Due to an influx of spam, we have had to impose restrictions on new accounts. Please see this wiki page for instructions on how to get full permissions. Sorry for the inconvenience.
Admin message
The migration is almost done, at least the rest should happen in the background. There are still a few technical difference between the old cluster and the new ones, and they are summarized in this issue. Please pay attention to the TL:DR at the end of the comment.
Displayport to HDMI 2.1 adapters no longer working correctly on 6.4.12 works on 6.4.11
I am using a DisplayPort to HDMI 2.1 adapter (specifically this one ) to workaround #1417 on Kernel versions before 6.4.12 I was able to successfully get 4k 120 full rgb with VRR enabled and working, however on 6.4.12 I get either a no signal error message on my tv, an invalid format error, or just a black screen.
Display(s): 1 LG C2 TV ( I do use 2 other LG monitors one with DP and the other with HDMI, but I was able to reproduce the issue with both disconnected)
Type of Display Connection: TV is connected with a DP -> HDMI 2.1 adapter
System information:
Distro name and Version: Garuda Linux
Kernel version: Linux version 6.4.12-arch1-1 (linux@archlinux) (gcc (GCC) 13.2.1 20230801, GNU ld (GNU Binutils) 2.41.0) #1 SMP PREEMPT_DYNAMIC Thu, 24 Aug 2023 00:38:14
Custom kernel: N/A
AMD official driver version: Mesa 23.1.6
How to reproduce the issue:
Connect an HDMI 2.1 display with a DP -> HDMI 2.1 adapter
@superm1 I can replicate the issue with my RX6700xt, arch linux, same display port adapter/cable but with LG C1.
After few kernel compilations git bisect claims this is the problematic commit:
commit dcfd5a3ecf91da0afc00ffa2ee0d20d65c45cfa5Author: Daniel Miess <daniel.miess@amd.com>Date: Tue Apr 25 14:02:02 2023 -0400drm/amd/display: Remove v_startup workaround for dcn3+[ Upstream commit 3a31e8b89b7240d9a17ace8a1ed050bdcb560f9e ][Why]Calls to dcn20_adjust_freesync_v_startup are no longerneeded as of dcn3+ and can cause underflow in some cases[How]Move calls to dcn20_adjust_freesync_v_startup up intovalidate_bandwidth for dcn2.xReviewed-by: Jun Lei <jun.lei@amd.com>Acked-by: Aurabindo Pillai <aurabindo.pillai@amd.com>Signed-off-by: Daniel Miess <daniel.miess@amd.com>Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>Signed-off-by: Alex Deucher <alexander.deucher@amd.com>Signed-off-by: Sasha Levin <sashal@kernel.org>.../gpu/drm/amd/display/dc/dml/dcn20/dcn20_fpu.c | 24 ++++++++++++++++++----
@superm1 ok test done ... i compiled linux-stable master at commit 7d2f353b2682dcfe5f9bc71e5b61d5b61770d98e - same issue
then i did git revert -n dcfd5a3ecf91da0afc00ffa2ee0d20d65c45cfa5
and that fixes it, so it seems like at least for me that commit is pretty much what breaks it ...
now that i have all this setup - let me know if you want me to test any variations of that commit or anything else.
just do cat /path/to/that/file |patch -p1 while in the root kernel source dir/checkout ... no clue if it will apply to drm-next, but it applies for me on 6.4.12
if you are messing with the arch PKGBUILD ...there is a way to tell it to apply a patch before it builds, but remember being somewhat confusing.
Also until a developer responds... I'm for sure not daily driving it with this patch ... it was just an exercise to confirm that it is indeed the commit causing the issue for me.
Thanks, @gyordanov. I just wanted to check if 6.5 fixes it for me (and I didn't want to sit for an hour and build a kernel, so I got prebuilt packages from chaotic aur). I'll try to
Thanks. We unfortunately can't revert as that commit fixes some other problems so it's trading one issue for another. I've raised it with some internal folks to look at how to cover both cases.
Just want to confirm that v6.6-rc1 fixed the issue of invalid format with this adapter (not the exact same adapter that @gyordanov uses, but likely the same hardware doing the DP -> HDMI conversion).
I had the same experience as @gyordanov, this initially worked on 6.4.11 and was broken by 6.4.12.
Using a RX 6800XT with an LG C1.
Update 1: Seems like when waking from "Screen energy saving" (KDE terminology, not sure what this translates to in kernel-land) I get the "Invalid format" message on the display. Switching to another TTY and then back again seems to resolve the issue.
Update 2: Seems like the screen energy saving issue does not trigger every time.
However, 6.6-rc1 has another issue leading to a GPU reset ("ring sdma3 timeout"). I'll check out 6.4.12 with reverted dcfd5a3ecf91da0afc00ffa2ee0d20d65c45cfa5 and see if that works.
Update 3: 6.4.12 with reverted dcfd5a3ecf91da0afc00ffa2ee0d20d65c45cfa5 works well, with the exception of "Invalid format" being displayed immediately upon login (i.e. I assume that's when kwin starts). This was solved by switching TTYs and back again.