Thanks Massimo, in fact the problem, as you say, was already known. I hope Apple is listening... Another very disturbing change is that the g2c library disappeared. I recuperated the one from 3.1, which seems to work, but is not very elegant. Also, it would be nice to have a shared version of this library. THis is needed if you mix shared libraries containing Fortran and C++. REgards, Fed
On Wed, 9 Jul 2003, Massimo Marino wrote: > Fed, > > on the Fink main page they report the problem. Apparently the assembler > coming with the new gcc 3.3 is broken or does not tolerate wrond > generated code which passed instead in 3.1 > > If you go to the Fink page and click on the 30 June Update on the > broken assembler you may read the report (same errors as yours). > Martin Costabel (from Fink) reports temporarily solving the problem by > installing again the gas from 3.1. > > Apple has been made aware of the issue but I am not sure whether the > problem has to be solved by them of by GNU in their compiler. > > Massimo > > On Wednesday, July 9, 2003, at 08:41 AM, Federico Carminati wrote: > > > Hello, > > I have an application in c++ that does not compile. gcc 3.1 fails > > with > > (internal compiler fault, bus error). I figured out that there was no > > point is submitting a bug report when a new version was out, so I > > moved to > > 3.3. This has two effects. First, as explained below, you have to > > /usr/sbin/gcc_select 3.1 every time you run fink. Second, if you > > compile > > the following with g77 (the one in fink, given that Apple still does > > not > > provide Fortran with the developer suite!!) > > > > subroutine test > > stop > > end > > > > you get > > > > /var/tmp//ccNyrfCQ.s:18:section difference relocatable subtraction > > expression, "LC0" minus "L1$pb" using a symbol at the end of section > > will > > not produce an assembly time constant > > > > /var/tmp//ccNyrfCQ.s:18:use a symbol with a constant value created > > with an > > assignment instead of the expression, L_const_sym = LC0 - L1$pb > > > > /var/tmp//ccNyrfCQ.s:17:section difference relocatable subtraction > > expression, "LC0" minus "L1$pb" using a symbol at the end of section > > will > > not produce an assembly time constant > > > > /var/tmp//ccNyrfCQ.s:17:use a symbol with a constant value created > > with an > > assignment instead of the expression, L_const_sym = LC0 - L1$pb > > > > This comes (I presume) from the new assembler generator installed when > > you > > update to 3.3, so even if you select 3.1 it fails in the same way. You > > cannot have a stop anymore in a subroutine, and it is plenty of > > those... > > So I am completely stuck. Ach! Help please. Fed > > > > On Fri, 27 Jun 2003, Massimo Marino wrote: > > > >> Apple has released a patch to the December 2002 Developer Tools which > >> includes gcc 3.3, their new compiler. (ADC Apple site) > >> > >> Caveat: Fink does not yet support compiling with gcc 3.3. In > >> addition, > >> it is important not to "mix and match" between compilers: all C++ > >> code > >> in fink packages needs to be compiled with the same compiler. > >> > >> For this reason, the Fink team recommends that if you update your > >> Developer Tools with the new patch, you should be careful to run > >> sudo gcc_select 3.1 (same warning on Apple doc concerning switching > >> GCC > >> versions) > >> > >> prior to any "fink build" or "fink install" commands. > >> > >> Massimo > >> > >> -- > >> Massimo Marino > >> NERSC Division - HPC Department > >> Lawrence Berkeley National Laboratory > >> On leave at CERN, CH, EP Division, Atlas experiment > >> phone: (+41) 22 767-1288 fax: (+41) 22 767-8350 Office: 32-R-C24 > >> alternate email: [EMAIL PROTECTED], [EMAIL PROTECTED] > >> > >> ---------------------------------------------------------------------- > >> MacOS X ListBox Web Archive : > >> https://wwwlistbox.cern.ch/earchive/support-macosx/ > >> ---------------------------------------------------------------------- > >> > > > > -- > > > > Federico Carminati > > > > ======================================================================= > > ======= > > || Federico Carminati || Telephone : +41.22.767.4959 > > || > > || CERN-EP || Fax : +41.22.767.9480 > > || > > || 1211 Geneva 23 || Mobile : +41.79.201.4843 > > || > > || Switzerland || > > || > > ======================================================================= > > ======= > > > > > > ---------------------------------------------------------------------- > > MacOS X ListBox Web Archive : > > https://wwwlistbox.cern.ch/earchive/support-macosx/ > > ---------------------------------------------------------------------- > > > > -- Federico Carminati ============================================================================== || Federico Carminati || Telephone : +41.22.767.4959 || || CERN-EP || Fax : +41.22.767.9480 || || 1211 Geneva 23 || Mobile : +41.79.201.4843 || || Switzerland || || ============================================================================== ------------------------------------------------------- This SF.Net email sponsored by: Parasoft Error proof Web apps, automate testing & more. Download & eval WebKing and get a free book. www.parasoft.com/bulletproofapps _______________________________________________ Fink-users mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/fink-users
