* [gentoo-science] Sage test failures on amd64
@ 2010-04-19 17:29 Christopher Schwan
2010-04-20 1:24 ` Steven Trogdon
0 siblings, 1 reply; 4+ messages in thread
From: Christopher Schwan @ 2010-04-19 17:29 UTC (permalink / raw
To: gentoo-science
Hi,
as some of us know there a lot of failures on amd64 when running Sage's set of
tests. Now I found out that pari caused some of these (I dont know exactly how
many, but I will do a full test run later), because it was compiled without "-
fno-strict-aliasing". I have written a bug report for that:
- https://bugs.gentoo.org/show_bug.cgi?id=316139
In short: Add -fno-strict-aliasing to your CFLAGS and recompile pari and there
should be less failing tests.
Cheers,
Christopher
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [gentoo-science] Sage test failures on amd64
2010-04-19 17:29 [gentoo-science] Sage test failures on amd64 Christopher Schwan
@ 2010-04-20 1:24 ` Steven Trogdon
2010-04-20 6:22 ` Christopher Schwan
0 siblings, 1 reply; 4+ messages in thread
From: Steven Trogdon @ 2010-04-20 1:24 UTC (permalink / raw
To: gentoo-science
[-- Attachment #1: Type: text/plain, Size: 4996 bytes --]
On 04/19/10 - 12:29:01, Christopher Schwan wrote:
> Hi,
>
> as some of us know there a lot of failures on amd64 when running
> Sage's set of
> tests. Now I found out that pari caused some of these (I dont know
> exactly how
> many, but I will do a full test run later), because it was compiled
> without "-
> fno-strict-aliasing". I have written a bug report for that:
>
> - https://bugs.gentoo.org/show_bug.cgi?id=316139
>
> In short: Add -fno-strict-aliasing to your CFLAGS and recompile pari
> and there
> should be less failing tests.
>
> Cheers,
>
> Christopher
>
Christopher,
This is significant! I'm listing test results here for comparison (pari
w/ USE=gmp):
Failures with CFLAGS="-march=k8 -O2 -pipe" and CXXFLAGS="${CFLAGS}" on
a single processor laptop:
sage -t -force_lib
"devel/sage/doc/en/constructions/plotting.rst"
sage -t -force_lib "devel/sage/sage/combinat/iet/strata.py"
sage -t -force_lib "devel/sage/sage/interfaces/gp.py"
sage -t -force_lib "devel/sage/sage/libs/pari/gen.pyx"
sage -t -force_lib "devel/sage/sage/symbolic/expression.pyx"
sage -t -force_lib
"devel/sage/sage/gsl/probability_distribution.pyx"
sage -t -force_lib "devel/sage/sage/lfunctions/lcalc.py"
sage -t -force_lib "devel/sage/sage/plot/plot.py"
sage -t -force_lib "devel/sage/sage/graphs/digraph.py"
sage -t -force_lib "devel/sage/sage/graphs/graph.py"
sage -t -force_lib "devel/sage/sage/misc/dist.py"
sage -t -force_lib "devel/sage/sage/misc/cython.py"
sage -t -force_lib "devel/sage/sage/misc/package.py"
sage -t -force_lib
"devel/sage/sage/functions/transcendental.py"
sage -t -force_lib "devel/sage/sage/functions/special.py"
sage -t -force_lib
"devel/sage/sage/modules/vector_real_double_dense.pyx"
sage -t -force_lib
"devel/sage/sage/modules/vector_double_dense.pyx"
sage -t -force_lib "devel/sage/sage/modular/modsym/space.py"
sage -t -force_lib "devel/sage/sage/matrix/matrix1.pyx"
sage -t -force_lib "devel/sage/sage/sets/set.py"
sage -t -force_lib "devel/sage/sage/numerical/optimize.py"
Failures with CFLAGS="-march=opteron -O2 -pipe" and
CXXFLAGS="${CFLAGS}" on a dual processor, dual core desktop:
sage -t -force_lib
"devel/sage/doc/en/constructions/plotting.rst"
sage -t -force_lib "devel/sage/sage/misc/cython.py"
sage -t -force_lib "devel/sage/sage/misc/package.py"
sage -t -force_lib "devel/sage/sage/misc/dist.py"
sage -t -force_lib "devel/sage/sage/symbolic/expression.pyx"
sage -t -force_lib "devel/sage/sage/lfunctions/sympow.py"
sage -t -force_lib "devel/sage/sage/lfunctions/lcalc.py"
sage -t -force_lib "devel/sage/sage/combinat/iet/strata.py"
sage -t -force_lib
"devel/sage/sage/combinat/words/suffix_trees.py" # Killed/crashed
sage -t -force_lib
"devel/sage/sage/combinat/posets/hasse_diagram.py" # Killed/crashed
sage -t -force_lib "devel/sage/sage/interfaces/gp.py"
sage -t -force_lib
"devel/sage/sage/categories/coxeter_groups.py" # Killed/crashed
sage -t -force_lib
"devel/sage/sage/categories/finite_coxeter_groups.py" # Killed/crashed
sage -t -force_lib "devel/sage/sage/matrix/matrix1.pyx"
sage -t -force_lib "devel/sage/sage/plot/plot.py"
sage -t -force_lib
"devel/sage/sage/modular/hecke/submodule.py"
sage -t -force_lib "devel/sage/sage/modular/abvar/abvar.py"
sage -t -force_lib "devel/sage/sage/modular/modsym/space.py"
sage -t -force_lib "devel/sage/sage/sets/set.py"
sage -t -force_lib
"devel/sage/sage/schemes/elliptic_curves/ell_rational_field.py"
sage -t -force_lib
"devel/sage/sage/modules/vector_double_dense.pyx"
sage -t -force_lib
"devel/sage/sage/modules/vector_real_double_dense.pyx"
sage -t -force_lib "devel/sage/sage/graphs/graph_list.py" #
Killed/crashed
sage -t -force_lib "devel/sage/sage/graphs/digraph.py"
sage -t -force_lib "devel/sage/sage/graphs/cliquer.pyx" #
Killed/crashed
sage -t -force_lib "devel/sage/sage/graphs/graph.py"
sage -t -force_lib
"devel/sage/sage/gsl/probability_distribution.pyx"
sage -t -force_lib "devel/sage/sage/numerical/optimize.py"
sage -t -force_lib "devel/sage/sage/geometry/polyhedra.py" #
Killed/crashed
sage -t -force_lib "devel/sage/sage/libs/pari/gen.pyx"
sage -t -force_lib
"devel/sage/sage/functions/transcendental.py"
sage -t -force_lib "devel/sage/sage/functions/special.py
All of the tests with "Killed/crashed" passed on the desktop when pari
was built without
"-fno-strict-aliasing".
Steve
[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [gentoo-science] Sage test failures on amd64
2010-04-20 1:24 ` Steven Trogdon
@ 2010-04-20 6:22 ` Christopher Schwan
2010-04-20 13:04 ` Steven Trogdon
0 siblings, 1 reply; 4+ messages in thread
From: Christopher Schwan @ 2010-04-20 6:22 UTC (permalink / raw
To: gentoo-science
On Tuesday 20 April 2010 03:24:11 Steven Trogdon wrote:
> On 04/19/10 - 12:29:01, Christopher Schwan wrote:
> > Hi,
> >
> > as some of us know there a lot of failures on amd64 when running
> > Sage's set of
> > tests. Now I found out that pari caused some of these (I dont know
> > exactly how
> > many, but I will do a full test run later), because it was compiled
> > without "-
> > fno-strict-aliasing". I have written a bug report for that:
> >
> > - https://bugs.gentoo.org/show_bug.cgi?id=316139
> >
> > In short: Add -fno-strict-aliasing to your CFLAGS and recompile pari
> > and there
> > should be less failing tests.
> >
> > Cheers,
> >
> > Christopher
>
> Christopher,
>
> This is significant! I'm listing test results here for comparison (pari
> w/ USE=gmp):
>
> Failures with CFLAGS="-march=k8 -O2 -pipe" and CXXFLAGS="${CFLAGS}" on
> a single processor laptop:
>
> sage -t -force_lib
> "devel/sage/doc/en/constructions/plotting.rst"
> sage -t -force_lib "devel/sage/sage/combinat/iet/strata.py"
> sage -t -force_lib "devel/sage/sage/interfaces/gp.py"
> sage -t -force_lib "devel/sage/sage/libs/pari/gen.pyx"
> sage -t -force_lib "devel/sage/sage/symbolic/expression.pyx"
> sage -t -force_lib
> "devel/sage/sage/gsl/probability_distribution.pyx"
> sage -t -force_lib "devel/sage/sage/lfunctions/lcalc.py"
> sage -t -force_lib "devel/sage/sage/plot/plot.py"
> sage -t -force_lib "devel/sage/sage/graphs/digraph.py"
> sage -t -force_lib "devel/sage/sage/graphs/graph.py"
> sage -t -force_lib "devel/sage/sage/misc/dist.py"
> sage -t -force_lib "devel/sage/sage/misc/cython.py"
> sage -t -force_lib "devel/sage/sage/misc/package.py"
> sage -t -force_lib
> "devel/sage/sage/functions/transcendental.py"
> sage -t -force_lib "devel/sage/sage/functions/special.py"
> sage -t -force_lib
> "devel/sage/sage/modules/vector_real_double_dense.pyx"
> sage -t -force_lib
> "devel/sage/sage/modules/vector_double_dense.pyx"
> sage -t -force_lib "devel/sage/sage/modular/modsym/space.py"
> sage -t -force_lib "devel/sage/sage/matrix/matrix1.pyx"
> sage -t -force_lib "devel/sage/sage/sets/set.py"
> sage -t -force_lib "devel/sage/sage/numerical/optimize.py"
>
> Failures with CFLAGS="-march=opteron -O2 -pipe" and
> CXXFLAGS="${CFLAGS}" on a dual processor, dual core desktop:
>
> sage -t -force_lib
> "devel/sage/doc/en/constructions/plotting.rst"
> sage -t -force_lib "devel/sage/sage/misc/cython.py"
> sage -t -force_lib "devel/sage/sage/misc/package.py"
> sage -t -force_lib "devel/sage/sage/misc/dist.py"
> sage -t -force_lib "devel/sage/sage/symbolic/expression.pyx"
> sage -t -force_lib "devel/sage/sage/lfunctions/sympow.py"
> sage -t -force_lib "devel/sage/sage/lfunctions/lcalc.py"
> sage -t -force_lib "devel/sage/sage/combinat/iet/strata.py"
> sage -t -force_lib
> "devel/sage/sage/combinat/words/suffix_trees.py" # Killed/crashed
> sage -t -force_lib
> "devel/sage/sage/combinat/posets/hasse_diagram.py" # Killed/crashed
> sage -t -force_lib "devel/sage/sage/interfaces/gp.py"
> sage -t -force_lib
> "devel/sage/sage/categories/coxeter_groups.py" # Killed/crashed
> sage -t -force_lib
> "devel/sage/sage/categories/finite_coxeter_groups.py" # Killed/crashed
> sage -t -force_lib "devel/sage/sage/matrix/matrix1.pyx"
> sage -t -force_lib "devel/sage/sage/plot/plot.py"
> sage -t -force_lib
> "devel/sage/sage/modular/hecke/submodule.py"
> sage -t -force_lib "devel/sage/sage/modular/abvar/abvar.py"
> sage -t -force_lib "devel/sage/sage/modular/modsym/space.py"
> sage -t -force_lib "devel/sage/sage/sets/set.py"
> sage -t -force_lib
> "devel/sage/sage/schemes/elliptic_curves/ell_rational_field.py"
> sage -t -force_lib
> "devel/sage/sage/modules/vector_double_dense.pyx"
> sage -t -force_lib
> "devel/sage/sage/modules/vector_real_double_dense.pyx"
> sage -t -force_lib "devel/sage/sage/graphs/graph_list.py" #
> Killed/crashed
> sage -t -force_lib "devel/sage/sage/graphs/digraph.py"
> sage -t -force_lib "devel/sage/sage/graphs/cliquer.pyx" #
> Killed/crashed
> sage -t -force_lib "devel/sage/sage/graphs/graph.py"
> sage -t -force_lib
> "devel/sage/sage/gsl/probability_distribution.pyx"
> sage -t -force_lib "devel/sage/sage/numerical/optimize.py"
> sage -t -force_lib "devel/sage/sage/geometry/polyhedra.py" #
> Killed/crashed
> sage -t -force_lib "devel/sage/sage/libs/pari/gen.pyx"
> sage -t -force_lib
> "devel/sage/sage/functions/transcendental.py"
> sage -t -force_lib "devel/sage/sage/functions/special.py
>
> All of the tests with "Killed/crashed" passed on the desktop when pari
> was built without
> "-fno-strict-aliasing".
You surely mean _with_ "-fno-strict-aliasing" ?
>
> Steve
Christopher
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [gentoo-science] Sage test failures on amd64
2010-04-20 6:22 ` Christopher Schwan
@ 2010-04-20 13:04 ` Steven Trogdon
0 siblings, 0 replies; 4+ messages in thread
From: Steven Trogdon @ 2010-04-20 13:04 UTC (permalink / raw
To: gentoo-science
[-- Attachment #1: Type: text/plain, Size: 883 bytes --]
On 04/20/10 01:22:30, Christopher Schwan wrote:
>
> You surely mean _with_ "-fno-strict-aliasing" ?
>
>>
>> Steve
>
> Christopher
>
Christopher,
To clarify, the test failures I listed were for the two computers when
"-fno-strict-aliasing" _was_ added to the CFLAGS in building pari. Of course,
I did this in the ebuild with append-flags. The failures for the second
computer with "Killed/crashed" at the end of the failed test like,
sage -t -force_lib "devel/sage/sage/graphs/cliquer.pyx" #Killed/crashed
passed on the same computer when pari was built in which
"-fno-strict-aliasing" was _not_ added to the CFLAGS. I will check again and
report if otherwise. With "-fno-strict-aliasing" added I get 21 failures on
the laptop and 32 failures on the desktop; with 7 of the failures on the
desktop of the type "Killed/crashed".
Steve
[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2010-04-20 14:02 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-04-19 17:29 [gentoo-science] Sage test failures on amd64 Christopher Schwan
2010-04-20 1:24 ` Steven Trogdon
2010-04-20 6:22 ` Christopher Schwan
2010-04-20 13:04 ` Steven Trogdon
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox