On Sat, 12 Jan 2013, Ben Gamari wrote:
rocon...@theorem.ca writes:
What version of GHC did you build?
This was building from the ghc-7.4 branch. That being said, after this
build finished I noticed that the resulting stage 2 compiler wasn't
invoking llc with the correct parameters, resulting in the same linker
error when doing builds.
This might have been due to the fact that I changed `build.mk` several
times mid-build. I'm currently waiting for a clean rebuild. However, I'm
now using a more specific configuration as passing `-optlc-float-abi=hard`
in `HC_SRC_OPTS` caused the stage 0 build to fail with,
...
*** LLVM Compiler:
'llc-3.0' '-O1' '-relocation-model=static' '/tmp/ghc25792_0/ghc25792_0.bc'
'-o' '/tmp/ghc25792_0/ghc25792_0.lm_s' '-float-abi=hard' '-mcpu=cortex-a9'
'-mattr=+v7,+vfp3,+d16' '-float-abi=hard'
llc-3.0: for the -float-abi option: may only occur zero or one times!
I'm not really sure where this second `-float-abi=hard` is coming from,
but using the following in `build.mk` works around the issue,
SRC_HC_OPTS = -H64m -Rghc-timing
GhcStage1HcOpts = -O -fvia-C
GhcStage2HcOpts = -O0 -fllvm -keep-llvm-files -debug -DDEBUG
-optc-mfloat-abi=hard -optc-mcpu=cortex-a9 -optlc-float-abi=hard
-optlc-mcpu=cortex-a9
GhcLibHcOpts = -O -fllvm -optc-mfloat-abi=hard -optc-mcpu=cortex-a9
-optlc-float-abi=hard -optlc-mcpu=cortex-a9
You've written "-optc-mfloat-abi=hard -optc-mcpu=cortex-a9" twice in your
GhcLibHcOpts.
--
Russell O'Connor <http://r6.ca/>
``All talk about `theft,''' the general counsel of the American Graphophone
Company wrote, ``is the merest claptrap, for there exists no property in
ideas musical, literary or artistic, except as defined by statute.''
_______________________________________________
Glasgow-haskell-users mailing list
Glasgow-haskell-users@haskell.org
http://www.haskell.org/mailman/listinfo/glasgow-haskell-users
