Hi Lyude,
On 3/4/20 11:36 PM, Lyude Paul wrote:
AMD's patch series for adding DSC support to the MST helpers unfortunately introduced a few regressions into the kernel that I didn't get around to fixing until just now. I would have reverted the changes earlier, but seeing as that would have reverted all of amd's DSC support
- everything that was done on top of that I realllllly wanted to avoid
doing that.
Anyway, this should fix everything as far as I can tell. Note that I don't have any DSC displays locally yet, so if someone from AMD could sanity check this I would appreciate it ♥.
Thank you for trying to fix this, unfortunately for me this is not fixed, with this series. My setup:
5.6-rc4 + your 3 patches (+ some unrelated patches outside of drm)
-Lenovo x1 7th gen + Lenovo TB3 dock gen 2 + 2 external 1920x1080@60 monitors connected to the 2 HDMI interfaces on the dock -System booted with the LID closed, so that the firmware/BIOS has already initialized both monitors when the kernel boots
This should be fairly easy to reproduce on a similar setup, other users are seeing similar problems when connecting more then 1 monitor to DP-MST docks, see e.g. :
https://bugzilla.redhat.com/show_bug.cgi?id=1809681 https://bugzilla.redhat.com/show_bug.cgi?id=1810070
Let me know if you want me to collect some drm.debug logs, I guess if you do, you want me to use drm.debug=0x114 ?
Regards,
Hans
Cc: Mikita Lipski mikita.lipski@amd.com Cc: Alex Deucher alexander.deucher@amd.com Cc: Sean Paul seanpaul@google.com Cc: Hans de Goede hdegoede@redhat.com
Lyude Paul (3): drm/dp_mst: Rename drm_dp_mst_is_dp_mst_end_device() to be less redundant drm/dp_mst: Don't show connectors as connected before probing available PBN drm/dp_mst: Rewrite and fix bandwidth limit checks
drivers/gpu/drm/drm_dp_mst_topology.c | 124 ++++++++++++++++++++------ 1 file changed, 96 insertions(+), 28 deletions(-)