On Wed, Nov 20, 2024 at 2:42 PM Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote: > I generally think that if you are sleeping relatively long, you may as > well sacrifice some precision for avoiding system stress so to speak, > so I've been considering something like flat 50 us for sleeps between > 1 and 5 ms and then 1% of the sleep duration for longer sleeps. What is the maximum rate of acpi_os_sleep() invocations? Assuming the reasoning for user-space timer-slack is sound @ fixed 50 usec, what logic supports acpi_os_sleep paying more timer slack delay than user space? What measurements can demonstrate the benefit of this proposed additional delay?
