Sage-2.9
William Stein
Sage 2.9 has been released. It is available at
http://sagemath.org/download.html
This is a *major* new release, with a huge number of tickets
closed, and the additional of both R and the ATLAS BLAS
as standard packages.
If you're an *expert* user or just brave, do
sage -upgrade
but this may yield your sage temporarily unusable -- if that does happen,
do email sage-support.
If you're a beginner install from source or wait for new
binaries to be posted in a few days (or less).
---------------
The following people contributed this release:
* Michael Abshoff
* Martin Albrecht
* Robert Bradshaw
* Tom Boothby
* Ondrej Certik
* Craig Citro
* F. Clark
* John Cremona
* Alyson Deines
* Burcin Erocal
* Jan Groenewald
* David Harvey
* Mike Hansen
* Josh Kantor
* Robert Miller
* Bobby Moretti
* Rich Morin
* Bill Page
* Willem Jan Palenstijn
* Yi Qiang
* David Roe
* William Stein
* Carl Witty
* Paul Zimmermann
Cheers,
Michael Abshoff (release chair), William Stein
* Major Features, New Spkgs and Bugfixes
We achieved most of the goals we set ourselves for 2.9. The major
changes compared to 2.8.15 are:
* Integration of PolyBoRi
* Integration of ATLAS BLAS/Lapack on non-OSX
* Integration of R 2.6.1 and rpy
* Update to FLINT 1.02
* Update to Symmetrica 2.0
* Much faster permutation arithmatic
* Very fast rubik's cube solvers
* Many build fixes applied and many bugs squashed
We closed a total of 110 tickets, for details see below or our
tracker at
http://www.sagetrac.org/sage_trac/query?status=closed&milestone=sage-2.9
* Known Issues with 2.9
#1137: Import gd fails to import on OSX 10.4 ppc. Any attempt
to fix this breaks other applications on different OSX
flavors. We hope to fix this in 2.9.1.
#1497: ATLAS fails to build on FC7, Dual core [known ATLAS
issue with workaround: disable power management]
* Upcoming Releases
The following releases are planned for the rest of the year:
* 2.9.1: chaired by Robert Miller, bug fixes, release
planned about a week from now
* 2.9.2: chaired by Michael Abshoff, bug fixes, release
planned about two weeks from now
2.9.2 will be the release we plan to distribute on DVD at the AMS
meeting, so we are shooting hard for an excellent release.
* Doctesting Coverage
For 2.9:
Overall weighted coverage score: 35.3%
Total number of functions: 17947
Compared to 2.8.15 this is an increase by 0.5%.
* About Sage (http://www.sagemath.org)
Sage is developed by volunteers and combines 71 open source packages.
It is available for download from sagemath.org and its mirrors in
source or binary form. If you have any questions and/or problems
please report them to the google groups sage-devel, sage-support,
sage-forum or sage-newbie. You can also drop by in #sage-devel in
freenode.
* Closed Tickets:
Merged in rc3:
#1529: William Stein: update install_scripts to also install R script.
#1530: William Stein: building tut.tex is currently broken
#1531: Michael Abshoff: fix doctest failure sage/calculus/calculus.py
#1532: Josh Kantor; Michael Abshoff: Error out with intelligent
message if ATLAS tune failed
Without tickets:
William Stein: update libpng.spkg
William Stein: don't let "sage -ba" start sage
William Stein: link libcsage.so against atlas on non-OSX
William Stein: link IML against atlas on non-OSX
Willaim Stein, Michael Abshoff: Various doctest fixes
Michael Abshoff: link LinBox against atlas on non-OSX
Michael Abshoff: remove forced check from FLINT.spkg
Merged in rc1/2:
#1258: Willem Jan Palenstijn, Robert Miller: additions and changes
to linear_codes
#1457: William Stein: "BUG: Rational.__pow__ called on a non-Rational"
#1464: Robert Miller: binary code canonical labels & automorphism
group generators
#1503: William Stein: formal function calls don't coerce correctly to
Mathematica
#1511: Robert Bradshaw: Export 3d objects in jmol format
#1527: William Stein: doctest R with "import rpy"
#1528: William Stein: make "sage -R" run R.
Merged in rc0:
#444: Robert Bradshaw: solve the rubik's cube fast!
#553: Mike Hansen, William Stein: calling of symbolic expressions is
sometimes ridiculous
#1077: Yi Qiang: DSage restarts two workers after timeout, various
other fixes and improvements
#1119: Martin Albrecht, Robert Bradshaw: EllipticCurve.random_element
for char=2
#1137: Robert Miller: matrix visualize_structure is completely broken
on OSX [later reverted]
#1162: Paul Zimmermann, Carl Witty: fix issues in RealField <-> RQDF
conversions
#1183: David Roe, William Stein, Alyson Deines: Residue fields are
broken. This patch set also fixes #1242 and #1185
#1232: Craig Citro: bug in modular symbols over GF(2)
#1237: David Roe: E.torsion_order() fails on curves with big
coefficients!
#1239: Robert Bradshaw, John Cremona: Wrap Simon's new gp two
descent code
#1393: Robert Miller: is_integral_domain may return incorrect answer
#1395: Mike Hansen: notebook(directory="foo/") misbehaves
#1401: Mike Hansen, William Stein: deprecate A[n] for n a matrix
#1404: F. Clark, William Stein: bug in %latex feature in the notebook)
#1407: William Stein: deciding that generators don't generate an order
in some extensions can be way way too slow.
#1423: William Stein: bug/issue in python.eval from sage
#1425: Mike Hansen, William Stein: wrong automatic simplification of
pow
#1442: Mike Hansen: fix wrong statement in calculus.py
#1447: Burcin Erocal: Polybori permissions issues [fixed in
polybori-0.1-r5.spkg]
#1454: William Stein: Change the prompt for sage -sh
#1460: William Stein, Carl Witty: bug in float( ... ) conversion in
calculus
#1461: William Stein: implement general subs_expr for calculus
expressions
#1472: Josh Kantor: gnuplotpy optional package doesn't work with numpy
[updated optional spkg]
#1473: Robert Bradschaw: make Java3D usable from command line as well
as notebook
#1480: William Stein: implement P.show() for mathematica elements --
nice mathematica graphics in the sage notebook!
#1491: William Stein: improve conversion from interface to sage
objects (i.e., the dot sage method)
#1494: William Stein: bug coercing from maximal order of cyclotomic
field into cyclotomic field
#1502: Mike Hansen: calculus -- bug in argument ordering for formal
functions
#1506: William Stein: NTL.spkg - dumb intentional error during the
build
#1507: William Stein: document how to use pylab / matlab style
plotting from sage
#1513: Michael Abshoff: FLINT install uses make -B, which isn't an
option on (slightly) older make versions
#1515: Robert Bradshaw: fix ParametricSurface bug
#1518: William Stein: cache computing order of elliptic curve over
finite field
#1519: William Stein: hg problem applying tickets via trac url: remove
characters after "?" from name
#1520: Michael Abshoff: prebuilt 2.9-alpha7 osx ppc binary has
hardcoded paths
Merged in alpha7:
#1340: Robert Bradshaw: %cython broken if directory name contains
spaces
#1398: Mike Hansen: Functions on integers which are not
object-oriented
#1462: William Stein: make sure that not to update the timestamp
cache if compilation fails
#1487: Michael Abshoff: ATLAS breaks with parallel make
#1488: William Stein: fix output of symbolic vectors
#1489: William Stein: serious bug in calculus maxima parsing
#1490: Michael Abhsoff: fix numerical noise doctest failure in
numerical/test.py
#1492: Yi Qiang: update sqlite.spkg to 3.5.3.p1 [this also closes
#1152, #1202]
#1493 Burcin Erocal: polybori doesn't free m4ri data on exit
#1497: Josh Kantor: Configure ATLAS properly for 32 bit user space
on 64 bit CPU
#1499: Michael Abshoff: fix import issues related to ATLAS BLAS on
non-OSX
#1500: William Stein: solve_mod -- implement solving modulo n in sage
#1501: Josh Kantor: udpate cvxopt to use ATLAS on Linux
Merged in alpha6:
#469: Burcin Erocal: Add BoolenPolynomialRing to global namespace.
With this patch PolyBoRi's integration should be finished
#823: Josh Kantor, Michael Abshoff: make atlas standard in Sage
#1430: Josh Kantor: sparse eigenvalues and splines [scipy sandbox]
#1434: Josh Kantor: cut sandbox directory out of scipy spkg
#1462: William Stein: speed up "sage -b" -- don't do anything cython
stuff if no .pyx, .pxd, or .pxi file changes
#1463: William Stein: polymake package has wrong cddlib,gmp versions
#1479: William Stein: Fix doctest failure fallout from #1296
Merged in alpha5:
#469: Burcin Erocal: partial merge of the PolyBoRi wrapper, updated
PolyBoRi.spkg [still some code to merge, so disable doctests
for now, BooleanRing is missing]
#1189: Ondrej Certik: SymPy <-> SAGE automatic conversion
#1235: William Stein: doctest fixes for 32 bit Linux
#1296: Robert Bradshaw: Fast permutation arithmatic [with bugfixes
and debugging by William Stein, Michael Abshoff, Carl Witty
and Mike Hansen - this was a real team effort :)]
#1394: Robert Miller: make plot documentation readable in graph.py
#1450: William Stein: update the error message for using maple/
mathmatica from sage
#1453: William Stein: fix cython dependency computation for new files
[later reverted, due to issues with #1366]
#1455: Michael Abshoff: Fix OSX doctest failures due to numerical
noise
Merged in alpha4:
No tickets, but build fixes for gfan and R by William Stein
Merged in alpha3:
#1366: Bobby Moretti, William Stein: speed up "sage -br" -- make it
cache the dependency diagram instead of computing it every
time [later reverted]
#1411: William Stein: pdate the README.txt file included in Sage
#1416: William Stein, Mike Hansen, Josh Kantor: make R standard
component
#1427: Josh Kantor: rpy doesn't build on osx 10.4
#1433: William Stein: Fix some documentation typos pointed out by
Rich Morin
#1436: Michael Abshoff, Bill Hart: Update FLINT.spkg to 1.02 relase
#1437: Michael Abshoff: fix spkg-check in FLINT to work "stand alone"
#1444: William Stein: fix some serious hard-coding issues that break
all binary installs
#1445: Mike Hansen: symmetrica's longints need to be converted
Merged in alpha2:
#469: Burcin Erocal: add PolyBoRi.spkg [integration patch still
missing]
#1164: Mike Hansen: Error in tutorial documentation about %hist
#1167: Mike Hansen: Spelling errors in documentation
#1182: Mike Hansen: update freetype.spkg to 2.3.5
#1235: William Stein: bug solving equations using maxima
#1243: Mike Hansen: Minor spelling mistakes in documentation
#1267: Mike Hansen: documentation for piecewise does not show up in
notebook
#1288: Mike Hansen: misformating of some of the reference manual in
live version; also out of date
#1324: Carl Witty: 2.8.14: doctest failure in sage/rings/real_rqdf.pyx
on FC6, x86-64
#1348: Mike Hansen: Incorporate Jan Groenewald's documentation fixes
#1378: Mike Hansen: add ssh-client (i.e., ssh-keygen) as a required
package in various places in the docs
#1412: Tom Boothby: creating directory in notebook cell #x makes
evaluation of cell #x hang
#1417: Mike Hansen: update symmetrica.spkg to 2.0
#1420: Mike Hansen: documentation typo
#1435: Michael Abshoff: Update FLINT.spkg to 1.01
Merged in alpha1:
#491: Michael Abshoff: gcc 4.3: fix givaro build due to ::memcpy
failure
#492: Craig Citro, David Harvey: gcc 4.3: fix ntl_wrap.cc in
sage_c_lib [this was some time ago]
#1260: Michael Abshoff, Bill Hart: update to FLINT 1.0 and clean
up spkg-install
#1403: John Cremona: Analog fixes for termination issues for
the other mwrank binaries
#1406: Mike Hansen, William Stein: bug in coercion of multivariate
polynomials
#1410: Michael Abshoff: fix leftovers from 1256, make mwrank
binaries link dynamically, fix cython doctest issue
Merged in alpha0:
#1002: Bill Page: Update clisp to the 2.43 release [later reverted]
#1256: Michael Abshoff: mwrank*.spkg now redundant, included in
cremona*.spkg
#1337: Carl Witty: technically incorrect code in integer_mod.pyx
#1403: John Cremona: mwrank has termination issues
#1405: Michael Abshoff, Carl Witty: gcc 4.3-snapshot: cremona
compile fixes.
Reviewers: Carl Witty, William Stein, Mike Hansen, Robert Miller,
Craig Citro, Michael Abshoff; apologies to anybody I
forgot.
Feeback: Jaap Spies, apologies to anybody I forgot.
--
William Stein
Associate Professor of Mathematics
University of Washington
http://wstein.org
Justin C. Walker
On Dec 16, 2007, at 8:04 AM, William Stein wrote:
>
> Hello,
>
> Sage 2.9 has been released. It is available at
>
> http://sagemath.org/download.html
>
> This is a *major* new release, with a huge number of tickets
> closed, and the additional of both R and the ATLAS BLAS
> as standard packages.
>
> If you're an *expert* user or just brave, do
>
> sage -upgrade
Chicken Little is my name :-}
Full build of 2.9 on Mac OS X, 10.4.11, Core 2 Duo:
real 93m42.212s
user 70m15.014s
sys 22m39.398s
To install gap, gp, singular, etc., scripts
in a standard bin directory, start sage and
type something like
sage: install_scripts('/usr/local/bin')
at the SAGE command prompt.
SAGE build/upgrade complete!
Testing underway.
Justin
--
Justin C. Walker, Curmudgeon-At-Large, Director
Institute for the Enhancement of the Director's Income
--------
The path of least resistance:
it's not just for electricity any more.
--------
Justin C. Walker
On Dec 16, 2007, at 8:04 AM, William Stein wrote:
>
> Hello,
>
> Sage 2.9 has been released. It is available at
>
> http://sagemath.org/download.html
Following up: Mac OS X, 10.4.11, Core 2 Duo:
All tests passed!
Total time for all tests: 2793.2 seconds
Great work!
Justin
--
Justin C. Walker, Curmudgeon-At-Large
Institute for the Absorption of Federal Funds
--------
If you're not confused,
You're not paying attention
--------
Justin C. Walker
On Dec 16, 2007, at 08:04 , William Stein wrote:
>
> Hello,
>
> Sage 2.9 has been released. It is available at
>
> http://sagemath.org/download.html
>
> This is a *major* new release, with a huge number of tickets
> closed, and the additional of both R and the ATLAS BLAS
> as standard packages.
Mac OS X, 10.5.1, Core Duo:
real 98m22.475s
user 84m34.309s
sys 18m7.910s
...
SAGE build/upgrade complete!
All tests passed!
Total time for all tests: 3066.8 seconds
Still seeing 'mwrank/genus2reduction' crashes during testing.
Justin
--
They said it couldn't be done, but sometimes,
it doesn't work out that way.
- Casey Stengel
--
mabshoff
On Dec 18, 8:17 pm, "Justin C. Walker" <jus...@mac.com> wrote:
> On Dec 16, 2007, at 08:04 , William Stein wrote:
>
>
>
> > Hello,
>
> > Sage 2.9 has been released. It is available at
>
> > http://sagemath.org/download.html
>
> > This is a *major* new release, with a huge number of tickets
> > closed, and the additional of both R and the ATLAS BLAS
> > as standard packages.
>
> Mac OS X, 10.5.1, Core Duo:
>
> real 98m22.475s
> user 84m34.309s
> sys 18m7.910s
> ...
>
> SAGE build/upgrade complete!
>
> All tests passed!
> Total time for all tests: 3066.8 seconds
>
> Still seeing 'mwrank/genus2reduction' crashes during testing.
ain't over till the fat lady sings] fix that a couple hours earlier. I
will test on gcc 4.3 on my end and then roll a new cremona.spkg you
might want to test then. All the really hard work was done by John, so
you should thank him :)
> Justin
Cheers,
Michael
John Cremona
>
>
> On Dec 16, 2007, at 08:04 , William Stein wrote:
>
> >
> > Hello,
> >
> > Sage 2.9 has been released. It is available at
> >
> > http://sagemath.org/download.html
> >
> > This is a *major* new release, with a huge number of tickets
> > closed, and the additional of both R and the ATLAS BLAS
> > as standard packages.
>
> Mac OS X, 10.5.1, Core Duo:
>
> real 98m22.475s
> user 84m34.309s
> sys 18m7.910s
> ...
>
> SAGE build/upgrade complete!
>
> All tests passed!
> Total time for all tests: 3066.8 seconds
>
> Still seeing 'mwrank/genus2reduction' crashes during testing.
What is that? mwrank has nothing related to genus 2 in it!! Or do
you mean crashes for each of mwrank and genus2 stuff? If so, what is
the mwrank problem (which trac#)?
John
>
> Justin
>
> --
> They said it couldn't be done, but sometimes,
> it doesn't work out that way.
> - Casey Stengel
> --
>
>
>
>
> >
>
--
John Cremona
Justin C. Walker
On Dec 18, 2007, at 12:00 , John Cremona wrote:
>
> On 18/12/2007, Justin C. Walker <jus...@mac.com> wrote:
>>
>>
>> On Dec 16, 2007, at 08:04 , William Stein wrote:
>>
>>>
>>> Hello,
>>>
>>> Sage 2.9 has been released. It is available at
>>>
>>> http://sagemath.org/download.html
>>>
>>> This is a *major* new release, with a huge number of tickets
>>> closed, and the additional of both R and the ATLAS BLAS
>>> as standard packages.
>>
>> Mac OS X, 10.5.1, Core Duo:
>>
>> real 98m22.475s
>> user 84m34.309s
>> sys 18m7.910s
>> ...
>>
>> SAGE build/upgrade complete!
>>
>> All tests passed!
>> Total time for all tests: 3066.8 seconds
>>
>> Still seeing 'mwrank/genus2reduction' crashes during testing.
>
> What is that? mwrank has nothing related to genus 2 in it!! Or do
> you mean crashes for each of mwrank and genus2 stuff? If so, what is
> the mwrank problem (which trac#)?
I opened Trac #1403 (mwrank has termination issues) a couple of weeks
ago. You/Michael closed it for 2.9. I have not yet looked at the
problem to know what causes it. 'genus2reduction' also blows up
during the 'make test' run, and I likewise have not yet checked into
that one.
Justin
--
Justin C. Walker, Curmudgeon-At-Large
Institute for the Absorption of Federal Funds
--------
Men are from Earth.
Women are from Earth.
Deal with it.
--------
mabshoff
On Dec 18, 9:28 pm, "Justin C. Walker" <jus...@mac.com> wrote:
> On Dec 18, 2007, at 12:00 , John Cremona wrote:
>
>
>
>
>
> > On 18/12/2007, Justin C. Walker <jus...@mac.com> wrote:
>
> >> On Dec 16, 2007, at 08:04 , William Stein wrote:
>
> >>> Hello,
>
> >>> Sage 2.9 has been released. It is available at
>
> >>> http://sagemath.org/download.html
>
> >>> This is a *major* new release, with a huge number of tickets
> >>> closed, and the additional of both R and the ATLAS BLAS
> >>> as standard packages.
>
> >> Mac OS X, 10.5.1, Core Duo:
>
> >> real 98m22.475s
> >> user 84m34.309s
> >> sys 18m7.910s
> >> ...
>
> >> SAGE build/upgrade complete!
>
> >> All tests passed!
> >> Total time for all tests: 3066.8 seconds
>
> >> Still seeing 'mwrank/genus2reduction' crashes during testing.
>
> > What is that? mwrank has nothing related to genus 2 in it!! Or do
> > you mean crashes for each of mwrank and genus2 stuff? If so, what is
> > the mwrank problem (which trac#)?
> I opened Trac #1403 (mwrank has termination issues) a couple of weeks
> ago. You/Michael closed it for 2.9.
Initially I thought it was two issues, so I "corrected" #1402 to a
"proper" bug report. After I realized that I had screwed up I opened
#1403. 2.9.1.alpha1, out in a couple hours should have a fix for
#1403. Keep an eye to #1402 for an updated cremona.spkg.
> I have not yet looked at the
> problem to know what causes it. 'genus2reduction' also blows up
> during the 'make test' run, and I likewise have not yet checked into
> that one.
might be a "crash at exit". Let us know if you find anything.
> Justin
>
Cheers,
Michael
John Cremona
issues in #1403 in my code and am making sure that he has those too,
since they did not get into 2.9. As for #1402, I cannot reproduce
that on my machines so have not fixed it. I *did* fix a different
memory problem which casued some earlier valgrind errors, and possibly
some other crashes, not on trac #1403.
John
On 18/12/2007, mabshoff
--
John Cremona
mabshoff
On Dec 18, 10:00 pm, "John Cremona" <john.crem...@gmail.com> wrote:
Hello,
> I'm not sure mabshoff and I are syncronised on this. I have fixed the
> issues in #1403 in my code and am making sure that he has those too,
> since they did not get into 2.9. As for #1402, I cannot reproduce
> that on my machines so have not fixed it. I *did* fix a different
> memory problem which casued some earlier valgrind errors, and possibly
> some other crashes, not on trac #1403.
>
provided me per email does anything about #1403 with gcc 4.3, but it
will take several hours until I have some results on this.
> John
>
Cheers,
Michael
mabshoff
I don't know if this is helpful, but on sage.math get the following
happens when running mwrank under valgrind with the curve from ticket
1402.
The issue goes away after I applied your initialization patch you send
me per email. The valgrind log is somewhat different from the one in
#1402, but is also involves NTL and happens in the same section of the
code. Accounting for different gccs and operating system your patch
might still have fixed #1402. Justin, could you please try
http://sage.math.washington.edu/home/mabshoff/cremona-20071124.p5.spkg
and report back if that fixes the issue for you?
Cheers,
Michael
mabshoff@sage:/tmp/Work-mabshoff/sage-2.9/local/bin$ ./valgrind --
tool=memcheck --leak-resolution=high ./mwrank
==9637== Memcheck, a memory error detector.
==9637== Copyright (C) 2002-2007, and GNU GPL'd, by Julian Seward et
al.
==9637== Using LibVEX rev 1791, a library for dynamic binary
translation.
==9637== Copyright (C) 2004-2007, and GNU GPL'd, by OpenWorks LLP.
==9637== Using valgrind-3.3.0.SVN, a dynamic binary instrumentation
framework.
==9637== Copyright (C) 2000-2007, and GNU GPL'd, by Julian Seward et
al.
==9637== For more details, rerun with: -v
==9637==
Program mwrank: uses 2-descent (via 2-isogeny if possible) to
determine the rank of an elliptic curve E over Q, and list a
set of points which generate E(Q) modulo 2E(Q).
and finally saturate to obtain generating points on the curve.
For more details see the file mwrank.doc.
For details of algorithms see the author's book.
Please acknowledge use of this program in published work,
and send problems to john.c...@gmail.com.
Version compiled on Dec 16 2007 at 09:53:10 by GCC 4.1.2 20061115
(prerelease) (Debian 4.1.1-21)
using base arithmetic option NTL_ALL (NTL bigints and multiprecision
floating point)
Using NTL multiprecision floating point with 15 decimal places.
Enter curve: [0,0,1,-1,0]
Curve [0,0,1,-1,0] : Basic pair: I=48, J=-432
disc=255744
2-adic index bound = 2
By Lemma 5.1(a), 2-adic index = 1
2-adic index = 1
One (I,J) pair
Looking for quartics with I = 48, J = -432
Looking for Type 2 quartics:
Trying positive a from 1 up to 1 (square a first...)
(1,0,-6,4,1) --trivial
Trying positive a from 1 up to 1 (...then non-square a)
Finished looking for Type 2 quartics.
Looking for Type 1 quartics:
Trying positive a from 1 up to 2 (square a first...)
(1,0,0,4,4) --nontrivial...(x:y:z) = (1 : 1 : 0)
Point = [0:0:1]
height = 0.0511114082399688
Rank of B=im(eps) increases to 1 (The previous point is on the egg)
Exiting search for Type 1 quartics after finding one which is globally
soluble.
Mordell rank contribution from B=im(eps) = 1
Selmer rank contribution from B=im(eps) = 1
Sha rank contribution from B=im(eps) = 0
Mordell rank contribution from A=ker(eps) = 0
Selmer rank contribution from A=ker(eps) = 0
Sha rank contribution from A=ker(eps) = 0
Rank = 1
Searching for points (bound = 8)...==9637== Conditional jump or move
depends on uninitialised value(s)
==9637== at 0x4B9E13E: qsieve::sift0(long, long, long, int)
(sieve_search.cc:903)
==9637== by 0x4B9E54A: qsieve::sift(long) (sieve_search.cc:829)
==9637== by 0x4B9EB8A: qsieve::search() (sieve_search.cc:1119)
==9637== by 0x4B91001: mw::search(NTL::RR, int, int) (mwprocs.cc:
724)
==9637== by 0x40927C: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x4073C1: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x5C074C9: (below main) (in /lib/libc-2.3.6.so)
==9637==
==9637== Conditional jump or move depends on uninitialised value(s)
==9637== at 0x4B9DAA5: qsieve::check_point(unsigned long, long,
long, long*, int) (sieve_search.cc:942)
==9637== by 0x4B9E157: qsieve::sift0(long, long, long, int)
(sieve_search.cc:905)
==9637== by 0x4B9E54A: qsieve::sift(long) (sieve_search.cc:829)
==9637== by 0x4B9EB8A: qsieve::search() (sieve_search.cc:1119)
==9637== by 0x4B91001: mw::search(NTL::RR, int, int) (mwprocs.cc:
724)
==9637== by 0x40927C: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x4073C1: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x5C074C9: (below main) (in /lib/libc-2.3.6.so)
==9637==
==9637== Conditional jump or move depends on uninitialised value(s)
==9637== at 0x4B9DACE: qsieve::check_point(unsigned long, long,
long, long*, int) (sieve_search.cc:942)
==9637== by 0x4B9E157: qsieve::sift0(long, long, long, int)
(sieve_search.cc:905)
==9637== by 0x4B9E54A: qsieve::sift(long) (sieve_search.cc:829)
==9637== by 0x4B9EB8A: qsieve::search() (sieve_search.cc:1119)
==9637== by 0x4B91001: mw::search(NTL::RR, int, int) (mwprocs.cc:
724)
==9637== by 0x40927C: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x4073C1: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x5C074C9: (below main) (in /lib/libc-2.3.6.so)
==9637==
==9637== Conditional jump or move depends on uninitialised value(s)
==9637== at 0x4B9DAD6: qsieve::check_point(unsigned long, long,
long, long*, int) (sieve_search.cc:951)
==9637== by 0x4B9E157: qsieve::sift0(long, long, long, int)
(sieve_search.cc:905)
==9637== by 0x4B9E54A: qsieve::sift(long) (sieve_search.cc:829)
==9637== by 0x4B9EB8A: qsieve::search() (sieve_search.cc:1119)
==9637== by 0x4B91001: mw::search(NTL::RR, int, int) (mwprocs.cc:
724)
==9637== by 0x40927C: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x4073C1: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x5C074C9: (below main) (in /lib/libc-2.3.6.so)
==9637==
==9637== Conditional jump or move depends on uninitialised value(s)
==9637== at 0x4B9DF1C: qsieve::check_point(unsigned long, long,
long, long*, int) (sieve_search.cc:930)
==9637== by 0x4B9E157: qsieve::sift0(long, long, long, int)
(sieve_search.cc:905)
==9637== by 0x4B9E54A: qsieve::sift(long) (sieve_search.cc:829)
==9637== by 0x4B9EB8A: qsieve::search() (sieve_search.cc:1119)
==9637== by 0x4B91001: mw::search(NTL::RR, int, int) (mwprocs.cc:
724)
==9637== by 0x40927C: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x4073C1: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x5C074C9: (below main) (in /lib/libc-2.3.6.so)
==9637==
==9637== Conditional jump or move depends on uninitialised value(s)
==9637== at 0x4B9DF5D: qsieve::check_point(unsigned long, long,
long, long*, int) (sieve_search.cc:930)
==9637== by 0x4B9E157: qsieve::sift0(long, long, long, int)
(sieve_search.cc:905)
==9637== by 0x4B9E54A: qsieve::sift(long) (sieve_search.cc:829)
==9637== by 0x4B9EB8A: qsieve::search() (sieve_search.cc:1119)
==9637== by 0x4B91001: mw::search(NTL::RR, int, int) (mwprocs.cc:
724)
==9637== by 0x40927C: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x4073C1: (within /tmp/Work-mabshoff/sage-2.9/local/bin/
mwrank)
==9637== by 0x5C074C9: (below main) (in /lib/libc-2.3.6.so)
done:
found points of rank 1
and regulator 0.0511114082399688
Processing points found during 2-descent...done:
now regulator = 0.0511114082399688
Saturating (bound = 100)...done:
points were already saturated.
Generator 1 is [0:-1:1]; height 0.0511114082399688
Regulator = 0.0511114082399688
The rank and full Mordell-Weil basis have been determined
unconditionally.
(20.1453 seconds)
Cheers,
Michael
John Cremona
functions, and those are fixed by what I sent you.
But the errors in what's listed in trac#1402 are in a different place
in my code, despite being called from roughly the same point in the
mwrank run.
The actual effect on the program in the uninitialized array elements
(as fixed by what I sent earlier this evening) could not cause
problems outside that chunk of code. Certain bit patterns used in a
sieve should have been initialized to 0s but instead were
uninitialized and hence effectively random. The worst effect of that
would be that a valid rational point might fail to be found, but it
could not cause invalid reads from memory.
I'll look again at the code where #1402 chokes which is in integer
factorization (pdivs() calls pdivs_pari() which -- before calling the
pari library -- calls pdivs_trial() to do trial division). But it
would still be interesting to see if the problem persists after the
other bug fix.
John
On 18/12/2007, mabshoff
<Michael...@fsmath.mathematik.uni-dortmund.de> wrote:
>
--
John Cremona
Justin C. Walker
On Dec 16, 2007, at 08:04 , William Stein wrote:
>
> Hello,
>
> Sage 2.9 has been released. It is available at
>
> http://sagemath.org/download.html
>
> This is a *major* new release, with a huge number of tickets
> closed, and the additional of both R and the ATLAS BLAS
> as standard packages.
>
> If you're an *expert* user or just brave, do
I got brave :-}
Upgrade of 2.8.15 on 10.5.1 (Core Duo):
real 74m23.156s
user 60m48.071s
sys 7m15.158s
All tests passed! (-j2)
Total time for all tests: 2763.2 seconds
Upgrade of 2.8.14 on 10.4.11 (Dual Quad-core Xeon):
real 119m38.975s
user 40m7.949s
sys 8m4.822s
All tests passed! (-j6)
Total time for all tests: 2519.5 seconds
Whuh! 119 minutes to upgrade! Normally, I build with "-j6" on the
Xeon, and it comes in around 61 (2.8.14)-73 (2.9) minutes... I
didn't do a parallel upgrade.
Justin
--
Justin C. Walker, Curmudgeon-at-Large
() The ASCII Ribbon Campaign
/\ Help Cure HTML Email
mabshoff
On Dec 19, 1:42 am, "Justin C. Walker" <jus...@mac.com> wrote:
> On Dec 16, 2007, at 08:04 , William Stein wrote:
>
>
>
> > Hello,
>
> > Sage 2.9 has been released. It is available at
>
> > http://sagemath.org/download.html
>
> > This is a *major* new release, with a huge number of tickets
> > closed, and the additional of both R and the ATLAS BLAS
> > as standard packages.
>
> > If you're an *expert* user or just brave, do
>
> I got brave :-}
>
> Upgrade of 2.8.15 on 10.5.1 (Core Duo):
> real 74m23.156s
> user 60m48.071s
> sys 7m15.158s
>
> All tests passed! (-j2)
> Total time for all tests: 2763.2 seconds
> Upgrade of 2.8.14 on 10.4.11 (Dual Quad-core Xeon):
> real 119m38.975s
> user 40m7.949s
> sys 8m4.822s
>
> All tests passed! (-j6)
> Total time for all tests: 2519.5 seconds
>
> Whuh! 119 minutes to upgrade! Normally, I build with "-j6" on the
> Xeon, and it comes in around 61 (2.8.14)-73 (2.9) minutes... I
> didn't do a parallel upgrade.
model doesn't have pre-tuned parameters for it, so tuning is very
expensive. Adding those settings to ATLAS is #1547.
> Justin
Cheers,
Michael