D8065: Ensure libinput connection thread uses real time scheduling

2017-10-31 Thread Martin Flöser
This revision was automatically updated to reflect the committed changes. Closed by commit R108:91d78daac49f: Ensure libinput connection thread uses real time scheduling (authored by graesslin). REPOSITORY R108 KWin CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D8065?vs=21011&id=2162

D8065: Ensure libinput connection thread uses real time scheduling

2017-10-23 Thread David Edmundson
davidedmundson accepted this revision. This revision is now accepted and ready to land. REPOSITORY R108 KWin BRANCH libinput-rr REVISION DETAIL https://phabricator.kde.org/D8065 To: graesslin, #kwin, #plasma, davidedmundson Cc: davidedmundson, anthonyfieroni, plasma-devel, kwin, ZrenBot,

D8065: Ensure libinput connection thread uses real time scheduling

2017-10-23 Thread David Edmundson
davidedmundson added a comment. Why didn't you just call gainRealTime again with reset_on_fork from the connection thread? By creating it without and then changing the main thread after we've created the input thread, we're also putting your DBus thread, and the thread used by the fake

D8065: Ensure libinput connection thread uses real time scheduling

2017-10-20 Thread Martin Flöser
graesslin updated this revision to Diff 21011. graesslin added a comment. Restricted Application edited projects, added Plasma; removed KWin. Use anonymous namespace REPOSITORY R108 KWin CHANGES SINCE LAST UPDATE https://phabricator.kde.org/D8065?vs=20131&id=21011 BRANCH libinput-rr R

D8065: Ensure libinput connection thread uses real time scheduling

2017-10-16 Thread Anthony Fieroni
anthonyfieroni added inline comments. INLINE COMMENTS > graesslin wrote in main_wayland.cpp:91 > Could you please explain why you think that this function should be static? > Please keep in mind that main_wayland.cpp is it's own translation unit. It does not have other purpose, i don't see it's

D8065: Ensure libinput connection thread uses real time scheduling

2017-10-16 Thread Martin Flöser
graesslin added inline comments. INLINE COMMENTS > anthonyfieroni wrote in main_wayland.cpp:91 > static Could you please explain why you think that this function should be static? Please keep in mind that main_wayland.cpp is it's own translation unit. REPOSITORY R108 KWin REVISION DETAIL

D8065: Ensure libinput connection thread uses real time scheduling

2017-10-16 Thread Anthony Fieroni
anthonyfieroni added inline comments. INLINE COMMENTS > main_wayland.cpp:91 > + > +void gainRealTime(RealTimeFlags flags = RealTimeFlags::DontReset) > +{ static REPOSITORY R108 KWin REVISION DETAIL https://phabricator.kde.org/D8065 To: graesslin, #kwin, #plasma Cc: anthonyfieroni, plasma-

D8065: Ensure libinput connection thread uses real time scheduling

2017-10-16 Thread Martin Flöser
graesslin added a comment. ping REPOSITORY R108 KWin REVISION DETAIL https://phabricator.kde.org/D8065 To: graesslin, #kwin, #plasma Cc: plasma-devel, kwin, bwowk, ZrenBot, progwolff, lesliezhai, ali-mohamed, hardening, jensreuterberg, abetts, sebas, apol, mart

D8065: Ensure libinput connection thread uses real time scheduling

2017-09-30 Thread Martin Flöser
graesslin created this revision. graesslin added reviewers: KWin, Plasma. Restricted Application added a project: KWin. Restricted Application added subscribers: kwin, plasma-devel. REVISION SUMMARY The SCHED_RESET_ON_FORK results in threads (not only processes) to have the scheduling policy r