I noticed that src:libfabric regressed on amd64 according to https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/libfabric.html
The difference looked odd and then I noticed that version 2.0.0-2 in
the official builds on arm64, ppc64, s390x and sparc64 currently (not
all builds are done) have the same issue. The Depends line in
libfabric-bin shows:
Depends: libc6 (>= 2.38), libfabric1 (>= 2.0.)
Note the cropped version for libfabric1.
* Paul Gevers <elbrus@debian.org>, 2025-03-15 10:15:
I noticed that src:libfabric regressed on amd64 according to
https://tests.reproducible-builds.org/debian/rb-pkg/unstable/amd64/libfabric.html
The difference looked odd and then I noticed that version 2.0.0-2 in
the official builds on arm64, ppc64, s390x and sparc64 currently (not
all builds are done) have the same issue. The Depends line in
libfabric-bin shows:
Depends: libc6 (>= 2.38), libfabric1 (>= 2.0.)
Note the cropped version for libfabric1.
libfabric1 has these in the symbols file:
fi_freeinfo@FABRIC_1.8 2.0.0
fi_getinfo@FABRIC_1.8 2.0.
Both of these symbols used by the *-bin package. The truncated and non-truncated versions are equal (in the sense of Policy §5.6.12). I
suppose that confuses dpkg-shlibdeps?
Sysop: | Keyop |
---|---|
Location: | Huddersfield, West Yorkshire, UK |
Users: | 481 |
Nodes: | 16 (2 / 14) |
Uptime: | 26:46:36 |
Calls: | 9,543 |
Calls today: | 3 |
Files: | 13,656 |
Messages: | 6,140,474 |