1 |
> -----Original Message----- |
2 |
> From: J. Roeleveld <joost@××××××××.org> |
3 |
> Sent: Monday, June 15, 2020 16:20 |
4 |
> To: gentoo-user@l.g.o |
5 |
> Subject: Re: [gentoo-user] clone root from HDD to SSD causes no video with |
6 |
> NVIDIA driver |
7 |
> |
8 |
> On Monday, June 15, 2020 9:56:39 AM CEST Raffaele BELARDI wrote: |
9 |
> > * From: Dale <rdalek1967@×××××.com> |
10 |
> > * Sent: Wednesday, June 10, 2020 08:02 |
11 |
> > * To: gentoo-user@l.g.o<mailto:gentoo- |
12 |
> user@l.g.o> |
13 |
> > * Subject: Re: [gentoo-user] clone root from HDD to SSD causes no video |
14 |
> > with NVIDIA driver |
15 |
> > |
16 |
> > * Raffaele BELARDI wrote: |
17 |
> > * nomodeset did not change anything, but adding EFI_FB to the kernel |
18 |
> > finally got me a functional console. |
19 |
> * But if I startx from there I am |
20 |
> > back again to the same point, no X, no console switching with CTR-ALT-Fn, * |
21 |
> > no crash in syslog, I have to SSH to get to a working shell. I'm not |
22 |
> > getting anywhere, I think I'll better install from stage3. |
23 |
> > |
24 |
> > * Odds are, if you start from stage3, you will get the same problem |
25 |
> > again unless you do something different. |
26 |
> * When I first stated using |
27 |
> > Gentoo, I didn't realize that one can restart a install pretty much |
28 |
> > anywhere in the install. * Starting over doesn't get you anything |
29 |
> > different if you repeat the same steps. |
30 |
> > Just to update: I tried all the hints received here with no luck. |
31 |
> > Since others on this list managed to get uefifb working with the |
32 |
> > NVIDIA driver I believe the problem could be my mobo/UEFI FW/GPU |
33 |
> > combination. I found some rather old posts ([1], [2]) supporting this |
34 |
> > hypothesis. For the moment I switched to nouveau. |
35 |
> |
36 |
> > Thanks again to all, |
37 |
> > |
38 |
> > raffaele |
39 |
> > |
40 |
> > [1] |
41 |
> > https://forums.developer.nvidia.com/t/uefi-nvidia-vga-console-complain |
42 |
> > ts/37 |
43 |
> > 690 |
44 |
> [2] |
45 |
> > https://forums.developer.nvidia.com/t/nvidia-devs-any-eta-on-fbdev-con |
46 |
> > sole-> |
47 |
> mode-setting-implementation/47043 |
48 |
> |
49 |
> Personally, I would not expect this to be related to mainboard firmware/bios |
50 |
> issues as I have not had any issues with efifb and nvidia-drivers on several |
51 |
> systems. |
52 |
|
53 |
I still have some hopes, I intend to give NVIDIA another try later. |
54 |
> |
55 |
> What is your kernel-commandline? |
56 |
> |
57 |
> Mine is really simple: |
58 |
> $ cat /proc/cmdline |
59 |
> root=/dev/nvme0n1p3 |
60 |
|
61 |
root=/dev/sdb5 ro quiet raid=noautodetect |
62 |
|
63 |
> |
64 |
> I get the following in my dmesg for "efifb": |
65 |
> ==== |
66 |
> [ 8.717047] efifb: probing for efifb |
67 |
> [ 8.717061] efifb: framebuffer at 0xd1000000, using 3072k, total 3072k |
68 |
> [ 8.717062] efifb: mode is 1024x768x32, linelength=4096, pages=1 |
69 |
> [ 8.717064] efifb: scrolling: redraw |
70 |
> [ 8.717065] efifb: Truecolor: size=8:8:8:8, shift=24:16:8:0 |
71 |
> [ 8.719748] fb0: EFI VGA frame buffer device |
72 |
> ==== |
73 |
|
74 |
Same here: |
75 |
[ 0.705019] efifb: probing for efifb |
76 |
[ 0.705029] efifb: framebuffer at 0xc0000000, using 3072k, total 3072k |
77 |
[ 0.705030] efifb: mode is 1024x768x32, linelength=4096, pages=1 |
78 |
[ 0.705030] efifb: scrolling: redraw |
79 |
[ 0.705031] efifb: Truecolor: size=8:8:8:8, shift=24:16:8:0 |
80 |
[ 0.705122] Console: switching to colour frame buffer device 128x48 |
81 |
[ 0.706608] fb0: EFI VGA frame buffer device |
82 |
|
83 |
> Which is nowhere near the real resolution my screen can handle, but for |
84 |
> emergencies, this is definitely sufficient. |
85 |
> |
86 |
> For completeness, these are the entries for nvidia: |
87 |
> ==== |
88 |
> $ dmesg | grep -i nvidia |
89 |
> [ 11.222893] nvidia: loading out-of-tree module taints kernel. |
90 |
> [ 11.222908] nvidia: module license 'NVIDIA' taints kernel. |
91 |
> [ 11.241368] nvidia-nvlink: Nvlink Core is being initialized, major device |
92 |
> number 240 |
93 |
> [ 11.241687] nvidia 0000:01:00.0: vgaarb: changed VGA decodes: |
94 |
> olddecodes=io+mem,decodes=none:owns=io+mem |
95 |
> [ 11.283229] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 440.82 |
96 |
> Wed Apr |
97 |
> 1 20:04:33 UTC 2020 |
98 |
> [ 11.287732] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for |
99 |
> UNIX platforms 440.82 Wed Apr 1 19:41:29 UTC 2020 |
100 |
> [ 11.289189] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver |
101 |
> [ 11.289191] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on |
102 |
> minor 0 |
103 |
> [ 11.861737] input: HDA NVidia HDMI/DP,pcm=3 as /devices/ |
104 |
> pci0000:00/0000:00:03.0/0000:01:00.1/sound/card1/input28 |
105 |
> [ 11.862152] input: HDA NVidia HDMI/DP,pcm=7 as /devices/ |
106 |
> pci0000:00/0000:00:03.0/0000:01:00.1/sound/card1/input29 |
107 |
> [ 11.979061] input: HDA NVidia HDMI/DP,pcm=8 as /devices/ |
108 |
> pci0000:00/0000:00:03.0/0000:01:00.1/sound/card1/input30 |
109 |
> [ 11.979134] input: HDA NVidia HDMI/DP,pcm=9 as /devices/ |
110 |
> pci0000:00/0000:00:03.0/0000:01:00.1/sound/card1/input31 |
111 |
> ==== |
112 |
|
113 |
I don't have these at the moment because I switched to noveau to stabilize the system, later I'll try again with NVIDIA. |
114 |
|
115 |
> On a side-note, anyone know how to prevent these sound-devices from |
116 |
> appearing? |
117 |
> I never use these on this system. |
118 |
> |
119 |
> -- |
120 |
> Joost |
121 |
> |
122 |
> |
123 |
> |