Bug 111876 - AMD 5700 XT / Navi - BenQ XL2420G - No EDID read with DP connector, HDMI works fine
Summary: AMD 5700 XT / Navi - BenQ XL2420G - No EDID read with DP connector, HDMI work...
Status: NEW
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/AMDgpu (show other bugs)
Version: DRI git
Hardware: x86-64 (AMD64) Linux (All)
: not set not set
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-10-01 14:11 UTC by Christopher Jordan
Modified: 2019-10-01 14:29 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg output when using DP (172.25 KB, text/plain)
2019-10-01 14:11 UTC, Christopher Jordan
no flags Details
dmesg output with HDMI connected (197.19 KB, text/plain)
2019-10-01 14:12 UTC, Christopher Jordan
no flags Details
lspci output (4.39 KB, text/plain)
2019-10-01 14:12 UTC, Christopher Jordan
no flags Details
pacman -Q output (28.88 KB, text/plain)
2019-10-01 14:13 UTC, Christopher Jordan
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Christopher Jordan 2019-10-01 14:11:22 UTC
Created attachment 145598 [details]
dmesg output when using DP

I recently built a new machine (specs below), but I'm having trouble using the display port connector with my old monitor (BenQ XL2420G) - I get the following EDID error:

[drm:dc_link_detect [amdgpu]] *ERROR* No EDID read.

Following instructions to manually use an EDID appears to do nothing (https://wiki.archlinux.org/index.php/kernel_mode_setting). I generated an EDID with the "read-edid" package after modprobe-ing i2c-dev and HDMI connected, as well as my old machine's nvidia card via "acquire EDID" in nvidia-settings.

I'm not sure how quirky other G-Sync or high-refresh-rate monitors are, but this one *only* allows a 144 Hz refresh rate with display port. However, using HDMI, everything appears to be fine. My Windows partition in the same machine appears quite happy to use DP at 144 Hz, so I'm particularly baffled as to why linux is struggling. After scouring the internet, it appears no one else has the same issue, so it is possible that this monitor is simply not playing nicely.

I have supplied my lspci output, as well as dmesg outputs with "drm.debug=0x1e log_buf_len=4M" for both DP and HDMI connected. I am running up-to-date archlinux, using linux-firmware from the testing repo (which should just be neatly providing the navi firmware in a non-AUR package) and a bunch of packages from the mesa-git repo. My pacman -Q is also attached.

Any help would be greatly appreciated! Many thanks in advance.

ASUS TUF Gaming X570-PLUS WiFi
AMD Ryzen 3900X
Powercolor Red Devil 5700 XT
Comment 1 Christopher Jordan 2019-10-01 14:12:17 UTC
Created attachment 145599 [details]
dmesg output with HDMI connected
Comment 2 Christopher Jordan 2019-10-01 14:12:35 UTC
Created attachment 145600 [details]
lspci output
Comment 3 Christopher Jordan 2019-10-01 14:13:06 UTC
Created attachment 145601 [details]
pacman -Q output
Comment 4 Christopher Jordan 2019-10-01 14:29:55 UTC
I forgot to mention - when using DP, after that EDID error, my screen displays a 640x480 tty. I can start X with it, but xrandr shows no other resolutions to use (probably because the KMS has no EDID info). I haven't been brave enough to manually generate and force EDIDs, but I figure that something else is wrong, as this all works fine under Windows.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.