Author gregory.p.smith
Recipients christian.heimes, collinwinter, gregory.p.smith, jyasskin, loewis, mark.dickinson, pitrou, schuppenies, vstinner
Date 2009-02-19.02:40:15
SpamBayes Score 3.33141e-07
Marked as misclassified No
Message-id <>
Here are the results from 32-bit x86 on core2 duo gcc 4.0.1 using 4000:

30-bit digits                     (14):  15719 ms
30-bit digits + optimizations (13+ops):  12490 ms
unpatched py3k                        :  13289 ms

(again, i had to manually add #define PYLONG_DIGIT_SIZE 30 to pyconfig.h 
for the longdigit13+optimizations patch).

and pybench runs on the same builds vs unpatched:

30-bit digits                     (14):  -1.4% (slightly slower)
30-bit digits + optimizations (13+ops):  -0.2% (insignificant)

My votes:

Obviously use the optimized version (but fix the configure stuff).

+0  for enabling it by default on 32bit builds.
+10 for enabling it by default on 64bit builds.
Date User Action Args
2009-02-19 02:40:18gregory.p.smithsetrecipients: + gregory.p.smith, loewis, collinwinter, mark.dickinson, pitrou, vstinner, christian.heimes, jyasskin, schuppenies
2009-02-19 02:40:17gregory.p.smithsetmessageid: <>
2009-02-19 02:40:16gregory.p.smithlinkissue4258 messages
2009-02-19 02:40:15gregory.p.smithcreate