[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-19 Thread danglin at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 John David Anglin danglin at gcc dot gnu.org changed: What|Removed |Added Status|WAITING |RESOLVED

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-17 Thread hubicka at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #11 from Jan Hubicka hubicka at gcc dot gnu.org --- Author: hubicka Date: Tue Sep 17 15:46:06 2013 New Revision: 202657 URL: http://gcc.gnu.org/viewcvs?rev=202657root=gccview=rev Log: PR middle-end/58329 * ipa-devirt.c

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-17 Thread hubicka at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 Jan Hubicka hubicka at gcc dot gnu.org changed: What|Removed |Added Status|UNCONFIRMED |WAITING Last

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-17 Thread hubicka at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #12 from Jan Hubicka hubicka at gcc dot gnu.org --- Author: hubicka Date: Tue Sep 17 16:07:21 2013 New Revision: 202658 URL: http://gcc.gnu.org/viewcvs?rev=202658root=gccview=rev Log: PR middle-end/58329 * ipa-devirt.c

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-08 Thread hubicka at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #10 from Jan Hubicka hubicka at gcc dot gnu.org --- OK, I made similar patch on trip to Pisa. I think correct is to return NULL and make callers handle it - it does not make sense to consider non-local label to be local alias. I will

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-07 Thread dave.anglin at bell dot net
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #9 from dave.anglin at bell dot net --- On 6-Sep-13, at 8:05 AM, hubicka at ucw dot cz wrote: perhaps we want to simply check for ASM_OUTPUT_DEF in symtab and refuse to use local aliases then? I think the other macros alow only

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-06 Thread hubicka at ucw dot cz
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #3 from Jan Hubicka hubicka at ucw dot cz --- I wouldn't be surprised. I don't have assembler output or preprocessed source yet. There is some alias support in gas for HP-UX but I believe it may not work when we have a call

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-06 Thread hubicka at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #4 from Jan Hubicka hubicka at gcc dot gnu.org --- *** Bug 58293 has been marked as a duplicate of this bug. ***

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-06 Thread rguenth at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 Richard Biener rguenth at gcc dot gnu.org changed: What|Removed |Added Target Milestone|--- |4.9.0

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-06 Thread dave.anglin at bell dot net
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #5 from dave.anglin at bell dot net --- On 6-Sep-13, at 3:54 AM, hubicka at ucw dot cz wrote: I hoped that the targets either do not have runtime interposition in dynamic libraries or they do have working notion of alias. The

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-06 Thread hubicka at ucw dot cz
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #6 from Jan Hubicka hubicka at ucw dot cz --- The PA-RISC HP-UX linker interposes import and export stubs in dynamic libraries. Whether there is a working notion of alias is somewhat unclear and involves digging into the

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-06 Thread danglin at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #8 from John David Anglin danglin at gcc dot gnu.org --- Created attachment 30755 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=30755action=edit Assembler output

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-06 Thread danglin at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #7 from John David Anglin danglin at gcc dot gnu.org --- Created attachment 30754 -- http://gcc.gnu.org/bugzilla/attachment.cgi?id=30754action=edit Preprocessed source

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-05 Thread hubicka at ucw dot cz
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #1 from Jan Hubicka hubicka at ucw dot cz --- Symbol has type data which is wrong for procedure label: Symbols from system_error.o: ValueInfo Type Scope ck HQIRCDSKLN xl reloc Name Data Unsat 0

[Bug ipa/58329] [4.9 Regression] ld: Invalid symbol type for plabel (.libs/libstdc++.lax/libc++11convenience.a/system_error.o, std::error_category::default_error_condition(int) const [clone .localalia

2013-09-05 Thread dave.anglin at bell dot net
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #2 from dave.anglin at bell dot net --- On 5-Sep-13, at 7:31 PM, hubicka at ucw dot cz wrote: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58329 --- Comment #1 from Jan Hubicka hubicka at ucw dot cz --- Symbol has type data which