Bug 103382 - Flickering / Artifacts on TTY with GTX 1060
Summary: Flickering / Artifacts on TTY with GTX 1060
Status: RESOLVED MOVED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/nouveau (show other bugs)
Version: unspecified
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Nouveau Project
QA Contact: Xorg Project Team
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2017-10-21 02:20 UTC by Lukas Sabota
Modified: 2019-12-04 09:32 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg output (62.56 KB, text/plain)
2017-10-21 02:20 UTC, Lukas Sabota
no flags Details
EDID for Samsung monitor (349 bytes, text/plain)
2017-10-21 02:20 UTC, Lukas Sabota
no flags Details
EDID for LG Ultrawide monitor (349 bytes, text/plain)
2017-10-21 02:21 UTC, Lukas Sabota
no flags Details
VBIOS (222.00 KB, application/octet-stream)
2017-10-21 02:23 UTC, Lukas Sabota
no flags Details
decoded edids (9.23 KB, text/plain)
2017-10-21 02:34 UTC, Ilia Mirkin
no flags Details
dmesg with debugging enabled (3.75 MB, text/plain)
2017-10-21 02:37 UTC, Lukas Sabota
no flags Details
Verbose dmesg output: Booting with only LG Ultrawide (3.68 MB, text/plain)
2017-10-21 03:00 UTC, Lukas Sabota
no flags Details
Verbose dmesg output: Booting only with Samsung 4K (3.92 MB, text/plain)
2017-10-21 03:02 UTC, Lukas Sabota
no flags Details

Description Lukas Sabota 2017-10-21 02:20:03 UTC
Created attachment 134957 [details]
dmesg output

I have been unable to get a working TTY when using nouveau on my GTX 1060.  I have been booting straight to TTY and the TTY flickers and blacks every few seconds.  When my Ultrawide monitor is detected as the primary monitor, I also see colored artifacts on the screen.  However, when I launch sway or a compositor, the flickering/artifacting mostly stops, but the compositor seems to be unstable.

I am testing with Arch Linux 64bit with Linux kernel 4.13.7, package version "1.0.15-2" for extra/xf86-video-nouveau which is current with Arch Linux extra at the moment.

I have attached the EDID information for both of my monitors obtained via /sys.  I have also attached `dmesg` output from booting into nouveau and running into this issue.

Here is a link to the VBIOS that I have copied from /sys: https://xannode.com/vbios.rom

Monitors connected via DisplayPort:
Samsung U28E590D 28-Inch UHD LED-Lit Monitor
LG Ultrawide Monitor
Comment 1 Lukas Sabota 2017-10-21 02:20:41 UTC
Created attachment 134958 [details]
EDID for Samsung monitor
Comment 2 Lukas Sabota 2017-10-21 02:21:00 UTC
Created attachment 134959 [details]
EDID for LG Ultrawide monitor
Comment 3 Lukas Sabota 2017-10-21 02:23:29 UTC
Created attachment 134960 [details]
VBIOS
Comment 4 Ilia Mirkin 2017-10-21 02:28:27 UTC
One interesting part here:

[    7.578995] nouveau 0000:01:00.0: disp: 0x00006465[0]: INIT_GENERIC_CONDITON: unknown 0x07
[    8.057232] nouveau 0000:01:00.0: disp: outp 01:0006:0f84: link rate unsupported by sink

which happens every few seconds. Probably related.
Comment 5 Ilia Mirkin 2017-10-21 02:34:26 UTC
Created attachment 134961 [details]
decoded edids

Ran the EDIDs through the latest edid-decode.
Comment 6 Lukas Sabota 2017-10-21 02:37:17 UTC
Created attachment 134962 [details]
dmesg with debugging enabled

I booted again with nouveau and additional kernel parameters to provide more verbose dmesg output.
Comment 7 Lukas Sabota 2017-10-21 03:00:50 UTC
Created attachment 134963 [details]
Verbose dmesg output:  Booting with only LG Ultrawide

On this run, I only had the LG Ultrawide monitor physically connected to the machine.  After the bootloader and a few (5-10) early console output lines, I lost video signal to the monitor.
Comment 8 Lukas Sabota 2017-10-21 03:02:04 UTC
Created attachment 134964 [details]
Verbose dmesg output: Booting only with Samsung 4K

In this test, I booted with only the Samsung 4K monitor attached.  There was no flickering or artifacting in tty and no visible problems in sway.
Comment 9 caguduzexi 2018-01-29 14:02:54 UTC
I wont recommend using/keeping the GP106 (GTX 1060). It cant ever run with free software: https://www.theregister.co.uk/2015/04/15/nvidia_gtx_900_linux_driver_roadbloack/
https://www.phoronix.com/scan.php?page=news_item&px=Nouveau-XDC2017
https://www.phoronix.com/scan.php?page=news_item&px=Nouveau-XDC2016-NVIDIA

Sell this crappy GP102 card away and go away from nvidia. Nvidia died with the 780ti card. Its the last end-user card that can be used normaly. Everything else is in some countries even a legal problem. Because the manufacturer (nvidia) blocks the users from beeing able to boot the software they want on THEIR hardware - happyly illegal in some countries. Hopefully some layer would sue the heck out of nvidia so that they would have to release the private signing key or close their doors.
Blocking the freedom of the users on such way should not be accepted by anyone.
Comment 10 Martin Peres 2018-01-29 17:35:41 UTC
(In reply to caguduzexi from comment #9)
> I wont recommend using/keeping the GP106 (GTX 1060). It cant ever run with
> free software:
> https://www.theregister.co.uk/2015/04/15/
> nvidia_gtx_900_linux_driver_roadbloack/
> https://www.phoronix.com/scan.php?page=news_item&px=Nouveau-XDC2017
> https://www.phoronix.com/scan.php?page=news_item&px=Nouveau-XDC2016-NVIDIA
> 
> Sell this crappy GP102 card away and go away from nvidia. Nvidia died with
> the 780ti card. Its the last end-user card that can be used normaly.
> Everything else is in some countries even a legal problem. Because the
> manufacturer (nvidia) blocks the users from beeing able to boot the software
> they want on THEIR hardware - happyly illegal in some countries. Hopefully
> some layer would sue the heck out of nvidia so that they would have to
> release the private signing key or close their doors.
> Blocking the freedom of the users on such way should not be accepted by
> anyone.

User banned
Comment 11 Martin Peres 2019-12-04 09:32:12 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/driver/xf86-video-nouveau/issues/376.


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.