Subject: make-array with initial-contents in gcl is slow
From: Rupert Swarbrick
Date: Wed, 18 Jan 2012 14:47:52 +0000
John Lapeyre <lapeyre.math122a at gmail.com> writes:
> In gcl make-array with :initial-contents from a list
> is poorly implemented in that the copying is O(n^2).
> A test on one machine shows that initializing a
> list of length of 5 10^4 takes 1 minute in gcl and a few ms in sbcl.
Have you reported this to the GCL project? It would be bizarre, to say
the least, for Maxima to ship with fixes to various lisp
implementations' basic algorithms.
Rupert
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 315 bytes
Desc: not available
URL: <http://www.math.utexas.edu/pipermail/maxima/attachments/20120118/83cc3d3e/attachment.pgp>;