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

Daniel Richard G. skunk at iSKUNK.ORG
Wed Aug 12 23:15:32 CEST 2009


On Wed, 2009 Aug 12 11:34+0200, Torbjorn Granlund wrote:
> 
> What do we actually want --host=none to mean?  Clearly, it should
> avoid the host specific code in mpn/CPUFAMILY.  This is accomplished
> by configure's playing with the various path variables.  Then
> longlong.h should have LONGLONG_STANDALONE defined, else it might make
> references that need mpn/CPUFAMILY.  NO_ASM is meant to be something
> different, which might do strictly more switching-off than
> LONGLONG_STANDALONE, but not necessarily so.
>
> I suppose this should be fixed in different ways depending on
> whether we want --host=none to use (standalone) assembly code from
> longlong.h or not.

For my part, I would be inclined to think it shouldn't... I interpret
--host=none as "generic Unix", which would imply no assembly code. Is
there a reason why one would want to use assembly in one place, as
opposed to all or none?

> Perhaps we should use more pedestrian mechanisms for disabling
> mpn/CPUFAMILY, like --disable-assembly and for also disabling
> longlong.h, --disable-all-assembly?  --host=none is a little strange.

It would have the advantage of being easily discoverable from "configure
--help". If there are to be two switches, what about
--disable-mpn-assembly and --disable-longlong-assembly? (Maybe "-asm"
instead of "-assembly", to save some typing? :)

>   I'd be happy to provide any further information on this, as well as
>   test potential fixes.
>
> There will be no potential fixes for the FAILures before somebody with
> access debugs it.  How about you?

Debugging Alpha assembly code is a bit beyond my ken :>

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.

If you'd like to try some "remote debugging via e-mail," I'd be happy to
oblige. (This is a system at my employer's facility, so arranging direct
access would be difficult.)


--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