This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: fixincl now run at every stage?


 > From: Jeffrey A Law <law@cygnus.com>
 >  
 >   In message <200001311838.NAA12263@caip.rutgers.edu>you write:
 >   > 
 >   > Scanning recent Makefile.in changes, I'm going to hazard a guess that
 >   > this was the culprit:
 >   > 
 >   >  > 2000-01-20  Zack Weinberg  <zack@wolery.cumb.org>
 >   >  >  
 >   >  >         * Makefile.in (fixinc.sh): Depend on specs.
 >   > 
 >   > Because "specs" is part of $(STAGESTUFF), We move "specs" into the
 >   > stageN dir between stages so it gets rebuilt.  Then fixinc.sh gets
 >   > rebuilt, which triggers the stmp-fixinc target, which reruns fixinc.sh
 >   > at every stage.
 >   > 
 >   > It may correct this to remove "specs" from $(STAGESTUFF).  But I'm not
 >   > sure what side effect that would have.
 >   > 
 >   > Incidentally, this problem also causes libgcc.a to be rebuilt at every
 >   > stage due to dependencies on the header fixes.
 > I suspect the specs files are in STAGESTUFF so that we can find them in the
 > stageX directories via -BstageX/.  If we don't move them, then it's likely
 > we won't find the spec file during the stage2/stage3 build.


If specs doesn't change between stages, we can simply copy it into the
stageN dir instead of moving it.  That's what we do for libgcc.a.


 >  
 > libgcc does need to get rebuilt at each stage since it can contain
 > langauge specific stuff which wouldn't have been built during stage1.

Well IIRC, up until recently libgcc was only built in stage1 and
stage2 (where it got lang dependent stuff,) but not in stage3.

I know this because the warnings for libgcc.c didn't show up in a
stage3 warn_summary until just recently.  (I don't know how this
stage3 ommision is done though.)


 > I'm not sure how best to address this problem.
 > jeff


The gcc Makefile makes my head hurt!  :-)

But perhaps doing "cp specs stageN" might be the right fix.

		--Kaveh
--
Kaveh R. Ghazi			Engagement Manager / Project Services
ghazi@caip.rutgers.edu		Qwest Internet Solutions

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]