Message 82404 - Python tracker

Message82404

Author	mark.dickinson
Recipients	christian.heimes, collinwinter, gregory.p.smith, jyasskin, loewis, mark.dickinson, pitrou, schuppenies, vstinner
Date	2009-02-18.03:32:10
SpamBayes Score	3.1832802e-05
Marked as misclassified	No
Message-id	<1234927934.51.0.693592754595.issue4258@psf.upfronthosting.co.za>
In-reply-to

Content
Thanks very much for the timings, Victor. Just out of interest, could you try the pydigits script with the +optimizations patch on 32-bit? As mentioned above, there's a significant (for 30-bit digits) problem with x_divrem: the inner loop does a 32 x 64-bit multiply when it should be doing a 32 x 32-bit multiply (the variable q is declared as twodigits, but always fits into a digit). This is fixed in the +optimizations patch, and pi_digits is heavy on the divisions, so I wonder whether this might make a difference.

Content

Thanks very much for the timings, Victor.

Just out of interest, could you try the pydigits script with the 
+optimizations patch on 32-bit?

As mentioned above, there's a significant (for 30-bit digits) problem 
with x_divrem:  the inner loop does a 32 x 64-bit multiply when it 
should be doing a 32 x 32-bit multiply  (the variable q is declared as 
twodigits, but always fits into a digit).  This is fixed in the 
+optimizations patch, and pi_digits is heavy on the divisions, so I
wonder whether this might make a difference.

History
Date	User	Action	Args
2009-02-18 03:32:14	mark.dickinson	set	recipients: + mark.dickinson, loewis, collinwinter, gregory.p.smith, pitrou, vstinner, christian.heimes, jyasskin, schuppenies
2009-02-18 03:32:14	mark.dickinson	set	messageid: <1234927934.51.0.693592754595.issue4258@psf.upfronthosting.co.za>
2009-02-18 03:32:11	mark.dickinson	link	issue4258 messages
2009-02-18 03:32:10	mark.dickinson	create