Jump to letter: [
4ABCDEFGHIJKLMNOPQRSTUVWXYZ
]
libgnat: GNU Ada 95 runtime shared libraries
| Name: | libgnat |
Vendor: | Scientific Linux |
| Version: | 3.2.3 |
License: | GPL |
| Release: | 56 |
URL: | http://gcc.gnu.org |
- Summary
- GNAT is a GNU Ada 95 front-end to GCC. This package includes shared libraries,
which are required to run programs compiled with the GNAT.
Changelog
- * Mon May 22 19:00:00 2006 Jakub Jelinek <jakub{%}redhat{*}com> 3.2.3-56
- add workaround for buggy programs that link in their own unwinder
and reexport it (#192259)
- make all globals in libgcc_eh.a hidden, so that newly (incorrectly)
linked programs can't reexport the unwinder
- * Tue Apr 25 19:00:00 2006 Jakub Jelinek <jakub{%}redhat{*}com> 3.2.3-55
- fix splitting of SSE conditional move patterns (Alexandre Oliva, #173224)
- fix unwinding through signal frames (#187810, PR other/26208, glibc BZ#300)
- fix for C++ overlapping stack temporaries (Ulrich Weigand, #186252, #187909,
PR middle-end/15054)
- fix crash in zero-initialization of mmx registers (Richard Henderson,
Alexandre Oliva, #180778)
- report unused parameter warnings even on optimized out functions/methods
(Alexandre Oliva, Jan Hubicka, #181894)
- * Thu Nov 17 18:00:00 2005 Jakub Jelinek <jakub{%}redhat{*}com> 3.2.3-54
- disable os_defines.h changes introduced in 3.2.3-45 also for
IBM xlC compiler (#146196, IT#75174)
- fix pushdecl_class_level for OVERLOADs (Mark Mitchell,
Alexandre Oliva, #156185)
- reword diagnostics about unresolved overloaded type
(Alexandre Oliva, #156291, IT#70101)
- fix x86_64 compiler hang when passing object with 3+ base classes by value
(Zak Kipling, PR target/18300, #164421, #171940, IT#76454)
- predefine __GNUC_RH_RELEASE__ macro to rpm's %{release} (Alexandre Oliva)
- weakref attribute support, use it in <bits/gthr.h> various C++ headers
are using (Alexandre Oliva, #165728, IT#73356)
- don't warn in templates about missing return if return type is dependent
on template parameters (Alexandre Oliva, PR c++/13106, #169044, IT#77857)
- ensure ppc64 keeps stack 16 byte aligned even with dynamic allocation
(alloca, VLAs), but don't rely on it in the generated code
(Alexandre Oliva, #169111, IT#76136)
- fix invalid CC clobberation on i?86/x86_64 (Eric Botcazou,
PR optimization/12799, #169654)
- don't assume hard frame pointer register is STACK_BOUNDARY aligned
if frame pointer isn't used and the register has been reused
for something else (Eric Botcazou, PR optimization/13041, #169845)