Re: bad package build of gcc10

[Jonathan Perkin <jperkin%mnx.io@localhost>]

* On 2024-10-18 at 23:40 BST, Greg Troxel wrote:

> and that seems to be the issue: //lib/. instead of /lib, leading to a
> REQUIRES that should have been a match to the PROVIDES and hence
> omittted.  Basically, the checking code is not using the same resolution
> rules that the dynamic linker does.

The current implementation parses ldd, see:

  requires=`(${PKGSRC_SETENV} ${LDD_ENV:U} ...

in mk/pkgformat/pkg/metadata.mk.  For the SmartOS builds I wrote a patch 
many years ago to do it via elfdump/NEEDED instead:

  https://github.com/TritonDataCenter/pkgsrc/commit/9eedc359894303e4cdc19eee6204c8c648a12543

which I've been using successfully since then, but that's obviously not 
generic so is unsuitable for committing.

> Does the theory that it should have been canonicalized (well, not RPATH,
> but the resulting lib) seem valid?

Yes, they should.

--
Jonathan Perkin   -   mnx.io   -   pkgsrc.smartos.org
Open Source Complete Cloud   www.tritondatacenter.com