Nicholas Piggin <npig...@gmail.com> writes: > We want to move away from using SMT prioroty updates for cpu_relax, and > use a 'wait' instruction which is similar to x86. As well as being a > much better fit for what everybody else uses and tests with, priority > nops are stateful which is nasty (interrupts have to consider they might > be taken at a different priority), and they're expensive to execute, > similar to a mtSPR which can effect other threads in the pipe. > > Signed-off-by: Nicholas Piggin <npig...@gmail.com> > --- > Unfortunately qemu TCG does not emulate pause_short properly and will > cause hangs.
That _really_ sucks for testing, being able to use qemu is a huge benefit. I can boot test multiple kernels per minute using qemu, vs multiple minutes per kernel using real hardware. > I have a patch for it but not merged yet. But if we tune > qspinlock code it would be best to do it with this patch. What's the urgency on this patch? Can we wait for the qemu change to land? I guess Qemu 8 is not due until next year? :/ cheers
