[PATCH] [LNT] Aggregate most recent runs
Yi.Kong at arm.com
Fri May 9 09:18:22 PDT 2014
Patch updated to conform code style.
Also attached, screenshot comparison of stdev estimation using previous
10 runs and aggregation of 8 runs on a Cortex-A9 board.
On Thu, 2014-05-08 at 11:17 +0100, Yi Kong wrote:
> Hi all,
> After previously discussion, I've implemented aggregation across most
> recent runs. This method significantly reduces noise when applied in
> combination of Mann-Whitney U test. It replaces standard deviation
> estimation feature, as it clearly performs better.
> On a Cortex-A9 board running one iteration per run, regressions can be
> easily spotted when the sliding window is greater than 8 runs and
> aggregated by median, however changes are usually only significant
> enough after 6 consecutive runs. On x86 machines with 10 iterations, the
> improvement is not very obvious.
> Yi Kong
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 12837 bytes
More information about the llvm-commits