'make ptestlong'
--
All tests passed!
--
Total time for all tests: 2293.2 seconds
cpu time: 14311.3 seconds
cumulative wall time: 17775.3 seconds
[jaap@vr
On Ubuntu 18.04 running on Core i7-6700HQ + 16 GB RAM:
- incremental build (-j8) from 9.0.rc0 OK
- all tests from ptestlong passed, except for a transient failure:
sage -t --long --warn-long 86.4 src/sage/modular/ssmod/ssmod.py # 2
doctests failed
This is the same error as that already reported f
On my Raspberry Pi 4B:
Total time for all tests: 37560.2 seconds
cpu time: 38702.7 seconds
cumulative wall time: 48875.0 seconds
make: *** [Makefile:178: ptestlong] Error 21
sage -t --long --warn-long 253.8 src/sage/rings/tests.py # Killed due to
segmentation fault
sage -t --long --warn-
MacBook Pro, Mojave 10.14.6, 'make' and 'make ptestlong' for Sage 9.0.rc1.
No problems!
make
> [sagelib-9.0.rc1] real 0m4.291s
> [sagelib-9.0.rc1] user 0m2.986s
> [sagelib-9.0.rc1] sys 0m1.135s
> real 2m26.975s
> user 2m2.529s
> sys 0m23.517s
> Sage build/upgrade complete!
make ptestlong
> All
Is that running with 4 threads and the 4GB model? 10h wall time is long but
just about fast enough that we could set up a buildbot
On Monday, December 30, 2019 at 11:08:57 PM UTC+1, Jaap Spies wrote:
>
> On my Raspberry Pi 4B:
> Total time for all tests: 37560.2 seconds
> cpu time: 387
4 threads but the 2GB model. I will have a 4GB model within a few days.
Testing with make ptestlong seems to be rather random. See my post on
sage-devel.
On Monday, December 30, 2019 at 11:39:36 PM UTC+1, Volker Braun wrote:
>
> Is that running with 4 threads and the 4GB model? 10h wall time is l
I'm surprised that you managed to build the documentation with 2gb, must
have been swapping quite a bit ;-)
On Monday, December 30, 2019 at 11:52:48 PM UTC+1, Jaap Spies wrote:
>
> 4 threads but the 2GB model. I will have a 4GB model within a few days.
> Testing with make ptestlong seems to be
Le mardi 31 décembre 2019 09:40:06 UTC+1, Frédéric Chapoton a écrit :
>
> I have problems when building R 3.6.2 on my (usually rather stable) Ubuntu
> patchbot.
>
Which version of Ubuntu? I could build R 3.6.2 without any trouble on
Ubuntu 18.04, with the system BLAS (i.e. the Ubuntu package lib
Welcome to Ubuntu 18.04.3 LTS (GNU/Linux 4.15.0-70-generic x86_64)
This fails in the same way, even after I installed libblas-dev.
F
Le mardi 31 décembre 2019 10:41:31 UTC+1, Eric Gourgoulhon a écrit :
>
> Le mardi 31 décembre 2019 09:40:06 UTC+1, Frédéric Chapoton a écrit :
>>
>> I have problem
On Tue, Dec 31, 2019 at 6:39 AM Volker Braun wrote:
>
> Is that running with 4 threads and the 4GB model? 10h wall time is long but
> just about fast enough that we could set up a buildbot
if it just needs to be an ARM 64, there are more powerful SoC boards, e.g.
https://www.seeedstudio.com/
you might have remains of unmerged tickets with spkg-configure for openblas
etc. Check that you don't have links to *blas.pc files in
local/lib/pkgconfig/
On Tue, 31 Dec 2019, 18:13 Frédéric Chapoton, wrote:
> Welcome to Ubuntu 18.04.3 LTS (GNU/Linux 4.15.0-70-generic x86_64)
>
> This fails in t
I had three files there : blas.pc, cblas.pc and openblas.pc. Removing them,
and then ./configure, and then make build : same problem, unchanged.
F
Le mardi 31 décembre 2019 11:36:21 UTC+1, Dima Pasechnik a écrit :
>
> you might have remains of unmerged tickets with spkg-configure for
> openblas
have you built openblas successfully?
On Tue, 31 Dec 2019, 18:52 Frédéric Chapoton, wrote:
> I had three files there : blas.pc, cblas.pc and openblas.pc. Removing
> them, and then ./configure, and then make build : same problem, unchanged.
>
> F
>
> Le mardi 31 décembre 2019 11:36:21 UTC+1, Dima
I have launched "sage -f openblas", and this is re-building many things.
Now building giac, so probably succedeed in building openblas.
F
Le mardi 31 décembre 2019 12:08:48 UTC+1, Dima Pasechnik a écrit :
>
> have you built openblas successfully?
>
> On Tue, 31 Dec 2019, 18:52 Frédéric Chapoton,
After 'sage -f openblas", I am back at the previous situation : I have the
three files blas.pc etc again in local/lib/pkgconfig. Building R still
fails in the same way.
F
Le mardi 31 décembre 2019 12:14:15 UTC+1, Frédéric Chapoton a écrit :
>
> I have launched "sage -f openblas", and this is re
I probably had a similar problem on _one_ of my three Linux Mint patchbots.
I then installed all recommended dependencies listed on
http://doc.sagemath.org/html/en/installation/source.html#linux-recommended-installation
as well as libpango1.0-dev . Then, IIRC, it worked again.
The libpan
It's probably
https://developer.r-project.org/Blog/public/2019/05/15/gfortran-issues-with-lapack/index.html
- and you have a slightly broken "new" gfortran...
Sorry for barking up the wrong tree.
IMHO, we should just use
https://trac.sagemath.org/ticket/27870 and
https://trac.sagemath.org/ticket/
So what would be a practical solution in my case ?
Le mardi 31 décembre 2019 16:18:34 UTC+1, Dima Pasechnik a écrit :
>
> It's probably
>
> https://developer.r-project.org/Blog/public/2019/05/15/gfortran-issues-with-lapack/index.html
>
> - and you have a slightly broken "new" gfortran...
>
> S
On Tue, Dec 31, 2019 at 11:23 PM Frédéric Chapoton wrote:
>
> So what would be a practical solution in my case ?
>
upgrade gfortran to a fixed version
(see https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90329)
perharps your Ubuntu version has a fix...
or downgrade to gfortran6
> Le mardi 31 déce
On Tue, Dec 31, 2019 at 11:32 PM Dima Pasechnik wrote:
>
> On Tue, Dec 31, 2019 at 11:23 PM Frédéric Chapoton
> wrote:
> >
> > So what would be a practical solution in my case ?
> >
>
on a developent box (thsi won't do for buildbots etc, naturally)
you can pull
https://trac.sagemath.org/ticket/
20 matches
Mail list logo
