On 01/24/2013 05:04 PM, Karel Gardas wrote:
On 01/24/13 04:50 PM, Stephen Paul Weber wrote:
Somebody claiming to be Nathan Hüsken wrote:
On 01/24/2013 04:28 PM, Stephen Paul Weber wrote:
Do you think it is specifically the 3.2 that made it work? Yes. With llvm version 3.1 I was only able to get an unregisterised build to work. http://hackage.haskell.org/trac/ghc/attachment/ticket/7621/unregistered-arm-...
?
Not exactly, see the patch here: http://www.haskell.org/pipermail/ghc-devs/2013-January/000118.html and the changes to compiler/llvmGen/LlvmCodeGen/Ppr.hs
Oh, man, the fact that I don't have that setting for QNX is probably not doing me any favours...
IIRC it's more about optimization, i.e. proper data type alignment.
How the heck am I supposed to figure out what that string should be? :(
Described here: http://llvm.org/docs/LangRef.html#data-layout -- for LLVM 3.3. General rule, I would start with what Linux requires and check from that.
That said... how did you get an unregisterised build to work with an LLVM backend? Everything I've seen in the code implied that the moment you are unregisteried, it uses via-C... Which is what my above patch is primarily about.
IMHO Nathan is using registerised build. Nathan correct me please if I'm wrong as I'd like to also give your Android work a try in the future (being Android user myself...)
Yes, that is correct. registerised with llvm 3.2.