Somebody claiming to be Stephen Paul Weber wrote:
I've decided to move on to the essential ARM port of my GHC cross-compiler. I got pretty far.

I also needed to install LLVM, since there is no native code generator for ARM. However, it seems like something in the build system is not properly aware that I should be building for ARM, since the errors I'm getting (see <http://pastie.org/pastes/5503341/text>) seem to contain x86 assembly.

This seems to have worked:

diff --git a/compiler/main/DriverPipeline.hs b/compiler/main/DriverPipeline.hs
index fd5694f..80a7b99 100644
--- a/compiler/main/DriverPipeline.hs
+++ b/compiler/main/DriverPipeline.hs
@@ -1331,9 +1331,9 @@ runPhase LlvmOpt input_fn dflags
     output_fn <- phaseOutputFilename LlvmLlc
     io $ SysTools.runLlvmOpt dflags
                ([ SysTools.FileOption "" input_fn,
                     SysTools.Option "-o",
-                    SysTools.FileOption "" output_fn]
+                    SysTools.FileOption "" output_fn, SysTools.Option 
"-mtriple=arm-unknown-nto-qnx8.0.0eabi"]
                 ++ optFlag
                 ++ [SysTools.Option tbaa]
                 ++ map SysTools.Option lo_opts)

Shouldn't we just always pass through the target triple?

--
Stephen Paul Weber, @singpolyma
See <http://singpolyma.net> for how I prefer to be contacted
edition right joseph

Attachment: signature.asc
Description: Digital signature

_______________________________________________
Cvs-ghc mailing list
Cvs-ghc@haskell.org
http://www.haskell.org/mailman/listinfo/cvs-ghc

Reply via email to