Hi Joachim, I think automatic regression notification at least to the author is a good idea. Probably I can do it in a nearest time. Unfortunately, right now I fail to get your code up. Also I believe it is a good style to check commits for regressions before pushing them. But maybe GHC community is less performance oriented.
25.07.2014, 01:41, "Joachim Breitner" <[email protected]>: > Hi Alexander, > > Am Donnerstag, den 24.07.2014, 19:22 +0400 schrieb Alexander Pakhomov: >> I've been informed you created http://ghcspeed-nomeata.rhcloud.com. I >> believe it is necessary for any serious perfomance work. >> Do you need some help? What have you done and what problems are >> unsolved? > thanks for your interest. > > One way to help is to monitor the page for regressions, and notify > whoever caused it. I’m doing that from time to time, relying on the > Latest Results summary. > > Then I don’t find the codespeed software to be perfect. The thread at > https://groups.google.com/forum/#!topic/codespeed/yY5-kPrcG94 discusses > some of the things I don’t like. So if you feel like hacking Python, > just hack away on it and I’ll happily accept pull requests at > https://github.com/nomeata/codespeed (branch ghc), or directly upstream. > > Eventually, the whole setup should be moved to some dedicated and > official hardware, but I’m not in a hurry with this. It’s been running > for just a week or two, and I’d like to observe how its behaving. > > Greetings, > Joachim > > PS: I prefer to discuss these things on the appropriate mailing list and > in public, if only for the archive. If you agree, simply reply to the > ghc-dev list (quoting in full) and I’ll read your reply there. > > -- > Joachim “nomeata” Breitner > [email protected] • http://www.joachim-breitner.de/ > Jabber: [email protected] • GPG-Key: 0xF0FBF51F > Debian Developer: [email protected] _______________________________________________ ghc-devs mailing list [email protected] http://www.haskell.org/mailman/listinfo/ghc-devs
