interesting error during make check for gmp-5.0.4 on SunOS5.8-i386

[Dennis Clarke]

> On Fri, 2 Mar 2012, Dennis Clarke wrote:
>
>> any thoughts?
>
> 386 is long dead, it makes no sense to compile for it. Solaris 8 (dead as
> well) probably didn't even run on it. Try 686 at least.

Well, from strictly a factual perspective, Solaris 8 is hardly dead. It
was put into vintage support by Sun a while back and as someone that
works in the enterprise server market I can tell you there are a bucket
load of big servers in banks, insurance, R&D, marketing etc that are
still running it. Probably because it never falls over.

Next point is that the ABI in Solaris 10 ( and yes, even 11 ) is
downwards compliant with Solaris 8. Anything that compiles and tests
clean on Solaris 8 is warranted, documented and absolutely going to
work flawlessly on Solaris 9 and Solaris 10 and Solaris 11. Thus
it seems reasonable, and in fact very efficient to compile and test
on Solaris 8.

Lastly, I have many many perfectly beautiful results on Solaris 8
for gmp/mpfr/mpc going back years and years and also these :

http://gcc.gnu.org/ml/gcc-testresults/2011-11/msg01044.html

http://gcc.gnu.org/ml/gcc-testresults/2011-07/msg00138.html

not as pretty but reasonable result :

http://gcc.gnu.org/ml/gcc-testresults/2011-07/msg00139.html

Those all use gmp/mpfr/mpc libs which all built perfectly.

The fly in the ointment is with x86_64 libs which must be
built on Solaris 10 of course however.

Dennis

-- 
--
http://pgp.mit.edu:11371/pks/lookup?op=vindex&search=0x1D936C72FA35B44B
+-------------------------+-----------------------------------+
| Dennis Clarke           | Solaris and Linux and Open Source |
| dclarke at blastwave.org   | Respect for open standards.       |
+-------------------------+-----------------------------------+