Messages in this thread | | | Subject | Re: [RFC 1/2] softirq: Defer net rx/tx processing to ksoftirqd context | From | Mike Galbraith <> | Date | Fri, 12 Jan 2018 06:23:08 +0100 |
| |
On Thu, 2018-01-11 at 12:22 -0800, Linus Torvalds wrote: > On Thu, Jan 11, 2018 at 12:16 PM, Eric Dumazet <edumazet@google.com> wrote: > > > > Note that when I implemented TCP Small queues, I did experiments between > > using a work queue or a tasklet, and workqueues added unacceptable P99 > > latencies, when many user threads are competing with kernel threads. > > Yes. > > So I think one solution might be to have a hybrid system, where we do > the softirq's synchronously normally (which is what you really want > for good latency). > > But then fall down on a threaded model - but that fallback case should > be per-softirq, not global. So if one softirq uses a lot of CPU time, > that shouldn't affect the latency of other softirqs. > > So maybe we could get rid of the per-cpu ksoftirqd entirely, and > replace it with with per-cpu and per-softirq workqueues?
How would that be better than what RT used to do, and I still do for my RT kernels via boot option, namely split ksoftirqd into per-softirq threads.
-Mike
| |