From: Felix Gruber <felgru@posteo.net>
To: "Ludovic Courtès" <ludovic.courtes@inria.fr>
Cc: 62842@debbugs.gnu.org
Subject: [bug#62842] [PATCH 00/11] gnu: maths: dune-*: Update to version 2.9.0
Date: Tue, 2 May 2023 20:18:13 +0000 [thread overview]
Message-ID: <cdfb119e-1794-a470-c6aa-836013bc51c8@posteo.net> (raw)
In-Reply-To: <877ctq99ay.fsf@gnu.org>
[-- Attachment #1: Type: text/plain, Size: 2386 bytes --]
Hi Ludo’,
On 5/2/23 15:01, Ludovic Courtès wrote:
> Felix Gruber <felgru@posteo.net> skribis:
>
>> This patchset updates all packages belonging to the Dune finite element
>> library to the latest release 2.9.0.
>
> There’s one test failure on aarch64-linux:
>
> --8<---------------cut here---------------start------------->8---
> 99/106 Test #99: eigenvaluestest ........................Subprocess aborted***Exception: 1.24 sec
> Eigenvalues of Rosser matrix: (-1020.05,0) (0.0980486,0) (-1.05342e-14,0) (1020.05,0) (1020,0) (1019.9,0) (1000,0) (1000,0)
> Eigenvalues of Rosser matrix: (-1020.05,0) (0.0980486,0) (-1.05342e-14,0) (1020.05,0) (1020,0) (1019.9,0) (1000,0) (1000,0)
> Eigenvalues of Rosser matrix: (-1020.05,0) (0.0980486,0) (-1.05342e-14,0) (1020.05,0) (1020,0) (1019.9,0) (1000,0) (1000,0)
> terminate called after throwing an instance of 'Dune::MathError'
> what(): MathError [testSymmetricFieldMatrix:/tmp/guix-build-dune-common-2.9.0.drv-0/dune-common-2.9.0/dune/common/test/eigenvaluestest.cc:143]: Vector computed by FMatrixHelp::eigenValuesVectors is not an eigenvector
>
> […]
> The following tests FAILED:
> 99 - eigenvaluestest (Subprocess aborted)
> Errors while running CTest
> make: *** [Makefile:74: test] Error 8
> --8<---------------cut here---------------end--------------->8---
>
> From <https://ci.guix.gnu.org/build/1228281/details>.
>
> Could you take a look and/or report it upstream?
Thanks for letting me know. I've looked into upstream issues/fixes for
this problem and it seems that these aarch64 test failures are a known
issue [1]. This bugreport mentions build failures in dune-common's
Debian package, which nowadays seems to be patched to disable the
failing tests [2].
[1]: https://gitlab.dune-project.org/core/dune-common/-/issues/262
[2]:
https://sources.debian.org/src/dune-common/2.9.0-2/debian/patches/disable-long-double-eigenvalue-checks.patch/
I've prepared a patch to apply Debian's patch for disabling the failing
tests to our dune-common package. You can find it attached to this
email. Can you test if it fixes the aarch64-linux builds (I don't have
access to an aarch64 machine)?
I hope that this would also fix the build failures for the other dune-*
packages on aarch64-linux, which are currently all failing because they
depend on the broken dune-common package.
Kind regards,
Felix
[-- Attachment #2: 0001-gnu-dune-common-Fix-build-on-aarch64-linux.patch --]
[-- Type: text/x-patch, Size: 3397 bytes --]
From 1001e15623108e6f92b89a1a2daacd43b5e2a123 Mon Sep 17 00:00:00 2001
From: Felix Gruber <felgru@posteo.net>
Date: Tue, 2 May 2023 21:59:30 +0200
Subject: [PATCH] gnu: dune-common: Fix build on aarch64-linux.
* gnu/packages/maths.scm (dune-common)[source]: Add patch to disable
tests that are failing on aarch64-linux.
* gnu/packages/patches/dune-common-disable-long-double-eigenvalue-checks.patch:
New file.
* gnu/local.mk (dist_patch_DATA): Register it.
---
gnu/local.mk | 1 +
gnu/packages/maths.scm | 4 ++-
...isable-long-double-eigenvalue-checks.patch | 28 +++++++++++++++++++
3 files changed, 32 insertions(+), 1 deletion(-)
create mode 100644 gnu/packages/patches/dune-common-disable-long-double-eigenvalue-checks.patch
diff --git a/gnu/local.mk b/gnu/local.mk
index 1476c730ba..68a40e7c6c 100644
--- a/gnu/local.mk
+++ b/gnu/local.mk
@@ -1056,6 +1056,7 @@ dist_patch_DATA = \
%D%/packages/patches/dovecot-trees-support-dovecot-2.3.patch \
%D%/packages/patches/dstat-fix-crash-when-specifying-delay.patch \
%D%/packages/patches/dstat-skip-devices-without-io.patch \
+ %D%/packages/patches/dune-common-disable-long-double-eigenvalue-checks.patch \
%D%/packages/patches/dune-grid-add-missing-include-cassert.patch \
%D%/packages/patches/dune-istl-fix-solver-playground.patch \
%D%/packages/patches/dvd+rw-tools-add-include.patch \
diff --git a/gnu/packages/maths.scm b/gnu/packages/maths.scm
index d7e2719b70..31ae3796fc 100644
--- a/gnu/packages/maths.scm
+++ b/gnu/packages/maths.scm
@@ -6805,7 +6805,9 @@ (define-public dune-common
version "/dune-common-" version ".tar.gz"))
(sha256
(base32
- "04pzk8q0bibci8z5xlwndhh3y3vs63mw7kad62lbzfwrr5121hrd"))))
+ "04pzk8q0bibci8z5xlwndhh3y3vs63mw7kad62lbzfwrr5121hrd"))
+ (patches (search-patches
+ "dune-common-disable-long-double-eigenvalue-checks.patch"))))
(build-system cmake-build-system)
(arguments
`(#:phases
diff --git a/gnu/packages/patches/dune-common-disable-long-double-eigenvalue-checks.patch b/gnu/packages/patches/dune-common-disable-long-double-eigenvalue-checks.patch
new file mode 100644
index 0000000000..cdab27b933
--- /dev/null
+++ b/gnu/packages/patches/dune-common-disable-long-double-eigenvalue-checks.patch
@@ -0,0 +1,28 @@
+--- a/dune/common/test/eigenvaluestest.cc
++++ b/dune/common/test/eigenvaluestest.cc
+@@ -340,20 +340,20 @@
+ testSymmetricFieldMatrix<double,200>();
+ testSymmetricFieldMatrix<float,4>();
+ testSymmetricFieldMatrix<float,200>();
+- testSymmetricFieldMatrix<long double,4>();
+- testSymmetricFieldMatrix<long double,200>();
++// testSymmetricFieldMatrix<long double,4>();
++// testSymmetricFieldMatrix<long double,200>();
+ #endif // HAVE_LAPACK
+
+ testSymmetricFieldMatrix<double,2>();
+ testSymmetricFieldMatrix<double,3>();
+ testSymmetricFieldMatrix<float,2>();
+ testSymmetricFieldMatrix<float,3>();
+- testSymmetricFieldMatrix<long double,2>();
+- testSymmetricFieldMatrix<long double,3>();
++// testSymmetricFieldMatrix<long double,2>();
++// testSymmetricFieldMatrix<long double,3>();
+
+ checkMultiplicity<double>();
+ checkMultiplicity<float>();
+- checkMultiplicity<long double>();
++// checkMultiplicity<long double>();
+
+ return 0;
+ }
--
2.39.2
next prev parent reply other threads:[~2023-05-02 20:19 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-14 19:25 [bug#62842] [PATCH 00/11] gnu: maths: dune-*: Update to version 2.9.0 Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 01/11] gnu: dune-common: Update to 2.9.0 Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 02/11] gnu: dune-geometry: " Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 03/11] gnu: dune-uggrid: " Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 04/11] gnu: dune-grid: " Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 05/11] gnu: dune-istl: " Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 06/11] gnu: dune-localfunctions: " Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 07/11] gnu: dune-alugrid: " Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 08/11] gnu: dune-subgrid: Update to 2.9.0-1-e83f3f9 Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 09/11] gnu: dune-typetree: Update to 2.9.0 Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 10/11] gnu: dune-functions: " Felix Gruber
2023-04-14 19:27 ` [bug#62842] [PATCH 11/11] gnu: dune-pdelab: Update to 2.8.0-0.d5dddb6 Felix Gruber
2023-05-02 10:13 ` bug#62842: [PATCH 00/11] gnu: maths: dune-*: Update to version 2.9.0 Ludovic Courtès
2023-05-02 13:01 ` [bug#62842] " Ludovic Courtès
2023-05-02 20:18 ` Felix Gruber [this message]
2023-05-04 9:35 ` Ludovic Courtès
2023-05-02 13:02 ` Ludovic Courtès
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=cdfb119e-1794-a470-c6aa-836013bc51c8@posteo.net \
--to=felgru@posteo.net \
--cc=62842@debbugs.gnu.org \
--cc=ludovic.courtes@inria.fr \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
Code repositories for project(s) associated with this external index
https://git.savannah.gnu.org/cgit/guix.git
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.