On 8/20/24 3:10 PM, David Wei wrote: > On 2024-08-19 16:28, Jens Axboe wrote: >> Waiting for events with io_uring has two knobs that can be set: >> >> 1) The number of events to wake for >> 2) The timeout associated with the event >> >> Waiting will abort when either of those conditions are met, as expected. >> >> This adds support for a third event, which is associated with the number >> of events to wait for. Applications generally like to handle batches of >> completions, and right now they'd set a number of events to wait for and >> the timeout for that. If no events have been received but the timeout >> triggers, control is returned to the application and it can wait again. >> However, if the application doesn't have anything to do until events are >> reaped, then it's possible to make this waiting more efficient. >> >> For example, the application may have a latency time of 50 usecs and >> wanting to handle a batch of 8 requests at the time. If it uses 50 usecs >> as the timeout, then it'll be doing 20K context switches per second even >> if nothing is happening. >> >> This introduces the notion of min batch wait time. If the min batch wait >> time expires, then we'll return to userspace if we have any events at all. >> If none are available, the general wait time is applied. Any request >> arriving after the min batch wait time will cause waiting to stop and >> return control to the application. > > I think the batch request count should be applied to the min_timeout, > such that: > > start_time min_timeout timeout > |--------------------|--------------------| > > Return to user between [start_time, min_timeout) if there are wait_nr > number of completions, checked by io_req_local_work_add(), or is it > io_wake_function()? Right, if we get the batch fulfilled, we should ALWAYS return. If we have any events and min_timeout expires, return. If not, sleep the full timeout. > Return to user between [min_timeout, timeout) if there are at least one > completion. Yes > Return to user at timeout always. Yes This should be how it works, and how I described it in the commit message. -- Jens Axboe
