| Bug ID | 95206 |
|---|---|
| Summary | Display port bandwidth regression |
| Product | DRI |
| Version | unspecified |
| Hardware | x86-64 (AMD64) |
| OS | Linux (All) |
| Status | NEW |
| Severity | normal |
| Priority | medium |
| Component | DRM/Radeon |
| Assignee | dri-devel@lists.freedesktop.org |
| Reporter | barneyman@gmx.de |
Hi, I have a mobile laptop (Dell Precision M4600) with an external monitor (Dell U2414H) connected via display port. Starting with mainline kernel 4.5.x I'm unable to connect the monitor with mode (1920x1080p), instead the highest available option is 1920x1080i (interlaced) which produces heavy flickering on the monitor and makes it unable to use at all. Forcing the monitor into 1920x1080p (using xrandr --newmode, --addmode and --mode) causes a total signal loss, however forcing the monitor into 1920x1080p@30hz makes the monitor available again. If I connect the monitor using DVI/HDMI 1080p@60hz works without any problems. So I think the problem might be that the maximum bandwidth of the display port is reduced compared to the previous version. Using kernel 4.4.8 the monitor is connected with 1080p using DP as expected again. So it's definitely a regression introduced in kernel 4.5. As a side note: the Ubuntu 16.04 kernel (4.4.0-21.37) also has this display port problem, but as far as I know the ubuntu devs have backported large portions of the 4.5 radeon patches. If you need any further information (dmesg, xrandr output), don't hesitate to ask... Kind regards, Christian
Please attach your xorg log and dmesg output. Can you bisect?
Created attachment 123350 [details]
dmesg kernel 4.5.2
Created attachment 123351 [details]
Xorg.0.log kernel 4.5.2
Created attachment 123352 [details]
xrandr output kernel 4.5.2
"bisecting" is a way to find out which commit caused a specific regression. This involves compiling the linux kernel from git and testing the compiled versions. If you can find out which commit is the culprit chances are pretty good that the problem can be fixed quickly. To learn more about bisecting I suggest seaching for "git bisect".
Ok, understood. Didn't know that feature. I'll give it a try in the next days.
Hi, I've bisected the problem down to the following commit to the mainline kernel (see below). Can I help with some further information? 092c96a8ab9d1bd60ada2ed385cc364ce084180e is the first bad commit commit 092c96a8ab9d1bd60ada2ed385cc364ce084180e Author: Alex Deucher <alexander.deucher@amd.com> Date: Thu Dec 17 10:23:34 2015 -0500 drm/radeon: fix dp link rate selection (v2) Need to properly handle the max link rate in the dpcd. This prevents some cases where 5.4 Ghz is selected when it shouldn't be. v2: simplify logic, add array bounds check Reviewed-by: Tom St Denis <tom.stdenis@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> :040000 040000 c27fa6e02534c3ef9f53c3c954fa936b1e414cb3 0be8f64ebe28383efa39ebce68247b148f4a78f2 M drivers
Created attachment 123611 [details] [review] add debugging output Can you attach the dmesg output with the attached debugging patch applied?<= /pre>
Created attachment 12=
3618 [details]
dmesg with debugging patch
attached dmesg output of mainline kernel 4.5.0 with your debugging patch
applied
Created attachment 123629 = [details] [review] possible fix This patch should fix it.
Your fix works like a charm. Thanks alot! Applied the fix to 4.5.0 mainline and the problem is gone.
May be causing https://bugzilla.redhat.com/show_bug.cgi?id=3D1402293
| What | Removed | Added |
|---|---|---|
| Status | NEW | RESOLVED |
| Resolution | --- | FIXED |
http://g= it.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=3Dff0bd44= 1bdfbfa09d05fdba9829a0401a46635c1
| What | Removed | Added |
|---|---|---|
| Resolution | FIXED | --- |
| Status | RESOLVED | REOPENED |
Sorry Alex, I wasn't clear. It's the fixed version that (I've now verified) is causing the Radeon HD6320 problem as described here https://b= ugzilla.redhat.com/show_bug.cgi?id=3D1402293
The (native) display resolution exhibiting the problem on the = Radeon HD6320 (AMD G-series T56N) is 1366x768.
The good and bad kernels give identical xrandr outputs: Screen 0: minimum 320 x 200, current 1366 x 768, maximum 8192 x 8192 XWAYLAND0 connected 1360x768+0+0 520mm x 320mm 1360x768 59.98*+
(In reply to mrj from = comment #15) > Sorry Alex, I wasn't clear. >=20 > It's the fixed version that (I've now verified) is causing the Radeon = HD6320 > problem as described here >=20 > http= s://bugzilla.redhat.com/show_bug.cgi?id=3D1402293 Please open a new bug as this one has been addressed.
| What | Removed | Added |
|---|---|---|
| Status | REOPENED | RESOLVED |
| Resolution | --- | FIXED |