[Gc] GC 6.6 on Solaris for Intel OK [Solaris for Intel?

Boehm, Hans hans.boehm at hp.com
Fri Apr 14 14:43:12 PDT 2006


You might try defining USE_GENERIC_PUSH_REGS, and perhaps changing the
implementation of GC_with_callee_saves_pushed (at least in 6.7, it maybe
elsewhere in 6.6) to use getcontext instead of setjmp.

My real goal is to get this to work reliably with 7.0.  It's not
completely inconceivable that it already works with the current CVS
sources, at least if you build with gcc.

I'm not going to have the time to do anything significant with the 6.X
solaris-specific thread support, if that turns out to be part of the
problem.  7.0 uses the generic pthread support for Solaris.

Hans

> -----Original Message-----
> From: gc-bounces at napali.hpl.hp.com 
> [mailto:gc-bounces at napali.hpl.hp.com] On Behalf Of jim marshall
> Sent: Friday, April 14, 2006 2:28 PM
> To: Alexander Petrossian (PAF)
> Cc: gc at napali.hpl.hp.com
> Subject: Re: [Gc] GC 6.6 on Solaris for Intel OK [Solaris for Intel?
> 
> I finally was able to get to the Solaris Intel machine again 
> (had some networking issues). I put 6.6 on it and got the same issue
> 
> Apparently failed to mark form some function arguments.
> Perhaps GC_push_regs was configured incorrectly?
> Test failed
> /bin/bash: line 1:  8998 Abort                   (core 
> dumped) ${dir}$tst
> FAIL: gctest
> ==================================
> 1 of 1 tests failed
> Please report to Hans.Boehm at hp.com
> ==================================
> 
> Any thoughts on what GC_push_regs should be configured as?
> 
> Thanks
> 
> Alexander Petrossian (PAF) wrote:
> 
> > Jim,
> > I'm using GC 6.6 on Solaris for Intel without troubles.
> > regretfully, can't test it right now, but it was some older 
> gcc that I 
> > used to compile it.
> >
> > If you need more details, please let me know.
> >
> > Alexander
> >
> > ----- Original Message ----- From: "jim marshall" 
> > <jim.marshall at wbemsolutions.com>
> > To: <gc at napali.hpl.hp.com>
> > Sent: Monday, April 10, 2006 7:49 PM
> > Subject: Re: [Gc] Solaris for Intel?
> >
> >
> >> Hello,
> >> I am surprised that no one has tried to run the GC on Solaris for 
> >> Intel. Anyone have any thoughts on how I might investigate 
> the problem?
> >>
> >> -Jim
> >>
> >> jim marshall wrote:
> >>
> >>> Hello,
> >>> Has anyone brought the GC (6.7) over to Solaris for Intel yet? I 
> >>> tried it but I get the output below from make check.
> >>>
> >>> System:
> >>> Solaris 10 for Intel (uname -a = SunOS Solaris-intel 5.10
> >>> Generic_118844-26 i86pc i386 i86pc)
> >>> AMD 1900+ cpu
> >>> 512mb RAM
> >>> gcc 3.3.2 (from sunfreeware.com)
> >>>  Reading specs from
> >>> /usr/local/lib/gcc-lib/i386-pc-solaris2.10/3.3.2/specs
> >>> Configured with: ../configure --with-as=/usr/ccs/bin/as 
> >>> --with-ld=/usr/ccs/bin/ld --disable-nls Thread model: posix gcc 
> >>> version 3.3.2
> >>>
> >>> GC configure was run like this:
> >>>  ./configure --prefix=/testdir/gc67/deb 
> >>> --exec-prefix=/testdir/gc67/deb --enable-gc-assertions 
> >>> --enable-full-debug --enable-static=no --enable-threads=posix
> >>>
> >>>
> >>> -bash-3.00$ make check
> >>> Making check in doc
> >>> make[1]: Entering directory `/testdir/gc67/gc6.7/doc'
> >>> make[1]: Nothing to be done for `check'.
> >>> make[1]: Leaving directory `/testdir/gc67/gc6.7/doc'
> >>> Making check in include
> >>> make[1]: Entering directory `/testdir/gc67/gc6.7/include'
> >>> make[1]: Nothing to be done for `check'.
> >>> make[1]: Leaving directory `/testdir/gc67/gc6.7/include'
> >>> make[1]: Entering directory `/testdir/gc67/gc6.7'
> >>> make  gctest
> >>> make[2]: Entering directory `/testdir/gc67/gc6.7'
> >>> /bin/bash ./libtool --mode=link gcc -fexceptions -g -O2  
> -o gctest 
> >>> test.o ./libgc.la -lpthread -ldl  gcc -fexceptions -g -O2 -o 
> >>> .libs/gctest test.o  ./.libs/libgc.so -lpthread -ldl 
> >>> -R/testdir/gc67/deb/lib creating gctest
> >>> make[2]: Leaving directory `/testdir/gc67/gc6.7'
> >>> make  check-TESTS
> >>> make[2]: Entering directory `/testdir/gc67/gc6.7'
> >>> Apparently failed to mark form some function arguments.
> >>> Perhaps GC_push_regs was configured incorrectly?
> >>> Test failed
> >>> /bin/bash: line 1: 17653 Abort                   (core dumped) 
> >>> ${dir}$tst
> >>> FAIL: gctest
> >>> ==================================
> >>> 1 of 1 tests failed
> >>> Please report to Hans.Boehm at hp.com
> >>> ==================================
> >>> make[2]: *** [check-TESTS] Error 1
> >>> make[2]: Leaving directory `/testdir/gc67/gc6.7'
> >>> make[1]: *** [check-am] Error 2
> >>> make[1]: Leaving directory `/testdir/gc67/gc6.7'
> >>> make: *** [check-recursive] Error 1
> >>>
> >>>
> >>> Thanks
> >>> -Jim
> >>> _______________________________________________
> >>> Gc mailing list
> >>> Gc at linux.hpl.hp.com
> >>> http://www.hpl.hp.com/hosted/linux/mail-archives/gc/
> >>>
> >>>
> >>>
> >> _______________________________________________
> >> Gc mailing list
> >> Gc at linux.hpl.hp.com
> >> http://www.hpl.hp.com/hosted/linux/mail-archives/gc/
> >>
> > _______________________________________________
> > Gc mailing list
> > Gc at linux.hpl.hp.com
> > http://www.hpl.hp.com/hosted/linux/mail-archives/gc/
> >
> >
> >
> _______________________________________________
> Gc mailing list
> Gc at linux.hpl.hp.com
> http://www.hpl.hp.com/hosted/linux/mail-archives/gc/
> 



More information about the Gc mailing list