbecker wrote:
I was carrying that
On 12/26/13, 8:30 AM, Frederic Weisbecker wrote:
On Thu, Dec 26, 2013 at 10:24:03AM -0500, David Ahern wrote:
On 12/26/13, 10:14 AM, Frederic Weisbecker wrote:
I was carrying that patch while working on perf-kvm-stat-live last
Fall. It does not solve the problem for live commands, so ended up
On 12/27/13, 1:14 AM, Masami Hiramatsu wrote:
Nowhere, since there is no terminal code for user
command interface.
Those memories are released when the program terminated.
I think it is just a waste of the time to free the memory
pieces which are not used(and allocated) repeatedly.
Or, is there
On 12/26/13, 10:14 AM, Frederic Weisbecker wrote:
I was carrying that patch while working on perf-kvm-stat-live last
Fall. It does not solve the problem for live commands, so ended up
dropping it and going with local (to the command) hacks. I still
think for live commands getting a perf_clock tim
On 12/26/13, 12:38 AM, Namhyung Kim wrote:
diff --git a/tools/perf/util/log.c b/tools/perf/util/log.c
new file mode 100644
index ..3838d49f82de
--- /dev/null
+++ b/tools/perf/util/log.c
@@ -0,0 +1,105 @@
+#include
+#include
+#include "util/debug.h"
+
+#define LINEMAP_GROW 128
+
+st
On 12/26/13, 12:41 AM, Masami Hiramatsu wrote:
And this shows the available variables at the given line of
the function.
# ./perf probe -x perf --vars map__load:8
Available variables at map__load:8
@
char* name
struct map* map
On 12/26/13, 9:15 AM, Arnaldo Carvalho de Melo wrote:
Right, but I can't apply that patch, as it makes 'perf stat
whatever-workload' to fail, as I realized when doing a demo to someone
interested in using perf ;-\
So for now I'm not applying that one.
right, so you want one with < 0 check or w
On 12/26/13, 12:41 AM, Masami Hiramatsu wrote:
@@ -180,7 +181,20 @@ static int opt_set_target(const struct option *opt, const
char *str,
else
return ret;
- params.target = str;
+ /* Expand given path to absolute path, except fo
On 12/24/13, 8:30 AM, Arnaldo Carvalho de Melo wrote:
Em Tue, Dec 24, 2013 at 09:53:42AM -0300, Arnaldo Carvalho de Melo escreveu:
>The thing to check is perf_evlist__{prepare,start}_workload notification
>errors using SIGUSR1, that we need to check for in the caller, and emit
>the message, no?
Masami:
If you do not have any problems with this version please Ack so Arnaldo
will pick it up.
Thanks,
David
On 12/20/13, 1:04 PM, David Ahern wrote:
When requesting a function from a userspace library the error message to
the user is less than helpful. e.g.,
perf probe -x /lib64
On 12/23/13, 2:37 PM, Arnaldo Carvalho de Melo wrote:
int perf_evlist__start_workload(struct perf_evlist *evlist)
{
if (evlist->workload.cork_fd > 0) {
char bf = 0;
int ret;
/*
* Remove the cork, let it rip!
On 12/23/13, 8:10 AM, Frederic Weisbecker wrote:
On Fri, Dec 20, 2013 at 10:09:53AM -0700, David Ahern wrote:
On 12/20/13, 5:27 AM, Joseph Schuchart wrote:
I know this comes late, but: As far as I can see, your change does not
preserve the logic of the code I suggested. The idea was to first
iki (https://perf.wiki.kernel.org/index.php/Todo).
I tried to continue the work started by David Ahern
(https://github.com/dsahern/linux.git branch:perf-config). This
version relies on the features detection tests available in
linux/tools/perf/config/feature-checks so as to enable some
configuration options (a
libpthread-2.14.90.so -a
'lock_full=__pthread_mutex_lock_full'
no symbols found in /lib64/libpthread-2.14.90.so, maybe install a debug
package?
Failed to find function in /lib64/libpthread-2.14.90.so. Perhaps it is a
local variable?
Error: Failed to add events. (-22)
Signed-off
On 12/20/13, 5:27 AM, Joseph Schuchart wrote:
I know this comes late, but: As far as I can see, your change does not
preserve the logic of the code I suggested. The idea was to first gather
all the maximum timestamps of all cpus (that is, the last timestamp seen
on each cpu) and then determine th
On 12/20/13, 1:52 PM, Dongsheng Yang wrote:
Hi Arnaldo:
Please help to review this patchset.
Changelog:
- v2:
* splite the first commit.
At best you are creating knowledge of running tasks in the VMs when the
perf session is started. That's it. That informatio
perf-stat should not try
to print stats.
With this change:
$ perf stat -v -- foo
Failed to start workload
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Stephane Eranian
---
tools/perf/builtin-stat.c | 11 +--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/tools/pe
lsyms] [g] __switch_to
7.91%:6471 [guest.kernel.kallsyms] [g] scheduler_tick
1.75%:6471 [guest.kernel.kallsyms] [g] native_apic_mem_write
0.21%:6471 [guest.kernel.kallsyms] [g] apic_timer_interrupt
Signed-off-by: Dongsheng Yang
Acked-by: D
] [unknown] [g] 0x8164f330
This bug was introduced by 207b57926 (perf kvm: Fix regression with guest
machine creation).
commit 207b5792696206663a38e525b9793644895bad3b
Author: David Ahern
Date: Sun Jul 1 16:11:37 2012 -0600
perf kvm: Fix regression with guest machine creation
On 12/19/13, 8:34 AM, Ingo Molnar wrote:
Btw., having to specify a '--guestmount /tmp/guestmount/' just to
achieve a natural feature (proper guest symbol resolution) is a poor
user interface, obviously.
Is there perhaps a predictable pattern as to where qemu (or libvirt)
puts a user's guest mou
On 12/17/13, 7:15 AM, Arnaldo Carvalho de Melo wrote:
Just tried it here, and on fedora 18 one doesn't even needs the
--color=always,
Default aliases:
$ alias
alias egrep='egrep --color=auto'
alias fgrep='fgrep --color=auto'
alias grep='grep --color=auto'
--
To unsubscribe from this list: sen
Commit-ID: 8d00be815c05ed0f0202f606bab4e54f98fd3b30
Gitweb: http://git.kernel.org/tip/8d00be815c05ed0f0202f606bab4e54f98fd3b30
Author: David Ahern
AuthorDate: Tue, 10 Dec 2013 21:35:38 -0700
Committer: Arnaldo Carvalho de Melo
CommitDate: Fri, 13 Dec 2013 10:30:21 -0300
perf tools
On 12/11/13, 5:36 AM, Alexander Shishkin wrote:
diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c
index a0c7c59..80817ec 100644
--- a/tools/perf/util/dso.c
+++ b/tools/perf/util/dso.c
@@ -446,6 +446,7 @@ struct dso *dso__new(const char *name)
dso->cache = RB_ROOT;
On 12/12/13, 10:26 AM, Ramkumar Ramachandra wrote:
Introduce
$ perf kvm --list-cmds
to dump a raw list of commands for use by the completion script. While
at it, refactor kvm_usage so that there's only one copy of the command
listing.
Cc: David Ahern
Cc: Arnaldo Carvalho de Melo
S
On 12/12/13, 12:05 PM, Arnaldo Carvalho de Melo wrote:
Well, if we can pass somehow the magic number of an executable mmap
in the PERF_RECORD_MMAP2 record, we would be able, together with the
data we already have in the perf.data header (uname in a live session),
to figure that out, no?
Sure,
On 12/12/13, 12:34 AM, Ramkumar Ramachandra wrote:
David Ahern wrote:
Why not make raw_dump a proper argument?
Sure, that'd work too. I was thinking of a minimal way to fix the
problem myself.
diff --git a/tools/perf/builtin-list.c b/tools/perf/builtin-list.c
index 011195e
On 12/12/13, 2:09 AM, Ramkumar Ramachandra wrote:
David Ahern wrote:
I don't like the command list being embedded in the completion script. It
will get outdated quickly.
I could introduce a
$ perf kvm --list-cmds
but that would mean a
printf("record report stat ..")
i
On 12/12/13, 5:05 AM, Adrian Hunter wrote:
diff --git a/tools/perf/util/dso.h b/tools/perf/util/dso.h
index 384f2d9..62680e1 100644
--- a/tools/perf/util/dso.h
+++ b/tools/perf/util/dso.h
@@ -91,6 +91,7 @@ struct dso {
u8 annotate_warned:1;
u8 sname_alloc:1;
On 12/11/13, 2:18 PM, Andi Kleen wrote:
David Ahern writes:
Are you looking up the current or next task? If the former why not use
sample->pid rather than parsing the sched_switch tracepoint?
The itrace stream doesn't have a pid field, and it needs the exact
time stamp of the switc
Mon Sep 17 00:00:00 2001
From: David Ahern
Date: Wed, 11 Dec 2013 14:00:20 -0700
Subject: [PATCH] perf list: Fix raw-dump arg
Ramkumar reported that perf list --raw-dump was broken by 44d742e.
Fix by making raw-dump a proper argument.
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Arnaldo Carva
On 12/11/13, 12:55 PM, Arnaldo Carvalho de Melo wrote:
Em Wed, Dec 11, 2013 at 12:30:40PM -0700, David Ahern escreveu:
On 12/11/13, 5:36 AM, Alexander Shishkin wrote:
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 55f3608..52fbfb6 100644
--- a/tools/perf/util
On 12/11/13, 12:50 PM, Arnaldo Carvalho de Melo wrote:
Em Wed, Dec 11, 2013 at 04:04:15PM +0530, Ramkumar Ramachandra escreveu:
Currently, there is no way to enumerate the subcommands under 'perf
kvm', so hardcode them.
[acme@ssdandy linux]$ perf kvm -h 2>&1 | grep usage
usage: perf kvm [] {
On 12/11/13, 5:37 AM, Alexander Shishkin wrote:
From: Adrian Hunter
If a file contains Instruction Tracing data then always allow
fields 'addr' and 'cpu' to be selected as options for perf
script. This is necessary because Instruction Trace decoding
may synthesize events with that information.
On 12/11/13, 5:36 AM, Alexander Shishkin wrote:
From: Adrian Hunter
Add an option to cause a selected event
to be opened always without a pid when
configured by perf_evsel__config().
This is needed when using the sched_switch
tracepoint to follow object code execution.
sched_switch occurs befo
On 12/11/13, 5:36 AM, Alexander Shishkin wrote:
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index 55f3608..52fbfb6 100644
--- a/tools/perf/util/machine.c
+++ b/tools/perf/util/machine.c
@@ -1412,3 +1412,29 @@ pid_t machine__get_thread_pid(struct machine *machine,
pid_t ti
On 12/11/13, 5:36 AM, Alexander Shishkin wrote:
From: Adrian Hunter
Add a function to get the pid from the tid.
This is needed when using the sched_switch
tracepoint to follow object code execution.
sched_switch identifies the thread but, to
find the process mmaps, we need the process
pid.
A
On 12/11/13, 5:36 AM, Alexander Shishkin wrote:
diff --git a/tools/perf/util/dso.c b/tools/perf/util/dso.c
index a0c7c59..80817ec 100644
--- a/tools/perf/util/dso.c
+++ b/tools/perf/util/dso.c
@@ -446,6 +446,7 @@ struct dso *dso__new(const char *name)
dso->cache = RB_ROOT;
Commit-ID: 3160565f0e005d2ec736ae25cf0a79988c0cbe71
Gitweb: http://git.kernel.org/tip/3160565f0e005d2ec736ae25cf0a79988c0cbe71
Author: David Ahern
AuthorDate: Wed, 4 Dec 2013 19:41:41 -0700
Committer: Arnaldo Carvalho de Melo
CommitDate: Thu, 5 Dec 2013 10:09:58 -0300
perf trace: Fix
Commit-ID: 9aca7f1792c5d2d5d367bbe5cfe204fe40517929
Gitweb: http://git.kernel.org/tip/9aca7f1792c5d2d5d367bbe5cfe204fe40517929
Author: David Ahern
AuthorDate: Wed, 4 Dec 2013 19:41:39 -0700
Committer: Arnaldo Carvalho de Melo
CommitDate: Thu, 5 Dec 2013 10:05:45 -0300
perf trace: Add
Commit 1902efe7f for the new comm infra added the wrong check for return
code on thread__set_comm. err == 0 is normal, so don't return at that
point unless err != 0.
Signed-off-by: David Ahern
Cc: Frederic Weisbecker
---
tools/perf/util/thread.c | 2 +-
1 file changed, 1 insertion(
On 12/11/13, 9:30 AM, Dongsheng Yang wrote:
@@ -1001,8 +1002,28 @@ static int data_init(int argc, const char **argv)
use_default = false;
}
} else if (perf_guest) {
- defaults[0] = "perf.data.host";
- defaults[1] = "perf.
On 12/9/13, 9:14 AM, Borislav Petkov wrote:
From: Borislav Petkov
Move debugfs.* to api/fs/. We have a common tools/lib/api/ place
where the Makefile lives and then we place the headers in subdirs.
For example, all the fs-related stuff goes to tools/lib/api/fs/ from
which we get libapikfs.a (ac
imespec ts;
clock_gettime(CLOCK_PERF, &ts);
Updated to 3.13
Signed-off-by: Pawel Moll
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
Cc: Arnaldo Carvalho de Melo
Cc: Mike Galbraith
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Stephane Eranian
Cc: J
On 12/10/13, 11:54 AM, Dongsheng Yang wrote:
Okey, David, I saw your commit 207b57926 (perf kvm: Fix regression with
guest machine creation) here. It is applied to fix a bug of SEGFAULT. I
have to say that it changed the behavior of report and lead to current bug.
Dongsheng, I appreciate your
de Melo
Cc: David Ahern
---
tools/perf/Documentation/perf-report.txt | 6 ++
tools/perf/Documentation/perf-script.txt | 6 ++
tools/perf/builtin-report.c | 19 ---
tools/perf/builtin-script.c | 13 ++---
4 files changed, 38
On 12/9/13, 11:49 AM, Arnaldo Carvalho de Melo wrote:
Em Mon, Dec 09, 2013 at 11:08:05AM -0700, David Ahern escreveu:
On 12/9/13, 11:07 AM, Arnaldo Carvalho de Melo wrote:
Em Sun, Dec 08, 2013 at 10:39:36AM -0700, David Ahern escreveu:
On 12/6/13, 4:56 AM, Arnaldo Carvalho de Melo wrote:
You
Required for off-box analysis to convert kernel addresses.
Signed-off-by: David Ahern
Cc: Jiri Olsa
---
tools/perf/Documentation/perf-diff.txt | 3 +++
tools/perf/builtin-diff.c | 2 ++
2 files changed, 5 insertions(+)
diff --git a/tools/perf/Documentation/perf-diff.txt
b/tools
On 12/9/13, 11:07 AM, Arnaldo Carvalho de Melo wrote:
Em Sun, Dec 08, 2013 at 10:39:36AM -0700, David Ahern escreveu:
On 12/6/13, 4:56 AM, Arnaldo Carvalho de Melo wrote:
You get correct results because the modules/kallsyms info you provided
matches the VM you're testing, i.e. the pid te
On 12/9/13, 3:02 AM, Jiri Olsa wrote:
On 12/1/13, 5:01 PM, David Ahern wrote:
The dwarf option for callstacks is getting stuck on the unwind and showing
repeated frames. For example,
qemu-system-x86 13759 [001] 287663.084957: syscalls:sys_enter_futex: uaddr:
0x7f770a958700, op: 0x0081
On 12/9/13, 10:12 AM, Dongsheng Yang wrote:
On 12/08/2013 10:42 PM, David Ahern wrote:
On 12/9/13, 8:20 AM, Dongsheng Yang wrote:
How about introduce an option named --guestpid? Then we can make the
usage of perf kvm
more clear:
* perf kvm --guestkallsyms --guestmodules --guestpid
[top
On 12/9/13, 8:20 AM, Dongsheng Yang wrote:
How about introduce an option named --guestpid? Then we can make the
usage of perf kvm
more clear:
* perf kvm --guestkallsyms --guestmodules --guestpid
[top|record|report]
This usage is for only one guest and will not resolve the
symbols fr
On 12/4/13, 12:39 AM, Masami Hiramatsu wrote:
> What would you like to do with perf probe? Direct address accessing for
> userspace is not a good way to do, because userspace is relocatable...
Masami, I am referring to the address within the executable which is
what perf-probe pushes to the kernel
ping.. Jiri, any thoughts on this?
On 12/1/13, 5:01 PM, David Ahern wrote:
The dwarf option for callstacks is getting stuck on the unwind and showing
repeated frames. For example,
qemu-system-x86 13759 [001] 287663.084957: syscalls:sys_enter_futex: uaddr:
0x7f770a958700, op: 0x0081, val
On 12/5/13, 10:47 AM, Arnaldo Carvalho de Melo wrote:
Sorry, some editing problem happened here... Now with proper subject
line.
- Arnaldo
Em Thu, Dec 05, 2013 at 05:25:50PM +0100, Borislav Petkov escreveu:
From: Borislav Petkov
Ok,
step one: carve out fs-related stuff into a libapikfs.a li
801 - 900 of 1863 matches
Mail list logo
