Re: [sage-devel] delay 8.8?

[E. Madison Bray]

On Fri, Jun 21, 2019 at 8:58 PM Vincent Delecroix
<20100.delecr...@gmail.com> wrote:
>
> Dear all,
>
> Even though we are at the "release candidate" stage of the release
> there are two pretty annoying blockers for 8.8
>
> https://trac.sagemath.org/ticket/28036
> https://trac.sagemath.org/ticket/28042
>
> Wouldn't it be reasonable to delay the stable release? Fixing bugs in
> the rc stage seems pretty fragile to me.

+1

Sometimes I have heard there is a conference or something for which a
release is needed.  If that is the case someone should speak up.
Otherwise I think these issues should be resolved if there is no
serious hurry.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAOTD34YaFpAPskc3P-ff%2BR0cbPXO6C2fCvukwu6Gq4LLj8yvaw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] delay 8.8?

[Vincent Delecroix]

Dear all,

Even though we are at the "release candidate" stage of the release
there are two pretty annoying blockers for 8.8

https://trac.sagemath.org/ticket/28036
https://trac.sagemath.org/ticket/28042

Wouldn't it be reasonable to delay the stable release? Fixing bugs in
the rc stage seems pretty fragile to me.

Best
Vincent

--
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/63b0dfde-120a-9169-a88d-3781a3bdb141%40gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] Bug(s) in scalar product over GF(p)

[Dima Pasechnik]

I have opened https://trac.sagemath.org/ticket/28042
This looks pretty bad...


On Fri, Jun 21, 2019 at 4:20 PM Clement Pernet  wrote:
>
> Hi,
>
> The following simple session triggers what I think is at least one bug, if 
> not two:
>
> sage: p=193379
> sage: K=GF(p)
> sage: a=K(1)
> sage: b=K(191495)
> sage: c=K(109320)
> sage: d=K(167667)
> sage: e=103937
> sage: a*c+b*d-e
> 102041
> sage: vector([a,b])*vector([c,d])-e
> -91339
> sage: -91339+e
> 12598
> sage: vector([a,b])*vector([c,d])
> 12599
>
>
> Namely, the scalar product of GF(p) elements followed by a subtraction is 
> negative (first bug).
>
> Then the negative value is off by one (second bug).
>
> Some weird coercion seem to happen between IntegerMod_int64 IntegerMod_int 
> and ZZ:
>
> sage: type(vector([a,b])[0])
> 
> sage: type(vector([a,b])*vector([c,d]))
> 
>
> Are these known bugs?


>
> Best.
>
> Clément
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/efa76bd3-53ba-e1af-5e84-b3ea428836de%40gmail.com.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq1VLLizWoHb2dafuO_T4OK1kjDkGgyFELdP_0%2BP0TUhKA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Bug(s) in scalar product over GF(p)

[Clement Pernet]

Hi,

The following simple session triggers what I think is at least one bug, if not 
two:

sage: p=193379
sage: K=GF(p)
sage: a=K(1)
sage: b=K(191495)
sage: c=K(109320)
sage: d=K(167667)
sage: e=103937
sage: a*c+b*d-e
102041
sage: vector([a,b])*vector([c,d])-e
-91339
sage: -91339+e
12598
sage: vector([a,b])*vector([c,d])
12599


Namely, the scalar product of GF(p) elements followed by a subtraction is 
negative (first bug).

Then the negative value is off by one (second bug).

Some weird coercion seem to happen between IntegerMod_int64 IntegerMod_int and 
ZZ:

sage: type(vector([a,b])[0])

sage: type(vector([a,b])*vector([c,d]))


Are these known bugs?

Best.

Clément

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/efa76bd3-53ba-e1af-5e84-b3ea428836de%40gmail.com.
For more options, visit https://groups.google.com/d/optout.

[sage-devel] Re: regression between 8.7 and 8.8.rc1 with operations on polygons with vertices in a quadratic field

[Simon King]

Hi Dima,

On 2019-06-21, Dima Pasechnik  wrote:
> Basically, this is a great illustration that pexpect interface (what's
> topcom using)
> is slow...

Question: Is that slowness OS dependent? I recall that when I was using
openSuse on my laptop, a couple of years ago, I needed to patch the OS
because otherwise the pexpect interface became extremely slow.

Best regards,
Simon

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/qeijjh%247fpq%241%40blaine.gmane.org.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] regression between 8.7 and 8.8.rc1 with operations on polygons with vertices in a quadratic field

[Sébastien Labbé]

> no, it does not, please see https://trac.sagemath.org/ticket/28037 
> (2nd  part of the 
> description). One must remove executables. 
> (all these $SAGE_LOCAL/bin/point2*) 
> if I remove them manually, everything gets back to normal. 
>

Thanks for creating the ticket and for that clean trick.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/b3b5236d-1759-4a6c-818c-5438e8ca9c0f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] regression between 8.7 and 8.8.rc1 with operations on polygons with vertices in a quadratic field

[Dima Pasechnik]

On Thu, Jun 20, 2019 at 9:53 PM Sébastien Labbé  wrote:
>
>
>>> Is there an easy command to uninstall topcom?
>>
>> make topcom-clean
>>
>> should work...
>>
>
> it seems to uninstall properly (topcom does not appear in the list of 
> optional packages), but the issue persists after uninstallation

no, it does not, please see https://trac.sagemath.org/ticket/28037
(2nd part of the description). One must remove executables.
(all these $SAGE_LOCAL/bin/point2*)
if I remove them manually, everything gets back to normal.

>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/908c6a80-62b6-4f47-8e5b-9e12d0158335%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq3iYeianSaT2Qtf3zm-DLzJVK2jbko7rgVGEcf_wf2MLw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] regression between 8.7 and 8.8.rc1 with operations on polygons with vertices in a quadratic field

[Dima Pasechnik]

On Fri, Jun 21, 2019 at 10:39 AM Sébastien Labbé  wrote:
>
> Ok so here is an easy way to reproduce the sleep problem:
>
> With topcom installed, I get a bunch of sleep calls when I compute the volume 
> of a triangle:
>
> sage: p = Polyhedron([(0,1/3),(0,1),(1/3,1)])
> sage: %prun p.volume()
>
>  750 function calls (749 primitive calls) in 0.216 seconds
>
>Ordered by: internal time
>
>ncalls  tottime  percall  cumtime  percall filename:lineno(function)
> 50.2010.0400.2010.040 {time.sleep}
> 30.0060.0020.0060.002 {posix.read}

yes, I can reproduce this.

I think it's stuff in src/sage/geometry/triangulation/point_configuration.py
that plays fast and loose with "engine" parameter.

Basically, this is a great illustration that pexpect interface (what's
topcom using)
is slow...

Let us fix this https://trac.sagemath.org/ticket/28037

> 10.0040.0040.0040.004 {posix.forkpty}
> 40.0010.0000.0010.000 {select.select}
> 10.0000.0000.0010.001 
> point_configuration.py:1291(volume)
> 20.0000.0000.2150.107 
> point_configuration.py:671(_TOPCOM_communicate)
> 10.0000.0000.2160.216 base.py:5201(volume)
>210.0000.0000.0000.000 {posix.lstat}
> 10.0000.0000.0100.010 ptyprocess.py:172(spawn)
> 10.0000.0000.0000.000 matrix_space.py:720(__call__)
> 10.0000.0000.0000.000 ptyprocess.py:152(__init__)
> 10.0000.0000.0000.000 
> matrix_space.py:432(__classcall__)
> 20.0000.0000.2140.107 
> point_configuration.py:597(_TOPCOM_exec)
> 10.0000.0000.2160.216 :1()
> 60.0000.0000.0000.000 expect.py:13(new_data)
> 40.0000.0000.0020.001 expect.py:91(expect_loop)
> 40.0000.0000.0000.000 contextlib.py:21(__exit__)
> 10.0000.0000.0000.000 {method 'close' of '_io.FileIO' 
> objects}
> 40.0000.0000.0000.000 ptyprocess.py:670(isalive)
> 10.0000.0000.0000.000 
> matrix_space.py:103(get_matrix_class)
> 40.0000.0000.0000.000 
> spawnbase.py:192(compile_pattern_list)
> 20.0000.0000.1010.050 pty_spawn.py:526(send)
>210.0000.0000.0000.000 posixpath.py:132(islink)
> 30.0000.0000.0000.000 base.py:2062(dim)
>
> --
> You received this message because you are subscribed to the Google Groups 
> "sage-devel" group.
> To unsubscribe from this group and stop receiving emails from it, send an 
> email to sage-devel+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-devel@googlegroups.com.
> Visit this group at https://groups.google.com/group/sage-devel.
> To view this discussion on the web visit 
> https://groups.google.com/d/msgid/sage-devel/2b4c27bd-ee66-4814-8502-bcf9889dd888%40googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/CAAWYfq2t7-9FnJ8zMThrndv-YXq66C%2B7Xpc9iE6FnY3oDuyHTg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] regression between 8.7 and 8.8.rc1 with operations on polygons with vertices in a quadratic field

[Sébastien Labbé]

Ok so here is an easy way to reproduce the sleep problem:

With topcom installed, I get a bunch of sleep calls when I compute the 
volume of a triangle:

sage: p = Polyhedron([(0,1/3),(0,1),(1/3,1)])
sage: %prun p.volume()

 750 function calls (749 primitive calls) in 0.216 seconds

   Ordered by: internal time

   ncalls  tottime  percall  cumtime  percall filename:lineno(function)
50.2010.0400.2010.040 {time.sleep}
30.0060.0020.0060.002 {posix.read}
10.0040.0040.0040.004 {posix.forkpty}
40.0010.0000.0010.000 {select.select}
10.0000.0000.0010.001 
point_configuration.py:1291(volume)
20.0000.0000.2150.107 
point_configuration.py:671(_TOPCOM_communicate)
10.0000.0000.2160.216 base.py:5201(volume)
   210.0000.0000.0000.000 {posix.lstat}
10.0000.0000.0100.010 ptyprocess.py:172(spawn)
10.0000.0000.0000.000 matrix_space.py:720(__call__)
10.0000.0000.0000.000 ptyprocess.py:152(__init__)
10.0000.0000.0000.000 
matrix_space.py:432(__classcall__)
20.0000.0000.2140.107 
point_configuration.py:597(_TOPCOM_exec)
10.0000.0000.2160.216 :1()
60.0000.0000.0000.000 expect.py:13(new_data)
40.0000.0000.0020.001 expect.py:91(expect_loop)
40.0000.0000.0000.000 contextlib.py:21(__exit__)
10.0000.0000.0000.000 {method 'close' of 
'_io.FileIO' objects}
40.0000.0000.0000.000 ptyprocess.py:670(isalive)
10.0000.0000.0000.000 
matrix_space.py:103(get_matrix_class)
40.0000.0000.0000.000 
spawnbase.py:192(compile_pattern_list)
20.0000.0000.1010.050 pty_spawn.py:526(send)
   210.0000.0000.0000.000 posixpath.py:132(islink)
30.0000.0000.0000.000 base.py:2062(dim)

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/2b4c27bd-ee66-4814-8502-bcf9889dd888%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] regression between 8.7 and 8.8.rc1 with operations on polygons with vertices in a quadratic field

[Sébastien Labbé]

>
>> perhaps a dodgy pip-installed module, installed as a dependency, or some 
>> other thing that turned that weird multiprocessing (mostly sleeping though) 
>> thing on.
>>
>
> On a fresh sage-8.8.rc2 clone, the following reproduces the problem 
> without any dodgy pip-installed module:
>
> sage -i topcom
> sage -t arXiv_1906_01104.rst
>

oups, you are right, I forgot there is one dodgy pip module involved to 
reproduce the issue:

sage -pip install slabbe
sage -i topcom
sage -t arXiv_1906_01104.rst

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/151d4324-8379-4f9b-83e5-4a98e7999e7c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-devel] regression between 8.7 and 8.8.rc1 with operations on polygons with vertices in a quadratic field

[Sébastien Labbé]

>
>
>> it seems to uninstall properly (topcom does not appear in the list of 
>> optional packages), but the issue persists after uninstallation
>>
>>
> perhaps a dodgy pip-installed module, installed as a dependency, or some 
> other thing that turned that weird multiprocessing (mostly sleeping though) 
> thing on.
>

On a fresh sage-8.8.rc2 clone, the following reproduces the problem without 
any dodgy pip-installed module:

sage -i topcom
sage -t arXiv_1906_01104.rst

topcom has only one dependency being cddlib which is a standard package.  
So the only other thing that turns that weird multiprocessing thing has to 
be caused by the installation of TOPCOM itself.
 

>
> I would also check if there is not something in ~/.sage/ that causes this 
> issue.
>

I am able to reproduce the issue even after doing:

mv .sage .sage-moved

So I do not think something in .sage causes the issue.

Sébastien

-- 
You received this message because you are subscribed to the Google Groups 
"sage-devel" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-devel+unsubscr...@googlegroups.com.
To post to this group, send email to sage-devel@googlegroups.com.
Visit this group at https://groups.google.com/group/sage-devel.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sage-devel/961632c4-2cb3-4bb4-b007-8dc9e997c144%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.