io_uring sets current->in_iowait when waiting for completions, which achieves two things: 1. Proper accounting of the time as iowait time 2. Enable cpufreq optimisations, setting SCHED_CPUFREQ_IOWAIT on the rq For block IO this makes sense as high iowait can be indicative of issues. But for network IO especially recv, we do not control when the completions happen. When doing network IO with the new min-wait feature that lets io_uring wait for a certain number of completions before returning, this manifests as high iowait time. Some user tooling attributes iowait time as 'CPU utilisation' time, so high iowait time looks like high CPU util even though the task is not scheduled and the CPU is free to run other tasks. This patchset adds a IOURING_ENTER_NO_IOWAIT flag that can be set on enter. If set, then current->in_iowait is not set. By default this flag is not set to maintain existing behaviour i.e. in_iowait is always set. Not setting in_iowait does mean that we also lose cpufreq optimisations above because in_iowait semantics couples 1 and 2 together. Eventually we will untangle the two so the optimisations can be enabled independently of the accounting. David Wei (3): io_uring: add IORING_ENTER_NO_IOWAIT flag io_uring: do not set no_iowait if IORING_ENTER_NO_WAIT io_uring: add IORING_FEAT_IOWAIT_TOGGLE feature flag include/uapi/linux/io_uring.h | 2 ++ io_uring/io_uring.c | 8 +++++--- io_uring/io_uring.h | 1 + 3 files changed, 8 insertions(+), 3 deletions(-) -- 2.43.5
