[Bug 200645] 4.18-rc regression bisected to e03fd3f30: amdgpu polaris11/rx460 only activates on one output/monitor of two

[bugzilla-daemon at bugzilla.kernel.org]

https://bugzilla.kernel.org/show_bug.cgi?id=200645

Duncan (1i5t5.duncan at cox.net) changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #277491|0                           |1
        is obsolete|                            |

--- Comment #6 from Duncan (1i5t5.duncan at cox.net) ---
Created attachment 277749
  --> https://bugzilla.kernel.org/attachment.cgi?id=277749&action=edit
xrandr output (working at 60 Hz)

The previous xrandr working output was with a setting on the 4K TV hooked up to
HDMI-A-0 set wrong, so @3840x2160 it'd only give me 30 Hz (tho that shouldn't
have mattered for this bug, since the problem is with DVI-D-0, not HDMI-A-0). 
This one has the TV set correctly so HDMI-A-0 will do 3840x2160 at full 60 Hz.

The bug's still there on rc8, with a revert of the commit to once again force
bpc <= 8 still fixing it (it's what I used to get this xrandr).  So at this
point it seems release will still be bugged.

Comparing to the tested-fixed bug @ freedesktop that bisected to the same
commit, the problem there was with the HDMI forced to 30 Hz @3840x2160, while
here it's with no display on a 1920x1080 DVI-D, with the 3840x2160 HDMI-A
behaving normally -- early-boot displays on the DVI-D and the HDMI-A at the
usual low res, but the DVI-D goes blank as soon as it switches to DRM/hi-res,
while the HDMI-A behaves entirely normally.  And once in X, xrandr says there's
no modes available for the (1920x1080-resolution) DVI-D, while again the
3840x2160 HDMI-A behaves entirely normally.

So perhaps a similar fix to the one on that bug (and now in tree) for HDMI, is
needed for DVI-D, tho it could be more than that as well, since (unlike the
freedesktop bug) the DVI-D is entirely blank and xrandr says it has no modes,
not even the low resolutions it was working with just fine in grub and early
boot.

-- 
You are receiving this mail because:
You are watching the assignee of the bug.