Hrmm. I wonder if that is in part caused by the inefficient use of sequence in https://github.com/ghc/nofib/blob/master/spectral/fibheaps/Main.lhs#L234 If that sequence . map is swapped for a traverse, what happens? -Edward On Mon, Sep 29, 2014 at 3:44 AM, Joachim Breitner wrote:

Hi,

Am Samstag, den 27.09.2014, 21:07 +0000 schrieb git@git.haskell.org:

...

commit f636faa7b2b7fc1d0663f994ad08f365d39a746d Author: Herbert Valerio Riedel Date: Sat Sep 27 22:55:19 2014 +0200

Set default-impl of `mapM`/`sequence` methods to `traverse`/`sequenceA`

This is made possible by the AMP, as we don't need the `WrappedMonad` helper for that anymore.

according to

http://ghcspeed-nomeata.rhcloud.com/changes/?rev=f636faa7b2b7fc1d0663f994ad08f365d39a746d&exe=2&env=nomeata%27s%20buildbot this has caused a 11% regression in fibheaps allocation:

http://ghcspeed-nomeata.rhcloud.com/timeline/?ben=nofib/allocs/fibheaps&env=1#/?exe=2&base=2+68&ben=nofib/allocs/fibheaps&env=1&revs=50&equid=on

Greetings, Joachim

-- Joachim “nomeata” Breitner mail@joachim-breitner.de • http://www.joachim-breitner.de/ Jabber: nomeata@joachim-breitner.de • GPG-Key: 0xF0FBF51F Debian Developer: nomeata@debian.org

_______________________________________________ ghc-devs mailing list ghc-devs@haskell.org http://www.haskell.org/mailman/listinfo/ghc-devs