Date: Fri, 31 Mar 2023 18:09:19 -0000 (UTC) From: "Peter 'PMc' Much" <pmc@citylink.dinoex.sub.org> To: freebsd-hackers@freebsd.org Subject: Re: Periodic rant about SCHED_ULE Message-ID: <slrnu2e8ef.5bc.pmc@disp.intra.daemon.contact> References: <a401e51a-250a-64a0-15cb-ff79bcefbf94@m5p.com> <8173cc7e-e934-dd5c-312a-1dfa886941aa@FreeBSD.org> <8cfdb951-9b1f-ecd3-2291-7a528e1b042c@m5p.com> <c3f5f667-ba0b-c40c-b8a6-19d1c9c63c5f@FreeBSD.org> <ZBtRJhNHluj5Nzyk@troutmask.apl.washington.edu> <CAGudoHEj%2BkoaYhkjzDE5KX9OsCno=X5M_E3z9uwg6Pg7dtqTsA@mail.gmail.com> <CAGudoHHxTT-Cn11zcFB3ZwF76UcRUv=QS28RLgzd=hVehTy0Kg@mail.gmail.com> <CAGudoHGoh30O-3O0jjwevDvP43-ykUt6JUDiwRNW918VZfybhA@mail.gmail.com> <CAGudoHEWfy61XSMhXdYOrKWVotuC0Kc6NSWiaaZCy6aQhbvXoQ@mail.gmail.com> <CAGudoHFPqz_LtsVNnz4P2gyKXz5Z8hU%2Bv6QYGizm4%2BDtZRn8Yg@mail.gmail.com> <CAGudoHGzBjXjXZFs%2BqZJUS-M6VeX5=LB2ifRLP7hFBZXPvqP7g@mail.gmail.com> <CAK7dMtAsBehP2cy6cn31Z%2BSo6T2Q_mpN5ibEmYMNPOkWQHk8FA@mail.gmail.com> <CAK7dMtAHOpq579Hb_Ar3e-VaYZGfawAxBMDoM0Zt0T_=UD5Jgw@mail.gmail.com> <CAGudoHGNRQudBUj-XCNb=AirOgWn%2Bs0mi96v3axK4ZZSSdybZA@mail.gmail.com> <a2740715-bb8f-8c16-04d0-c84e2e6c9622@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On 2023-03-30, Matthias Andree <mandree@FreeBSD.org> wrote: > Am 30.03.23 um 20:50 schrieb Mateusz Guzik: > >> Anyhow after sending the above e-mail an actual solution hit me: the X >> server can tell the kernel what processes connect to it over the unix >> socket, which again very well may be good enough. >> >> In the reports I got I found pulseaudio, this one may need to be >> patched in a similar manner. > > Pulseaudio shouldn't need such stuff since it would either try to go > into a real-time scheduling group (which I don't think SCHED_ULE would > have), or do a (privileged) renice -11 to have considerable priority > over non-privileged stuff. At least on Linux, that is. > >From what I get, it seems most people with professional audio are using JACK, and JACK runs in rtprio by default: UID PID PPID C PRI NI VSZ RSS MWCHAN STAT TT TIME COMMAND 1100 1404 1403 2 -51 0 153508 71816 sigwait I<s - 1:28.97 /usr/local/bin/jackd -r With the other aspects, I think there are two different issues. One issue is that compute jobs do not always get their fair timeshare - this is concerning servers with batch compute load: I need such tasks to complete within precalculated time, and that does not always happen now. The other is about interactivity scoring etc., and that concerns mainly desktop systems, and I think that is a bit more difficult to get proper.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?slrnu2e8ef.5bc.pmc>