Hi Ben,
what's wrong is your build.mk file. I'm using:
GhcLibWays = v
SplitObjs = NO
HADDOCK_DOCS = NO
BUILD_DOCBOOK_HTML = NO
BUILD_DOCBOOK_PS = NO
BUILD_DOCBOOK_PDF = NO
here and it's working well. Last GHC/LLVM/HEADs testsuite results are:
OVERALL SUMMARY for test run started at Thu Oct 13 22:56:00 CEST 2011
3014 total tests, which gave rise to
7257 test cases, of which
0 caused framework failures
2696 were skipped
4300 expected passes
145 expected failures
1 unexpected passes
115 unexpected failures
Karel
On 10/14/11 06:34 AM, Ben Gamari wrote:
On Thu, 13 Oct 2011 20:38:31 +0200, Karel Gardas<[email protected]>
wrote:
-fvia-C means you are building unregisterised build. Perhaps, you might
try registerised build or even that was your intention? For this you
will need to apply my submitted patch or do this yourself...
I must say, I think I must be doing something wrong. For my this latest
build I didn't intend on doing an unregisterised build yet I am still
getting this same error:
"inplace/bin/ghc-stage1" -H64m -O0 -Rghc-timing -package-name base-4.4.0.0
-hide-all-packages -i -ilibraries/base/. -ilibraries/base/dist-install/build
-ilibraries/base/dist-install/build/autogen -Ilibraries/base/dist-install/build
-Ilibraries/base/dist-install/build/autogen -Ilibraries/base/include
-optP-DOPTIMISE_INTEGER_GCD_LCM -optP-include
-optPlibraries/base/dist-install/build/autogen/cabal_macros.h -package ghc-prim-0.2.0.0
-package integer-gmp-0.3.0.0 -package rts-1.0 -package-name base -XHaskell98 -XCPP -O
-fasm -no-user-package-conf -rtsopts -odir libraries/base/dist-install/build -hidir
libraries/base/dist-install/build -stubdir libraries/base/dist-install/build -hisuf hi
-osuf o -hcsuf hc -c libraries/base/./Foreign/Marshal/Alloc.hs -o
libraries/base/dist-install/build/Foreign/Marshal/Alloc.o
on the commandline:
Warning: Compiler has no native codegen, so ignoring -fasm
/tmp/ghc430_0/ghc430_0.hc:469:1:
warning: built-in function ‘free’ declared as non-function
/tmp/ghc430_0/ghc430_0.hc: In function
‘base_ForeignziMarshalziAlloc_zdwa_entry’:
/tmp/ghc430_0/ghc430_0.hc:544:2:
warning: conflicting types for built-in function ‘malloc’
/tmp/ghc430_0/ghc430_0.hc: In function ‘s1LZ_ret’:
/tmp/ghc430_0/ghc430_0.hc:845:2:
warning: conflicting types for built-in function ‘realloc’
/tmp/ghc430_0/ghc430_0.hc: In function
‘base_ForeignziMarshalziAlloc_zdwccall_entry’:
/tmp/ghc430_0/ghc430_0.hc:992:2:
error: ‘free’ redeclared as different kind of symbol
/tmp/ghc430_0/ghc430_0.hc:469:1:
note: previous declaration of ‘free’ was here
As you see, -fasm is passed to the stage 1 compiler yet it seems to
think that I have no native codegen so it seems that the stage 1 build
didn't realize it could build the LLVM backend. My LLVM bin/ directory
is in $PATH, so why would this be the case? Where in the configuration
process does LLVM detection take place? Searching in config.log doesn't
turn up anything. I am on GHC commit
2a9f42095cb2d9aace991b11bf052d12ca654ef8.
Cheers,
- Ben
[1] Build.mk
SRC_HC_OPTS = -H64m -O0 -Rghc-timing
GhcStage1HcOpts = -O -fvia-C
GhcStage2HcOpts = -O0 -fasm
GhcLibHcOpts = -O -fasm
GhcLibWays = v
SplitObjs = NO
_______________________________________________
Cvs-ghc mailing list
[email protected]
http://www.haskell.org/mailman/listinfo/cvs-ghc
