On Tue, Dec 14, 2021 at 12:55 PM Peter Zijlstra [off-list ref] wrote:
Hi,
This is actually tested code; but still missing the SMP wake-to-idle machinery.
I still need to think about that.
Thanks, Peter!
At a first glance, your main patch does not look much smaller than
mine, and I thought the whole point of re-doing it was to throw away
extra features and make things smaller/simpler...
Anyway, I'll test your patchset over the next week or so and let you
know if anything really needed is missing (other than waking an idle
server if there is one on a worker wakeup; this piece is definitely
needed).
I'll post my test-hack as a reply, but basically it does co-operative and
preemptive UP-like user scheduling.
Patches go on top of tip/master as they rely on the .fixup removal
recently merged in tip/x86/core.
Also, I still need to audit a bunch of mm code, because I'm not sure things are
actually as well behaved as this code supposes they are.