On 1/16/24 9:13 AM, Keith Busch wrote: > On Mon, Jan 15, 2024 at 02:53:55PM -0700, Jens Axboe wrote: >> If the block plug gets flushed, eg on preempt or schedule out, then >> we invalidate the cached clock. > > There must be something implicitly happening that I am missing. Where is > the 'cur_time' cached clock invalidated on a plug flush? It's not just yet, and it's also missing on preemption. I have those in my current tree. The current test doesn't hit any of these, so it didn't impact the testing. -- Jens Axboe
