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: Build Failure for 2.96 (20000731) on sparc-sun-sunos4.1.4


I believe this should be fixed by the patch, Alexandre sent yesterday:

  From: Alexandre Oliva <aoliva@redhat.com>
  Sender: aoliva@guarana.lsd.ic.unicamp.br
  Organization: GCC Team, Red Hat
  To: Manfred Hollstein <manfred.h@gmx.net>
  Cc: gcc-patches@gcc.gnu.org
  Subject: Re: Accept blanks as language separators
  Date: 01 Aug 2000 13:56:24 -0300

  Indeed.  Here's a patch that presumably handles LANGUAGES correctly.
  Ok to install, if it works?  (I haven't tested it, because I've got to
  leave in a couple of minutes; I'd appreciate if someone else could
  give it a try while I'm away)

  >from  Alexandre Oliva  <aoliva@redhat.com>
	* configure.in (CHILL_FOR_TARGET, CXX_FOR_TARGET): Convert blanks to
	commas in $LANGUAGES.

BTW, Jason Merrill approved it in the meantime, so it may have been
checked in already.

Cheers, manfred

On Wednesday, 2 August 2000, 11:20:45 -0400, ghazi@caip.rutgers.edu wrote:

>  > "make bootstrap" fails on sparc-sun-sunos4.1.4:
>  > 
>  > ...
>  > make[2]: Entering directory
>  > `/home3/egcs-sunos/sparc-sun-sunos4.1.4/libio'
>  > if [ x"" != x ] && [ ! -d pic ]; then \
>  >   mkdir pic; \
>  > else true; fi
>  > touch stamp-picdir
>  > rootme=`pwd`/ ; export rootme; \
>  > CC="/home3/egcs-sunos/gcc/xgcc -B/home3/egcs-sunos/gcc/
>  > -B/afs/btv.ibm.com/data/n37v/software/gnu-sunos/egcs/sparc-sun-sunos4.1.4/bin/
>  > -B/afs/btv.ibm.com/data/n37v/software/gnu-sunos/egcs/sparc-sun-sunos4.1.4/lib/
>  > -isystem
>  > /afs/btv.ibm.com/data/n37v/software/gnu-sunos/egcs/sparc-sun-sunos4.1.4/include
>  > -I. -I../../../egcs/libio"; export CC; \
>  > CXX=" -I. -I../../../egcs/libio -nostdinc++ -nostdinc++ -g -O2";
>  > export CXX; \
>  > CONFIG_NM="nm"; export CONFIG_NM; \
>  > /bin/sh ../../../egcs/libio/gen-params LIB_VERSION=2.8.0
>  > >tmp-params.h
>  > ../../../egcs/libio/gen-params: -I.: not found
>  > gen-params: could not compile dummy.C with  -I. -I../../../egcs/libio
>  > -nostdinc++ -nostdinc++ -g -O2
>  > make[2]: *** [_G_config.h] Error 1
>  > make[2]: Leaving directory
>  > `/home3/egcs-sunos/sparc-sun-sunos4.1.4/libio'
>  > make[1]: *** [all-target-libio] Error 2
>  > make[1]: Leaving directory `/home3/egcs-sunos'
>  > make: *** [bootstrap] Error 2
> 
> David,
> 
> >From the output above and by examining libio/{Makefile.in,gen-params},
> the error appears to occur because CXX is not set properly before
> being called to compile some c++ code.  Inside libio/Makefile.in we
> have:
> 
>  > _G_config.h: ${srcdir}/gen-params
>  > 	rootme=`pwd`/ ; export rootme; \
>  > 	CC="$(CC) $(CINCLUDES)"; export CC; \
>  > 	CXX="$(CXX) $(CXXINCLUDES) $(NOSTDINC) $(CXXFLAGS)"; export CXX; \
>  > 	CONFIG_NM="$(NM)"; export CONFIG_NM; \
>  > 	$(SHELL) ${srcdir}/gen-params LIB_VERSION=$(VERSION) $(G_CONFIG_ARGS) >tmp-params.h
>  > 	mv tmp-params.h _G_config.h
> 
> Note that if CXX is empty, it'll choke just like the error you and I
> both see.
> 
> I'm not sure why this changed or why it only happens to trigger on
> sunos4, but given the timing my unconfirmed guess is that its
> something Alexandre did at the top level directory Makefile.in and/or
> configure* stuff.  (Though I could be wrong.)
> 
> Alexandre, would you please take a look?  Maybe libio should be using
> CXX_FOR_TARGET?  But if so, why did it used to work?
> 
> 		Thanks,
> 		--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]