On 03/21/2017 11:21 PM, Arnaldo Carvalho de Melo wrote:
Em Tue, Mar 21, 2017 at 11:14:07AM -0300, Arnaldo Carvalho de Melo escreveu:
Em Tue, Mar 21, 2017 at 07:20:20AM +0900, Taeung Song escreveu:
And,
I tested by perf-stat on the same situation as below.
$ perf stat -e
On 03/21/2017 11:21 PM, Arnaldo Carvalho de Melo wrote:
Em Tue, Mar 21, 2017 at 11:14:07AM -0300, Arnaldo Carvalho de Melo escreveu:
Em Tue, Mar 21, 2017 at 07:20:20AM +0900, Taeung Song escreveu:
And,
I tested by perf-stat on the same situation as below.
$ perf stat -e
Hi,
On 03/21/2017 11:21 PM, Arnaldo Carvalho de Melo wrote:
Em Tue, Mar 21, 2017 at 11:14:07AM -0300, Arnaldo Carvalho de Melo escreveu:
Em Tue, Mar 21, 2017 at 07:20:20AM +0900, Taeung Song escreveu:
And,
I tested by perf-stat on the same situation as below.
$ perf stat -e
Hi,
On 03/21/2017 11:21 PM, Arnaldo Carvalho de Melo wrote:
Em Tue, Mar 21, 2017 at 11:14:07AM -0300, Arnaldo Carvalho de Melo escreveu:
Em Tue, Mar 21, 2017 at 07:20:20AM +0900, Taeung Song escreveu:
And,
I tested by perf-stat on the same situation as below.
$ perf stat -e
Em Tue, Mar 21, 2017 at 11:14:07AM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Tue, Mar 21, 2017 at 07:20:20AM +0900, Taeung Song escreveu:
> > And,
> > I tested by perf-stat on the same situation as below.
> >
> > $ perf stat -e "{cycles,page-faults,branch-misses}" ./old > 6623856
>
>
Em Tue, Mar 21, 2017 at 11:14:07AM -0300, Arnaldo Carvalho de Melo escreveu:
> Em Tue, Mar 21, 2017 at 07:20:20AM +0900, Taeung Song escreveu:
> > And,
> > I tested by perf-stat on the same situation as below.
> >
> > $ perf stat -e "{cycles,page-faults,branch-misses}" ./old > 6623856
>
>
Em Tue, Mar 21, 2017 at 07:20:20AM +0900, Taeung Song escreveu:
> And,
> I tested by perf-stat on the same situation as below.
>
> $ perf stat -e "{cycles,page-faults,branch-misses}" ./old6623856
Please always try to spell out all the steps needed to get to some
result, for instance, in
Em Tue, Mar 21, 2017 at 07:20:20AM +0900, Taeung Song escreveu:
> And,
> I tested by perf-stat on the same situation as below.
>
> $ perf stat -e "{cycles,page-faults,branch-misses}" ./old6623856
Please always try to spell out all the steps needed to get to some
result, for instance, in
And,
I tested by perf-stat on the same situation as below.
$ perf stat -e "{cycles,page-faults,branch-misses}" ./old sum' is zero..
util/annotate.c:1660~1661
1660h = annotation__histogram(notes, evidx + k);
1661src_line->samples[k].percent = 100.0 * h->addr[i] / h->sum;
And,
I tested by perf-stat on the same situation as below.
$ perf stat -e "{cycles,page-faults,branch-misses}" ./old sum' is zero..
util/annotate.c:1660~1661
1660h = annotation__histogram(notes, evidx + k);
1661src_line->samples[k].percent = 100.0 * h->addr[i] / h->sum;
Hi Arnaldo :)
Here the perf.data is,
https://www.dropbox.com/s/nr4nnv8g3cipluf/perf.data?dl=1=1
I tested as below.
$ perf record -e "{cycles,page-faults,branch-misses}" ./old /dev/null | grep -i nan | head -3
29.04-nan1.52 old_pack_knapsack.c:34
28.27-nan0.00
Hi Arnaldo :)
Here the perf.data is,
https://www.dropbox.com/s/nr4nnv8g3cipluf/perf.data?dl=1=1
I tested as below.
$ perf record -e "{cycles,page-faults,branch-misses}" ./old /dev/null | grep -i nan | head -3
29.04-nan1.52 old_pack_knapsack.c:34
28.27-nan0.00
Em Mon, Mar 20, 2017 at 11:56:55AM +0900, Taeung Song escreveu:
> Currently perf-annotate with --print-line can print
> -nan(0x8) because of division by zero
> when calculating percent.
>
> So if a sum of samples is zero, skip calculating percent.
Tried to reproduce it here,
Em Mon, Mar 20, 2017 at 11:56:55AM +0900, Taeung Song escreveu:
> Currently perf-annotate with --print-line can print
> -nan(0x8) because of division by zero
> when calculating percent.
>
> So if a sum of samples is zero, skip calculating percent.
Tried to reproduce it here,
Currently perf-annotate with --print-line can print
-nan(0x8) because of division by zero
when calculating percent.
So if a sum of samples is zero, skip calculating percent.
Before:
$ perf annotate --stdio -l
Sorted summary for file /home/taeung/workspace/a.out
Currently perf-annotate with --print-line can print
-nan(0x8) because of division by zero
when calculating percent.
So if a sum of samples is zero, skip calculating percent.
Before:
$ perf annotate --stdio -l
Sorted summary for file /home/taeung/workspace/a.out
16 matches
Mail list logo