384.69 (and every other version post-364.19) Doesn't find laptop display

Testing with 381.22 on the same machine where I gave up and went back to 364 on previous versions (https://devtalk.nvidia.com/default/topic/1003432/linux/375-381-09-980m-hangs-system-if-external-monitor-is-connected/), the driver does not appear to find the internal laptop screen as a valid display device.

The machine will boot to a black screen, and I can ctrl-alt-f1 to a VTY and see the terminal session fine, I just can’t access the xorg sessions. I can even log into the black screen by typing my password blindly, but the desktop will of course be black after I hear it startup since the driver’s not outputing anything.

Using prime-select nvidia:

36.358] (**) NVIDIA(0): Option "IgnoreDisplayDevices" "CRT"
[    36.359] (**) NVIDIA(0): Enabling 2D acceleration
[    36.741] (--) NVIDIA(0): Valid display device(s) on GPU-0 at PCI:1:0:0
[    36.741] (--) NVIDIA(0):     DFP-0
[    36.741] (--) NVIDIA(0):     DFP-1
[    36.741] (--) NVIDIA(0):     DFP-2
[    36.741] (--) NVIDIA(0):     DFP-3
[    36.741] (--) NVIDIA(0):     DFP-4
[    36.758] (II) NVIDIA(0): NVIDIA GPU GeForce GTX 980M (GM204-B) at PCI:1:0:0 (GPU-0)
[    36.758] (--) NVIDIA(0): Memory: 8388608 kBytes
[    36.758] (--) NVIDIA(0): VideoBIOS: 84.04.85.00.1c
[    36.758] (II) NVIDIA(0): Detected PCI Express Link width: 16X
[    36.759] (--) NVIDIA(GPU-0): DFP-0: disconnected
[    36.759] (--) NVIDIA(GPU-0): DFP-0: Internal TMDS
[    36.759] (--) NVIDIA(GPU-0): DFP-0: 165.0 MHz maximum pixel clock
[    36.759] (--) NVIDIA(GPU-0): 
[    36.759] (--) NVIDIA(GPU-0): DFP-1: disconnected
[    36.759] (--) NVIDIA(GPU-0): DFP-1: Internal DisplayPort
[    36.759] (--) NVIDIA(GPU-0): DFP-1: 960.0 MHz maximum pixel clock
[    36.759] (--) NVIDIA(GPU-0): 
[    36.759] (--) NVIDIA(GPU-0): DFP-2: disconnected
[    36.759] (--) NVIDIA(GPU-0): DFP-2: Internal TMDS
[    36.759] (--) NVIDIA(GPU-0): DFP-2: 165.0 MHz maximum pixel clock
[    36.759] (--) NVIDIA(GPU-0): 
[    36.759] (--) NVIDIA(GPU-0): DFP-3: disconnected
[    36.759] (--) NVIDIA(GPU-0): DFP-3: Internal DisplayPort
[    36.759] (--) NVIDIA(GPU-0): DFP-3: 960.0 MHz maximum pixel clock
[    36.759] (--) NVIDIA(GPU-0): 
[    36.759] (--) NVIDIA(GPU-0): DFP-4: disconnected
[    36.759] (--) NVIDIA(GPU-0): DFP-4: Internal TMDS
[    36.759] (--) NVIDIA(GPU-0): DFP-4: 165.0 MHz maximum pixel clock
[    36.759] (--) NVIDIA(GPU-0): 
[    36.759] (==) NVIDIA(0): 
[    36.759] (==) NVIDIA(0): No modes were requested; the default mode "nvidia-auto-select"
[    36.759] (==) NVIDIA(0):     will be used as the requested mode.
[    36.759] (==) NVIDIA(0): 
[    36.759] (--) NVIDIA(0): No enabled display devices found; starting anyway because
[    36.759] (--) NVIDIA(0):     AllowEmptyInitialConfiguration is enabled

meanwhile, if I prime-select intel, it can find the internal display fine and I can log in and use the machine normally:

382.589] (II) LoadModule: "intel"
[   382.589] (II) Loading /usr/lib/xorg/modules/drivers/intel_drv.so
[   382.592] (II) Module intel: vendor="X.Org Foundation"
[   382.592]    compiled for 1.18.3, module version = 2.99.917
[   382.592]    Module class: X.Org Video Driver
[   382.592]    ABI class: X.Org Video Driver, version 20.0
[   382.592] (II) intel: Driver for Intel(R) Integrated Graphics Chipsets:
        i810, i810-dc100, i810e, i815, i830M, 845G, 854, 852GM/855GM, 865G,
        915G, E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM,
        Pineview G, 965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33,
        GM45, 4 Series, G45/G43, Q45/Q43, G41, B43
[   382.593] (II) intel: Driver for Intel(R) HD Graphics: 2000-6000
[   382.593] (II) intel: Driver for Intel(R) Iris(TM) Graphics: 5100, 6100
[   382.593] (II) intel: Driver for Intel(R) Iris(TM) Pro Graphics: 5200, 6200, P6300
[   382.594] (II) intel(0): Using Kernel Mode Setting driver: i915, version 1.6.0 20161121
[   382.594] (II) intel(0): SNA compiled: xserver-xorg-video-intel 2:2.99.917+git20160706-1ubuntu1 (Timo Aaltonen <tjaalton@debian.org>)
[   382.594] (II) intel(0): SNA compiled for use with valgrind
[   382.597] (--) intel(0): gen9 engineering sample
[   382.597] (--) intel(0): CPU: x86-64, sse2, sse3, ssse3, sse4.1, sse4.2, avx, avx2; using a maximum of 4 threads
[   382.597] (II) intel(0): Creating default Display subsection in Screen section
        "Default Screen Section" for depth/fbbpp 24/32
[   382.597] (==) intel(0): Depth 24, (--) framebuffer bpp 32
[   382.597] (==) intel(0): RGB weight 888
[   382.597] (==) intel(0): Default visual is TrueColor
[   382.597] (**) intel(0): Option "Backlight" "intel_backlight"
[   382.597] (II) intel(0): Output eDP1 has no monitor section
[   382.644] (**) intel(0): Found backlight control interface intel_backlight (type 'raw') for output eDP1
[   382.644] (II) intel(0): Enabled output eDP1

I don’t have the xserver-xorg-nouveau drivers installed at all, so I’m not sure what would block the 381 drivers from seeing the display device. I’ve tried with prime as well as manually setting up xorg and setting the nogpumanager kernel parameter.

Backleveling to 364 (and switching bios to discrete mode), the panel can be seen again.

[   247.369] (--) NVIDIA(GPU-0): SDC (DFP-1): connected
[   247.369] (--) NVIDIA(GPU-0): SDC (DFP-1): Internal DisplayPort
[   247.369] (--) NVIDIA(GPU-0): SDC (DFP-1): 960.0 MHz maximum pixel clock

Here’s the info from 364 with both the internal and external screen connected (381 says there are 0 display devices). The internal screen is the 4k g-sync monitor:

19:14:34 evil@sager ~» sudo nvidia-xconfig --query-gpu-info 
Number of GPUs: 1

GPU #0:
  Name      : GeForce GTX 980M
  UUID      : GPU-e2f980da-ea7e-4335-6ae3-41ae731aed6d
  PCI BusID : PCI:1:0:0

  Number of Display Devices: 2

  Display Device 0 (TV-0):
      EDID Name             : Ancor Communications Inc ASUS MX299
      Minimum HorizSync     : 26.000 kHz
      Maximum HorizSync     : 83.000 kHz
      Minimum VertRefresh   : 24 Hz
      Maximum VertRefresh   : 76 Hz
      Maximum PixelClock    : 240.000 MHz
      Maximum Width         : 2560 pixels
      Maximum Height        : 1080 pixels
      Preferred Width       : 2560 pixels
      Preferred Height      : 1080 pixels
      Preferred VertRefresh : 60 Hz
      Physical Width        : 670 mm
      Physical Height       : 280 mm

  Display Device 1 (TV-1):
      EDID Name             : SDC
      Minimum HorizSync     : 133.193 kHz
      Maximum HorizSync     : 133.193 kHz
      Minimum VertRefresh   : 59 Hz
      Maximum VertRefresh   : 59 Hz
      Maximum PixelClock    : 526.910 MHz
      Maximum Width         : 3840 pixels
      Maximum Height        : 2160 pixels
      Preferred Width       : 3840 pixels
      Preferred Height      : 2160 pixels
      Preferred VertRefresh : 60 Hz
      Physical Width        : 340 mm
      Physical Height       : 190 mm

Testing today with the newer 384.47 driver shows that 384.47 does not see the display in msyhbrid mode either. Forcing the state to connected in the xorg.conf does not seem to help (shows connected, can’t read EDID).

With the video configured for discrete mode in the BIOS (the same way I’m using 364.19 right now), 384.47 does find the display, but still isn’t usable. Everything in Xorg.0.log looks good but the laptop boots to a black screen (cannot alt-tab to a terminal), and the following can be found in the syslog:

Jul 13 13:38:37 sager kernel: [   45.273352] nvidia-modeset: WARNING: GPU:0: Lost display notification (0:0x00000000); continuing.
Jul 13 13:38:40 sager kernel: [   48.025945] nvidia-modeset: ERROR: GPU:0: Idling display engine timed out: 0x0000957d:0:0

Here’s some more info about the display that I got while running the 364.19 driver with debug mode enabled:

[    56.967] (--) NVIDIA(GPU-0): SDC (DFP-1): connected
[    56.968] (--) NVIDIA(GPU-0): SDC (DFP-1): Internal DisplayPort
[    56.968] (--) NVIDIA(GPU-0): SDC (DFP-1) Name Aliases:
[    56.968] (--) NVIDIA(GPU-0):   DFP
[    56.968] (--) NVIDIA(GPU-0):   DFP-1
[    56.968] (--) NVIDIA(GPU-0):   DPY-1
[    56.968] (--) NVIDIA(GPU-0):   DP-0
[    56.968] (--) NVIDIA(GPU-0):   DPY-EDID-0b18b22d-cfc6-0444-01f5-177cb2304044
[    56.968] (--) NVIDIA(GPU-0):   DP-0
[    56.968] (--) NVIDIA(GPU-0): SDC (DFP-1): 960.0 MHz maximum pixel clock
[    56.968] (--) NVIDIA(GPU-0): 
[    56.968] (--) NVIDIA(GPU-0): --- EDID for SDC (DP-0) ---
[    56.968] (--) NVIDIA(GPU-0): EDID Version                 : 1.3
[    56.968] (--) NVIDIA(GPU-0): Manufacturer                 : SDC
[    56.968] (--) NVIDIA(GPU-0): Monitor Name                 : SDC
[    56.968] (--) NVIDIA(GPU-0): Product ID                   : 0x434b
[    56.968] (--) NVIDIA(GPU-0): 32-bit Serial Number         : 0x00000000
[    56.968] (--) NVIDIA(GPU-0): Serial Number String         : 
[    56.968] (--) NVIDIA(GPU-0): Manufacture Date             : 2014, week 0
[    56.968] (--) NVIDIA(GPU-0): DPMS Capabilities            : Standby Suspend Active Off
[    56.968] (--) NVIDIA(GPU-0): Input Type                   : Digital
[    56.968] (--) NVIDIA(GPU-0): Prefer first detailed timing : Yes
[    56.968] (--) NVIDIA(GPU-0): Supports GTF                 : No
[    56.968] (--) NVIDIA(GPU-0): Maximum Image Size           : 340 mm x 190 mm
[    56.968] (--) NVIDIA(GPU-0): Valid HSync Range            : 133.2 kHz - 133.2 kHz
[    56.968] (--) NVIDIA(GPU-0): Valid VRefresh Range         : 60.0 Hz - 60.0 Hz
[    56.968] (--) NVIDIA(GPU-0): EDID maximum pixel clock     : 526.9 MHz
[    56.968] (--) NVIDIA(GPU-0): 
[    56.968] (--) NVIDIA(GPU-0): Detailed Timings:
[    56.968] (--) NVIDIA(GPU-0):   3840 x 2160 @ 60 Hz
[    56.968] (--) NVIDIA(GPU-0):     Pixel Clock      : 526.91 MHz
[    56.968] (--) NVIDIA(GPU-0):     HRes, HSyncStart : 3840, 3888
[    56.968] (--) NVIDIA(GPU-0):     HSyncEnd, HTotal : 3920, 3956
[    56.968] (--) NVIDIA(GPU-0):     VRes, VSyncStart : 2160, 2162
[    56.968] (--) NVIDIA(GPU-0):     VSyncEnd, VTotal : 2167, 2220
[    56.968] (--) NVIDIA(GPU-0):     H/V Polarity     : -/-
[    56.968] (--) NVIDIA(GPU-0):     Image Size       : 344 mm x 194 mm
[    56.968] (--) NVIDIA(GPU-0): 
[    56.968] (--) NVIDIA(GPU-0): --- End of EDID for SDC (DP-0) ---
[    56.968] (--) NVIDIA(GPU-0): 
[    56.968] (--) NVIDIA(GPU-0): 
[    56.968] (--) NVIDIA(GPU-0): Raw EDID bytes:
[    56.968] (--) NVIDIA(GPU-0): 
[    56.968] (--) NVIDIA(GPU-0):   00 ff ff ff ff ff ff 00  4c 83 4b 43 00 00 00 00
[    56.968] (--) NVIDIA(GPU-0):   00 18 01 03 95 22 13 78  ea ce 85 9e 5b 4c 94 26
[    56.968] (--) NVIDIA(GPU-0):   17 50 54 00 00 00 01 01  01 01 01 01 01 01 01 01
[    56.968] (--) NVIDIA(GPU-0):   01 01 01 01 01 01 d3 cd  00 74 f0 70 3c 80 30 20
[    56.968] (--) NVIDIA(GPU-0):   25 00 58 c2 10 00 00 19  00 00 00 0f 00 00 00 00
[    56.968] (--) NVIDIA(GPU-0):   00 00 00 00 00 37 82 05  46 00 00 00 00 fe 00 53
[    56.968] (--) NVIDIA(GPU-0):   41 4d 53 55 4e 47 0a 20  4c 83 46 4c 00 00 00 fe
[    56.968] (--) NVIDIA(GPU-0):   00 31 35 36 46 4c 30 32  2d 31 30 31 0a 20 00 2b

As usual, back-leveling everything to 364.19 gets it working again.

same here i can confirm this issue.
this has been an issue for a long time now.
Clevo p651re notebook with gtx-970m

Sounds like we have similar hardware. My system is a Sager NP8658S (same as a Clevo P650RG). Does your laptop also use the SDC 3840x2160 display panel?

No i’v got a fullHD display panel. but i think they are wired up the same way.

Tested with 384.59 at lunch. Same issues. :(

Jul 26 12:23:29 sager kernel: [    1.501000] nvidia-nvlink: Nvlink Core is being initialized, major device number 247
Jul 26 12:23:29 sager kernel: [    1.505586] NVRM: loading NVIDIA UNIX x86_64 Kernel Module  384.59  Wed Jul 19 23:53:34 PDT 2017 (using threaded interrupts)
Jul 26 12:23:29 sager kernel: [    1.509341] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms  384.59  Wed Jul 19 23:46:42 PDT 2017
Jul 26 12:23:29 sager kernel: [    1.512613] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
Jul 26 12:23:29 sager kernel: [    2.885075] nvidia-uvm: Loaded the UVM driver in 8 mode, major device number 246
Jul 26 12:23:29 sager kernel: [   29.188257] input: HDA NVidia HDMI/DP,pcm=3 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input20
Jul 26 12:23:29 sager kernel: [   29.188316] input: HDA NVidia HDMI/DP,pcm=7 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input21
Jul 26 12:23:29 sager kernel: [   29.188433] input: HDA NVidia HDMI/DP,pcm=8 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input22
Jul 26 12:23:29 sager kernel: [   29.188522] input: HDA NVidia HDMI/DP,pcm=9 as /devices/pci0000:00/0000:00:01.0/0000:01:00.1/sound/card1/input23
Jul 26 12:23:30 sager nvidia-persistenced: Verbose syslog connection opened
Jul 26 12:23:30 sager nvidia-persistenced: Now running with user ID 123 and group ID 131
Jul 26 12:23:30 sager nvidia-persistenced: Started (6952)
Jul 26 12:23:30 sager systemd[1]: Starting NVIDIA Persistence Daemon...
Jul 26 12:23:30 sager systemd[1]: Started NVIDIA Persistence Daemon.
Jul 26 12:23:30 sager nvidia-persistenced: device 0000:01:00.0 - registered
Jul 26 12:23:30 sager nvidia-persistenced: Local RPC service initialized
Jul 26 12:23:30 sager kernel: [   77.373601] nvidia-modeset: Allocated GPU:0 (GPU-e2f980da-ea7e-4335-6ae3-41ae731aed6d) @ PCI:0000:01:00.0
...
Jul 26 12:23:37 sager kernel: [   84.442354] nvidia-modeset: WARNING: GPU:0: Lost display notification (0:0x00000000); continuing.
Jul 26 12:23:40 sager kernel: [   87.194444] nvidia-modeset: ERROR: GPU:0: Idling display engine timed out: 0x0000957d:0:0

Xorg.0.log (154 KB)

Same issue with 384.69. Machine BIOS is set to just use discrete chipset, grub set for nogpumanager and using my already working (with 364) nvidia-settings-created xorg.conf:

Sep  7 08:18:16 sager kernel: [    1.616633] nvidia-nvlink: Nvlink Core is being initialized, major device number 244
Sep  7 08:18:16 sager kernel: [    1.617743] nvidia 0000:01:00.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=none:owns=io+mem
Sep  7 08:18:16 sager kernel: [    1.621371] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms  384.69  Wed Aug 16 19:39:44 PDT 2017
Sep  7 08:18:16 sager kernel: [    1.622945] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver
Sep  7 08:18:16 sager kernel: [    1.624335] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0
Sep  7 08:18:16 sager kernel: [    3.101867] nvidia-uvm: Loaded the UVM driver in 8 mode, major device number 243
Sep  7 08:18:17 sager nvidia-persistenced: Verbose syslog connection opened
Sep  7 08:18:17 sager nvidia-persistenced: Now running with user ID 123 and group ID 131
Sep  7 08:18:17 sager nvidia-persistenced: Started (6306)
Sep  7 08:18:17 sager nvidia-persistenced: device 0000:01:00.0 - registered
Sep  7 08:18:17 sager nvidia-persistenced: Local RPC service initialized
Sep  7 08:18:17 sager kernel: [   35.674414] nvidia-modeset: Allocated GPU:0 (GPU-e2f980da-ea7e-4335-6ae3-41ae731aed6d) @ PCI:0000:01:00.0
Sep  7 08:18:24 sager kernel: [   42.865740] nvidia-modeset: WARNING: GPU:0: Lost display notification (0:0x00000000); continuing.
Sep  7 08:18:27 sager kernel: [   45.713922] nvidia-modeset: ERROR: GPU:0: Idling display engine timed out: 0x0000957d:0:0

Tested with both 4.4 and 4.12 kernels and various ACPI kernel settings, same result. Above was using:

GRUB_CMDLINE_LINUX_DEFAULT="nogpumanager rdblacklist=nouveau nouveau.modeset=0 pcie_aspm=force acpi_osi=! acpi_osi=\"Windows 2009\""

Nouveau drivers are not installed, but still blacklisted from back when it was.

Reverting to 364.19 works fine, as usual.