On Feb 18, 11:16 am, William Stein <wst...@gmail.com> wrote:
> This is about 3.3.rc2.
>
> 1. On my macbookpro os x 10.5 intel,
> sage -t -long "devel/sage/sage/misc/package.py"
> **********************************************************************
> File "/Users/wstein/build/build/sage-3.3.rc2/devel/sage/sage/misc/package.py",
> line 50:
> sage: sage.misc.package.install_all_optional_packages(dry_run=True)
> Expected:
> Installing ...
> []
> Got:
> Using SAGE Serverhttp://www.sagemath.org//packages
> http://www.sagemath.org//packages/optional/list-->
> /Users/wstein/build/build/sage-3.3.rc2/tmp/list
> [Errno socket error] (8, 'nodename nor servname provided, or not known')
>
> SOLUTION: This was caused by the networking being down during this
> test. These tests should be marked # optional, since doctesting sage
> *must* not require an external network connection. This is now
> http://trac.sagemath.org/sage_trac/ticket/5309
Ok.
> 2. Numerical noise (on OS X 10.5 Intel):
> sage -t -long "devel/sage/sage/schemes/elliptic_curves/sha_tate.py"
> **********************************************************************
> File
> "/Users/was/build/sage-3.3.rc2/devel/sage/sage/schemes/elliptic_curves/sha_
> tate.py",
> line 88:
> sage: [sha.an_numerical(prec) for prec in xrange(30,100,10)] # long time
> Expected:
> [1.0000000,
> 1.0000000000,
> 1.0000000000000,
> 1.0000000000000000,
> 1.0000000000000000000,
> 1.0000000000000000000000,
> 1.0000000000000000000000000]
> Got:
> [0.99999969, 1.0000000000, 1.0000000000000, 1.0000000000000000,
> 1.0000000000000000000, 1.0000000000000000000000,
> 1.0000000000000000000000000]
> **********************************************************************
> 1 items had failures:
> 1 of 11 in __main__.example_1
> ***Test Failed*** 1 failures.
#5303
> 3. A segfault on OS X 10.5 32-bit Intel (bsd.math.washington.edu):
>
> sage -t -long "devel/sage/sage/schemes/elliptic_curves/ell_finite_field.py"
> A mysterious error (perphaps a memory error?) occurred, which may have
> crashed doctest.
> [22.7 s]
>
> Doing this again with --verbose yields:
>
> ----------------------------------------------
> Trying:
> for p in prime_range(Integer(10000)): #long time
> (~20s)###line 1206:_sage_ >>> for p in prime_range(10000):
> #long time (~20s)
> if p != Integer(389):
> G=E.change_ring(GF(p)).abelian_group()
> Expecting nothing
>
> error: no more memory
> System 5120k:5120k Appl 4681k/438k Malloc 4095k/0k Valloc 1024k/438k
> Pages 161/95 Regions 2:2
>
> halt 14
> [33.3 s]
> exit code: 1024
> ----------------------------------------------
>
> Having this fail on a box with 8GB RAM is unacceptable (even if it is 32-bit).
This has been #3760 for a while. IIRC that doctest passes with Sage
build in 64 bit mode on OSX, but it would not surprised me if this is
pari related.
> 4. Numerical noise on 32-bit centos:
> sage -t -long "devel/sage/sage/graphs/graph_plot.py"
> **********************************************************************
> File "/space/wstein/farm/sage-3.3.rc2/devel/sage/sage/graphs/graph_plot.py",
> line 108:
> sage: g.get_pos()
> Expected:
> {0: [6.123233995736766e-17, 1.0],
> 1: [-0.95105651629515353, 0.30901699437494751],
> 2: [-0.58778525229247325, -0.80901699437494734],
> 3: [0.58778525229247292, -0.80901699437494756],
> 4: [0.95105651629515364, 0.30901699437494717]}
> Got:
> {0: [6.1230317691118863e-17, 1.0], 1: [-0.95105651629515353,
> 0.30901699437494751], 2: [-0.5877852522
> 9247325, -0.80901699437494734], 3: [0.58778525229247292,
> -0.80901699437494756], 4: [0.95105651629515364,
> 0.30901699437494717]}
> **********************************************************************
> 1 items had failures:
> 1 of 8 in __main__.example_2
> ***Test Failed*** 1 failures.
> For whitespace errors, see the file
> /space/wstein/farm/sage-3.3.rc2/tmp/.doctest_graph_plot.py
#5302
> 5. Numerical noise on 32-bit Debian:
> sage -t -long "devel/sage/sage/graphs/graph.py"
> **********************************************************************
> File "/space/wstein/farm/sage-3.3.rc2/devel/sage/sage/graphs/graph.py",
> line 5814:
> sage: G.get_pos()
> Expected:
> {0: [-0.81..., -0.32...],
> 1: [-0.49..., 0.53...],
> 2: [0.04..., 0.96...],
> 3: [0.00..., 0.01...], 4: [0.17..., -0.71...], 5: [-0.47...,
> 0.06...], 6: [0.35..., -0.17...], 7: [0.54..., 0.50...],
> 8: [-0.30..., -0.57...],
> 9: [0.95..., -0.28...]}
> Got:
> {0: [-0.80997637536460398, -0.35675046789870352], 1:
> [-0.51384897577569877, 0.53512146351622436], 2:
> [0.042631304307322225, 0.97335632168802844], 3:
> [-0.0030513258274788989, 0.025544483144631966], 4: [0.2
> 0670420775241458, -0.73773340681618726], 5: [-0.48699482654804005,
> 0.075150016159533609], 6: [0.35416487
> 310979794, -0.17356258407013175], 7: [0.54561819770567421,
> 0.51479320869014911], 8: [-0.2958763202878703
> 1, -0.5780855204646983], 9: [0.96062924092848434, -0.27783351394884659]}
> **********************************************************************
> 1 items had failures:
#5302
> 6. A serious-looking weird error on 64-bit Debian (debian64 vm). This
> does *not* happen in Debian Lenny, but does in Debian 4.0, so I'm
> willing to let it slide (?):
>
> sage -t -long "devel/sage/sage/graphs/graph_bundle.py"
> libpng error: Image width or height is zero in IHDR
> **********************************************************************
> File "/space/wstein/farm/sage-3.3.rc2/devel/sage/sage/graphs/graph_bundle.py",
> line 163:
> sage: B.plot()
> Exception raised:
> Traceback (most recent call last):
> File "/space/wstein/farm/sage-3.3.rc2/local/bin/ncadoctest.py",
> line 1231, in run_one_test
> self.run_one_example(test, example, filename, compileflags)
> File "/space/wstein/farm/sage-3.3.rc2/local/bin/sagedoctest.py",
> line 38, in run_one_example
> OrigDocTestRunner.run_one_example(self, test, example,
> filename, compileflags)
> File "/space/wstein/farm/sage-3.3.rc2/local/bin/ncadoctest.py",
> line 1172, in run_one_example
> compileflags, 1) in test.globs
> File "<doctest __main__.example_5[5]>", line 1, in <module>
> B.plot()###line 163:
> sage: B.plot()
> File "sage_object.pyx", line 92, in
> sage.structure.sage_object.SageObject.__repr__
> (sage/structure/sage_object.c:1082)
> File
> "/space/wstein/farm/sage-3.3.rc2/local/lib/python2.5/site-packages/sage/plo
> t/plot.py",
> line 714, in _repr_
> self.show()
> File
> "/space/wstein/farm/sage-3.3.rc2/local/lib/python2.5/site-packages/sage/plo
> t/plot.py",
> line 1068, in show
> hgridlinesstyle=hgridlinesstyle)
> File
> "/space/wstein/farm/sage-3.3.rc2/local/lib/python2.5/site-packages/sage/plo
> t/plot.py",
> line 1373, in save
> canvas.print_figure(filename, dpi=dpi)
> File
> "/space/wstein/farm/sage-3.3.rc2/local/lib/python2.5/site-packages/matplotl
> ib/backend_bases.py",
> line 1453, in print_figure
> **kwargs)
> File
> "/space/wstein/farm/sage-3.3.rc2/local/lib/python2.5/site-packages/matplotl
> ib/backends/backend_agg.py",
> line 334, in print_png
> filename_or_obj, self.figure.dpi)
> RuntimeError: Error building image
> **********************************************************************
> 1 items had failures:
> 1 of 6 in __main__.example_5
I hit that on another box and I am clueless at this point. Note than
in the MPL update we completely disabled the tkagg backend for now
since it constantly would throw uncaught exceptions. This bug might be
unrelated though.
Jason: any thought?
Another related piece of info: We upgraded libpng, but one would
expect this to fail on every box then. Strange.
> 7. Mandriva 64-bit. There's a segfault while building Python:
>
> gcc -pthread -shared -L/space/wstein/farm/sage-3.3.rc2/local/lib -I.
> -IInclude -I./Include -I/space/wstein/farm/sage-3.3.rc2/local/include
> build/temp.linux-x86_64-2.5/space/wstein/farm/sage-3.3.rc2/spkg/build/pytho
> n-2.5.2.p8/src/Modules/readline.o
> -L/usr/lib/termcap -L/space/wstein/farm/sage-3.3.rc2/local/lib
> -L/usr/local/lib -lreadline -ltermcap -o
> build/lib.linux-x86_64-2.5/readline.so
> /bin/sh: line 1: 26906 Segmentation fault CC='gcc -pthread'
> LDSHARED='gcc -pthread -shared' OPT='-DNDEBUG -g -fwrapv -O3 -Wall
> -Wstrict-prototypes' ./python -E ./setup.py build
> make[2]: *** [sharedmods] Error 139
> make[2]: Leaving directory
> `/space/wstein/farm/sage-3.3.rc2/spkg/build/python-2.5.2.p8/src'
> Error building Python.
>
> ---
>
> IIRC this goes away if I try to build it directly without using ssh
> and nohup. I upgraded Mandriva last time we saw this and that did not
> fix it. Adding a warning to the README.txt under "known problems"
> would be a good idea, if we can't fix this.
Yeah, this one is odd. You might want to consider making your build
farm script to execute shell scripts/command inside screen session.
Cheers,
Michael
> ---
>
> IMHO, the sooner we call an absolute 100% feature freeze on this
> release and going into 100% "stabilize and release" the better.
>
> William
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send email to sage-devel@googlegroups.com
To unsubscribe from this group, send email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
