[GHC] #10800: vector-0.11 compile time increased substantially with 7.10.1
GHC
ghc-devs at haskell.org
Thu Mar 24 10:17:24 UTC 2016
#10800: vector-0.11 compile time increased substantially with 7.10.1
-------------------------------------+-------------------------------------
Reporter: bgamari | Owner: bgamari
Type: bug | Status: new
Priority: highest | Milestone: 8.0.1
Component: Compiler | Version: 7.10.2
Resolution: | Keywords:
Operating System: Unknown/Multiple | Architecture:
Type of failure: Compile-time | Unknown/Multiple
performance bug | Test Case:
Blocked By: | Blocking:
Related Tickets: | Differential Rev(s):
Wiki Page: |
-------------------------------------+-------------------------------------
Comment (by erikd):
From IRC:
{{{
<bgamari> erikd, IIRC I have a more minimal testcase as well
<bgamari> the problem centers around zip
<bgamari> when things are inlined has changed
<bgamari> which results in a large number of stream fusion primitives that
can't be resolved until late in compilation
<bgamari> which produces a large amount of code
<bgamari> erikd, that should at least be a hint
<bgamari> unfortunately I ran out of time before identifying the root
cause
<bgamari> erikd, https://gist.github.com/bgamari/2ca0f7d2a4e0b6447aee
<bgamari> erikd, that's my minimized testcase
<erikd> thanks, i'll check it out
<bgamari> It helps to look at the -dverbose-core2core output
<bgamari> comparing between the relevant compiler versions
<bgamari> erikd, this utility may come in handy for turning the
verbose-core2core output into something less unwieldy
<bgamari> https://github.com/bgamari/ghc-utils/blob/master/split-
core2core.py
}}}
--
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/10800#comment:14>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
More information about the ghc-tickets
mailing list