Messages in this thread | | | Date | Mon, 8 Aug 2016 16:04:51 +0200 | From | Peter Zijlstra <> | Subject | Re: [RFC PATCH 3/4] Make deadline max-heap faster avoiding real swaps on heapify ops. |
| |
On Tue, Jul 19, 2016 at 11:44:52AM +0200, Tommaso Cucinotta wrote: > This change achieves up to 10% of speed-up for cpudl_set() calls, as > measured with a andomly generated workload of 1K,10K,100K random heap > insertions and deletions (75% cpudl_set() calls with is_valid=1 and > 25% with is_valid=0), and randomly generated cpu IDs, with up to 256 > CPUs, as measured on an Intel Core2 Duo.
Changelog fails to explain _what_ the change is.
Some details on the change in implementation and why this is faster would be good.
| |