[GHC] #8095: TypeFamilies painfully slow

GHC ghc-devs at haskell.org
Fri Mar 8 12:14:45 UTC 2019


#8095: TypeFamilies painfully slow
-------------------------------------+-------------------------------------
        Reporter:  MikeIzbicki       |                Owner:  goldfire
            Type:  bug               |               Status:  new
        Priority:  high              |            Milestone:  8.8.1
       Component:  Compiler (Type    |              Version:  7.6.3
  checker)                           |
      Resolution:                    |             Keywords:  TypeFamilies
Operating System:  Unknown/Multiple  |         Architecture:
 Type of failure:  Compile-time      |  Unknown/Multiple
  performance bug                    |            Test Case:
      Blocked By:                    |             Blocking:
 Related Tickets:  #5321, #11598,    |  Differential Rev(s):  Phab:D3752,
  #12506, #13386                     |  Phab:D4766
       Wiki Page:                    |
-------------------------------------+-------------------------------------

Comment (by mwu):

 The observation about stack behavior is interesting, I was curious to what
 degree it affects our builds. I have built and tested the stack's fork
 with binary interface support (many thanks to hussein.aitlahcen!) on our
 codebase ([https://github.com/luna/luna]). I have observed visible
 improvements both in build time (32 min vs 39 min) and memory usage
 (peaking at 12.3 GB instead of 14.2 GB) — however it is still very slow.
 With `-fomit-interface-pragmas` everything builds in little over 4 minutes
 but the binary performance is much worse.

 So I'm afraid GHC-side solution still remains very much needed. Is there
 still a chance that the fix will land in GHC 8.8?

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8095#comment:64>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list