Andreas Tille, on 2020-11-12 18:35:28 +0100: > On Thu, Nov 12, 2020 at 05:45:44PM +0100, Étienne Mollier wrote: > > While working on updating the libmaus2 to version 2.0.762, the > > program routine-update caught the version 2.0.764, which failed > > to build, apparently due to a new dependency not yet part of > > Debian, the brand new headers-only libsecrecy: > > > > https://gitlab.com/german.tischler/libsecrecy/ > > > > I'm finishing the update of libmaus2 to version 2.0.762, > > although it is already outdated, because the current version > > 2.0.743 in unstable is too old for the newer version of its > > biobambam2 reverse dependency. > > Sounds pretty sensible. > > > I am also not sure of the time > > it would take to get the libsecrecy into the archive given that > > the freeze is two months from now. > > Given that ftpmaster is soooooo fast since some time I have no > doubt we will manage this. If you want to start a new package > (which looks pretty straightforward from a short look) you can > do so. Otherwise I might have some spare cycles tomorrow. I haven't done that many new packages; I suppose it shouldn't hurt that I give a try to this one. :) > > I pushed changes on Salsa for further review, as I'm not > > entirely sure yet that the library can be released as is: > > > > https://salsa.debian.org/med-team/libmaus2 > > > > I'm still running additionnal builds to catch eventual > > regressions on arm64 and ppc64el, just in case. > > Very cool. Please let us know the result. The build and build-time tests targeting arm64 went quite well. The build went well on ppc64el, but build-time test crashed on testdnarank.sh and testRank.sh at the same steps as riscv64[1], itanium[2] and mips64el[3] (that is, "testing CacheLineRank" and "running short tests", respectively). [1] https://buildd.debian.org/status/fetch.php?pkg=libmaus2&arch=riscv64&ver=2.0.743%2Bdfsg-1&stamp=1598631353&raw=0 [2] https://buildd.debian.org/status/fetch.php?pkg=libmaus2&arch=ia64&ver=2.0.743%2Bdfsg-1&stamp=1598627935&raw=0 [3] https://buildd.debian.org/status/fetch.php?pkg=libmaus2&arch=mips64el&ver=2.0.743%2Bdfsg-1&stamp=1598625102&raw=0 There is one difference though, the test now crashes with a segmentation fault, instead of a full report with a backtrace: [V] testing CacheLineRank: qemu: uncaught target signal 11 (Segmentation fault) - core dumped ./testRank.sh: line 7: 474960 Segmentation fault ../src/testRank Exiting with return code 139 FAIL testRank.sh (exit status: 139) [...] [V] running short tests... qemu: uncaught target signal 11 (Segmentation fault) - core dumped ./testdnarank.sh: line 7: 474972 Segmentation fault ../src/testdnarank Exiting with return code 139 FAIL testdnarank.sh (exit status: 139) At least, the build on i386 went through, so there are chances #934619 can be addressed with a new update at least. I started an armhf build just to see... Kind Regards, -- Étienne Mollier <etienne.mollier@mailoo.org> Fingerprint: 8f91 b227 c7d6 f2b1 948c 8236 793c f67e 8f0d 11da Sent from /dev/pts/7, please excuse my verbosity.
Attachment:
signature.asc
Description: PGP signature