Re: [sage-release] Sage 6.10.beta4 released

2015-11-17 Thread John Cremona 
On 13 November 2015 at 00:29, Rob Beezer  wrote:
> On SageMathCloud, built from source, passes long tests ("ptestlong"), except
>
> src/sage/schemes/elliptic_curves/ell_number_field.py
>
> timed out first time, and passed when run independently.

Can you say which test timed out?   Some of the longer ones may be my fault...

John

>
>
> Development Images avaiable at:
> https://cloud.sagemath.com/projects/53b77207-8614-4086-a032-432af4b4cdbd/files/sage-dev-images/
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-release" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-release+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-release@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-release.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Sage 6.10.beta4 released

2015-11-17 Thread Justin C. Walker 

On Nov 11, 2015, at 22:13 , Volker Braun wrote:

> As always, you can get the latest beta version from the "develop" git 
> branch. Alternatively, the self-contained source tarball is at 
> http://www.sagemath.org/download-latest.html

Built from tarball, w/o problems, on two OS X systems: 10.6.8 (Dual 6-core 
Xeons), 10.10.5 (Quad-core Core i7).

Testing ('ptestlong') on 10.10.5 completed w/o complaints.

On 10.6.8, the first attempt caused a system lock-up, resulting in a forced 
reboot after 30 minutes of thumb-twiddling.  The second attempt completed with 
an "unhandled floating point exception" while testing 
src/sage/categories/classical_crystals.py.

This was not repeatable.  In addition, the second test froze the system for a 
short while.  I suspect this old horse (the machine, not me) is hurting in some 
way.  These problems have been with me for only a short while, but they occur 
with every full test run.

Justin

--
Justin C. Walker, Curmudgeon at Large
Director
Institute for the Enhancement of the Director's income
---
--
They said it couldn't be done, but sometimes,
it doesn't work out that way.
  - Casey Stengel
--



-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Sage 6.10.beta4 released

2015-11-17 Thread John Cremona 
On 17 November 2015 at 12:38, Rob Beezer  wrote:
> Hi John,
>
> The log for this is a bit long.  This link should be publicly accessible.
> If not, let me know and I'll send the log to you as an attachment off-list.
>
> https://cloud.sagemath.com/53b77207-8614-4086-a032-432af4b4cdbd/raw/sage-dev-images/sage-6.10.beta4/logs/ptestlong.log
>

Thanks, I could read that with no trouble and the offending test was
as I guessed one of the isogeny class tests.  I put in several since
there are many different things that can happen, but these only happen
for fields of quite large degree, hence slow.  (But desirable -- no
other package can do this!)

The (worst) offender is

File "src/sage/schemes/elliptic_curves/ell_number_field.py", line
2852, in 
sage.schemes.elliptic_curves.ell_number_field.EllipticCurve_number_field.reduction.isogeny_class
Warning, slow doctest:
CL = EL.isogeny_class(); len(CL) # long time (~21s)
Test ran for 898.04 s
Timed out

and it's a mystery why your test ran for 898s compared to my comment
of about 21s.  When I ran that just now on Sage-6.9 it took 2m 53s,
i.e. 173s, which I'm sure people will say is too long for a doctest,
but a lot less than 898s!

We (or I) should probably mark this as # not tested in view of this.
Even E.isogeny_class() now takes 21s when it used to take half that.
This is not exactly a regression;  but since those tests were first
written I added code which returns an optimal model of each curve in
the isogeny class, and it is that step which takes a long time over a
large degree field.

Sorry.

John

> HTH,
> Rob
>
>
>
>
> On 11/17/2015 06:13 AM, John Cremona wrote:
>>
>> On 13 November 2015 at 00:29, Rob Beezer  wrote:
>>>
>>> On SageMathCloud, built from source, passes long tests ("ptestlong"),
>>> except
>>>
>>> src/sage/schemes/elliptic_curves/ell_number_field.py
>>>
>>> timed out first time, and passed when run independently.
>>
>>
>> Can you say which test timed out?   Some of the longer ones may be my
>> fault...
>>
>> John
>>
>>>
>>>
>>> Development Images avaiable at:
>>>
>>> https://cloud.sagemath.com/projects/53b77207-8614-4086-a032-432af4b4cdbd/files/sage-dev-images/
>>>
>>> --
>>> You received this message because you are subscribed to the Google Groups
>>> "sage-release" group.
>>> To unsubscribe from this group and stop receiving emails from it, send an
>>> email to sage-release+unsubscr...@googlegroups.com.
>>> To post to this group, send email to sage-release@googlegroups.com.
>>> Visit this group at http://groups.google.com/group/sage-release.
>>> For more options, visit https://groups.google.com/d/optout.
>>
>>
>
> --
> You received this message because you are subscribed to the Google Groups
> "sage-release" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-release+unsubscr...@googlegroups.com.
> To post to this group, send email to sage-release@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-release.
> For more options, visit https://groups.google.com/d/optout.

-- 
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.

Re: [sage-release] Sage 6.10.beta4 released

2015-11-17 Thread Rob Beezer 

Dear John,

It feels to me like SMC has been a bit slow on occassion the past few days.  But 
I have not looked carefully at the load since I'm not sure what indicators to 
examine.  My project is on a members-only server, and reasonably powerful, 
though maybe I should give it more "CPU shares" than the bare minimum.


Perhaps William has some insights on the situation.  I just added one of my 4.5 
remaining CPU shares to that project - perhaps that will help.


I personally do not mind the slow-down (or timeouts), as I just build these for 
others to copy (and hence they don't need to run initial tests).  So I wouldn't 
be in a rush to limit these tests if they are behaving as expected on 
"standalone" hardware.


Rob

On 11/17/2015 11:16 AM, John Cremona wrote:

On 17 November 2015 at 12:38, Rob Beezer  wrote:

Hi John,

The log for this is a bit long.  This link should be publicly accessible.
If not, let me know and I'll send the log to you as an attachment off-list.

https://cloud.sagemath.com/53b77207-8614-4086-a032-432af4b4cdbd/raw/sage-dev-images/sage-6.10.beta4/logs/ptestlong.log



Thanks, I could read that with no trouble and the offending test was
as I guessed one of the isogeny class tests.  I put in several since
there are many different things that can happen, but these only happen
for fields of quite large degree, hence slow.  (But desirable -- no
other package can do this!)

The (worst) offender is

File "src/sage/schemes/elliptic_curves/ell_number_field.py", line
2852, in 
sage.schemes.elliptic_curves.ell_number_field.EllipticCurve_number_field.reduction.isogeny_class
Warning, slow doctest:
 CL = EL.isogeny_class(); len(CL) # long time (~21s)
Test ran for 898.04 s
 Timed out

and it's a mystery why your test ran for 898s compared to my comment
of about 21s.  When I ran that just now on Sage-6.9 it took 2m 53s,
i.e. 173s, which I'm sure people will say is too long for a doctest,
but a lot less than 898s!

We (or I) should probably mark this as # not tested in view of this.
Even E.isogeny_class() now takes 21s when it used to take half that.
This is not exactly a regression;  but since those tests were first
written I added code which returns an optimal model of each curve in
the isogeny class, and it is that step which takes a long time over a
large degree field.

Sorry.

John


HTH,
Rob




On 11/17/2015 06:13 AM, John Cremona wrote:


On 13 November 2015 at 00:29, Rob Beezer  wrote:


On SageMathCloud, built from source, passes long tests ("ptestlong"),
except

src/sage/schemes/elliptic_curves/ell_number_field.py

timed out first time, and passed when run independently.



Can you say which test timed out?   Some of the longer ones may be my
fault...

John




Development Images avaiable at:

https://cloud.sagemath.com/projects/53b77207-8614-4086-a032-432af4b4cdbd/files/sage-dev-images/

--
You received this message because you are subscribed to the Google Groups
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.





--
You received this message because you are subscribed to the Google Groups
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.




--
You received this message because you are subscribed to the Google Groups 
"sage-release" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to sage-release+unsubscr...@googlegroups.com.
To post to this group, send email to sage-release@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-release.
For more options, visit https://groups.google.com/d/optout.