(2014/02/10 13:02), Masami Hiramatsu wrote:
> (2014/02/09 23:37), Ingo Molnar wrote:
>>
>> * Masami Hiramatsu wrote:
>>
>>> I guess the second reason is why the stap takes so long time to set
>>> probes. stap tries to register kprobes without disabled flag, that
>>> means we enables thousands
(2014/02/10 13:02), Masami Hiramatsu wrote:
(2014/02/09 23:37), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
I guess the second reason is why the stap takes so long time to set
probes. stap tries to register kprobes without disabled flag, that
means we
(2014/02/09 23:37), Ingo Molnar wrote:
>
> * Masami Hiramatsu wrote:
>
>> I guess the second reason is why the stap takes so long time to set
>> probes. stap tries to register kprobes without disabled flag, that
>> means we enables thousands of probes (overheads).
>>
>> So the similar thing
Hi -
> > So the similar thing happens when we enables events as below;
> >
> > # for i in /sys/kernel/debug/tracing/events/kprobes/* ; do date; echo 1 >
> > $i; done
> > Wed Jan 29 10:44:50 UTC 2014
> > ...
> >
> > I tried it and canceled after 4 min passed. It enabled about 17k
> >
* Masami Hiramatsu wrote:
> I guess the second reason is why the stap takes so long time to set
> probes. stap tries to register kprobes without disabled flag, that
> means we enables thousands of probes (overheads).
>
> So the similar thing happens when we enables events as below;
>
> #
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
I guess the second reason is why the stap takes so long time to set
probes. stap tries to register kprobes without disabled flag, that
means we enables thousands of probes (overheads).
So the similar thing happens when we enables
Hi -
So the similar thing happens when we enables events as below;
# for i in /sys/kernel/debug/tracing/events/kprobes/* ; do date; echo 1
$i; done
Wed Jan 29 10:44:50 UTC 2014
...
I tried it and canceled after 4 min passed. It enabled about 17k
events and slowed down
(2014/02/09 23:37), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
I guess the second reason is why the stap takes so long time to set
probes. stap tries to register kprobes without disabled flag, that
means we enables thousands of probes (overheads).
So
Hi Ingo,
(2013/12/20 19:46), Ingo Molnar wrote:
>
> * Masami Hiramatsu wrote:
>
>> (2013/12/20 17:20), Ingo Molnar wrote:
>>>
>>> * Masami Hiramatsu wrote:
>>>
> But a closer look indicates that the insertion of kprobes is
> taking about three (!!) orders of magnitude longer than
Hi Ingo,
(2013/12/20 19:46), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
(2013/12/20 17:20), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
But a closer look indicates that the insertion of kprobes is
taking about three
(2013/12/23 22:04), Ingo Molnar wrote:
>
> * Masami Hiramatsu wrote:
>
>>> I can reproduce one weirdness, with just 13 probes added, 'perf probe
>>> -l' [which should really be 'perf probe list'!] executes very slowly:
>>>
>>> # perf stat --null --repeat 3 perf probe -l
>>>
>>> Performance
* Masami Hiramatsu wrote:
> > I can reproduce one weirdness, with just 13 probes added, 'perf probe
> > -l' [which should really be 'perf probe list'!] executes very slowly:
> >
> > # perf stat --null --repeat 3 perf probe -l
> >
> > Performance counter stats for 'perf probe -l' (3 runs):
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
I can reproduce one weirdness, with just 13 probes added, 'perf probe
-l' [which should really be 'perf probe list'!] executes very slowly:
# perf stat --null --repeat 3 perf probe -l
Performance counter stats for 'perf
(2013/12/23 22:04), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
I can reproduce one weirdness, with just 13 probes added, 'perf probe
-l' [which should really be 'perf probe list'!] executes very slowly:
# perf stat --null --repeat 3 perf probe -l
(2013/12/20 22:40), Frank Ch. Eigler wrote:
> Hi -
>
> mingo wrote:
>> [...]
>> For example a hash table (hashed by probe address) could be used in
>> addition to the list, to speed up basic operations.
>
> In the past, when this sort of behavior popped up, it was due to
> machine-wide
(2013/12/20 19:46), Ingo Molnar wrote:
>
> * Masami Hiramatsu wrote:
>
>> (2013/12/20 17:20), Ingo Molnar wrote:
>>>
>>> * Masami Hiramatsu wrote:
>>>
> But a closer look indicates that the insertion of kprobes is
> taking about three (!!) orders of magnitude longer than before, as
(2013/12/20 19:46), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
(2013/12/20 17:20), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
But a closer look indicates that the insertion of kprobes is
taking about three (!!) orders
(2013/12/20 22:40), Frank Ch. Eigler wrote:
Hi -
mingo wrote:
[...]
For example a hash table (hashed by probe address) could be used in
addition to the list, to speed up basic operations.
In the past, when this sort of behavior popped up, it was due to
machine-wide halt/sync operations
Hi -
mingo wrote:
> [...]
> For example a hash table (hashed by probe address) could be used in
> addition to the list, to speed up basic operations.
In the past, when this sort of behavior popped up, it was due to
machine-wide halt/sync operations being done too eagerly. At one
point, the
* Masami Hiramatsu wrote:
> (2013/12/20 17:20), Ingo Molnar wrote:
> >
> > * Masami Hiramatsu wrote:
> >
> >>> But a closer look indicates that the insertion of kprobes is
> >>> taking about three (!!) orders of magnitude longer than before, as
> >>> judged by the rate of increase of 'wc
(2013/12/20 17:20), Ingo Molnar wrote:
>
> * Masami Hiramatsu wrote:
>
>>> But a closer look indicates that the insertion of kprobes is
>>> taking about three (!!) orders of magnitude longer than before, as
>>> judged by the rate of increase of 'wc -l
>>> /sys/kernel/debug/kprobes/list'.
>>
* Masami Hiramatsu wrote:
> > But a closer look indicates that the insertion of kprobes is
> > taking about three (!!) orders of magnitude longer than before, as
> > judged by the rate of increase of 'wc -l
> > /sys/kernel/debug/kprobes/list'.
>
> Right, because kprobes are not designed
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
But a closer look indicates that the insertion of kprobes is
taking about three (!!) orders of magnitude longer than before, as
judged by the rate of increase of 'wc -l
/sys/kernel/debug/kprobes/list'.
Right, because
(2013/12/20 17:20), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
But a closer look indicates that the insertion of kprobes is
taking about three (!!) orders of magnitude longer than before, as
judged by the rate of increase of 'wc -l
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
(2013/12/20 17:20), Ingo Molnar wrote:
* Masami Hiramatsu masami.hiramatsu...@hitachi.com wrote:
But a closer look indicates that the insertion of kprobes is
taking about three (!!) orders of magnitude longer than before,
Hi -
mingo wrote:
[...]
For example a hash table (hashed by probe address) could be used in
addition to the list, to speed up basic operations.
In the past, when this sort of behavior popped up, it was due to
machine-wide halt/sync operations being done too eagerly. At one
point, the
(2013/12/20 5:46), Frank Ch. Eigler wrote:
>
> Hi, Masami -
>
>
> masami.hiramatsu.pt wrote:
>
>> Here is the version 6 of NOKPROBE_SYMBOL series. :)
>> [...]
>
> Some preliminary results from building these on top of tip/master on
> x86-64.
>
> # stap -te "probe kprobe.function("*") {}"
>
Hi, Masami -
masami.hiramatsu.pt wrote:
> Here is the version 6 of NOKPROBE_SYMBOL series. :)
> [...]
Some preliminary results from building these on top of tip/master on
x86-64.
# stap -te "probe kprobe.function("*") {}"
starts up OK, without crashes, which looks like great progress.
Hi,
Here is the version 6 of NOKPROBE_SYMBOL series. :)
This includes small updates and introducing nokprobe_inline
macro to prevent probing on the static/inlined small
functions since NOKPROBE_SYMBOL will inhibit inlining
by referring function address.
This macro is more self-describing than
Hi,
Here is the version 6 of NOKPROBE_SYMBOL series. :)
This includes small updates and introducing nokprobe_inline
macro to prevent probing on the static/inlined small
functions since NOKPROBE_SYMBOL will inhibit inlining
by referring function address.
This macro is more self-describing than
Hi, Masami -
masami.hiramatsu.pt wrote:
Here is the version 6 of NOKPROBE_SYMBOL series. :)
[...]
Some preliminary results from building these on top of tip/master on
x86-64.
# stap -te probe kprobe.function(*) {}
starts up OK, without crashes, which looks like great progress. But a
(2013/12/20 5:46), Frank Ch. Eigler wrote:
Hi, Masami -
masami.hiramatsu.pt wrote:
Here is the version 6 of NOKPROBE_SYMBOL series. :)
[...]
Some preliminary results from building these on top of tip/master on
x86-64.
# stap -te probe kprobe.function(*) {}
starts up OK,
32 matches
Mail list logo