NetBSD-Users archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: OT - I have a NUC



Hello to everyone.


I'm a NetBSD user. I've installed NetBSD 9.2. My monitor turns off every time with multiple graphic cards connected,but only one is used and chosen as primary on the BIOS,this :


000:02:0: Intel UHD Graphics 630 (VGA display, revision 0x02)

My motherboard is the AORUS pro and my CPU is I9-9900k.


On my PC I have 3 graphic cards :


000:02:0: Intel UHD Graphics 630 (VGA display, revision 0x02)
001:00:0: NVIDIA product 1e04 (VGA display, revision 0xa1) : (NVIDIA RTX 2080ti)
002:00:0: NVIDIA GeForce GTX 1060 3GB (VGA display, revision 0xa1)


NetBSD always chooses the graphic card n. 2 even if it is not chosen as primary on the BIOS. Even if I have recompiled the kernel with DRM_MAX_RESOLUTION uncommented in the kernel config.

Below u can see how I have configured everything :

Inside the folder /etc/X11/xorg.conf.d I've created a file called "xorg-modesetting.conf" with this content inside :

Section "Device"
Identifier "Card0"
Driver "modesetting"
BusID "pci:0:2:0"
EndSection

boot.cfg ->


menu=Boot normally:rndseed /var/db/entropy-file;gop 4;boot menu=Boot single user:rndseed /var/db/entropy-file;boot -s menu=Boot DEVELOPMENT kernel:rndseed /var/db/entropy-file;gop 4;boot netbsd -v -x menu=Drop to boot prompt:prompt #userconf=disable i915drmkms* userconf=disable nouveau* default=1 timeout=5 clear=1

With BusID "pci:0:2:0" added, things are like this with the Intel DRM driver:


[ 1.000000] [ 377.000000] nouveau* disabled ... [ 1.015531] i915drmkms0 at pci0 dev 2 function 0: Intel UHD Graphics 630 (rev. 0x02) ... [ 10.221872] kern.module.path=/stand/amd64/9.99.82/modules [ 10.231872] kern info: [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). [ 10.231872] kern info: [drm] Driver supports precise vblank timestamp query. [ 10.231872] kern info: [drm] Failed to find VBIOS tables (VBT) [ 10.231872] allocated pic msi9 type edge pin 0 level 6 to cpu1 slot 2 idt entry 98 [ 10.231872] i915drmkms0: interrupting at msi9 vec 0 (i915drmkms0) [ 10.231872] kern info: [drm] Finished loading DMC firmware i915/kbl_dmc_ver1_04.bin (v1.4) [ 10.501871] kern info: [drm] failed to retrieve link info, disabling eDP [ 10.521872] kern info: [drm] Initialized i915 1.6.0 20200114 for i915drmkms0 on minor 0 [ 12.071867] i915drmkms0: info: [drm] Cannot find any crtc or sizes [ 12.081866] kern info: [drm] DRM_I915_DEBUG enabled [ 12.081866] kern info: [drm] DRM_I915_DEBUG_GEM enabled [ 14.231859] i915drmkms0: info: [drm] Cannot find any crtc or sizes ...


No intelfb0 framebuffer is created, and so we have in the Xorg.log file:


[   320.124] (EE) open /dev/dri/card0: No such file or directory
[   320.124] (WW) Falling back to old probe method for modesetting
[   320.124] (EE) open /dev/dri/card0: No such file or directory
[   320.124] (WW) VGA arbiter: cannot open kernel arbiter, no multi-card support
[   320.124] (EE) Screen 0 deleted because of no matching config section.
[   320.125] (II) UnloadModule: "modesetting"
[   320.125] (EE) Device(s) detected, but none match those in the config file.


X is still loading the nouveau video driver for some reason.


If I choose to boot NetBSD with the Nvidia card using this xorg.conf file :


Section "Device" Identifier "Card0" Driver "modesetting" BusID "pci:1:0:0" EndSection

The nouveau DRM driver didn't configure either on my Nvidia cards, so only genfb is running on all 3:


[ 1.000000] [ 373.000000] i915drmkms* disabled ... [ 1.012984] genfb0 at pci1 dev 0 function 0: NVIDIA product 1e04 (rev. 0xa1) [ 1.012984] genfb0: framebuffer at 0x40000000, size 1920x1080, depth 32, stride 7680 [ 1.012984] genfb0: shadow framebuffer enabled, size 8100 KB [ 1.012984] wsdisplay0 at genfb0 kbdmux 1: console (default, vt100 emulation) [ 1.012984] wsmux1: connecting to wsdisplay0 [ 1.012984] drm at genfb0 not configured ... [ 1.012984] genfb1 at pci2 dev 0 function 0: NVIDIA GeForce GTX 1060 3GB (rev. 0xa1) ... [ 1.012984] genfb2 at pci0 dev 2 function 0: Intel UHD Graphics 630 (rev. 0x02)


And, in Xorg.log:


[    53.577] (EE) open /dev/dri/card0: Operation not supported by device
[    53.577] (WW) Falling back to old probe method for modesetting
[    53.577] (EE) open /dev/dri/card0: Operation not supported by device
[    53.577] (EE) No devices detected.
The screen doesn't turn off because its running on genfb (no Nvidia DRM drivers have taken over!); and X doesnt start because the modesetting driver needs DRM kernel drivers. It won't run on top of genfb.
[     9.576848] kern info: [drm] failed to retrieve link info, disabling eDP
[    11.136843] i915drmkms0: info: [drm] Cannot find any crtc or sizes
[    13.286834] i915drmkms0: info: [drm] Cannot find any crtc or sizes

This is why the screen turns off. The DRM driver can't find any monitors or the EDID info. returned by the monitor is wrong or not read properly. This is the situation:

Nouveau disabled; i915drmkms enabled, but not finding any monitors.

Xorg runs, modesetting driver loaded (nouveau_drv also loaded, don't know why), but can't run because no monitors found. Nouveau_drv won't run because the nouveau DRM driver is disabled.

I think that you will find the log below interesting. It has been taken after booting the kernel with -v -x. I've pressed Power off; then I've interrupted the bootloader; at the prompt I typed boot -v -x.


https://pastebin.ubuntu.com/p/rBwpMMjtZ6/


this is the relevant part : [ 103.410755] QUAGGA drm:netbsd:intel_dp_aux_xfer+0x9fa ELAND dp_aux_ch timeout status 0x7d4003ff


If u want to have more informations,there is a long troubleshooting that can be reached at the link below :


https://www.unitedbsd.com/d/582-trying-to-install-the-driver-for-my-main-gputhe-intel-uhd-graphics-630i915/66


Il giorno lun 11 ott 2021 alle ore 21:53 Thomas D. Dean <tomdean%wavecable.com@localhost> ha scritto:
On 10/11/21 10:10 AM, Rhialto wrote:
> On Sun 10 Oct 2021 at 17:07:03 -0400, Bob Bernstein wrote:
>> On Sun, 10 Oct 2021, Rhialto wrote:
>>
>>>> ahcisata0 port 0: PHY offline
>>
>>> This seems to be the point where things start failing.
>>
>> Am I being told that the two tiny little connectors from the SSD's "carrier"
>> (for lack of the correct term) that plug into what I suppose might be called
>> the thing's mainboard need to be reseated?
>
> Definitely check the cabling. I don't have an example handy of what it
> looks like when a device is connected, but powered off. But so far it
> looks like the port thinks that no cable + device is connected. This may
> still be a software problem. It could also be that the device is
> connected but not powered properly. For that it usually has a separate
> cable, with a connector similar to the SATA cable, but a different
> width. Sometimes the two connectors may be combined in one big one, I
> suppose, but in that case I expect to cables connected to such a
> connector, because one goes to the motherboard and the other to the
> power supply
On my NUC, there is a connector between the SATA connector and the SATA
power connector.

Tom Dean


--
Mario.


Home | Main Index | Thread Index | Old Index