On Fri, 09 Jan, at 04:58:35PM, Peter Zijlstra wrote: > On Fri, Jan 09, 2015 at 03:24:42PM +0000, Matt Fleming wrote: > > > > Right, but now we need a "steal limit", so we know when to stop stealing > > active RMIDs. > > > > (cqm_max_rmid + 1) / 4 ? > > > > I guess any limit is better than no limit. > > Yeah, that'll work, when the free+limbo count is 1/4th the total we > should stop pulling more plugs.
We actually stop pulling plugs as soon as there's a single RMID available to use as the rotation RMID. When we find one, we don't wait any longer to do any kind of batching (though if we go from zero available RMIDs, to 5 RMIDs in one loop, we'll pickup all 5 RMIDs). -- Matt Fleming, Intel Open Source Technology Center -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/