[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2011-09-08 Thread bergner at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #12 from Peter Bergner 2011-09-08 18:09:02 UTC --- Author: bergner Date: Thu Sep 8 18:08:53 2011 New Revision: 178703 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=178703 Log: gcc/ Backport from mainline 2011-08-23

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2011-08-23 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 Jakub Jelinek changed: What|Removed |Added Status|ASSIGNED|RESOLVED Resolution|

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2011-08-23 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #10 from Jakub Jelinek 2011-08-23 13:23:29 UTC --- Author: jakub Date: Tue Aug 23 13:23:22 2011 New Revision: 177988 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=177988 Log: PR c++/46862 * class.c (finish_struct_1):

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2011-08-23 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #9 from Jakub Jelinek 2011-08-23 13:20:11 UTC --- Author: jakub Date: Tue Aug 23 13:20:04 2011 New Revision: 177987 URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=177987 Log: PR c++/46862 * class.c (finish_struct_1): I

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2011-08-22 Thread pthaugen at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #8 from Pat Haugen 2011-08-22 18:44:01 UTC --- I've verified the proposed patch fixes the testcase from bug 50102.

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2011-08-22 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #7 from Jakub Jelinek 2011-08-22 11:26:04 UTC --- Created attachment 25075 --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25075 gcc47-pr46862.patch Untested fix.

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2011-08-22 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 Jakub Jelinek changed: What|Removed |Added CC||pthaugen at gcc dot gnu.org --- Comment #

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2010-12-14 Thread janis at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 Janis Johnson changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed|

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2010-12-09 Thread janis at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #4 from Janis Johnson 2010-12-09 18:50:13 UTC --- This doesn't have anything to do with libdecnumber, it's about how std::decimal::decimalxx are defined in libstdc++. Instead of an ICE with a nice message it would be more appropriate

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2010-12-09 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #3 from Jakub Jelinek 2010-12-09 18:46:28 UTC --- This got reported to us in https://bugzilla.redhat.com/show_bug.cgi?id=657606 but no details about whether it was pristine or modified libdecnumber and which version thereof. GCC shou

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2010-12-09 Thread janis at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #2 from Janis Johnson 2010-12-09 18:42:13 UTC --- TYPE_TRANSPARENT_AGGR is set in begin_class_definition in semantics.c for the classes std::decimal::decimal{32|64|128}. It assumes that they are defined as in the GNU Standard C++ Libr

[Bug c++/46862] ICE on std::decimal::decimal32 without any fields

2010-12-09 Thread jakub at gcc dot gnu.org
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=46862 --- Comment #1 from Jakub Jelinek 2010-12-09 11:25:08 UTC --- Not sure if we should error out if a TYPE_TRANSPARENT_AGGR doesn't in the end have any fields, or simply mangle them normally if first_field returns NULL, but we shouldn't ICE on it.