#1910: Native Code gen miscompiles double2Int# / float2Int# on x86_32
----------------------+-----------------------------------------------------
Reporter: int-e | Owner:
Type: bug | Status: new
Priority: normal | Milestone: 6.8.2
Component: Compiler | Version: 6.8.1
Severity: normal | Resolution:
Keywords: | Difficulty: Easy (1 hr)
Testcase: | Architecture: x86
Os: Multiple |
----------------------+-----------------------------------------------------
Changes (by int-e):
* os: MacOS X => Multiple
Comment:
improved patch note: It is important to load the word stored by fnstcw
with a word sized (16 bit) load, otherwise there's a stall and a noticable
performance hit.
The code is now on par with gcc's in terms of speed and 5 bytes shorter.
The speed measurements were done on an Athlon XP processor.
Oh the bug applies to Linux too.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/1910#comment:3>
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
