On 16/11/2024 12:21, Alastair McKinstry wrote:
On 07/11/2024 11:31, Emilio Pozuelo Monfort wrote:
On 01/11/2024 11:43, Emilio Pozuelo Monfort wrote:
On 22/10/2024 08:44, Alastair McKinstry wrote:
On 21/10/2024 13:25, Drew Parsons wrote:
All uploads for this transition are now done.
We're waiting for openmpi to be fixed so petsc and reverse dependencies can
build.
Build is done for amd64 (and i386).
For that reason I suggest triggering the last binNMUs, for dolfinx-mpc and
fenicsx-performance-tests. They should be able to (re)build successfully
for amd64 (and i38 now6), and will wait their turn on the other arches.
Drew
I've uploaded a version to experimental (5.0.5-4) that fixes the RC bug and
preps the libopenmpi40 (cxx issue) transition. It needs more testing
(especially the transition)
Can you check the openmpi autopkgtests pass? They are currently failing in
unstable, see [1]. Also a test rebuild of rdeps would be nice before ack'ing
this.
Any news on this? We need to make progress here in order to untangle many
transitions that got stuck due to openmpi.
I've done an upload of openmpi-5.0.5-6 to unstable this morning that fixes the
autopkgtest failures, and some build failures discovered in a rebuild for
libopenmpi40.
I will be uploading rebuilds of failing packages later today/tomorrow.
As of now, libopenmpi40 does not trigger any (new) openmpi-related failures.
Summary to follow after uploads.
I have been filing bugs, doing NMUs and requesting removal for NBS binaries, and
we're at a point I think where we can get rid of openmpi on armel/armhf/i386.
However, the autopkgtest regressions are still an issue for openmpi migration to
testing, and that I think is blocked on the libopenmpi40 transition if I
understood things correctly.
Have you been able to test-rebuild the rdeps against the new libopenmpi40? Where
are we with that?
Cheers,
Emilio
