Message263730
| Author |
lemburg |
| Recipients |
brett.cannon, lemburg, pitrou, r.david.murray, vstinner, yselivanov |
| Date |
2016年04月19日.11:01:23 |
| SpamBayes Score |
-1.0 |
| Marked as misclassified |
Yes |
| Message-id |
<57160FF3.8090203@egenix.com> |
| In-reply-to |
<1461063141.34.0.82021694339.issue26058@psf.upfronthosting.co.za> |
| Content |
On 19.04.2016 12:52, STINNER Victor wrote:
>
> As usual, I'm very skeptical on the pybench results which almost look like noise. I don't understand how my change can make any operation *faster*, whereas some benchmarks are faster with the patch...
This can easily happen as a result of different memory layout, but is
very much dependent on the machine architecture, CPU, memory type, etc.
> Dict microbenchmarks:
>
> DictCreation: 38ms 36ms +4.8% 39ms 37ms +3.9%
> DictWithFloatKeys: 40ms 40ms -0.8% 40ms 40ms -0.4%
> DictWithIntegerKeys: 33ms 31ms +7.2% 33ms 31ms +7.6%
> DictWithStringKeys: 29ms 28ms +0.4% 29ms 29ms +0.7%
> SimpleDictManipulation: 59ms 59ms -0.4% 59ms 59ms -0.4%
Only dict creation and the integer keys benchmark results are relevant.
Could you perhaps check what's causing these slowdowns ? |
|