On Tue, 30 Apr 2019 16:20:47 +0200 Peter Zijlstra <peterz@xxxxxxxxxxxxx> wrote: > > > > Sure, but that's a completely different issue. We would need to solve > > the 'emulate' call for batch mode first. > > I don't see a problem there; when INT3 happens; you bsearch() the batch > array to find the one you hit, you frob that into the percpu variables > and do the thing. Or am I loosing the plot again? I may need to tweak Linus's patch slightly, but I may be able to get it to work with the batch mode. -- Steve
