#14058: Weakly reference binary operation codomains
-----------------------------------+----------------------------------------
Reporter: robertwb | Owner: rlm
Type: enhancement | Status: needs_review
Priority: major | Milestone: sage-5.7
Component: memleak | Resolution:
Keywords: | Work issues:
Report Upstream: N/A | Reviewers:
Authors: Robert Bradshaw | Merged in:
Dependencies: #12313 | Stopgaps:
-----------------------------------+----------------------------------------
Comment (by SimonKing):
By the way, it seems that the "insufficient" caching (not enough to keep
C() alive) has no influence on performance, at least not if computations
are done in a close cycle:
{{{
sage: %attach /home/simon/SAGE/work/memleak/testcoercion.py
sage: OA = A()
sage: OB = B()
sage: a = OA(1)
sage: b = OB(1)
sage: def test(x,y):
....: for _ in xrange(2*10^5):
....: z = x*y
....:
sage: %time test(a,a)
CPU times: user 4.49 s, sys: 0.00 s, total: 4.49 s
Wall time: 4.49 s
sage: %time test(a,b)
CPU times: user 8.99 s, sys: 0.11 s, total: 9.11 s
Wall time: 9.13 s
sage: c = a*b
sage: print c.parent()
C
sage: %time test(a,b)
CPU times: user 8.85 s, sys: 0.03 s, total: 8.88 s
Wall time: 8.89 s
}}}
In the first run of test(a,b), there is the possibility that `parent(a*b)`
becomes created repeatedly, as I have demonstrated above. In the second
run, `parent(a*b)` is kept alive, by keeping a pointer to the result of
`a*b`. There is no significant difference in performance. Of course, we
also see that multiplication within one parent is faster.
Is there a way to get the patchbots report significant regressions in some
examples? Because I am sure that some parts of Sage (schemes and combinat
are typical candidates) rely on a strong coercion cache.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/14058#comment:18>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sage-trac?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
