A relevant trac ticket and wiki page can be found here:
http://hackage.haskell.org/trac/ghc/ticket/3557
http://hackage.haskell.org/trac/ghc/wiki/VectorComputing
Vivian seems to now have partial patches for #3557. I haven't reviewed
the discussion in the ticket or the patches in particular but it'
The main thing about FFI support is calling conventions, and LLVM supports
a variety of them, so if your LLVM code is using a calling convention
GHC supports, you should be able to manage it (there probably isn't any
toolchain support though, so you'll have to build the LLVM bits manually
and then
I'm curious: as I understand it, one of the purposes of the FFI is to allow
bindings to "close to the metal" languages. I'm curious what would be
involved in letting the FFI provide bindings to LLVM code in Haskell.
While I'm at it, I'll mention the most recent reason that this thought
occurred t
On Tue, Mar 29, 2011 at 10:12:15PM -0400, wren ng thornton wrote:
>
> 2695 total tests, which gave rise to
>14978 test cases, of which
>0 caused framework failures
>12589 were skipped
>
> 2302 expected passes
> 74 expected failures
>0 unexpected passes
>
Hi,
could it be that the llvm version used for the hackage buildbot is too
old?
The newest version of my 'Biobase' library does not compile there:
http://hackage.haskell.org/packages/archive/Biobase/0.3.1.0/logs/failure/ghc-7.0
and this seems to be due to llvm being too old:
http://comments.gmane.
On 30/03/2011 03:12, wren ng thornton wrote:
On 3/28/11 10:33 PM, Jens Petersen wrote:
FYI testsuite results:
[...]
8 unexpected failures on x86:
DoParamM(normal)
T3064(normal)
T3330a(normal)
T3738(normal)
T4316(ghci)
T4801(normal)
break024(ghci)
space_leak_001(normal)
FYI, testsuite results f
