This way we don't need RANLIB anymore, making it a bit simpler for the meson build to generate Makefile.global for PGXS compatibility. FreeBSD, NetBSD, OpenBSD, the only platforms where we didn't use AROPT=crs, all have supported the 's' option for a long time. On macOS we ran ranlib after installing a static library. This was added a long time ago, in 58ad65ec2def. I cannot reproduce an issue in more recent macOS versions. This is removed now. Based on discussion with Tom, I left the 'touch' at the end of static libraries generation, added in 826eff57c4c, in place. While it looks like current versions of Apple's ar/ranlib don't need it, it was needed not too long ago. Reviewed-by: Tom Lane <tgl@sss.pgh.pa.us> Discussion: https://postgr.es/m/20221005200710.luvw5evhwf6clig6@awork3.anarazel.de
When building a shared library with exports.txt there's no need to build an intermediary static library, we can just pass -Wl,-bE:... when generating the .so. When building a shared library without exports.txt, there's no need to call mkldexport.sh to export all symbols, because all symbols are exported anyway, and we don't need the export file on the import side (like we do for postgres.imp). This makes building .so's on aix a lot more similar to building on other platforms. In particular, we don't create and remove a .a of the same name but different contents anymore. Discussion: https://postgr.es/m/20220820174213.d574qde4ptwdzoqz@awork3.anarazel.de
Not passing -shared to gcc when building a shared library triggers linking to the wrong libgcc (libgcc.a instead of libgcc_s.a) and prevents emitting correct unwind information. It's somewhat surprising that this hasn't caused known problems so far. Doing so requires adding path to libgcc to libpath, or linking statically to libgcc - as the latter increases .so size substantially (for not entirely obvious reasons), shared linking seems preferrable. It likely is worth building executables with -shared-libgcc too, but I've not done that here. Discussion: https://postgr.es/m/20220820174213.d574qde4ptwdzoqz@awork3.anarazel.de
Reviewed-By: Thomas Munro <thomas.munro@gmail.com> Discussion: https://postgr.es/m/20220807012914.ydz73yte6j3coulo@awork3.anarazel.de
Move DLSUFFIX from makefiles into header files for all platforms. Move the DLSUFFIX assignment from src/makefiles/ to src/templates/, have configure read it, and then substitute it into Makefile.global and pg_config.h. This avoids the need for all makefile rules that need it to locally set CPPFLAGS. It also resolves an inconsistent setup between the two Windows build systems. Reviewed-by: Andres Freund <andres@anarazel.de> Discussion: https://www.postgresql.org/message-id/2f9861fb-8969-9005-7518-b8e60f2bead9@enterprisedb.com
Rather than define ld_library_path_ver with a big nested $(if), just put the overriding values in the makefiles for the relevant ports. Also add a variable for port makefiles to append their own stuff to with_temp_install, and use it to set LD_LIBRARY_PATH_RPATH=1 on FreeBSD which is needed to make LD_LIBRARY_PATH override DT_RPATH if DT_RUNPATH is not set (which seems to depend in unpredictable ways on the choice of compiler, at least on my system). Backpatch for the benefit of anyone doing regression tests on FreeBSD. (For other platforms there should be no functional change.)
It's important for link commands to list *.o input files before -l switches for libraries, as library code may not get pulled into the link unless referenced by an earlier command-line entry. This is certainly necessary for static libraries (.a style). Apparently on some platforms it is also necessary for shared libraries, as reported by Donald Dong. We often put -l switches for within-tree libraries into LDFLAGS, meaning that link commands that list *.o files after LDFLAGS are hazardous. Most of our link commands got this right, but a few did not. In particular, places that relied on gmake's default implicit link rule failed, because that puts LDFLAGS first. Fix that by overriding the built-in rule with our own. The implicit link rules in src/makefiles/Makefile.* for single-.o-file shared libraries mostly got this wrong too, so fix them. I also changed the link rules for the backend and a couple of other places for consistency, even though they are not (currently) at risk because they aren't adding any -l switches to LDFLAGS. Arguably, the real problem here is that we're abusing LDFLAGS by putting -l switches in it and we should stop doing that. But changing that would be quite invasive, so I'm not eager to do so. Perhaps this is a candidate for back-patching, but so far it seems that problems can only be exhibited in test code we don't normally build, and at least some of the problems are new in HEAD anyway. So I'll refrain for now. Donald Dong and Tom Lane Discussion: https://postgr.es/m/CAKABAquXn-BF-vBeRZxhzvPyfMqgGuc74p8BmQZyCFDpyROBJQ@mail.gmail.com
Building a shlib on AIX requires use of the mkldexport.sh script, but we failed to install that, preventing its use from non-source-tree contexts. Also, Makefile.aix had the wrong idea about where to find the installed copy of the postgres.imp symbol file used by AIX. Per report from John Pierce. Patch all the way back, since this has been broken since the beginning of PGXS.
Per recommendation from Peter. Neither choice is bulletproof, but this is the existing style and it does help prevent unexpected environment variable substitution.
Per buildfarm results.