On Sun, 2018-06-24 at 18:40 +0200, Michael Banck wrote:
> Hi Drew,
>
> On Sun, Sep 10, 2017 at 01:48:09PM +0800, Drew Parsons wrote:
> > nwchem cannot build on its own, since changes are made to some
> > upstream files outside of the debian/patches system.
...
> > web
On Sat, 23 Jun 2018 12:54:19 +0300 Adrian Bunk wrote:
> ...
> odepkg_octsolver_mebdfdae.cc:42:10: fatal error: config.h: No such
file or directory
> #include
> ^~
The patch is trivial. Replace all with .
config.h was already deprecated, and is now removed in octave 4.4.0.
On Thu, 28 Jun 2018 20:57:52 +0200 =?iso-8859-1?Q?S=E9bastien?=
Villemot wrote:
> >
> > The patch is trivial. Replace all with .
> >
> > config.h was already deprecated, and is now removed in octave
4.4.0.
>
> Yes, but then you get many numerical test failures, which are not
trivial at
> all t
On Tue, 2018-07-03 at 15:25 +0100, Carnë Draug wrote:
> On 3 July 2018 at 14:59, Andreas Tille wrote:
> > Hi,
> >
> > the bug log of #790196[1] shows that rasmol in its current state
> > will
> > not be distributable with Buster. Upstream is dead so if we want
> > to
> > keep it in Debian we hav
On Tue, 10 Jul 2018 19:57:57 +0200 Anton Gladky >
> it looks like the version 3.1.1.real-1 introduces the regression in
autopkgtest.
>
Not just the autopkgtest, it's fouling up petsc and dolfin too ;(
On Wed, 11 Jul 2018 13:29:54 +0800 Drew Parsons
wrote:
>
> Not just the autopkgtest, it's fouling up petsc and dolfin too ;(
Might be worth mentioning though, openmpi 3.1.1.real-1 has been with us
over a week now, but the problem with the dolfin build (unable to build
with petsc,
Package: libopenmpi-dev
Version: 3.1.1.real-2
Severity: grave
Justification: renders package unusable
Did something get missed in 3.1.1.real-2? It still depends on
libpmix2 (e.g.
/usr/lib/x86_64-linux-gnu/openmpi/lib/openmpi3/mca_pmix_pmix2x.so),
and petsc tests fail, reporting:
cd src/snes/ex
On Thu, 12 Jul 2018 22:35:04 +0800 Drew Parsons
wrote:
>
> and petsc tests fail, reporting:
Never mind petsc, a simpler test will do:
$ mpirun -np 2 date
[grendel:03595] mca_base_component_repository_open: unable to open
mca_pmix_pmix2x:
/usr/lib/x86_64-linux-gnu/openmpi/lib/op
On Mon, 2018-07-16 at 13:54 +0200, Rafael Laboissière wrote:
>
> I tried to build the package using the tip of the Mercurial
> repository
> for the odepkg package [1]. After some tweaks in the documentation,
> I
> was able to compile the source code, but the unit tests fail
> miserably.
>
> I
On Tue, 2018-07-17 at 00:30 +0800, Drew Parsons wrote:
> On Mon, 2018-07-16 at 13:54 +0200, Rafael Laboissière wrote:
> >
> > I tried to build the package using the tip of the Mercurial
> > repository
>
Weird though, https://wiki.octave.org/Odepkg points at
http
Package: libreoffice-common
Version: 1:5.3.0~alpha1-1
Severity: serious
Justification: Policy 6.2
The new libreoffice 5.3 in experimental fails to install.
libreoffice-common fails during preinst attempting to remove
directories which are not there:
Preparing to unpack .../libreoffice-common_1%3
On 2019-02-26 14:54, Andreas Tille wrote:
I just pushed
+ceres-solver (1.14.0-4) UNRELEASED; urgency=medium
...
to Git. Is there any reason not to upload this and simply fix #883619.
Looks like a pretty low hanging fruit to fix an RC bug and save the
package for Buster that I can not imagine
On Tue, 28 Feb 2017 10:46:53 + Santiago Vila
wrote:
> Package: src:mpi4py
> Version: 2.0.0-2
...
>
> I tried to build this package in stretch with "dpkg-buildpackage -A"
> but it failed:
...
> MPI.Win.Create(None, 1, MPI.INFO_NULL, MPI.COMM_SELF).Free()
> File "MPI/Win.pyx", line 73, in
On Wed, 2017-03-08 at 11:30 +0100, Andreas Tille wrote:
> Hi Drew,
>
> ... your suggestion to
> turn an error about "invalid window" into a warning seems to be an
> apropriate action in this case.
Thanks Andreas.
> May be the bug could be reopened later with lower severity to make
> sure
> the i
On Wed, 2017-03-08 at 14:19 +0100, Andreas Tille wrote:
>
> Perfect. I was just wondering whether closing + reopening is
> somehow
> easier to handle technically since testing migration requires closing
> an RC bug - but yes, we both mean the same in the end.
That's a fair point. For the sake o
.UTF-8, LC_CTYPE=en_AU.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Init: systemd (via /run/systemd/system)
>From dd69f5f6bafeb518cd236ca167aff5233ab0daeb Mon Sep 17 00:00:00 2001
From: Drew Parsons
Date: Fri, 10 Mar 2017 11:53:14 +0800
Subject: [PATCH 1/2] NMU: fix FTBFS by tre
On Wed, 08 Feb 2017 14:25:12 +0100 =?ISO-8859-1?Q?S=E9bastien?=
Villemot wrote:
>
> I don't know what's the proper solution to this issue. The
circularity
> can easily be dealt with (using patchelf for closing the loop in
> DT_NEEDED entries).
>
> The problem is rather that the Cblacs_pinfo symb
On Tue, 21 Mar 2017 10:50:29 +0800 Drew Parsons
wrote:
>
> That document is from 1997 though. The MPI standard has moved
through
> 2 major versions since then. But blacs remains unchanged.
Further, BLACS is now part of (packaged with) scalapack, see
http://netlib.org/scalapack/scalap
Package: im-config
Version: 0.36-1
Severity: critical
Justification: breaks unrelated software
im-config currently generates an error at login:
/usr/lib/gdm3/gdm-x-session[6333]: /etc/gdm3/Xsession: 59:
/usr/share/im-config/data/21_ibus.rc: Syntax error: "fi" unexpected (expecting
"then")
Thi
Source: pybind11
Followup-For: Bug #917722
This looks like more fallout from the new numpy. Bad numpy, bad.
Package: libopenmpi3
Version: 3.1.3-8
Severity: serious
Justification: Policy 8.6
openmpi 3.1.3-8 is triggering test failure on a wide range of client
packages, including combblas, dolfin and family, petsc and family,
superlu-dist.
dolfin and petsc (etc) leave a record of the fault,
e.g. petsc
h
Source: python-pbcore
Followup-For: Bug #917670
Control: title -1 FTBFS: dh_auto_test: numpy 1.16 arraysetops dtype error
Control: forwarded -1 https://github.com/PacificBiosciences/pbcore/issues/120
Confirmed, still occurs with pbcore 1.6.5.
Bug filed upstream.
Drew
Control: tag -1 pending
Hello,
Bug #917670 in python-pbcore reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:
https://salsa.debian.org/med-team/python-pbcore/commit/9f8a0d4c84e5f544ecfbe5
Package: spyder
Version: 3.3.2+dfsg1-1
Followup-For: Bug #919000
Control: severity -1 normal
It's hardly critical. Please don't overinflate bug severities.
https://www.debian.org/Bugs/Developer#severities
On 2019-01-12 10:08, Adrian Bunk wrote:
https://www.debian.org/Bugs/Developer#severities
"serious" is the severity that is commonly used for missing
dependencies.
"Grave" could also be justified, "makes the package mostly unusable".
Drew
Package: libpmix2
Version: 3.1.2-1
Severity: grave
Justification: renders package unusable
The new version of libpmix2 is confusing client packages,
breaking CI tests in many packages, e.g. for petsc:
run SNES testex19
Possible error running C/C++ src/snes/examples/tutorials/ex19 with 1 MPI
Package: release.debian.org
Severity: serious
User: release.debian@packages.debian.org
Usertags: binnmu
dolfin has started to FTBFS, see
https://tests.reproducible-builds.org/debian/rbuild/unstable/amd64/dolfin_2018.1.0.post1-13.rbuild.log.gz
The dolfin build error comes from mpi4py. The dol
On Thu, 2017-01-05 at 14:10 +0100, Santiago Vila wrote:
> found 802706 3.7.3+dfsg1-3
> thanks
>
> Hi.
>
> Sorry for the reopening but this is happening again in stretch.
> (I built this package 200 times, and it failed 200 times).
>
> Build logs available here:
>
> https://people.debian.org/~sa
Package: texlive-bibtex-extra
Version: 2016.20170118-1
Severity: grave
Justification: renders package unusable
The new version of biblatex in this package cannot be used with biber.
That makes this biblatex unusable.
Output looks like:
$ biber project
INFO - This is Biber 2.6
INFO - Logfile is 'p
On Fri, 20 Jan 2017 18:22:01 +0100 Matthias Klose
wrote:
>
> The severity of this report is likely to be raised before the
release,
> so that the gcc-5 package can be removed for the release.
>
"Likely to be raised..." ?? The severity is already serious!
Drew
Package: libscotchmetis-dev
Version: 7.0.3-1
Followup-For: Bug #1036576
Awkward. Thanks for catching it.
Package: python3-duecredit
Followup-For: Bug #1030408
Control: forwarded -1 https://github.com/duecredit/duecredit/pull/184
This bug is affecting unrelated packages, confirming the Severity:serious.
For instance the new mp_api module tries to read its version using
pkg_resources.get_distribution,
Control: tag -1 pending
Hello,
Bug #1029690 in python-scipy reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:
https://salsa.debian.org/python-team/packages/scipy/-/commit/5b51c3695ddc2750
On 2023-02-09 15:02, Andreas Tille wrote:
Control: tags -1 help
Am Thu, Feb 09, 2023 at 12:50:45AM +0100 schrieb Santiago Vila:
Package: src:umap-learn
Version: 0.4.5+dfsg-3
Severity: serious
Tags: ftbfs
Dear maintainer:
During a rebuild of all packages in bookworm, your package failed to
bu
On 2023-02-09 20:37, Andreas Tille wrote:
Hi again,
Am Thu, Feb 09, 2023 at 05:52:13PM +0100 schrieb Andreas Tille:
> That said, the np.matrix discrepancy was only just fixed recently, in
> December, by
> https://github.com/lmcinnes/umap/pull/946
I've updated the packaging, added the PR, fixe
Source: xtensor
Followup-For: Bug #1030775
Control: tags -1 wontfix moreinfo
I think this is not a bug. xtensor is a header-only library. The
builds you're seeing with -march=native are in the tests only, not
compiled into the binary package.
We could make some effort to prevent the tests runnin
Source: pymatgen
Followup-For: Bug #1031705
The situation is not entirely clear. That is, there is another
timeout in
https://ci.debian.net/data/autopkgtest/unstable/amd64/p/pymatgen/31542441/log.gz
but in this case in test_cif.py. test_babel.py passed.
Does it mean there are two separate tests
Source: pymatgen
Followup-For: Bug #1031705
Apart from that one case of test_cif.py, every other timeout occurs in
test_babel.py. If it were a 3rd test being held up and resolving too
late for other tests to pass, then the timeout would occur more
randomly.
So looks like the 7th item (test_local
Package: opendrop
Version: 3.3.1-4
Severity: grave
Tags: patch upstream
Justification: renders package unusable
debian patch sundials_nvector_API6.patch enabled opendrop to build
against sundials 6.
Nevertheless it still fails when attempting to analysis interfacial
tension measurements:
[ARKOD
Source: xrayutilities
Version: 1.7.4-1
Severity: serious
Justification: debci
python3-xrayutilities has an inconsistent Dependency on h5py
It declares python3-h5py-serial [alpha, hppa, m68k, sparc64, x32], in
other words is not declared for most architectures.
But xrayutilities needs h5py in io/h
Source: statsmodels
Version: 0.13.5+dfsg-4
Severity: serious
Justification: debci
scipy 1.10 is triggering a problem with tests using int64 on 32 bit
arches (armhf, i386, armel)
Affected tests are in test_discrete.py:
TestDiscretizedGamma.test_basic
TestDiscretizedExponential::test_basic
TestDisc
On 2023-02-01 22:40, Stefano Rivera wrote:
Hi Drew (2023.01.31_18:48:32_-0400)
xrayutilities does have Build-Depends: python3-h5py. Evidentally
dh-python3 isn't able to determine the correct dependency for h5py
(likely it gets confused by python3-h5py-serial). Until that's fixed,
python3-xrayu
Control: reopen 1030220
h5py 3.7.0-4 is uploaded fixing the source of the problem, but
xrayutilities will need to be rebuilt to use the new pydist definitions.
Source: scipy
Followup-For: Bug #1037854
The errors in the build log refer to pythran, not scipy.
Control: tag -1 pending
Hello,
Bug #1037823 in pythran reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:
https://salsa.debian.org/python-team/packages/pythran/-/commit/3dedbc4004f93215593
Source: dolfin
Version: 2019.2.0~legacy20230609.8b85e9d.ar-2
Severity: serious
Justification: Policy 8.6.2
Control: affects -1 python3-mshr
The accumulation of git changes and gcc-13 fixes has apparently
introduced an ABI inconsistency in libdolfin.so, such that mshr can no
longer be imported in p
Source: mplcursors
Version: 0.5.2-1
Severity: serious
Justification: debci
mplcursors Build-Depends:libopenblas0
This is bad for several reasons.
Firstly, OpenBLAS is not available on all architectures. In particular
it's not available on armel (and mipsel), and therefore debci testing
on armel
Actually looking more closely at the source, mplcursors doesn't even use
BLAS at all.
In that case just remove the libopenblas0 dependency altogether (both
Build-Depends: and Depends: for the binary package).
Source: python-dmsh
Followup-For: Bug #1027230
I can't reproduce this error. debci tests are passing fine on all
architectures.
Source: petsc4py
Followup-For: Bug #1026346
Control: affects 1027044 src:petsc4py
As far as I can tell, this error is the python/numpy verson of
"missing symbol" error after a library update with ABI bump. A
similar error was found in many packages depending on numpy during the
recent upgrade to
Source: matplotlib
Version: 3.6.2-3
Severity: serious
Justification: debci
Control: forwarded -1 https://github.com/matplotlib/matplotlib/issues/24842
Control: block 1027170 by -1
mplot3d.axes3d.quiver() accepts boths args and kwargs in its function
definition.
Both args (any remaining after the
severity 1026421 important
thanks
The version of rclone in testing is not directly intended to be built on
stable, hence this bug is not Severity: serious (it builds fine in
unstable and testing).
Severity: serious is intended to mark packages which are not fit for
release in testing (the fu
Package: coinor-libipopt-dev
Version: 3.11.9-2.1+b2
Severity: serious
Justification: causes memory leaks
IPOPT causes memory leaks due to use of not initialized memory with
MUMPS >= 5.1.0.
The problem affects idyntree, see
https://github.com/robotology/idyntree/issues/456
The latest version of i
Source: satpy
Version: 0.16.1-1
Severity: serious
Justification: FTBFS
Control: affects -1 src:python-scipy
satpy now FTBFS (and fails tests) because of the recent upgrade of
python3-pygac to 1.1.0-2.
test_basic_check_satpy (satpy.tests.test_config.TestCheckSatpy)
Test 'check_satpy' basic functio
Package: libpmi-pmix-dev
Version: 3.1.4~rc2-1
Severity: serious
Justification: breaks MPI client tests
Hi Alistair and Debian science,
I'm upgrading dolfin/fenics to 2019.1.0 alongside HYPRE 2.16.0, PETSc
3.11.3, pybind11 2.3.0. The tests of the libraries have passed, but
the build is failing ba
On 2019-08-04 19:04, Drew Parsons wrote:
I'm upgrading dolfin/fenics to 2019.1.0 alongside HYPRE 2.16.0, PETSc
3.11.3, pybind11 2.3.0. The tests of the libraries have passed, but
the build is failing badly with an MPI error:
Run C++ regressions tests (serial)
Test project /home/pro
On 2019-08-04 19:17, Drew Parsons wrote:
On 2019-08-04 19:04, Drew Parsons wrote:
Probably this is a clue: libdolfin.so should link to libpetsc.so. But
it is missing in the new libdolfin.so. That means pmix is not the
problem, so I'll close this bug once I can confirm.
Yes, dol
On 2019-08-04 19:04, Drew Parsons wrote:
I'm upgrading dolfin/fenics to 2019.1.0 alongside HYPRE 2.16.0, PETSc
3.11.3, pybind11 2.3.0. The tests of the libraries have passed, but
the build is failing badly with an MPI error:
Run C++ regressions tests (serial)
Test project /home/pro
On 2019-08-04 19:19, Drew Parsons wrote:
Probably this is a clue: libdolfin.so should link to libpetsc.so. But
it is missing in the new libdolfin.so. That means pmix is not the
problem, so I'll close the bug against pmix once I can confirm.
CMake logs confirm the that petsc test
Source: pygalmesh
Severity: important
Tags: moreinfo,unreproducible
Followup-For: Bug #933848
I'm quite confused by this bug report. pygalmesh is configured to
build using pybuild not cmake, and it does build that way without
cmake (whether built via 'fakeroot debian/rules binary' or
'dpkg-buildp
Package: python3-pytest
Version: 4.6.4-1
Severity: serious
Justification: breaks autopkgtest tests
Control: affects -1 + src:apipkg src:betamax src:ccdproc src:chardet
Control: affects -1 + src:dask src:django-axes src:doit src:drms
Control: affects -1 + src:fiat src:mpi4py src:pandas src:pygalmesh
On 2019-08-05 16:37, Novy, Ondrej wrote:
Tags: wontfix
On Mon, 05 Aug 2019 11:43:39 +0800 Drew Parsons
wrote:
This upgrade should be treated as a Transition, uploading the
package to
experimental first.
We are not using Transitions for Python upgrades because almost every
module upgrade
On 2019-08-07 03:41, Novy, Ondrej wrote:
Hi,
Drew Parsons píše v Út 06. 08. 2019 v 17:25 +0800:
e.g what to do about
...
"/usr/lib/python3/dist-packages/flaky/flaky_pytest_plugin.py", line
272,
in call_runtest_hook
INTERNALERROR> when=when,
INTERNALERROR> Typ
On 2019-08-08 15:57, Ondřej Nový wrote:
I see, coming from flaky. Thanks Ondrej.
this should be fixed now with python-flaky 3.6.1-1.
Thanks again Ondrej. That fixes the mpi4py 3.0.2-10 tests.
Drew
Package: src:golang-github-anacrolix-ffprobe
Followup-For: Bug #934375
Control: reassign -1 src:golang-github-anacrolix-missinggo
The bug is in github.com/anacrolix/missinggo/leaktest
i.e. missinggo
-- System Information:
Debian Release: bullseye/sid
APT prefers unstable
APT policy: (500, 'u
Control: tag -1 pending
Hello,
Bug #934375 in golang-github-anacrolix-missinggo reported by you has been fixed
in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:
https://salsa.debian.org/go-team/packages/golang-github-
Source: petsc
Followup-For: Bug #987618
ACK. Thanks for finding this, Andreas. I'll upload once the current
update (-3) is migrated.
Drew
Source: golang-go.opencensus
Followup-For: Bug #988250
Control: forwarded -1
https://github.com/census-instrumentation/opencensus-go/issues/1259
Prevented from emailing cont...@bugs.debian.org directly because of
https://github.com/roundcube/roundcubemail/issues/6438#issuecomment-835782075
Source: golang-go.opencensus
Followup-For: Bug #988250
Control: severity -1 normal
debian patch skip_flaky_test.patch skips the flaky test in 0.22.4-2
Hence downgrading severity to normal.
Ignores the problem, doesn't actually fix it, so not closing the bug.
Package: grace
Version: 1:5.1.25-8
Severity: grave
Justification: renders package unusable
On a clean installation, grace fails to launch:
$ xmgrace
X Error of failed request: BadValue (integer parameter out of range for
operation)
Major opcode of failed request: 12 (X_ConfigureWindow)
Val
Package: evdi-dkms
Version: 1.6.4+dfsg-1
Severity: grave
Justification: renders package unusable
evdi-dkms is currently unable to build the kernel module.
/var/lib/dkms/evdi/1.6.4+dfsg/build/make.log reports:
CC [M] /var/lib/dkms/evdi/1.6.4+dfsg/build/evdi_gem.o
/var/lib/dkms/evdi/1.6.4+dfsg/
The drmP.h build error can be addressed with
https://gitweb.frugalware.org/frugalware-current/raw/master/source/lib-extra/evdi/kernel-5.5-test.patch
from
https://aur.archlinux.org/packages/evdi-git/#comment-744589
But then it hits the fatal error: linux/reservation.h: No such file or
direc
Probably evdi 1.7.0 will help, released just 3 hours ago.
forwarded 979290 https://github.com/aragilar/pytest-mpi/issues/14
thanks
There is a proposed fix in PR12,
https://github.com/aragilar/pytest-mpi/pull/12
Package: libucx0
Version: 1.10.0~rc1-2
Severity: serious
Justification: debci
Our next round of whack-a-mole comes from the new UCX.
pmix 4.0.0-3 seems to have fixed the pmix error from bug#979744.
debci tests next report a problem with UCX, with
openmpi 4.1.0-5
pmix 4.0.0-3
ucx 1.10.0~rc1-
Package: libopenmpi3
Version: 4.1.0-5
Followup-For: Bug #979041
There's evidence this libfabric bug is not fully fixed.
pytest-mpi (0.4-3) is failing tests:
A process has executed an operation involving a call
to the fork() system call to create a child process.
As a result, the libfabr
Source: numba
Followup-For: Bug #972246
X-Debbugs-Cc: Andreas Tille
Hi Andreas, check the upstream bug. The fix was merged only 4 days
ago, 0.52.0 does not have it.
We could try backporting PR#6579 to 0.52.0. Maybe it will work but
there is a reasonable chance that it won't.
Package: libopenmpi3
Version: 4.1.0-6
Followup-For: Bug #979041
Control: reopen 979041
We need to reopen this bug unfortunately. The libfabric
(RDMAV_FORK_SAFE) issue is still live in python MPI applications.
You can see it in pytest-mpi tests as reported previously,
or in a rebuild of mpi4py, e.
On 2021-01-15 23:14, Alastair McKinstry wrote:
Ugh. Thanks Drew.
What are the contents of /etc/openmpi/openmpi-mca-params.conf on the
node?
Does a simple hello world (see Debian/tests/hello* ) work without
errors in the environment ?
Hi Alastair, sorry for the delay replying to these questi
Lucas Nussbaum wrote:
Source: dolfin
Version: 2019.2.0~git20200629.946dbd3+lfs-4
...
During a rebuild of all packages in sid, your package failed to build
on amd64.
Hi Lucas, there have been various test failures associated with the
recent openmpi upgrade. They seem to be resolved now. debc
Source: mpi4py
Followup-For: Bug #976404
mpi4py tests seem to have started passing stably on ppc64el after the
upgrade of libpmix2 from 3.2.0-1 to 3.2.1-1
(other test errors after that are other known openmpi problems,
recently fixed, not these timeout errors)
Tests are now passing reliably in b
Package: python3-scipy
Version: 1.6.1-1
Severity: serious
Tags: upstream
Justification: FTBFS
Control: forwarded -1 https://github.com/scipy/scipy/issues/13585
Control: affects -1 src:pandas src:qutip
scipy 1.6.1 marketed itself as "bug-fix only", but in fact introduced
a change in the API handlin
Control: tag -1 pending
Hello,
Bug #983404 in python-scipy reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:
https://salsa.debian.org/python-team/packages/scipy/-/commit/5acfe43c748bc487f
Source: petsc4py
Followup-For: Bug #972385
When the petsc4py tests fail, it seems to correlate with
mpiexec -n 48 python3 test_mat_cg.py
When the tests pass, they pass with
mpiexec -n 2 python3 test_mat_cg.py
Evidently the MPI tests are calling on the full number of available
CPUs. Th
On 2020-10-21 18:29, Adrian Bunk wrote:
On Wed, Oct 21, 2020 at 10:21:10AM +, Debian Bug Tracking System
wrote:
...
mdtraj (1.9.4-6) unstable; urgency=medium
.
* debian patch check_sse2.patch checks if system supports SSE2
(i.e. if system is x86 or amd64) before compiling with -mss
On 2020-10-21 20:56, Adrian Bunk wrote:
On Wed, Oct 21, 2020 at 06:43:56PM +0800, Drew Parsons wrote:
I see. Would the i386 issue be the cause of the segfault in
test_image_molecules?
https://ci.debian.net/data/autopkgtest/testing/i386/m/mdtraj/7659912/log.gz
...
Doesn't look SSE re
forwarded 972385 https://gitlab.com/petsc/petsc4py/-/issues/7
thanks
The problem is not as simple as the kspsolve demo being too small to
distribute over 48 CPUs. The failure happens also with 5 processes.
Seems the problem is that the kspsolve demo needs the number of
processors to be a power of 2. Not 5, not 48.
On 2020-11-03 07:33, Rebecca N. Palmer wrote:
As this bug cannot immediately be fixed, numba may be removed from
testing to unblock the python3.9 transition (#966426).
...
pandas will drop its python3-numba build/autopkgtest-dependency
(#973589). Other reverse dependencies that can do so (dol
Source: dolfin
Followup-For: Bug #973767
This is a strange new error to be sure.
I suspect it's related to the SLEPc upgrade, there seem to be some
inconsistencies in the new slepc configuration. I'm looking into it.
If it's not slepc, then it might be some new behaviour from an openmpi
upgrade.
Source: dolfin
Followup-For: Bug #973767
Yeah, I'm more certain SLEPc is the problem. In the new release they
abolishing SLEPc.pc, renaming it slepc.pc. Since pkg-config is
case-sensitive, the unexpected change is causing problems.
Notice how the reported error is only with dolfin64. Because t
Package: sasview-doc
Followup-For: Bug #980701
The cutting point is:
> reading sources... [ 16%] dev/sasview-api/sas.qtgui.UnitTesting
> qt.qpa.xcb: could not connect to display
> qt.qpa.plugin: Could not load the Qt platform plugin "xcb" in "" even though
> it was found.
> This application fai
Package: paraview
Version: 5.9.0-1
Severity: grave
Justification: renders package unusable
Thanks for packaging the new paraview release. The latest version of
python-meshio should be able to run with it.
paraview is segfaulting (SIGBABRT) at start up though,
$ paraview
Warning: Ignoring XDG_SE
Package: paraview
Followup-For: Bug #981378
Control: reassign 981378 meshio-tools 4.3.5-2
Aha, the segfault disappears if meshio-tools is not installed. It
must be related to the meshio plugin. I'll transfer this bug report
to python-meshio.
Package: meshio-tools
Followup-For: Bug #981378
Control: severity -1 important
Control: forwarded -1
https://gitlab.kitware.com/paraview/paraview/-/issues/20412
Probably severity grave is overegging it.
paraview and meshio-tools won't necessarily be installed at the same
time. paraview will work
Control: tag -1 pending
Hello,
Bug #974560 in mpi4py reported by you has been fixed in the
Git repository and is awaiting an upload. You can see the commit
message below and you can check the diff of the fix at:
https://salsa.debian.org/science-team/mpi4py/-/commit/1ee01a245b7097a46da95489a5e962
On 2020-11-16 00:07, Adrian Bunk wrote:
On Mon, Nov 02, 2020 at 11:33:05PM +, Rebecca N. Palmer wrote:
...
pandas will drop its python3-numba build/autopkgtest-dependency
(#973589).
Other reverse dependencies that can do so (dolfinx??) should do the
same if
they wish to stay in testing.
On 2020-11-16 01:27, Adrian Bunk wrote:
On Mon, Nov 16, 2020 at 12:54:53AM +0800, Drew Parsons wrote:
On 2020-11-16 00:07, Adrian Bunk wrote:
>
> It seems numba might get fixed in time for bullseye:
> https://github.com/numba/numba/issues/6345#issuecomment-721738251
Yes, they&#
Source: python-meshio
Followup-For: Bug #975092
Control: forwarded -1 https://github.com/nschloe/meshio/issues/970
Already reported upstream.
Can Nico's application for DD be accelerated? He'll need access to 32
bit porterboxes to get to the bottom of the bug.
Drew
Control: forwarded 975470 https://github.com/nschloe/meshplex/issues/90
Yeah, issue #90 already raised upstream.
Drew
1 - 100 of 700 matches
Mail list logo
