4.3.1 test failures on alphaev56-dec-osf4.0g (Tru64)

Daniel Richard G. skunk at iSKUNK.ORG
Thu Aug 13 02:08:30 CEST 2009


On Wed, 2009 Aug 12 23:46+0200, Torbjorn Granlund wrote:
> 
> I would suspect people use --host=none to work around some problems
> with the assembly code.  Such problems are not uncommon, since people
> keep inventing new relocation limitations.
>
> I don't see such problems with inline assembly.  Killing inline
> assembly will make GMP perform dreadfully.

Ah, so there is good reason to assert two different categories of
assembly code, then.

>   I tried building with CFLAGS="-O0 -g2 -DNO_ASM" and --host=none,
>   with both cc(1) and GCC---but everything passes, so I can't get at
>   the problem in straight C.
>
> I suspect the assembler might be the culprit.  If I understood your
> previous messages correctly, the crash happens in invert_limb.asm,
> which has a couple of relocations.  Either the m4 used is buggy (that
> is surprisingly common!) or the assembler is buggy or has new
> relocations limitations.  (I tried a gcc 4.4 cross compiler tergeted
> for alpha-dec-osf5, and it generates relocs identical to the ones I
> use in invert_limb.asm.  Also, I am 100% sure the file used to work
> with OSF in the past, with both gcc and cc.)

Would it help if I provided a complete build tree? Comparing the
compiled assembly to known-good builds, and perhaps even somehow linking
and executing some .o files on Linux/Alpha for testing?


--Daniel


-- 
NAME = Daniel Richard G.     _\|/_    Remember, skunks
MAIL = skunk at iSKUNK.ORG     (/o|o\) _- don't smell bad---
MAIL+= skunk at alum.MIT.EDU   < (^),>     it's the people who
WWW  = (not there yet!)      /   \      annoy us that do!


More information about the gmp-bugs mailing list