#5327: INLINABLE pragma and newtypes prevents inlining
---------------------------------+------------------------------------------
Reporter: reinerp | Owner:
Type: bug | Status: new
Priority: normal | Milestone: 7.4.1
Component: Compiler | Version: 7.1
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: Unknown/Multiple | Failure: Runtime performance bug
---------------------------------+------------------------------------------
Comment(by simonpj@…):
commit 29edeadbc7d974232f67cdc4fd05283b35721ae0
{{{
Author: Simon Peyton Jones <[email protected]>
Date: Mon Jul 25 09:33:09 2011 +0100
Re-engineer exprIsConApp_maybe (fixes Trac #5327)
The problem with #5327 was like this:
let x = I# 0 `cast` co1
in ...(case x `cast` co2 of I# y -> blah)...
The two casts cancelled out, but exprIsConApp_maybe couldn't see
that. This patch makes it simpler, faster, and more effective.
(Incidentally, usually 'x' would be inlined, in #5327 it wasn't
because of an INLINEABLE pragma and the lone-variable thing.
Instead of fiddling with alrady-delicate code, I just made
exprIsConApp_maybe better.)
compiler/coreSyn/CoreUnfold.lhs | 153
+++++++++++++++++++--------------------
1 files changed, 76 insertions(+), 77 deletions(-)
}}}
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/5327#comment:2>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler
_______________________________________________
Glasgow-haskell-bugs mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/glasgow-haskell-bugs
