6.4.2 under solaris

[Duncan Coutts]

On Tue, 2006-05-23 at 11:52 +0200, Christian Maeder wrote:
> Christian Maeder wrote:
> > I'm currently running the ghc-regression test (with stage1/ghc-inplace), 
> > but that does not look good (although it does not hang)
> 
> here is the frustrating summary:
> 
> OVERALL SUMMARY for test run started at Mon May 22 18:30:30 CEST 2006
>      1365 total tests, which gave rise to
>      4157 test cases, of which
>         0 caused framework failures
>       580 were skipped
> 
>      3051 expected passes
>        51 expected failures
>         0 unexpected passes
>       475 unexpected failures

This may be the same as a problem I found recently on Sparc Linux.

I built ghc-6.4.2 on Linux Sparc and had all the GHCi tests fail too. I
found that it was because my gcc was using the V8+ Sparc ABI and GHCi up
'til now only accepted V7 ABI .o files. So starting up GHCi gave an
error when trying to link/load HSbase.o. It complained of an unknown
architecture. I submitted a patch recently to fix this by adding a case
for EM_SPARC32PLUS in addition to the existing EM_SPARC.

You can check which Sparc ABI is needed by using the 'file' utility.
On my Linux Sparc machine it reports:

file /usr/lib/ghc-6.4.2/HSbase.o
/usr/lib/ghc-6.4.2/HSbase.o: ELF 32-bit MSB relocatable, SPARC32PLUS, V8
+ Required, version 1 (SYSV), not stripped

Duncan