Source: bedtools Version: 2.31.0+dfsg-1 Severity: serious Tags: ftbfs Justification: ftbfs
Dear Maintainer, While making a routine metadata update on a Debian patch against bedtools, Salsa CI caught a build failure on i386[1] which I could reproduce multiple times in sid and in testing with the unmodified version of the package. [1]: https://salsa.debian.org/med-team/bedtools/-/jobs/4891166 The relevant part of the build log shows the following differences in the test suite of the "intersect" tool: intersect.t22.p...0a1 > chr1 0 30 one_block_one_exon_30bp 40 - 0 30 0,0,0 1 30, 0, chr1 0 100 exon1 1 + 30 fail intersect.t22.q...1a2 > chr1 80 110 one_block_one_exon_20bp 40 - 80 110 0,0,0 1 30, 0, chr1 0 100 exon1 1 + 20 fail The tests are started from test/intersect/test-intersect.sh. I'm not sure what change caused the build failure to appear. The issue is not affecting amd64, nor armhf, which suggest something very i386 specific. I have not checked closely the other CPU architectures yet. The relevant upstream change that introduced this code begins to date back from quite some time ago[2] and the package builds with -ffloat-store for a while, so this may be caused by a dependency, or a compiler change. [2]: https://github.com/arq5x/bedtools2/commit/9d22ccb24f258553b0eff31e689b09563227331b Hope this helps pinpointing what's up, Étienne. -- System Information: Debian Release: trixie/sid APT prefers unstable APT policy: (500, 'unstable') Architecture: amd64 (x86_64) Foreign Architectures: i386 Kernel: Linux 6.5.0-3-amd64 (SMP w/12 CPU threads; PREEMPT) Locale: LANG=C.UTF-8, LC_CTYPE=C.UTF-8 (charmap=UTF-8), LANGUAGE not set Shell: /bin/sh linked to /usr/bin/dash Init: systemd (via /run/systemd/system) LSM: AppArmor: enabled -- .''`. Étienne Mollier <emoll...@debian.org> : :' : gpg: 8f91 b227 c7d6 f2b1 948c 8236 793c f67e 8f0d 11da `. `' sent from /dev/pts/7, please excuse my verbosity `- on air: Condition Red - Let It All Come Out
signature.asc
Description: PGP signature