Thanks Masami for detailed review.
Please see my comments below.
On Friday 17 March 2017 02:35 PM, Masami Hiramatsu wrote:
> Hi Ravi,
>
> (I avoided to review parser part since it may go to yacc in next version)
>
> On Tue, 14 Mar 2017 20:36:54 +0530
> Ravi Ba
On Wednesday 15 March 2017 06:46 PM, Arnaldo Carvalho de Melo wrote:
> Em Wed, Mar 15, 2017 at 01:03:31PM +0100, Jiri Olsa escreveu:
>> On Tue, Mar 14, 2017 at 08:36:54PM +0530, Ravi Bangoria wrote:
>>> +++ b/tools/perf/builtin-record.c
>>> +static int record__parse
2,69,72,403 cycles
CPU1 2,02,08,327 instructions # 0.34 insn per cycle
CPU2 73,17,123 instructions # 0.12 insn per cycle
1.000989132 seconds time elapsed
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/Documentation/perf-stat.txt | 3
On Wednesday 08 March 2017 02:07 PM, Masami Hiramatsu wrote:
> On Wed, 8 Mar 2017 12:29:07 +0530
> Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com> wrote:
>
>> '*ntevs' contains number of elements present in 'tevs' array. If
>> there are no elements in array, 'tevs
Thanks Masami for detailed review.
Please see my comments below.
On Saturday 18 March 2017 04:43 AM, Masami Hiramatsu wrote:
> On Tue, 14 Mar 2017 20:36:55 +0530
> Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com> wrote:
>
>> }
>> }
>>
>
On Saturday 18 March 2017 04:44 AM, Masami Hiramatsu wrote:
> The title of this patch seems not correct.
Oops. Sorry about that. Actual title is:
"perf/sdt: Remove stale warning"
Thanks for reporting,
Ravi
> On Tue, 14 Mar 2017 20:36:58 +0530
> Ravi B
Hi Masami,
On Thursday 16 March 2017 03:21 PM, Masami Hiramatsu wrote:
> On Tue, 14 Mar 2017 20:36:51 +0530
> Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com> wrote:
>
>> Changes in v5:
>> - Patch 2/7 is new. New option introduced in this patch helps to pass
>>
: 8e9f69371536 ("powerpc/kprobe: Don't emulate store when kprobe stwu r1")
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
History:
Commit 8e9f69371536 ("powerpc/kprobe: Don't emulate store when kprobe
stwu r1") fixed exception frame corruption for 32 bit
On Tuesday 28 March 2017 09:10 PM, Arnaldo Carvalho de Melo wrote:
> Em Wed, Mar 29, 2017 at 12:29:29AM +0900, Masami Hiramatsu escreveu:
>> Hi Arnaldo,
>>
>> please pull this, I've already acked to this series.
> I did it 15 minutes ago, running build tests on it now.
Hi Arnaldo,
Thanks for
: be96f63375a1 ("powerpc: Split out instruction analysis part of
emulate_step()")
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
Reviewed-by: Naveen N. Rao <naveen.n@linux.vnet.ibm.com>
---
History:
Commit 8e9f69371536 ("powerpc/kprobe: Don't emulate
Thanks Balbir for the review,
On Tuesday 11 April 2017 02:25 PM, Balbir Singh wrote:
> On Tue, 2017-04-11 at 10:38 +0530, Ravi Bangoria wrote:
>> If we set a kprobe on a 'stdu' instruction on powerpc64, we see a kernel
>> OOPS:
>>
>> [ 1275.165932] Bad ke
perf trace supports --no-syscalls option but it's not listed in
the man page. (Though, I see an example using --no-syscalls in
EXAMPLES section.)
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/Documentation/perf-trace.txt | 3 ++-
1 file changed, 2 inse
Hi Arnaldo, Thanks for pulling the patch.
On Thursday 13 April 2017 07:21 PM, Arnaldo Carvalho de Melo wrote:
> Em Thu, Apr 13, 2017 at 11:32:12AM +0530, Ravi Bangoria escreveu:
>> perf trace supports --no-syscalls option but it's not listed in
>> the man page. (Though, I see a
: https://lkml.org/lkml/2017/3/6/565
[1] https://lkml.org/lkml/2017/2/7/59
[2] https://lkml.org/lkml/2016/5/3/810
Hemant Kumar (1):
perf/sdt: Directly record SDT events with 'perf record'
Ravi Bangoria (6):
perf/sdt: Introduce util func is_sdt_event()
perf tool: Add option macro
Factor out the SDT event name checking routine as is_sdt_event().
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/parse-events.h | 20
tools/perf/util/probe-event.c | 9 +
2 files changed, 21 insertions(+), 8 deletions(-)
Perf was showing warning if user tries to record sdt event without
creating a probepoint. Now we are allowing direct record on sdt
events, remove this stale warning/hint.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/lib/api/fs/tracing_path.c | 17 --
:*
Warning: Recording on 2 occurrences of sdt_libpthread:*
$ sudo ./perf evlist
sdt_libpthread:mutex_release_3
sdt_libpthread:mutex_release_1
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/probe-event.c
.
Hint: Please use 'perf probe -d sdt_libpthread:*' to allow record on all
events.
$ sudo ./perf evlist
sdt_libpthread:mutex_release_3
sdt_libpthread:mutex_release_2
sdt_libpthread:mutex_release_1
sdt_libpthread:mutex_release
Signed-off-by: Ravi Bangoria <ravi.ba
:0xb126
$ sudo ./perf record -a -e sdt_libpthread:mutex_release
Matching event(s) from uprobe_events:
sdt_libpthread:mutex_release 0xb126@/usr/lib64/libpthread-2.24.so
Use 'perf probe -d ' to delete event(s).
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.
Error: sdt_libpthread:mutex_entry_1 not found in the cache
invalid or unsupported event: 'sdt_libpthread:mutex_entry_1'
Signed-off-by: Hemant Kumar <hem...@linux.vnet.ibm.com>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/builtin-record.c| 27 ++
Add an option macro that is the same as OPT_CALLBACK_OPTARG except
that the argument is not optional.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/lib/subcmd/parse-options.h | 4
1 file changed, 4 insertions(+)
diff --git a/tools/lib/subcmd/parse-optio
Thanks Masami for the review,
On Wednesday 08 March 2017 03:13 PM, Masami Hiramatsu wrote:
> On Wed, 8 Mar 2017 12:29:08 +0530
> Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com> wrote:
>
>> I don't see any user of this function. This function was being copied
>> to
Perf was showing warning if user tries to record sdt event without
creating a probepoint. Now we are allowing direct record on sdt
events, remove this stale warning/hint.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/lib/api/fs/tracing_path.c | 17 --
Error: sdt_libpthread:mutex_entry_1 not found in the cache
invalid or unsupported event: 'sdt_libpthread:mutex_entry_1'
Signed-off-by: Hemant Kumar <hem...@linux.vnet.ibm.com>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/builtin-record.c| 24
with 'perf record'
Ravi Bangoria (6):
perf/sdt: Introduce util func is_sdt_event()
perf/sdt: Allow recording of existing events
perf/sdt: Clean uprobe_events when event(out of multiple events)
parsing fails
perf/sdt: Warn when number of events recorded are not equal to cached
events
:*
Warning: Recording on 2 occurrences of sdt_libpthread:*
$ sudo ./perf evlist
sdt_libpthread:mutex_release_3
sdt_libpthread:mutex_release_1
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/probe-event.c | 58 +-
tools/perf/util/probe-e
:0xb126
$ sudo ./perf record -a -e sdt_libpthread:mutex_release
Matching event(s) from uprobe_events:
sdt_libpthread:mutex_release 0xb126@/usr/lib64/libpthread-2.24.so
Use 'perf probe -d ' to delete event(s).
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.
Factor out the SDT event name checking routine as is_sdt_event().
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/parse-events.h | 12
tools/perf/util/probe-event.c | 9 +
2 files changed, 13 insertions(+), 8 deletions(-)
diff
User may ask for multiple events in the same record command like,
perf record -a -e sdt_1:* -e sdt_2:*
If sdt_1:* events are already added to uprobe_events and sdt_2:*
event parsing fails, clean sdt_1:* events from uprobe_events.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.
.
Hint: Please use 'perf probe -d sdt_libpthread:*' to allow record on all
events.
$ sudo ./perf evlist
sdt_libpthread:mutex_release_3
sdt_libpthread:mutex_release_2
sdt_libpthread:mutex_release_1
sdt_libpthread:mutex_release
Signed-off-by: Ravi Bangoria <ravi.ba
On Tuesday 07 February 2017 08:25 AM, Masami Hiramatsu wrote:
> On Thu, 2 Feb 2017 16:41:38 +0530
> Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com> wrote:
>
>> The v5 patchset for sdt marker argument support for x86 [1] has
>> couple of issues. For example, it
I don't see any user of this function. This function was being copied
to tools/perf/builtin-probe.c by commit b02137cc6550 ("perf probe: Move
print logic into cmd_probe()"). Since then it has became stale.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools
ached events")
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/probe-event.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 28fb62c..4f9d6ee 100644
--- a/tools/perf/ut
x it by changing parameters of get_target_map().
Fixes: 801bc8193463 ("perf probe: Allow placing uprobes in alternate
namespaces.")
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/arch/powerpc/util/sym-handling.c | 2 +-
1 file changed, 1 inserti
ed-by: Anton Blanchard <an...@samba.org>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/annotate.c | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/tools/perf/util/annotate.c b/tools/perf/util/annotate.c
index 683f834..a031c4d
Hi Kim, Arnaldo,
On Thursday 01 June 2017 07:59 PM, Kim Phillips wrote:
> v2: addressing Arnaldo's clean-this-up comment: merged commit texts
> between Ravi's original, author changed in lieu of adding something I
> hadn't known nor thought about possibly doing: Originally-by: Ravi.
> Ravi, if
241631] [c000bbaabdf0] [c010be70]
call_usermodehelper_exec_async+0x170/0x210
[ 131.241955] [c000bbaabe30] [c000bae0]
ret_from_kernel_thread+0x5c/0x7c
Fix it by setting abi to PERF_SAMPLE_REGS_ABI_NONE when userspace
pt_regs are not set.
Signed-off-by: Ravi Bangoria <r
Hi Michael,
On Friday 16 June 2017 10:42 AM, Michael Ellerman wrote:
>
> That function (perf_get_regs_user()) didn't exist until 4.7, ie:
>
> ed4a4ef85cf5 ("powerpc/perf: Add support for sampling interrupt register
> state")
>
> So there must be something else going on.
>
> I'll hold off on
Thanks Naveen,
On Thursday 15 June 2017 08:57 PM, Naveen N. Rao wrote:
> Hmm... are you sure it's the same issue? The above commit only went into
> v4.7, which means we weren't able to use --call-graph=dwarf till v4.7.
Yes sorry. It's from v4.7 onwards.
-Ravi
Works like a charm with Milian's patch.
Acked-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
Note:
I still see very minor differences between libunwind and libdw. Also, second
last
function gets repeated two times in every callchain but it can be fixed later
on.
Otherwise all look
Hi Mark,
On Tuesday 13 June 2017 05:14 PM, Mark Wielaard wrote:
> I see the same on very short runs. But when doing a slightly longer run,
> even just using ls -lahR, which does some more work, then I do see user
> backtraces. They are still missing for some of the early samples though.
> It is
en for binaries that do not have CFI information. In addition,
>> libunwind is always preferred to libdw by the build machinery so this
>> cannot introduce regressions on machines that have both libunwind and
>> libdw installed.
>>
>> Cc: a...@kernel.org
>>
=0x8184e848 -l -d --no-show-raw -S -C
./vmlinux(test) 2>/dev/null|grep -v ./vmlinux(test):|expand'
Fix it by surrounding filename in double quotes.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/annotate.c | 2 +-
1 file changed, 1 insertion(+), 1
; the name of the DSO containing symbols also assumed no spaces existed
>> in the pathname, so grep also parsed these as extra positional
>> arguments.
> This one didn't apply to tip/perf/urgent, so I tried another patch
> addressing the same issue, from Ravi Bangoria, and it
On Friday 05 May 2017 06:52 PM, Ravi Bangoria wrote:
>
> On Friday 05 May 2017 06:27 PM, Arnaldo Carvalho de Melo wrote:
>> Em Wed, May 03, 2017 at 12:12:23PM +, Adam Stylinski escreveu:
>>> The path fed to the argument for objdump used in the perf-annotate
>>>
linux]$ git log | grep -i originally-by: | wc -l
> 58
> [acme@jouet linux]$
>
> Originally-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
>
> Ravi?
>
> I'm trying to catch up on my patch queue, so haven't read this
> thoroughly to have an idea if this is fair or OK,
x6c
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
Reviewed-by: Kamalesh Babulal <kamal...@linux.vnet.ibm.com>
---
arch/powerpc/perf/core-book3s.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/arch/powerpc/perf/core-book3s.c b/arch/powerpc/perf/core-book3s.
<pozdne...@gmail.com>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/callchain.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index be09d77..6d7f645 100644
---
<pozdne...@gmail.com>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
Changes in v2:
- Remove unnecessary checks for 'map'
tools/perf/util/callchain.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/tools/perf/util/callchain.c b/too
<pozdne...@gmail.com>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/callchain.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/tools/perf/util/callchain.c b/tools/perf/util/callchain.c
index 510b513..6d5a483 100644
---
e elf
sections") in binutils.
2. When the sample is from hypervisor. Hypervisor symbols can not
be resolved within guest and thus thread__find_addr_map() fails for
such symbols. Fix this by ignoring hypervisor symbols in the test.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux
e elf
sections") in binutils.
2. When the sample is from hypervisor. Hypervisor symbols can not
be resolved within guest and thus thread__find_addr_map() fails for
such symbols. Fix this by ignoring hypervisor symbols in the test.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.co
)[0x7f7fd8571830]
./perf(_start+0x29)[0x427a59]
This is because functions xyarray__max_x() and xyarray__max_y()
are returning incorrect values, causing crash while accessing
xyarray.
Fixes: d74be4767367 ("perf xyarray: Save max_x, max_y")
Signed-off-by: Ravi Bangoria &
^
Fix it by guarding function under HAVE_AUXTRACE_SUPPORT.
Fixes: 47e5a26a916b ("perf script: Fix --per-event-dump for auxtrace synth
evsels")
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/buil
On 11/14/2017 06:34 PM, Arnaldo Carvalho de Melo wrote:
Em Tue, Nov 14, 2017 at 08:55:40AM +0530, Ravi Bangoria escreveu:
There are many instructions, esp on powerpc, whose mnemonics are
longer than 6 characters. Using precision limit causes truncation
of such mnemonics.
Fix this by removing
On 11/20/2017 08:13 PM, Jin Yao wrote:
@@ -76,6 +97,17 @@ static struct rb_node *saved_value_new(struct rblist *rblist
__maybe_unused,
return >rb_node;
}
+static void saved_value_delete(struct rblist *rblist __maybe_unused,
+ struct rb_node *rb_node)
+{
the patch can be optimized a bit.
Please see my comments.
Looks ok, but you forgot to add lkml to the CC list, I'm doing that now
and also adding Ravi Bangoria @ IBM, which sent yet another of those
lines that would match your regexp.
Ravi & IBM crew, please see if we can proceed with Will's re
From: Madhavan Srinivasan <ma...@linux.vnet.ibm.com>
This will enable parsing JSON files on Power9 DD2.1.
Signed-off-by: Madhavan Srinivasan <ma...@linux.vnet.ibm.com>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/pmu-events/arch/powerpc/ma
r 0m0.587s
> sys 0m1.530s
> [root@s35lp76 perf]# fgrep '__ins__find sorted:1 nr_instructions:'
> /tmp/xxxaa | tail -1
> __ins__find sorted:1 nr_instructions:56 ins:0x3f406570
> [root@s35lp76 perf]#
>
> The number of different s390x branch/jump/cal
On 11/28/2017 01:26 PM, Thomas Richter wrote:
> The command 'perf annotate' parses the output of objdump and also
> investigates the comments produced by objdump. For example the
> output of objdump produces (on x86):
>
> 23eee: 4c 8b 3d 13 01 21 00 mov 0x210113(%rip),%r15
>
On 11/29/2017 08:33 PM, Thomas-Mich Richter wrote:
> On 11/29/2017 02:24 PM, Ravi Bangoria wrote:
>>
>> On 11/28/2017 01:26 PM, Thomas Richter wrote:
>>> The command 'perf annotate' parses the output of objdump and also
>>> investigates the comments
as memory address for imc events. While
grouping these two types of events, collect_events() tries to
interpret imc 'event_base' as a flag, which causes a corruption
resulting in a crash.
Consider only those events which belongs to 'perf_hw_context' in
collect_events().
Signed-off-by: Ravi
of various distibutions of Linux having to
> wait for a new release of the kernel's perf tool to be built with
> these trivial patches is inconvenient.
>
Tested-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
Thanks,
Ravi
Hi Balbir,
Sorry was away for few days.
On 12/14/2017 05:54 PM, Balbir Singh wrote:
> On Tue, Dec 12, 2017 at 11:29 PM, Ravi Bangoria
> <ravi.bango...@linux.vnet.ibm.com> wrote:
>> It may very well happen that branch instructions recorded by
>> bhrb entries already get
,-1
xscvdp vs1,vs1
add. r10,r10,r11
After:
li r11,-1
xscvdpsxds vs1,vs1
add. r10,r10,r11
Reported-by: Donald Stence <dste...@us.ibm.com>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/annotate.c | 18 +-
1
Hi Jiri,
On 11/14/2017 03:01 PM, Jiri Olsa wrote:
On Mon, Nov 13, 2017 at 09:14:38PM +0100, Jiri Olsa wrote:
On Mon, Nov 13, 2017 at 09:16:20PM +0530, Ravi Bangoria wrote:
Hi Jiri,
This patch seems to be causing segfault with "perf top --stdio".
Steps to reproduce:
1. start
Hi Jiri,
This patch seems to be causing segfault with "perf top --stdio".
Steps to reproduce:
1. start "perf top --stdio" in one terminal
2. run some simple workload in another terminal, let it get finished.
3. annotate function from previous workload in perf top (press 'a'
followed by 's')
Hi Jiri,
On 11/15/2017 07:34 PM, Jiri Olsa wrote:
On Tue, Nov 14, 2017 at 11:29:54AM +0100, Jiri Olsa wrote:
On Tue, Nov 14, 2017 at 03:45:27PM +0530, Ravi Bangoria wrote:
Hi Jiri,
On 11/14/2017 03:01 PM, Jiri Olsa wrote:
On Mon, Nov 13, 2017 at 09:14:38PM +0100, Jiri Olsa wrote:
On Mon
n": "Data TLB Miss page size 16G"
}
-]
\ No newline at end of file
+]
^^^ This 'newline thing' is unnecessary change. Can you please
resend the patch without it.
Otherwise, Ack-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
Thanks,
Ravi
On 12/08/2017 09:54 PM, Masami Hiramatsu wrote:
> On Thu, 7 Dec 2017 18:17:51 +0530
> Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com> wrote:
>
>>
>> On 12/07/2017 12:49 PM, Masami Hiramatsu wrote:
>>> Hi,
>>>
>>> Here is the 2nd vers
Hi Masami,
On 12/06/2017 08:36 AM, Masami Hiramatsu wrote:
> [4/5] Find versioned symbol from map without wildcard.
>This allows user to specify a symbol without version
>suffix nor wildcard. Anyway, at this point we have
>no way to specify version suffix, because parser
>
tml)
>
> The first version is here; https://lkml.org/lkml/2017/12/5/1124
>
> Here is the updates;
>
> [3/5] Add a description in Documentation/perf-probe.txt
> [5/5] Add a description and examples in Documentation/perf-probe.txt
>
> Thank you,
Ack-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
[c0027964] performance_monitor_exception+0x54/0x70
[c0009ba4] performance_monitor_common+0x114/0x120
Fix this by deferefencing them safely.
Suggested-by: Naveen N. Rao <naveen.n@linux.vnet.ibm.com>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com&g
@<fun+0>
int arg
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.vnet.ibm.com>
---
tools/perf/util/probe-event.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index b7aaf9b..0e04015
art', 'offset' becomes 0,
which is valid for normal symbols but *invalid* for zero length
symbols and thus updating h->addr[offset] causes memory corruption.
Fix this by adding one dummy element for zero length symbols.
Fixes: edee44be5919 ("perf annotate: Don't throw error for zero leng
Hi Masami,
On 05/04/2018 07:51 PM, Ravi Bangoria wrote:
>
>>> +}
>>> +
>>> +static void sdt_increment_ref_ctr(struct trace_uprobe *tu)
>>> +{
>>> + struct uprobe_map_info *info;
>>> +
>>> + uprobe_down_write_dup
On 05/08/2018 09:12 PM, Masami Hiramatsu wrote:
> On Thu, 3 May 2018 14:24:04 +0530
> Ravi Bangoria <ravi.bango...@linux.ibm.com> wrote:
>
>> Hi Masami,
>>
>> On 04/23/2018 12:32 PM, Masami Hiramatsu wrote:
>>> On Tue, 17 Apr 2018 09
Hi Masami,
On 05/07/2018 09:26 PM, Masami Hiramatsu wrote:
> On Mon, 7 May 2018 13:51:21 +0530
> Ravi Bangoria <ravi.bango...@linux.ibm.com> wrote:
>
>> Hi Masami,
>>
>> On 05/04/2018 07:51 PM, Ravi Bangoria wrote:
>>>>> +}
>>>>> +
&
Warn perf buildid-cache --purge-all failures in non verbose mode.
Ex,
$ sudo chown root:root /home/ravi/.debug -R
$ ./perf buildid-cache -P
Error: Permission denied.
Suggested-by: Masami Hiramatsu <mhira...@kernel.org>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.ibm.com>
org>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.ibm.com>
---
tools/perf/builtin-buildid-cache.c | 8 ++--
1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/tools/perf/builtin-buildid-cache.c
b/tools/perf/builtin-buildid-cache.c
index 7a7403913b57..ef6b3cc8d57d 100644
Hi Masami,
On 05/04/2018 10:18 AM, Masami Hiramatsu wrote:
>> +void uprobe_down_write_dup_mmap(void)
>> +{
>> +percpu_down_write(_mmap_sem);
>> +}
>> +
>> +void uprobe_up_write_dup_mmap(void)
>> +{
>> +percpu_up_write(_mmap_sem);
>> +}
>> +
> I'm not sure why these hunks are not done in
Arnaldo,
We already have a binary offset handy in perf code but there is no way to
dump it with perf script. We can derive it from symname+symoff but that's
a manual work.
Will it be good to have a '-F binoff' option?
Ravi
On 05/18/2018 01:29 AM, Arnaldo Carvalho de Melo wrote:
> Em Thu, May
Thanks Oleg for the review,
On 05/24/2018 09:56 PM, Oleg Nesterov wrote:
> On 04/17, Ravi Bangoria wrote:
>>
>> @@ -941,6 +1091,9 @@ typedef bool (*filter_func_t)(struct uprobe_consumer
>> *self,
>> if (ret)
>> goto err_buffer;
&g
: stvx : PASS
emulate_step_test: lxvd2x : FAIL
emulate_step_test: stxvd2x: FAIL
Fixes: https://github.com/linuxppc/linux/issues/148
Reported-by: Michael Ellerman <m...@ellerman.id.au>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.ibm.com>
---
Note: V
: lxvd2x : PASS (!CPU_FTR_VSX)
emulate_step_test: stxvd2x: PASS (!CPU_FTR_VSX)
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.ibm.com>
---
arch/powerpc/lib/test_emulate_step.c | 21 +++--
1 file changed, 15 insertions(+), 6 deletions(-)
diff --git a/arch/p
Replace 'op->type & INSTR_TYPE_MASK' expression with GETTYPE(op->type)
macro.
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.ibm.com>
---
arch/powerpc/include/asm/sstep.h | 2 ++
arch/powerpc/kernel/align.c | 2 +-
arch/powerpc/lib/sstep.c | 6 +++---
3
Reported-by: Michael Ellerman <m...@ellerman.id.au>
Signed-off-by: Ravi Bangoria <ravi.bango...@linux.ibm.com>
---
arch/powerpc/lib/sstep.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c
index db6bba259d91..23b7ddf
a new macro to simplify code a bit.
2nd patch fixes the kernel crash when VSX is not supported
or disabled.
3rd patch fixes emulate_step() tests
Ravi Bangoria (3):
powerpc/sstep: Introduce GETTYPE macro
powerpc/sstep: Fix kernel crash if VSX is not present
powerpc/sstep: Fix emulate_step
libc-2.26.so
> +8.83% _init+236 [.] _initls
> ...
>
> Reported-by: Ravi Bangoria
> Signed-off-by: Sandipan Das
Acked-by: Ravi Bangoria
Hi Oleg,
On 06/08/2018 10:06 PM, Oleg Nesterov wrote:
> Hello,
>
> I am travelling till the end of the next week, can't read this version
> until I return. Just one question,
>
> On 06/06, Ravi Bangoria wrote:
>>
>> 1. One of the major reason was the deadlock
Hi Masami,
>>> Hmm, it sounds simple... maybe we can increment refctr in
>>> install_breakpoint/
>>> remove_breakpoint?
>>
>> Not really, it would be simpler if I can put it inside install_breakpoint().
>> Consider an mmap() case. Probed instruction resides in the text section
>> whereas
>>
2600
H_SEND_CRQ 77 1762 7240 2447
Signed-off-by: Ravi Bangoria
---
.../perf/scripts/python/bin/powerpc-hcalls-record | 2 +
.../perf/scripts/python/bin/powerpc-hcalls-report | 2 +
tools/perf/scripts/python/powerpc-hcalls.py
correctness issue. I'm working on
a fix for this.
[1] https://lkml.org/lkml/2018/4/17/23
[2] https://lkml.org/lkml/2018/5/25/111
[3] https://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation
[4] https://github.com/iovisor/bcc/issues/327#issuecomment-200576506
Ravi Bangoria (7
Hi Paul,
On 06/06/2018 08:23 PM, Paul Clarke wrote:
> On 06/05/2018 12:50 PM, Arnaldo Carvalho de Melo wrote:
>> From: Ravi Bangoria
>>
>> Add python script to show hypervisor call statistics. Ex,
>>
>> # perf record -a -e "{powerpc:hcall_entry,powe
Hi Thomas,
On 05/24/2018 07:26 PM, Thomas Richter wrote:
> @@ -95,7 +98,7 @@ int test__session_topology(struct test *test
> __maybe_unused, int subtest __maybe
> {
> char path[PATH_MAX];
> struct cpu_map *map;
> - int ret = -1;
> + int ret;
This is failing for me:
ls 2282 1031.73: 25 cpu-clock:uhH: 7effe4b3a650
Signed-off-by: Ravi Bangoria
Fixes: a14390fde64e ("perf script: Allow creating per-event dump files")
---
tools/perf/builtin-script.c | 14 ++
1 file changed, 14 insertions(+)
diff --git a/tools/perf/builtin-script.c
record -o - ls | ./perf script
Segmentation fault (core dumped)
After:
# ./perf record -o - ls | ./perf script
ls 7031 4392.099856: 25 cpu-clock:uhH: 7f5e0ce7cd60
ls 7031 4392.100355: 25 cpu-clock:uhH: 7f5e0c706ef7
Signed-off-by: Ravi Bangoria
Fixes: 57b5de463925 ("perf r
Few fields are missing in a perf script -F hint. Add them.
Signed-off-by: Ravi Bangoria
---
tools/perf/builtin-script.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
index a31d7082188e..f3fefbcc4503 100644
First patch fixes perf output field hint. Second and third fixes
crash when used in a piped mode.
v2 changes:
- [PATCH 3/3] Changed as suggested by Arnaldo
v1: https://lkml.org/lkml/2018/6/20/538
Ravi Bangoria (3):
perf script: Add missing output fields in a hint
perf script: Fix crash
Hi Srikar,
On 07/02/2018 08:24 PM, Srikar Dronamraju wrote:
> * Ravi Bangoria [2018-06-28 10:52:08]:
>
>> Reference counter gate the invocation of probe. If present,
>> by default reference count is 0. Kernel needs to increment
>> it before tracing the probe an
201 - 300 of 1366 matches
Mail list logo
