On Aug 15, 5:48 am, Simon King wrote:
> Anyway, I'd be glad if someone could review #715, #11521, #12313 and
> #13370...
I'm trying! In the process, I think I might have found a useful
debugtool. In sage.structure.coerce_dict, I have added:
cpdef refcache():
"""
for debug reasons only! A
On 2012-08-15, Simon King wrote:
> I have created #13370...
... which now needs review.
It fixes one leak that has not be fixed in #715, #12215, #11521 and
#12313, and it improves some benchmark for computing echelon forms
(which should be interesting, given the original reason for this
thread).
Hi!
I have created #13370. My plan is to remove all calls to is_Field(R) and
replace them by "R in Fields()". This mainly concerns sage/schemes.
Using is_Field will be deprecated. A non-deprecated version will be
called _is_Field, and will (still) be used in Fields().__contains__.
The important po
PS:
On 2012-08-15, Simon King wrote:
> Alternatively, one could try to use the category framework in a nasty
> way: If "R in Fields()" returns True, then R.category() is refined to a
> sub-category of Fields(). In that way, is_Field(R) would only be called
> in the first instance of "R in Fields(
Hi Nils,
On 2012-08-15, Nils Bruin wrote:
> I think I have identified this mysterious object. I think it's
>
> sage/rings/ring.pyx:1926
>
> cdef dict _is_Field_cache = {}
>
> as introduced in
>
> http://trac.sagemath.org/sage_trac/ticket/11900
>
> with author listed Simon King :-)
Oops.
> DONT
> #The mysterious object is now L[0] and the next line shows that it
> doesn't really have a reference (the only one being the list L which
> we have just created)
I think I have identified this mysterious object. I think it's
sage/rings/ring.pyx:1926
cdef dict _is_Field_cache = {}
as introduce
Hi Nils,
On 2012-08-15, Nils Bruin wrote:
> Interesting data point: The echelonization isn't important. It's the
> creation of a non-square matrix space that seems to cement the
> integermodring in memory.
That is interesting indeed (and surprising to me, because I'd rather
expect that the categ
Interesting data point: The echelonization isn't important. It's the
creation of a non-square matrix space that seems to cement the
integermodring in memory. This has the same effect as before (you can
get the object L[0] as above)
import gc, collections
import sage.matrix.matrix_space as matrix_s
Hi Nils,
On 2012-08-15, Nils Bruin wrote:
> Continuing ... I think the patches made some difference. Previously, I
> saw a bunch of MatrixSpaces on the heap and they seem to have
> disappeared.
That's good!
> However, there are a whole bunch of IntegerModRings still around.
Like this?
sage:
Continuing ... I think the patches made some difference. Previously, I
saw a bunch of MatrixSpaces on the heap and they seem to have
disappeared. However, there are a whole bunch of IntegerModRings still
around. A very conspicuous structure is an ordinary dictionary that
has IntegerModRings as keys
On Aug 14, 2:07 am, Simon King wrote:
> #12215, #11521 and #12313 need review, #715 has a positive review but
> depends on #11521. Do these patches solve the memory leak you observed?
No they don't! I'm surprised. There is apparently yet something else
that keeps these objects alive. I've tried
Hi Nils,
On 2012-08-14, Nils Bruin wrote:
> On Aug 14, 2:07 am, Simon King wrote:
>
>> #12215, #11521 and #12313 need review, #715 has a positive review but
>> depends on #11521. Do these patches solve the memory leak you observed?
>
> I wasn't able to merge 715 on 5.3beta2. We should probably a
On Aug 14, 2:07 am, Simon King wrote:
> #12215, #11521 and #12313 need review, #715 has a positive review but
> depends on #11521. Do these patches solve the memory leak you observed?
I wasn't able to merge 715 on 5.3beta2. We should probably agree on
the version on which this should go. Are you
Hi Nils,
On 2012-08-14, Nils Bruin wrote:
> It looks there is quite some work on this:
>
> http://trac.sagemath.org/sage_trac/ticket/12215
> http://trac.sagemath.org/sage_trac/ticket/11521
> http://trac.sagemath.org/sage_trac/ticket/715
> http://trac.sagemath.org/sage_trac/ticket/12313
>
> Let me
It looks there is quite some work on this:
http://trac.sagemath.org/sage_trac/ticket/12215
http://trac.sagemath.org/sage_trac/ticket/11521
http://trac.sagemath.org/sage_trac/ticket/715
http://trac.sagemath.org/sage_trac/ticket/12313
Let me know if there is something I can help with to move this a
15 matches
Mail list logo
