Missing graphs with >4 Executables?

8 views
Skip to first unread message

Steffen Schwigon

unread,
Dec 1, 2011, 8:02:06 AM12/1/11
to code...@googlegroups.com
Hi!

Somehow my codespeed instance can not render the timeline of a single
benchmark with > 4 executables.

If I unselect some executables to have only <=4 it works.

The overview page with timelines for all benchmarks works with >4
executables.

Try it here:

http://speed.perlformance.net/timeline/
http://speed.perlformance.net/timeline/#/?exe=20,18,21,19,14&base=none&ben=%28F%29Fib&env=4&revs=50&equid=off

Anyone experienced this?
Where should I look to fix this?

I run codespeed from latest git.

Thanks.

Kind regards,
Steffen
--
Steffen Schwigon <s...@renormalist.net>
Dresden Perl Mongers <http://dresden-pm.org/>

Steffen Schwigon

unread,
Dec 1, 2011, 10:47:07 AM12/1/11
to code...@googlegroups.com
Steffen Schwigon <s...@renormalist.net> writes:
> Hi!
>
> Somehow my codespeed instance can not render the timeline of a single
> benchmark with > 4 executables.
>
> If I unselect some executables to have only <=4 it works.
>
> The overview page with timelines for all benchmarks works with >4
> executables.
>
> Try it here:
>
> http://speed.perlformance.net/timeline/
> http://speed.perlformance.net/timeline/#/?exe=20,18,21,19,14&base=none&ben=%28F%29Fib&env=4&revs=50&equid=off

Update obervations:

* It's not about how many Executables you choose in the navigation but
how many of them actually provide values.

- If in a benchmark only <=4 executables provide data, then it works.

- However, limiting them via "Show the last 10 results" does not help.


* At the time of this writing it can be seen here:

http://speed.perlformance.net/timeline/#/?exe=20,22,18,21,19,14&base=none&ben=%28F%29Prime.crypt_primes&env=4&revs=50&equid=off

Miquel Torres

unread,
Dec 1, 2011, 3:00:06 PM12/1/11
to code...@googlegroups.com
Hi,

it is a bug introduced when branches where added.

I filed an issue with a detailed description here:
https://github.com/tobami/codespeed/issues/86

I'll write to this thread when it is fixed, thanks for the detailed report!

Cheers


2011/12/1 Steffen Schwigon <s...@renormalist.net>:

Reply all
Reply to author
Forward
0 new messages