Re: Broken ghc-7.0.3/vector combination?

On 21/04/2011 12:29, Felipe Almeida Lessa wrote:

On Thu, Apr 21, 2011 at 8:08 AM, Simon Marlow wrote:

...

Right, it could be related to this. However this change was made to eliminate some causes of NaNs, see:

http://hackage.haskell.org/trac/ghc/ticket/4914

So I'm very depressed if it managed to introduce NaNs somehow.

Could someone make a ticket for this, with the smallest test case found so far please?

So in principle the LLVM backend should be fine?

Yes, also compiling with -msse2 on x86 should be fine. Cheers, Simon