Re: GDB 18.0.90 available for testing
Eli Zaretskii <[email protected]>
| Newsgroups | gmane.comp.gdb.patches |
|---|---|
| Message-ID | <[email protected]> |
> Date: Thu, 06 Aug 2026 19:12:40 +0100 > From: Andrew Burgess <[email protected]> > > Hello, > > I have just finished creating the gdb-18.0.90 pre-release. > It is available for download at the following location: > > https://sourceware.org/pub/gdb/snapshots/branch/gdb-18.0.90.tar.xz > > A gzip'ed version is also available: gdb-18.0.90.tar.gz. > > Please give it a test if you can and report any problems you might find. I've built this pretest for native MinGW debugging on MS-Windows, and encountered the following minor issues: 1. When configured with --enable-binary-file-formats='coff,xcoff,elf,macho', the build fails during final linking of GDB: CXXLD gdb.exe d:/usr/bin/../lib/gcc/mingw32/9.2.0/../../../../mingw32/bin/ld.exe: machoread.o: in function `macho_check_dsym': d:\gnu\gdb-18.0.90\gdb/machoread.c:738:(.text+0xb16): undefined reference to `bfd_mach_o_lookup_command' d:/usr/bin/../lib/gcc/mingw32/9.2.0/../../../../mingw32/bin/ld.exe: d:\gnu\gdb-18.0.90\gdb/machoread.c:757:(.text+0xbe6): undefined reference to `bfd_mach_o_lookup_command' collect2.exe: error: ld returned 1 exit status This is because bfd/mach-o.c is not compiled for some reason, although the configure script specified 'macho' as a binary format to support. I see that when I built GDB 17, I used mach-o, not macho. But the configure --help screen says to use 'macho': --enable-binary-file-formats=FORMATS enable support for selected file formats (default 'all') available formats: coff, elf, macho, xcoff, all (I fixed this by reconfiguring without 'macho'.) 2. Recompilation of libsframe/sframe.c emits warnings: CC libsframe_la-sframe.lo In file included from ./../libctf/swap.h:23, from sframe.c:27: ./../libctf/config.h:127: warning: "PACKAGE" redefined 127 | #define PACKAGE "libctf" | In file included from sframe.c:20: config.h:66: note: this is the location of the previous definition 66 | #define PACKAGE "libsframe" | In file included from ./../libctf/swap.h:23, from sframe.c:27: ./../libctf/config.h:133: warning: "PACKAGE_NAME" redefined 133 | #define PACKAGE_NAME "libctf" | In file included from sframe.c:20: config.h:72: note: this is the location of the previous definition 72 | #define PACKAGE_NAME "libsframe" | In file included from ./../libctf/swap.h:23, from sframe.c:27: ./../libctf/config.h:136: warning: "PACKAGE_STRING" redefined 136 | #define PACKAGE_STRING "libctf 1.2.0" | In file included from sframe.c:20: config.h:75: note: this is the location of the previous definition 75 | #define PACKAGE_STRING "libsframe BFD_VERSION" | In file included from ./../libctf/swap.h:23, from sframe.c:27: ./../libctf/config.h:139: warning: "PACKAGE_TARNAME" redefined 139 | #define PACKAGE_TARNAME "libctf" | In file included from sframe.c:20: config.h:78: note: this is the location of the previous definition 78 | #define PACKAGE_TARNAME "libsframe" | In file included from ./../libctf/swap.h:23, from sframe.c:27: ./../libctf/config.h:145: warning: "PACKAGE_VERSION" redefined 145 | #define PACKAGE_VERSION "1.2.0" | In file included from sframe.c:20: config.h:84: note: this is the location of the previous definition 84 | #define PACKAGE_VERSION "BFD_VERSION" | In file included from ./../libctf/swap.h:23, from sframe.c:27: ./../libctf/config.h:173: warning: "VERSION" redefined 173 | #define VERSION "1.2.0" | In file included from sframe.c:20: config.h:112: note: this is the location of the previous definition 112 | #define VERSION "BFD_VERSION" | This is because sframe.c includes libctf/swap.h, which includes libctf/config.h, which defines these macros for libctf, not for libsframe. Strangely, the initial compilation didn't produce these warnings, but when I needed to recompile sframe.c later, it did. Other than that, the MinGW GDB builds okay, passes all the self-tests ("maint selftest"), and seems to work fine. Thanks.