Executing on host: /xxx/gnu/gcc-3.4/objdir/gcc/testsuite/../g++ -B/xxx/gnu/gcc-3 .4/objdir/gcc/testsuite/../ /xxx/gnu/gcc-3.4/gcc/gcc/testsuite/g++.dg/rtti/tinfo 1.C -nostdinc++ -I/xxx/gnu/gcc-3.4/objdir/hppa2.0w-hp-hpux11.00/libstdc++-v3/in clude/hppa2.0w-hp-hpux11.00 -I/xxx/gnu/gcc-3.4/objdir/hppa2.0w-hp-hpux11.00/libs tdc++-v3/include -I/xxx/gnu/gcc-3.4/gcc/libstdc++-v3/libsupc++ -I/xxx/gnu/gcc-3. 4/gcc/libstdc++-v3/libsupc++ -I/xxx/gnu/gcc-3.4/gcc/libstdc++-v3/include/backwar d -I/xxx/gnu/gcc-3.4/gcc/libstdc++-v3/testsuite -fmessage-length=0 -ansi -peda ntic-errors -Wno-long-long -S -o tinfo1.s (timeout = 300) PASS: g++.dg/rtti/tinfo1.C (test for excess errors) FAIL: g++.dg/rtti/tinfo1.C scan-assembler _ZTIP9CTemplateIhE: PASS: g++.dg/rtti/tinfo1.C scan-assembler-not .globl[ \t]+_ZTIP9CTemplateIhE PASS: g++.dg/rtti/tinfo1.C scan-assembler-not .section[^\n\r]* _ZTIP9CTemplateIhE[^\n\r]* Looking at tinfo1.s, I see: .IMPORT _ZTIP9CTemplateIhE,DATA ldil LR'_ZTIP9CTemplateIhE,%r19 ldo RR'_ZTIP9CTemplateIhE(%r19),%r26 .CALL ARGW0=GR bl _ZNKSt9type_info4nameEv,%r2 ldil LR'_ZTIP9CTemplateIhE,%r19 ldo RR'_ZTIP9CTemplateIhE(%r19),%r26 .CALL ARGW0=GR bl _ZNKSt9type_info4nameEv,%r2 .align 4 _ZTIP9CTemplateIhE .IMPORT _ZTVN10__cxxabiv119__pointer_type_infoE,DATA .word _ZTVN10__cxxabiv119__pointer_type_infoE+8 .word _ZTSP9CTemplateIhE .word 8 .word _ZTI9CTemplateIhE .weak _ZTSP9CTemplateIhE .EXPORT _ZTSP9CTemplateIhE,DATA This problem was introduced by the fix for PR c++/16276.
I have a semi fix for it, as it also wrong on powerpc-apple-darwin, the rest of the fix is for other targets which define MAKE_DECL_ONE_ONLY also define MAKE_DECL_NON_ONE_ONLY with right definition.
Here is the first part of the patch which fixes the darwin issue: <http://gcc.gnu.org/ml/gcc-patches/ 2004-07/msg00878.html>, fixing the other target issues is easier and HJL suggested one approach but I think it is wrong and each target which defines its own MAKE_DECL_ONE_ONLY should also define its own MAKE_DECL_NON_ONE_ONLY.
Mine, I will do the second part tommorrow (Monday).
Patch posted here: <http://gcc.gnu.org/ml/gcc-patches/2004-07/msg01275.html>.
I am just reverting the patch per Mark's request.
Subject: Bug 16475 CVSROOT: /cvs/gcc Module name: gcc Branch: gcc-3_4-branch Changes by: pinskia@gcc.gnu.org 2004-07-12 22:26:39 Modified files: gcc/cp : ChangeLog rtti.c Log message: 2004-07-12 Andrew Pinski <apinski@apple.com> PR c++/16475 Revert: 2004-07-07 H.J. Lu <hongjiu.lu@intel.com> PR c++/16276 * rtti.c (emit_tinfo_decl): Turn off DECL_ONE_ONLY if typeinfo is not public. Patches: http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.3892.2.131&r2=1.3892.2.132 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/rtti.c.diff?cvsroot=gcc&only_with_tag=gcc-3_4-branch&r1=1.176.4.4&r2=1.176.4.5
Subject: Bug 16475 CVSROOT: /cvs/gcc Module name: gcc Changes by: pinskia@gcc.gnu.org 2004-07-12 22:28:54 Modified files: gcc/cp : ChangeLog rtti.c Log message: 2004-07-12 Andrew Pinski <apinski@apple.com> PR c++/16475 Revert: 2004-07-07 H.J. Lu <hongjiu.lu@intel.com> PR c++/16276 * rtti.c (emit_tinfo_decl): Turn off DECL_ONE_ONLY if typeinfo is not public. Patches: http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/ChangeLog.diff?cvsroot=gcc&r1=1.4193&r2=1.4194 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/cp/rtti.c.diff?cvsroot=gcc&r1=1.187&r2=1.188
Fixed.