On 2019-04-13 17:16, Paul Gevers wrote:
Hi Drew,
97cd175 also had memory fixes (BigInt) which did not apply cleanly. So
I've removed those parts of the patch (in debian patch
fix_race_condition_97cd175.patch)
There's quite some white-space changes (and one comment) in one of the
patches. For a next time I would appreciate it if those were removed as
it makes reviewing easier.
I was trying to keep as much of the upstream commit as possible, but
fair call, if I'm decimating the patch anyway then might as well go
whole-hog.
I've uploaded the 2 patches as hypre 2.15.1-5
They build and pass tests. We'll need to run a barrage of tests to
see
if they fixing the intermittent test failures.
As with python-scipy, I ran a bunch of tests (total to 21, all in
unstable). I'll unblock it.
If it doesn't help, the next option is to take the full transition to
hypre 2.16.0.
It's too late in the release to consider this option for buster. Please
don't upload to unstable until we release buster.
Indications are positive, 22/22 tests passed.
Drew
--
debian-science-maintainers mailing list
[email protected]
https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-science-maintainers
