#4914: FPU initialization required again
---------------------------------+------------------------------------------
Reporter: aruiz | Owner: simonmar
Type: bug | Status: new
Priority: high | Milestone: 7.2.1
Component: Compiler (NCG) | Version: 7.0.1
Keywords: | Testcase:
Blockedby: | Difficulty:
Os: Unknown/Multiple | Blocking:
Architecture: x86 | Failure: Incorrect result at runtime
---------------------------------+------------------------------------------
Comment(by aruiz):
Thanks for your help! I am now checking the tag word and also if a NaN is
produced (calling frstor after fsave to let the computation go on in the
original state). In an extremely large number of cases the stack is empty
and there is no problem. But in a few rare cases the stack is non empty
and then the computation produces a wrong NaN!
Two examples of unexpected tag word:
{{{
tag : 255 31 = 11 11 11 11 00 01 11 11
tag : 252 63 = 11 11 11 00 00 11 11 11
}}}
This can be observed in the hmatrix quickCheck tests. I have also found a
small deterministic test case not using quickCheck, but it still requires
the hmatrix package. Curiously, when I try to simplify the problem
defining locally some functions the problem disappears.
Note also that the problem disappears if cabal install with {{{--disable-
optimization}}}.
I will simplify the testcase as much as possible.
--
Ticket URL: <http://hackage.haskell.org/trac/ghc/ticket/4914#comment:12>
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
