[Bug java/11933] Parallel bootstrap failure linking gij

["lucier at math dot purdue dot edu" <gcc-bugzilla at gcc dot gnu dot org>]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=11933



------- Additional Comments From lucier at math dot purdue dot edu  2003-08-24 20:46 -------
Subject: Re:  Parallel bootstrap failure linking gij

> ------- Additional Comments From neroden at gcc dot gnu dot org  
> 2003-08-24 20:15 -------
> Not a regression, really, since parallel bootstrap was mostly disabled 
> in the Bad Old Days.  Retargeting.

It's a good thing I'm not Linus or I'd say you were smoking crack;-).  
(Re his recent comments about SCO.)

I've been testing solely with parallel builds for years, it's necessary 
on slowaris because of the generally slow processor speeds and slow 
system calls, coupled with having four-processor servers that you can 
bang on with 10-40 processes at a time if necessary.  And it's always 
worked, modulo bugs. All previous parallel build problems have been 
fixed.

This one arose since the beginning of this year.  It is interfering 
with my efforts to set up a regression tester here at Purdue on some 
recently-acquired 64-processor servers.

It is most certainly a regression.

I'll work with you if you like to try to narrow down when the bug 
arose, but the other multilib bug is preventing this, since it occurs 
earlier and with fewer parallel processes (three will trigger it pretty 
regularly).

Brad