Hi:
On 3/13/13 3:42 AM, chenggang wrote:
---
tools/perf/builtin-stat.c |2 +-
tools/perf/tests/open-syscall-tp-fields.c |2 +-
tools/perf/util/event.c | 12 +-
tools/perf/util/evlist.c |2 +-
tools/perf/util/evsel.c
On 3/13/13 3:42 AM, chenggang wrote:
diff --git a/tools/perf/Makefile b/tools/perf/Makefile
index a2108ca..7f3f066 100644
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
@@ -209,7 +209,8 @@ BASIC_CFLAGS = \
-Iutil \
-I. \
-I$(TRACE_EVENT_DIR) \
- -D_LARGEFILE64_S
On 3/13/13 3:42 AM, chenggang wrote:
tools/perf/util/evlist.c |4 +++-
tools/perf/util/evsel.c | 19 +--
tools/perf/util/evsel.h |5 -
tools/perf/util/header.c | 28 ++--
tools/perf/util/header.h |3 ++-
5 files changed, 44 ins
One component of tracking perf_clock timestamps to time-of-day is
updates to xtime by a user or ntpd. To that end this patch adds
tracepoints to the timekeeping code as suggested by Thomas:
https://lkml.org/lkml/2011/3/2/186
Signed-off-by: David Ahern
Cc: Thomas Gleixner
Cc: John Stultz
: David Ahern
Cc: Thomas Gleixner
Cc: John Stultz
Cc: Steven Rostedt
---
include/trace/events/timekeeping.h | 51
kernel/time/timekeeping.c | 11
2 files changed, 62 insertions(+)
create mode 100644 include/trace/events/timekeeping.h
diff
On 2/5/13 9:05 AM, Jiri Olsa wrote:
Adding -D option into the rwtop record command line, to get
immediate response when there's event in the buffer.
Seems like this will dramatically increase CPU usage of perf-record as
well as increase the overhead of each event with a process wakeup. How
ab
ble use case no longer fails.
The first problem is much more complicated and will be the subject of a
future patch.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
Cc: Jiri Olsa
Cc: Namhyung Kim
---
tools/perf/per
On 2/1/13 7:18 AM, Pawel Moll wrote:
8<---
From 2ad51a27fbf64bf98cee190efc3fbd7002819692 Mon Sep 17 00:00:00 2001
From: Pawel Moll
Date: Fri, 1 Feb 2013 14:03:56 +
Subject: [PATCH] perf: Add ioctl to return current time value
To co-relate user space events with the perf events stream
a cur
On 4/2/13 2:22 AM, Paul Bolle wrote:
tools/perf/Makefile |5 +
1 file changed, 5 insertions(+)
create mode 100644 tools/perf/Pconfig
All current Kconfig files can be found with git ls-files "*Kconfig*".
This file won't. Why does it need to be named differently?
It does not. I wil
On 4/2/13 3:32 AM, Paul Bolle wrote:
On Mon, 2013-04-01 at 21:54 -0600, David Ahern wrote:
--- a/tools/perf/Pconfig
+++ b/tools/perf/Pconfig
@@ -1,3 +1,14 @@
+config ARCH
+string
+option env="ARCH"
Why does this copy init/Kconfig?
+config X86
+def_bool y if A
On 4/2/13 3:37 AM, Paul Bolle wrote:
On Mon, 2013-04-01 at 21:54 -0600, David Ahern wrote:
--- a/tools/perf/Makefile
+++ b/tools/perf/Makefile
[...]
@@ -602,6 +590,16 @@ ifdef CONFIG_LIBELF
endif
else
+ifdef CONFIG_DWARF
+$(error CONFIG_DWARF depends on LIBELF)
+endif
On 4/3/13 3:17 AM, Stephane Eranian wrote:
I haven't done any specific testing with either approach yet. The goal is to
use this perf timestamp to correlate user level events to hardware
events recorded
by the kernel. I would assume there would be situations where those user events
could be on th
On 4/3/13 8:00 AM, Stephane Eranian wrote:
What's the advantage of changing apps -- like the JIT compiler -- to emit
perf based timestamps versus having perf emit existing timestamps? ie.,
monotonic and realtime clocks already have vdso mappings for userspace with
well known performance character
Hi
On 1/31/13 2:08 AM, chenggang@gmail.com wrote:
From: chenggang@gmail.com
This patch depends on a prev patch: https://lkml.org/lkml/2013/1/29/47
If the engineers want to analyze the direct io behavior of some applications
without source code, perf tools with some appropriate tracepoi
On 1/31/13 10:25 AM, Arnaldo Carvalho de Melo wrote:
From: Arnaldo Carvalho de Melo
Removing one more memory leak found with valgrind.
If you are spending time one memory consumption by perf-top one thing to
look at is handling of terminated threads. Currently, terminated threads
are stuck
On 1/24/13 11:27 AM, Arnaldo Carvalho de Melo wrote:
Em Fri, Jan 18, 2013 at 01:51:24PM -0600, Tom Zanussi escreveu:
This patchset fixes various problems I hit while doing a run-through
of the existing 'perf script' scripts.
Also, sometimes when running rwtop I get a 'failed to read event
heade
On 2/25/13 11:39 AM, Oleg Nesterov wrote:
To clarify, I am not trying to review this patch, I'd like to ask
the question...
On 02/25, Jiri Olsa wrote:
Currently the -C option does not work for record command,
because of the targets mismatch when synthesizing threads.
Fixing this by using prop
er of nodes in every row are same.
The interface to append and shrink a exist xyarray is provided.
1) xyarray__append()
append a column for all rows.
2) xyarray__remove()
remove a column for all rows.
Cc: David Ahern
Cc: Peter Zijlstra
Cc: Paul Mackerras
Cc: Ingo Molnar
Cc: Arnaldo Car
On 2/26/13 2:41 AM, chenggang wrote:
---8<---
diff --git a/tools/perf/util/event.c b/tools/perf/util/event.c
index 5cd13d7..91d2848 100644
--- a/tools/perf/util/event.c
+++ b/tools/perf/util/event.c
@@ -327,8 +327,8 @@ int perf_event__synthesize_thread_map(struct perf_tool
*tool,
err =
Dongsoo reported that perf-kvm was segfaulting. The bug was introduced by
commit 3786063 which moved file_name from struct perf_kvm which was
initialized on the stack to an uninitialized stack variable. Fix by
initializing.
Needs to be applied to 3.7 and 3.8 as well.
Signed-off-by: David Ahern
On 2/27/13 9:31 PM, Xiao Guangrong wrote:
On 02/28/2013 11:29 AM, David Ahern wrote:
Dongsoo reported that perf-kvm was segfaulting. The bug was introduced by
commit 3786063 which moved file_name from struct perf_kvm which was
initialized on the stack to an uninitialized stack variable. Fix by
On 2/28/13 7:24 AM, Arnaldo Carvalho de Melo wrote:
Just fast forwarded my perf/urgent to tip/perf/urgent, checked and the
fix is there as pointed out by Xiao.
hmm hanging out in your perf/core branch rather than perf/urgent.
David
--
To unsubscribe from this list: send the line "unsubscri
On 2/26/13 2:41 AM, chenggang wrote:
diff --git a/tools/perf/builtin-record.c b/tools/perf/builtin-record.c
index 774c907..13112c6 100644
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -31,6 +31,8 @@
#include
#include
+#define MMAP(e, y) (*(struct perf_mmap *)xyarra
On 2/18/13 8:16 AM, Stephane Eranian wrote:
Hi,
I think the advantage of the ioctl() is that is reuses existing infrastructure.
The downside is that to get the timestamp you need at a minimum:
uint64_t get_perf_timestamp(void)
{
struct perf_event_attr attr;
uint64_t ts = 0;
int fd;
Signed-off-by: David Ahern
---
tools/perf/util/trace-event-parse.c | 13 -
tools/perf/util/trace-event.h |1 -
2 files changed, 14 deletions(-)
diff --git a/tools/perf/util/trace-event-parse.c
b/tools/perf/util/trace-event-parse.c
index 8450bec..4454835 100644
--- a
Signed-off-by: David Ahern
---
tools/perf/util/machine.c | 22 +++---
tools/perf/util/machine.h |1 -
2 files changed, 11 insertions(+), 12 deletions(-)
diff --git a/tools/perf/util/machine.c b/tools/perf/util/machine.c
index efdb38e..c5e3b12 100644
--- a/tools/perf/util
Signed-off-by: David Ahern
---
tools/perf/util/trace-event-parse.c | 24
tools/perf/util/trace-event.h |3 ---
2 files changed, 27 deletions(-)
diff --git a/tools/perf/util/trace-event-parse.c
b/tools/perf/util/trace-event-parse.c
index 3aabcd6..8450bec
Hi Arnaldo:
Going through my backlog of perf cleanup patches. These remove unused
functions and then makes a sole callee static.
David Ahern (4):
perf: remove unused print_event function
perf: remove unused print_trace_event function
perf: remove unused perf_session__remove_thread
perf
Signed-off-by: David Ahern
---
tools/perf/util/session.c | 12
tools/perf/util/session.h |1 -
2 files changed, 13 deletions(-)
diff --git a/tools/perf/util/session.c b/tools/perf/util/session.c
index bd85280b..ab265c2 100644
--- a/tools/perf/util/session.c
+++ b/tools/perf
On 3/13/13 5:19 AM, Namhyung Kim wrote:
-static const char *output_file = "trace.info";
+static const char *output_file = "perf.data";
static int output_fd;
Not really needed. Only use is a couple of die() calls. Since the file
name is not used on the open the die message would be misleading
Hi Arnaldo:
I believe this has been fixed but I don't see it in your tree (perf/core
or perf/urgent) or in the tip tree:
builtin-annotate.c: In function ‘hists__find_annotations’:
builtin-annotate.c:161:4: error: duplicate case value
builtin-annotate.c:154:4: error: previously used here
make:
On 3/13/13 1:56 PM, Arnaldo Carvalho de Melo wrote:
Em Wed, Mar 13, 2013 at 08:50:35AM -0600, David Ahern escreveu:
I believe this has been fixed but I don't see it in your tree
(perf/core or perf/urgent) or in the tip tree:
builtin-annotate.c: In function ‘hists__find_annotations’:
bu
On 8/28/12 12:35 AM, Andrew Jones wrote:
This is the second user. As I said, move the stats code from
builtin-stat.c. Or are you suggesting to duplicate the code once,
and then refactor the second (really the third) time it's needed?
gah, I misunderstood your comment. I see your point now.
Dav
On 8/29/12 12:00 AM, Suzuki K. Poulose wrote:
The following commit:
author David Ahern
Tue, 31 Jul 2012 04:31:33 + (22:31 -0600)
committer Arnaldo Carvalho de Melo
Fri, 3 Aug 2012 13:39:51 + (10:39 -0300)
commit ee8dd3ca43f151d9fbe1edeef68fb8a77eb9f047
causes
perf does not have networking related functionality, and the inclusion of
these headers is one of the causes of compile failures for Android:
https://lkml.org/lkml/2012/8/23/316
https://lkml.org/lkml/2012/8/28/293
So, remove them.
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Frederic
On 8/28/12 4:33 PM, Irina Tirdea wrote:
Some systems (e.g. Android) define in their libc types.h
__le16, __be16, etc. Since perf is wrapping
with a local version, we need to define this
constants in the local version too.
Errors in Android:
In file included from bionic/libc/include/netinet/in.h
On 8/28/12 3:59 PM, Irina Tirdea wrote:
Wrapping asm/byteorder.h will also replace the glibc header,
not only the kernel one. asm/byteorder.h further includes
architecture dependent headers that define endianess.
Some systems (e.g. Android) need constant definitions that
depend on endianess (e.g
On 8/28/12 4:02 PM, Irina Tirdea wrote:
On some systems (e.g. Android), ALIGN is defined in
system headers as ALIGN(p). The definition of ALIGN used
in perf takes 2 parameters: ALIGN(x,a). This leads to
redefinition conflicts.
Redefinition error on Android:
In file included from util/include/lin
On 8/28/12 4:06 PM, Irina Tirdea wrote:
on_exit() is only available in new versions of glibc.
Using on_exit on Android leads to errors at compile time.
Replacing on_exit with its more portable version atexit.
This leads to using a global variable since on_exit supports
sending a parameters while
In addition to Andrew's comment about making the stats struct and
functions generic...
On 8/27/12 3:51 AM, Dong Hao wrote:
---8<---
+static void exit_event_decode_key(struct event_key *key, char decode[20])
+{
+ const char *exit_reason = get_exit_reason(key->key);
+
+ snprintf(deco
ith the suitable function
provided by the user.
Signed-off-by: Suzuki K Poulose
Cc: David Ahern
---
tools/perf/util/intlist.c |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/tools/perf/util/intlist.c b/tools/perf/util/intlist.c
index fd530dc..77c504f 100644
--- a/to
remove_node(rblist, pos);
}
free(rblist);
}
Acked-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.
On 9/3/12 2:48 AM, don wrote:
于 2012年08月31日 02:29, David Ahern 写道:
In addition to Andrew's comment about making the stats struct and
functions generic...
Yes. :-)
On 8/27/12 3:51 AM, Dong Hao wrote:
---8<---
+static void exit_event_decode_key(struct event_key *key, char
d
On 8/22/12 8:37 AM, chenggang qin wrote:
> From: Chenggang Qin
>
> While we use "perf top -p 'pid'" to monitor the symbols of specified
> processes, some new threads would be created by the monitored processes
> during "perf top" is running. In current version, these new threads and
> their symbo
On 9/4/12 4:32 AM, Maciek Borzecki wrote:
When analyzing perf data from hosts of other architecture than one of the local
host it's useful to call objdump that is part of a toolchain for that
architecture. Instead of calling regular objdump, call one that user specified
in command line.
Why not
On 9/11/12 3:20 AM, Robert Richter wrote:
On 10.09.12 10:40:16, David Ahern wrote:
--- a/tools/perf/builtin-record.c
+++ b/tools/perf/builtin-record.c
@@ -294,6 +294,11 @@ try_again:
perf_evsel__name(pos));
rc = -err
On 9/11/12 8:01 AM, Robert Richter wrote:
Ok, wrong question. Better would have been: Did you run it on a
non-pebs Intel machine of an non-ibs AMD machine?
Intel: yes. VM for example. All the servers I have now are Nehalem or
better - ie., with a PEBS.
AMD: no. I do not have any AMD-based s
On 9/12/12 8:40 AM, Andi Kleen wrote:
- while(!done) sleep(1);
+ char piddir[40];
+ if (target.pid && access("/proc", X_OK) == 0)
+ snprintf(piddir, sizeof piddir, "/proc/%d",
atoi(target.pid));
else path has piddir not set. Also
On 9/11/12 9:11 AM, Robert Richter wrote:
On 11.09.12 08:32:55, David Ahern wrote:
My guess would be /usr/include/bits/errno.h:
/* Linux has no ENOTSUP error code. */
# define ENOTSUP EOPNOTSUPP
Ok, so ENOTSUP is actually the same as EOPNOTSUPP. Since the syscall
returns a EOPNOTSUPP, I
Hopefully this wraps up the precise mode-exclude_guest dependency.
I'm sure someone will let me know if I screwed up the attribution
in the second patch.
David Ahern (2):
perf tool: precise mode requires exclude_guest
perf tool: give user better message if precise is not supported
x27;precise' request may not be supported. Try removing 'p' modifier
sleep: Terminated
v2: softened message to 'may not be' supported per Robert's suggestion
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Robert Richter
---
tools/perf/built
PU
supports IBS profiling.
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Robert Richter
Cc: Gleb Natapov
Cc: Avi Kivity
Link: https://lkml.org/lkml/2012/7/9/264
---
tools/perf/util/parse-events.c |3 +++
1 file changed, 3 insertions(+)
diff --git a/tools/perf/util/parse-e
virtual machine. Therefore we have to disable PEBS on VT-x
enter and re-enable on VT-x exit, enforcing a strict exclude_guest.
This patch enforces exclude_guest kernel side.
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Robert Richter
Cc: Gleb Natapov
Cc: Avi Kivity
Link: h
On 9/12/12 8:38 PM, Namhyung Kim wrote:
Hi David,
On Wed, 12 Sep 2012 09:16:29 -0600, David Ahern wrote:
From: Peter Zijlstra
Per Peter:
"Intel PEBS in VT-x context uses the DS address as a guest linear address,
even though its programmed by the host as a host linear address.
static const char * const kvm_usage[] = {
-"perf kvm [] {top|record|report|diff|buildid-list}",
+"perf kvm [] {top|record|report|diff|buildid-list|stat}",
NULL
};
The usage for the report/record sub commands of stat is never shown. e.g.,
$ perf kvm stat
--> shows help for perf
On 9/13/12 7:45 AM, Arnaldo Carvalho de Melo wrote:
Em Wed, Sep 12, 2012 at 10:56:44PM -0600, David Ahern escreveu:
static const char * const kvm_usage[] = {
+"perf kvm [] {top|record|report|diff|buildid-list|stat}",
The usage for the report/record sub commands of stat is n
On 9/13/12 9:58 AM, Andi Kleen wrote:
From: Andi Kleen
When counting a process with perf stat -p check if the process died
and exit collection if yes.
v2: Add more checks, handle non -p again. Handle /proc not there.
v3: Handle multi pid case. Fix non /proc error path
Signed-off-by: Andi Kleen
Hopefully this wraps up the precise mode-exclude_guest dependency.
I'm sure someone will let me know if I screwed up the attribution
in the second patch.
David Ahern (2):
perf tool: precise mode requires exclude_guest
perf tool: give user better message if precise is not supported
PU
supports IBS profiling.
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Robert Richter
Cc: Gleb Natapov
Cc: Avi Kivity
Link: https://lkml.org/lkml/2012/7/9/264
---
tools/perf/util/parse-events.c |3 +++
1 file changed, 3 insertions(+)
diff --git a/tools/perf/util/parse-e
x27;precise' request may not be supported. Try removing 'p' modifier
sleep: Terminated
v2: softened message to 'may not be' supported per Robert's suggestion
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Robert Richter
---
tools/perf/built
virtual machine. Therefore we have to disable PEBS on VT-x
enter and re-enable on VT-x exit, enforcing a strict exclude_guest.
This patch enforces exclude_guest kernel side.
Signed-off-by: David Ahern
Cc: Ingo Molnar
Cc: Peter Zijlstra
Cc: Robert Richter
Cc: Gleb Natapov
Cc: Avi Kivity
Link: h
On 7/5/12 7:02 AM, Jovi Zhang wrote:
From 91fbcca37a2c9979083f4b9a6fa9c1875fc2886f Mon Sep 17 00:00:00 2001
From: Jovi Zhang
Date: Fri, 6 Jul 2012 00:05:40 +0800
Subject: [PATCH] perf: add /proc/perf_events file for dump perf events info
This new /proc/perf_events file is used for real time d
On 9/13/12 11:43 PM, Ingo Molnar wrote:
v2: softened message to 'may not be' supported per Robert's suggestion
Well, either it's supported on this machine or it's not - why
does the text have to be so unsure about it?
Because EOPNOTSUPP is returned for more than just precise mode. We
cannot
On 9/14/12 5:36 AM, Ingo Molnar wrote:
Well, then that is useful information we *lost*, and that
situation needs to be improved on the ABI side: an expanded
error code present in the event structure, copied back to
user-space on errors, or so.
(Alternatively, a special event channel just to pass
On 9/13/12 8:56 PM, Xiao Guangrong wrote:
We'd better keep the record/report function, sometimes, we can only get
perf.data
from the customers whose machine can not be reached for us.
Especially, other tracepoints are also interesting for us when the customers
encounter
the performance issue,
On 9/14/12 9:39 AM, Andrew Jones wrote:
I recently tried 'perf script rwtop', and it immediately failed with
'failed to read event header'. Running it through strace I found that the
when rwtop.pl is reading from the pipe, and gets one of it's alarms, that
the ERESTARTSYS seems to confuse it - c
Cleans up the builtin-kvm code in preparation for the live mode.
No functional changes; only code movement.
Signed-off-by: David Ahern
Cc: Dong Hao
Cc: Runzhen Wang
Cc: Xiao Guangrong
Cc: Ingo Molnar
---
tools/perf/builtin-kvm.c | 460 +-
1 file
So this is still not a totally robust ordering but it's still
better than what we had before.
The only way to have a deterministic and solid ordering will be to use
per cpu perf.data files.
Reported-by: Stephane Eranian
Signed-off-by: Frederic Weisbecker
Signed-off-by: David Ahern
Cc: Peter Zi
$ /tmp/pbuild/perf report -i perf.data --kallsyms kallsyms
Error:
Please install objdump for i686.
You can add it to PATH, set CROSS_COMPILE or override the default using
--objdump.
And worse it refuses to run without it. If I was running the annotate
command I could understand the request --
On 11/1/12 7:11 PM, Namhyung Kim wrote:
From f0a9d6303f83452c8b6f81081abae8fdf9c81778 Mon Sep 17 00:00:00 2001
From: Namhyung Kim
Date: Fri, 2 Nov 2012 09:48:17 +0900
Subject: [PATCH] perf tools: Use normalized arch name for searching objdump
path
David reported that perf report for i686 targ
On 11/3/12 2:07 PM, Hannes Frederic Sowa wrote:
On Fri, Sep 7, 2012 at 8:04 AM, tip-bot for Suzuki K. Poulose
wrote:
Reported-by: Ananth N. Mavinakayanahalli
Signed-off-by: Suzuki K. Poulose
Acked-by: David Ahern
Cc: Ananth N Mavinakayanahalli
Cc: David Ahern
Cc: Frederic Weisbecker
Cc
. To prevent similar artifacts, normalize arch
name when comparing host and file architectures.
Reported-by: David Ahern
Cc: David Ahern
Cc: Irina Tirdea
Signed-off-by: Namhyung Kim
---
tools/perf/arch/common.c | 40 +---
1 file changed, 33 insertions(
nto a struct in order to reduce number of
those unused arguments.
This is a mechanical change and does not intend a functional change.
Cc: David Ahern
Cc: Irina Tirdea
Signed-off-by: Namhyung Kim
---
tools/perf/builtin-annotate.c | 2 +-
tools/perf/builtin-report.c | 4 ++--
time doens't make sense.
Reported-by: David Ahern
Cc: Irina Tirdea
Signed-off-by: Namhyung Kim
---
tools/perf/builtin-report.c| 9 ++---
tools/perf/builtin-top.c | 3 ++-
tools/perf/ui/browsers/hists.c | 22 --
tools/perf/util/hist.h | 7 +
On 7/24/12 12:03 PM, Arnaldo Carvalho de Melo wrote:
Em Tue, Jul 24, 2012 at 11:28:48AM -0600, David Ahern escreveu:
On 7/24/12 10:15 AM, Robert Richter wrote:
On AMD cpus precise_ip maps to IBS, which does not support hardware
options as perfctrs do. Thus, following attr flags are not
5a7ed29 fixed up perf-record but not perf-top. Similar argument
holds for it -- fallback to PMU only if it does not exist and handle
invalid attributes separately.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Frederic Weisbecker
-ag -- sleep 1
v2: simplified to saving the command in parse_options per Stephane's suggestion
Signed-off-by: David Ahern
CC: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Stephane Eranian
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/bu
directory name to
a pid. For example if guestmount contains a directory with the name 1foo, atoi
converts it to a pid of 1 and a machine is created with a pid of 1. This is
wrong; this directory really should be ignored. Use strtol to do that.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc
Some refactoring so that strlist and intlist have a common base - rblist.
>From there use intlist to only show guest messages once.
David Ahern (4):
perf tool: introducing rblist
perf tool: change strlist to use the new rblist
perf tool: introduce intlist
perf kvm top: limit guest ker
Built on rblist - like strlist. Used in the next patch.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/Makefile |2 +
tools/perf/util/intlist.c | 101
'perf kvm top' shows a continual flurry of:
Can't find guest [5201]'s kernel information
if it can't find the guest info and with a lot of VMs running a user
has no chance of reading them all. Limit message to once per guest.
Signed-off-by: David Ahern
Cc: Arnaldo C
rblist is the rbtree based code from strlist. It will be the common
code for strlist and the to-be-introduced intlist.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/Makefile
Replaces the direct use of rbtree code with the rblist API. In the end
the patch is a no-op on strlist functionality; the API for strlist is
not changed, only its implementaton.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Jiri Olsa
Cc: Namhyung Kim
Cc
On 9/20/12 4:13 PM, Irina Tirdea wrote:
From: Irina Tirdea
Shell path /bin/sh is hardcoded in various places in perf. Android has a
different folder structure and does not have /bin/sh.
Set the shell path at compile time in the Makefile by setting PERF_SHELL_PATH.
By default it is set to /bin/
On 9/21/12 1:47 AM, Pekka Enberg wrote:
diff --git a/tools/perf/util/symbol.c b/tools/perf/util/symbol.c
index e2e8c69..eb671d5 100644
--- a/tools/perf/util/symbol.c
+++ b/tools/perf/util/symbol.c
@@ -1051,7 +1051,7 @@ int dso__load(struct dso *dso, struct map *map,
symbol_filter_t filter)
On 9/13/12 2:59 PM, David Ahern wrote:
Hopefully this wraps up the precise mode-exclude_guest dependency.
I'm sure someone will let me know if I screwed up the attribution
in the second patch.
David Ahern (2):
perf tool: precise mode requires exclude_guest
perf tool: give user b
I like the idea, but can't checkout the patch - does not apply to
Arnaldo's latest perf/core branch. mind rebasing?
David
--
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.or
On 10/26/12 8:54 AM, Ingo Molnar wrote:
perf/core improvements:
. perf inject changes to allow showing where a task sleeps, from Andrew Vagin.
. Makefile improvements from Namhyung Kim.
These are really useful: there used to be a couple of seconds of
wait time at the beginning of every perf b
On 10/26/12 9:39 AM, Namhyung Kim wrote:
hm, thats the unique sample_type issue again ;) Once we set
PERF_SAMPLE_READ for event or group, we need to set it for
all other events in session, otherwise the report fails
Sorry, I don't understand why we need to set it for all events. Just
setting i
On 10/26/12 10:25 AM, Namhyung Kim wrote:
2012-10-26 (금), 10:14 -0600, David Ahern:
On 10/26/12 9:39 AM, Namhyung Kim wrote:
hm, thats the unique sample_type issue again ;) Once we set
PERF_SAMPLE_READ for event or group, we need to set it for
all other events in session, otherwise the report
This patch set consolidates all of the open counters loops into a
single common one.
Rebased to latest perf/core. Added more hooks to ui stubs file.
David Ahern (9):
perf python: add ui stubs file
perf top: make use of perf_record_opts
perf evlist: introduce open counters method
perf top
stdio based implementations of ui_ based functions for the python
library. Needed for patch 3 - consolidating open counters method.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/util/python-ext-sources
Changes top code to use the perf_record_opts struct. Stepping stone to
consolidating the open counters code.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/builtin-top.c | 84
Superset of the open counters code in perf-top and perf-record -
combining retry handling and error handling. Should be functionally
equivalent.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/util/evlist.c
Remove open counters code with all the retry and error handling in
favor of the new perf_evlist__open_counters method which is based
on the top code.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/builtin
Remove open counters code with all the retry and error handling in
favor of the new perf_evlist__open_counters method which is based
on the existing code.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf
This is required for perf-stat to use perf_evlist__open_counters.
And move opts to a stack variable.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/builtin-stat.c | 167
perf-stat continues on some open failures. top and record
opt out to maintain existing behavior.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/builtin-record.c |2 +-
tools/perf/builtin-top.c|2
Removes a lot of duplicated code moving to the common
open method.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/builtin-stat.c | 103 ++---
1 file changed, 22
No longer needed with perf-stat converted to perf_evlist__open_counters.
Signed-off-by: David Ahern
Cc: Arnaldo Carvalho de Melo
Cc: Ingo Molnar
Cc: Frederic Weisbecker
Cc: Peter Zijlstra
---
tools/perf/util/evsel.c |6 --
tools/perf/util/evsel.h |2 --
2 files changed, 8
1 - 100 of 1868 matches
Mail list logo