[Octave-bug-tracker] [bug #62358] Performance of Matrix Multiplication significantly slower in 7.1.0 compared to 6.1.0

[Markus Mützel]

Follow-up Comment #10, bug #62358 (project octave):

Thanks for bisecting.

@jwe: Do we have an idea why using custom allocators would cause a performance
decrease? Is there a way to get back to the previous performance?
Would it make a difference if we had a member in ArrayRep that we'd use as the
allocator for the data in that class instead of the ArrayRep class itself
being an allocator class?

-  class ArrayRep
+  class ArrayRep : public Alloc




    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?62358>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/