AMDGPU DC Gets A Final Batch Of Changes Before Linux 4.15

[nuetzel]

This latest set of patches appear to have caused a regression from what I can tell; Sometimes, when the monitors go into standby mode after inactivity and you try to wake them up again in a multimonitor configuration, some might not come back again unless you do something to force a re-init (like replugging the video cable, for instance), restart X, or reboot.

Good finding!
Not 'only' sometimes, during _all_ transitions into 'standby mode' I get 'signal lost' with my RX580.
I've to re-init with replugging the video cable (after blind switching back to login screen (kdm_greet)) or reboot.
Will report to AMD's Alex Deucher and one of the powerplay devs in a minute (under the above links).
Thanks,
Dieter

[Masush5]

Yeah, the recent patches also broke output to my 4k tv with my RX560 again, logspam about vblank timeouts and eventual full machine hangs when it's connected (runs fine as long i only enable one 1080 screen). Will try to bisect it tomorrow.

[pal666]

the last times, it did not reach Linus

actually, last time linus said he is willing to accept it

[smitty3268]

actually, last time linus said he is willing to accept it

To clarify, last time around Linus said he trusted his subsystem maintainers, and so if Dave said it was ready to go he'd accept it. The flip side of that is that if Dave said it wasn't ready (which he did) then Linus would also accept that.

[ResponseWriter]

To clarify, last time around Linus said he trusted his subsystem maintainers, and so if Dave said it was ready to go he'd accept it. The flip side of that is that if Dave said it wasn't ready (which he did) then Linus would also accept that.

Was that before or after TinyDRM?