[GHC] #13208: Do two-phase inlining in simpleOptPgm

GHC ghc-devs at haskell.org
Wed Mar 6 10:36:32 UTC 2019


#13208: Do two-phase inlining in simpleOptPgm
-------------------------------------+-------------------------------------
        Reporter:  lukemaurer        |                Owner:  (none)
            Type:  bug               |               Status:  closed
        Priority:  normal            |            Milestone:
       Component:  Compiler          |              Version:  8.1
      Resolution:  fixed             |             Keywords:  JoinPoints
Operating System:  Unknown/Multiple  |         Architecture:
                                     |  Unknown/Multiple
 Type of failure:  None/Unknown      |            Test Case:
                                     |  deSugar/should_compile/T13208
      Blocked By:                    |             Blocking:
 Related Tickets:                    |  Differential Rev(s):
       Wiki Page:                    |  https://gitlab.haskell.org/ghc/ghc/merge_requests/394
-------------------------------------+-------------------------------------

Comment (by simonpj):

 > I agree that comment:2 was addressed, but what about comment:1? It seems
 to me that this was not addressed and was the original reason for leaving
 the ticket open.

 Yes, you are right.

 * I'd like to get rid of `Note [Beta redexes]` in Core Lint.

 * The change to `simpleOptPgm` makes that more feasible.

 * But we do worker/wrapper join points, and currently w/w generates
 exactly these kind of beta redexes (contrary, I think, to the claim that
 it was only `simpleOptPgm`)

 I'm not sure it matters all that much, but yes we should leave the ticket
 open

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/13208#comment:8>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler


More information about the ghc-tickets mailing list