On Thu, 2014-07-31 at 18:16 +0200, Jirka Hladky wrote:
Peter, I'm seeing regressions for
SINGLE SPECjbb instance for number of warehouses being the same as total
number of cores in the box.
Example: 4 NUMA node box, each CPU has 6 cores => biggest regression is
for 24 warehouses.
By looking at your graph, that's around a 10% difference.
So I'm not seeing anywhere near as bad a regression on a 80-core box.
Testing single with 80 warehouses, I get:
Avg: 695385.94 bops
tip/master + patch (NUMA_SCALE/8 variant):
Avg: 707792.022 bops
So both these are pretty similar, however, when reverting, on avg we
increase the amount of bops a mere ~4%:
tip/master + reverted:
Avg: 738171.84 bops
Are there perhaps any special specjbb options you are using?