[Bug gold/13359] gold internal error in relocate_tls, at gold/x86_64.cc:3187

2011-12-19 Thread cvs-commit at gcc dot gnu.org
http://sourceware.org/bugzilla/show_bug.cgi?id=13359 --- Comment #4 from cvs-commit at gcc dot gnu.org cvs-commit at gcc dot gnu.org 2011-12-19 21:14:45 UTC --- CVSROOT:/cvs/src Module name:src Branch: binutils-2_22-branch Changes by:i...@sourceware.org2011-12-19 21:14:40

[Bug gold/13023] gold misinterprets dot assignments in sections

2011-12-19 Thread cvs-commit at gcc dot gnu.org
http://sourceware.org/bugzilla/show_bug.cgi?id=13023 --- Comment #7 from cvs-commit at gcc dot gnu.org cvs-commit at gcc dot gnu.org 2011-12-19 21:14:46 UTC --- CVSROOT:/cvs/src Module name:src Branch: binutils-2_22-branch Changes by:i...@sourceware.org2011-12-19 21:14:40

[Bug gold/13245] PREVAILING_DEF reported too often.

2011-12-19 Thread cvs-commit at gcc dot gnu.org
http://sourceware.org/bugzilla/show_bug.cgi?id=13245 --- Comment #16 from cvs-commit at gcc dot gnu.org cvs-commit at gcc dot gnu.org 2011-12-19 21:14:52 UTC --- CVSROOT:/cvs/src Module name:src Branch: binutils-2_22-branch Changes by:i...@sourceware.org2011-12-19 21:14:40

[Bug gold/13249] gold creates incorrect RELATIVE dynamic relocation for local IFUNC GOT entries

2011-12-19 Thread cvs-commit at gcc dot gnu.org
http://sourceware.org/bugzilla/show_bug.cgi?id=13249 --- Comment #4 from cvs-commit at gcc dot gnu.org cvs-commit at gcc dot gnu.org 2011-12-19 21:14:52 UTC --- CVSROOT:/cvs/src Module name:src Branch: binutils-2_22-branch Changes by:i...@sourceware.org2011-12-19 21:14:40

[Bug gold/13163] internal error in set_segment_offsets, at ../../../binutils/gold/layout.cc:3050

2011-12-19 Thread cvs-commit at gcc dot gnu.org
http://sourceware.org/bugzilla/show_bug.cgi?id=13163 --- Comment #6 from cvs-commit at gcc dot gnu.org cvs-commit at gcc dot gnu.org 2011-12-19 21:14:51 UTC --- CVSROOT:/cvs/src Module name:src Branch: binutils-2_22-branch Changes by:i...@sourceware.org2011-12-19 21:14:40

[Bug gold/13365] Fold functions that only jump to some other one

2011-12-19 Thread ian at airs dot com
http://sourceware.org/bugzilla/show_bug.cgi?id=13365 --- Comment #1 from Ian Lance Taylor ian at airs dot com 2011-12-19 23:30:39 UTC --- Of course one can only do this if there are no comparisons of function pointers. And that means that it can only be done if the symbols are not visible

[Bug gold/13366] Allow to pack some functions together

2011-12-19 Thread ian at airs dot com
http://sourceware.org/bugzilla/show_bug.cgi?id=13366 --- Comment #1 from Ian Lance Taylor ian at airs dot com 2011-12-19 23:31:56 UTC --- I think it will be relatively difficult for the linker to safely and reliably detect cases where this packing is valid. The packing may only be done if the

[Bug gold/13521] New: ld.gold prefers unversioned symbol over default version

2011-12-19 Thread jrnieder at gmail dot com
http://sourceware.org/bugzilla/show_bug.cgi?id=13521 Bug #: 13521 Summary: ld.gold prefers unversioned symbol over default version Product: binutils Version: 2.23 (HEAD) Status: NEW Severity: normal

[Bug gold/13521] ld.gold prefers unversioned symbol over default version

2011-12-19 Thread jrnieder at gmail dot com
http://sourceware.org/bugzilla/show_bug.cgi?id=13521 --- Comment #1 from Jonathan Nieder jrnieder at gmail dot com 2011-12-20 07:14:36 UTC --- Created attachment 6120 -- http://sourceware.org/bugzilla/attachment.cgi?id=6120 DSO with versioned and unversioned lzma_code symbol (Debian liblzma5

[Bug gold/13521] ld.gold prefers unversioned symbol over default version

2011-12-19 Thread jrnieder at gmail dot com
http://sourceware.org/bugzilla/show_bug.cgi?id=13521 --- Comment #2 from Jonathan Nieder jrnieder at gmail dot com 2011-12-20 07:17:03 UTC --- Created attachment 6121 -- http://sourceware.org/bugzilla/attachment.cgi?id=6121 output from ld.gold -o output.gold test.o liblzma.so.5.0.0, which