T3064 failures

[Joachim Breitner]

Hi,

Am Donnerstag, den 16.10.2014, 23:15 -0400 schrieb David Feuer:
> I don't know what's going on, but T3064 is giving some substantial
> performance trouble, making all the validations fail:
> 
> max_bytes_used value is too high:
>     Expected    T3064(normal) max_bytes_used: 13251728 +/-20%
>     Lower bound T3064(normal) max_bytes_used: 10601382 
>     Upper bound T3064(normal) max_bytes_used: 15902074 
>     Actual      T3064(normal) max_bytes_used: 17472560 
>     Deviation   T3064(normal) max_bytes_used:     31.9 %
> bytes allocated value is too high:
>     Expected    T3064(normal) bytes allocated:  385145080 +/-5%
>     Lower bound T3064(normal) bytes allocated:  365887826 
>     Upper bound T3064(normal) bytes allocated:  404402334 
>     Actual      T3064(normal) bytes allocated: 1372924224 
>     Deviation   T3064(normal) bytes allocated:      256.5 %
> 

I can’t observe the bytes allocated too high on ghc-speed:

    Expected    T3064(normal) max_bytes_used: 18744992 +/-20%
    Lower bound T3064(normal) max_bytes_used: 14995993 
    Upper bound T3064(normal) max_bytes_used: 22493991 
    Actual      T3064(normal) max_bytes_used: 19011008 
    Deviation   T3064(normal) max_bytes_used:      1.4 %
    Expected    T3064(normal) peak_megabytes_allocated: 42 +/-20%
    Lower bound T3064(normal) peak_megabytes_allocated: 33 
    Upper bound T3064(normal) peak_megabytes_allocated: 51 
    Actual      T3064(normal) peak_megabytes_allocated: 42 
    Expected    T3064(normal) bytes allocated: 385145080 +/-5%
    Lower bound T3064(normal) bytes allocated: 365887826 
    Upper bound T3064(normal) bytes allocated: 404402334 
    Actual      T3064(normal) bytes allocated: 383504280 
    Deviation   T3064(normal) bytes allocated:      -0.4 %


http://ghcspeed-nomeata.rhcloud.com/timeline/?exe=2&base=2%2B68&ben=tests%2Falloc%2FT3064&env=1&revs=50&equid=on

Are you sure you build with the exact same settings as validate would
do?

Also interesting that according to
https://phabricator.haskell.org/harbormaster/
only builds of DRs are failing, the builds of recent commits to master
succeed. Are these built with slightly different settings, or is that
just by accident?

Greetings,
Joachim

-- 
Joachim Breitner
  e-Mail: mail at joachim-breitner.de
  Homepage: http://www.joachim-breitner.de
  Jabber-ID: nomeata at joachim-breitner.de

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: This is a digitally signed message part
URL: <http://www.haskell.org/pipermail/ghc-devs/attachments/20141017/3e10e0d7/attachment.sig>