Bug 98091 - DisplayPort config working for Win7/64 but not Linux on same H/W
Summary: DisplayPort config working for Win7/64 but not Linux on same H/W
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: XOrg git
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Intel GFX Bugs mailing list
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-10-05 09:23 UTC by Clive
Modified: 2016-10-08 19:15 UTC (History)
1 user (show)

See Also:
i915 platform: SKL
i915 features: display/DP


Attachments

Description Clive 2016-10-05 09:23:37 UTC
First-time poster, inexperienced with bug reporting. Apologies in advance, etc. 

Running a dual-booting system based on an Intel 6700T Skylake on top of an Asus Z170i Pro Motherboard (16Gb RAM, 1Tb Samsung SSD). 

3 Monitors, all Dell U2415W (24", 1920x1200, 60Hz). 

Configure monitors using DisplayPort cables, daisy-chained with single output from PC to each monitor in turn (following Dell DP setup guide) and with Windows 7/64, this configuration works flawlessly, with excellent screen response and flicker-free stability.

Reboot machine into Mint Linux 17.3 (Rosa) 64-Bit and all 3 screens will remain dark. 

Current Mint setup detailed below... Entirely happy to provide more information, experiment, etc, if required. 

inxi -c5 -Fxxz
System: Host: SkyLake Kernel: 3.19.0-32-generic x86_64 (64 bit gcc: 4.8.2)
Desktop: Cinnamon 2.8.8 (Gtk 3.10.8~8+qiana) dm: mdm
Distro: Linux Mint 17.3 Rosa
Machine: Mobo: ASUSTeK model: Z170I PRO GAMING v: Rev X.0x
Bios: American Megatrends v: 0701 date: 11/16/2015
CPU: Quad core Intel Core i7-6700T (-HT-MCP-) cache: 8192 KB
flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 22464
clock speeds: min/max: 800/2801 MHz 1: 800 MHz 2: 800 MHz
3: 800 MHz 4: 800 MHz 5: 1900 MHz 6: 1100 MHz 7: 900 MHz
8: 1200 MHz
Graphics: Card: Intel Sky Lake Integrated Graphics
bus-ID: 00:02.0 chip-ID: 8086:1912
Display Server: X.Org 1.17.1 drivers: intel (unloaded: fbdev,vesa)
Resolution: 1920x1200@60.0hz, 1920x1200@60.0hz
GLX Renderer: Mesa DRI Intel Skylake DT GT2
GLX Version: 3.0 Mesa 10.5.9 Direct Rendering: Yes
Audio: Card Intel Sunrise Point-H HD Audio
driver: snd_hda_intel bus-ID: 00:1f.3 chip-ID: 8086:a170
Sound: Advanced Linux Sound Architecture v: k3.19.0-32-generic
Network: Card-1: Intel Ethernet Connection (2) I219-V
driver: e1000e v: 2.3.2-k bus-ID: 00:1f.6 chip-ID: 8086:15b8
IF: eth0 state: up speed: 1000 Mbps duplex: full mac: <filter>
Card-2: Qualcomm Atheros QCA6174 802.11ac Wireless Network Adapter
bus-ID: 04:00.0 chip-ID: 168c:003e
IF: N/A state: N/A mac: N/A
Drives: HDD Total Size: 1000.2GB (3.5% used)
ID-1: /dev/sda model: Samsung_SSD_850 size: 1000.2GB serial: {###############}
Partition: ID-1: / size: 16G used: 719M (5%) fs: ext4 dev: /dev/sda5
ID-2: /tmp size: 7.4G used: 18M (1%) fs: ext4 dev: /dev/sda10
ID-3: /usr size: 16G used: 5.8G (41%) fs: ext4 dev: /dev/sda7
ID-4: /var size: 16G used: 1.1G (8%) fs: ext4 dev: /dev/sda9
ID-5: /home size: 16G used: 4.9G (35%) fs: ext4 dev: /dev/sda8
ID-6: swap-1 size: 16.53GB used: 0.00GB (0%) fs: swap dev: /dev/sda6
RAID: System: supported: N/A
No RAID devices: /proc/mdstat, md_mod kernel module present
Unused Devices: none
Sensors: System Temperatures: cpu: 29.8C mobo: 27.8C
Fan Speeds (in rpm): cpu: N/A
Info: Processes: 236 Uptime: 2:14 Memory: 918.2/15932.9MB
Init: Upstart v: 1.12.1 runlevel: 2 default: 2 Gcc sys: 4.8.4
Client: Shell (bash 4.3.111 running in gnome-terminal) inxi: 2.2.28
Comment 1 Jani Nikula 2016-10-05 12:15:32 UTC
(In reply to Clive from comment #0)
> System: Host: SkyLake Kernel: 3.19.0-32-generic x86_64 (64 bit gcc: 4.8.2)

Sorry, that kernel version is *way* too old for Skylake, and you'll likely also need a much newer userspace for DP MST as well.
Comment 2 Clive 2016-10-05 12:21:25 UTC
Jani,

Firstly, thank you *so much* for such a rapid response. I am sorry for wasting your time with what may be a non-bug... 

As this is my "main" system, I will take the opportunity to refresh both Windows (to 10) and Mint (to 18.0, Sarah) to see if this addresses the concern. I will gladly report back as soon as I am able to report results. 

If it is not too much of an imposition, may I ask you if Mint 18.0 Kernel (4.4) will be appropriate to attempt DisplayPort daisy-chaining? 

Thanks in advance - and sorry for the silly questions...

Thanks

Clive
Comment 3 Jani Nikula 2016-10-05 12:32:54 UTC
(In reply to Clive from comment #2)
> If it is not too much of an imposition, may I ask you if Mint 18.0 Kernel
> (4.4) will be appropriate to attempt DisplayPort daisy-chaining? 

It might just work... however there are *plenty* of other important fixes for Skylake all the way to the latest 4.8 kernel released a couple of days ago.
Comment 4 Clive 2016-10-08 19:15:50 UTC
Jani, everyone... 

Please find below an (unsuccessful) status update. 

1. Swapped SSDs on the 6700T in order to be able to perform a clean and non-destructive installation of Mint 18.0 (Sarah) 64-bit, kernel 4.4. 

2. apt-get update; apt-get upgrade

3. Followed these simple and effective instructions to upgrade Sarah to the 4.8 kernel:-

https://mintguide.org/system/686-install-linux-kernel-4-8-stable-on-linux-mint.html

4. Attempted to implement 3-screen daisy-chaining, observing the following results... [ using Mint Sarah 18.0/kernel 4.8/Cinnamon desktop ]

4.1. At login, all 3 screens operate successful mirroring mode - output is clear, flicker-free, with no shadowing, artifacts or vsync issues.

4.2. After login, screen 0 remains active, screens 1 & 2 go dark.

4.3. Using the Cinnamon "Display" applet, observe that parameter for mirroring is activated and the detected screens shows just a single monitor. 

4.4. Deselect "Mirroring" and the detected screen graphic changes to show all 3 monitors [hardware potentially now recognised].

4.5 Select "Apply"; the Display applet quits, but screens 1 & 2 remain dark.

4.6 After approximately 10 seconds, the display Applet returns to the screen, once more in mirror mode. 

4.7 Various permutations attempted without success. 

4.8 Quick check of system configuration via "uname -r" and confirmation that 4.8 kernel is running...

5. For a final check, leave the cabling in daisy-chain configuration, swap back to the Windows 7/64 / Mint 17.3 dual-boot SSD and boot directly in to Windows. Observe triple-screen, operating flawlessly... 


Observations and conclusions...
There is a visible improvement in the functionality with the 4.8 kernel, with the display remaining. Trying to "push" a Cinnamon window across from display 0 to displays 1 or 2 results in the application window sliding off-screen, as though the OS "knows" that there is desktop real estate present, just that it is not rendering on screen... 

Offer
With the ready availability of a built 4.8 kernel system, I am happy to try config changes, patches, etc, if that would help. 

Thank you again for the kind advice - you certainly guided me in the right direction, even if the ultimate destination is still just out of reach...


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.