Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220813230324.e038b10bf7f6f29b010382f9@telenet.be> Date & Time: Sat, 13 Aug 2022 23:03:24 +0200 Martin /Nightowl/ Byttebier <nightowl@telenet.be> has written:

Op Sun, 14 Aug 2022 05:50:17 +0900 Masaru Nomiya <nomiya@galaxy.dti.ne.jp> schreef:

[...] MN> > Please show your kernel parameters.

cat /proc/cmdline BOOT_IMAGE=/boot/vmlinuz-5.19.0-1-default root=UUID=88da3879-fdf1-4cd5-a691-9346a9411953 pcie_aspm=off rd.driver.blacklist=nouveau mitigations=auto

How about adding the following parameter? nvidia-drm.modeset=1 Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "Tim Cook, the C.E.O. of Apple, said earlier this year that he would not let his nephew join social networks. Bill Gates banned cellphone until his children were teenagers, and Melinda Gates wrote that she wished they had waited even longer. Steve Jobs would not let his young children near iPads." -- The New York Times --

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220814102935.b086dbf1bfe2c4f86fb0df28@telenet.be> Date & Time: Sun, 14 Aug 2022 10:29:35 +0200 Martin /Nightowl/ Byttebier <nightowl@telenet.be> has written:

Op Sun, 14 Aug 2022 06:13:39 +0900 Masaru Nomiya <nomiya@galaxy.dti.ne.jp> schreef:

...

Hello,

In the Message;

Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220813230324.e038b10bf7f6f29b010382f9@telenet.be> Date & Time: Sat, 13 Aug 2022 23:03:24 +0200

Martin /Nightowl/ Byttebier <nightowl@telenet.be> has written:

...

Op Sun, 14 Aug 2022 05:50:17 +0900 Masaru Nomiya <nomiya@galaxy.dti.ne.jp> schreef:

[...] MN> > Please show your kernel parameters.

...

cat /proc/cmdline BOOT_IMAGE=/boot/vmlinuz-5.19.0-1-default root=UUID=88da3879-fdf1-4cd5-a691-9346a9411953 pcie_aspm=off rd.driver.blacklist=nouveau mitigations=auto

How about adding the following parameter?

nvidia-drm.modeset=1

With nvidia-drm.modeset=1 set vt 7 error is gone. But the plymouth booting screen still halts after I915 0000:00:02.0: enabling device (0006 -> 0007)

It's hard to judge from the information you provided, is there the following kernel parameter; nomodeset Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "She continues to deeply divide opinion, nearly a decade after her death in 2013, due to her policies of privatisation, breaking the power of trade unions, and selling off public housing." -- World Is One News --

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220814120205.0ac47ee4283b2f54807ce61f@telenet.be> Date & Time: Sun, 14 Aug 2022 12:02:05 +0200 Martin /Nightowl/ Byttebier <nightowl@telenet.be> has written:

Op Sun, 14 Aug 2022 18:05:11 +0900 Masaru Nomiya <nomiya@galaxy.dti.ne.jp> schreef:

[...]

...

It's hard to judge from the information you provided, is there the following kernel parameter;

nomodeset

With nomodeset the Plymouth boot screen works like before.

Good.

I wonder why I need to set this?

The nomodeset parameter instructs the kernel to not load video drivers and use BIOS modes instead, until X is loaded.

Prior to kernel 15.18.9-2 there was no problem at all.

This in itself is a mystery, and I have no idea why、at all. Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ Think. -- The IBM slogan --

Op Sun, 14 Aug 2022 07:18:28 -0400 Felix Miata <mrmazda@earthlink.net> schreef:

Please boot without using nomodeset, and following Ctrl-Alt-F3 and logging in on resulting vtty3, run the following:

cat /var/log/Xorg.0.log | susepaste

and report here the resulting URI. While logged into X after booting using nomodeset, please run the following:

CTRL+ALT+F3 doesn't work. I get No DP Signal from your device and the monitor goes into standby modus. Anyway this is the result https://paste.opensuse.org/96511644

inxi -Gaz

Graphics: Device-1: Intel RocketLake-S GT1 [UHD Graphics 750] vendor: Dell driver: N/A alternate: i915 arch: Gen12.1 process: Intel 10nm built: 2020-21 bus-ID: 00:02.0 chip-ID: 8086:4c8a class-ID: 0300 Device-2: NVIDIA GA104 [GeForce RTX 3060 Ti] vendor: Dell driver: nvidia v: 515.65.01 alternate: nouveau,nvidia_drm non-free: 515.xx+ status: current (as of 2022-06) arch: Ampere process: TSMC n7 (7nm) built: 2020-22 pcie: gen: 1 speed: 2.5 GT/s lanes: 16 link-max: gen: 4 speed: 16 GT/s bus-ID: 02:00.0 chip-ID: 10de:2486 class-ID: 0300 Display: x11 server: X.Org v: 21.1.4 with: Xwayland v: 22.1.3 compositor: kwin_x11 driver: X: loaded: nvidia gpu: nvidia display-ID: :0 screens: 1 Screen-1: 0 s-res: 2560x1440 s-dpi: 122 s-size: 533x302mm (20.98x11.89") s-diag: 613mm (24.12") Monitor-1: DP-2 res: 2560x1440 hz: 60 dpi: 123 size: 527x296mm (20.75x11.65") diag: 604mm (23.8") modes: N/A OpenGL: renderer: NVIDIA GeForce RTX 3060 Ti/PCIe/SSE2 v: 4.6.0 NVIDIA 515.65.01 direct render: Yes

Instead of nomodeset it is probable that nosimplefb=1 is needed. <https://bugzilla.opensuse.org/show_bug.cgi?id=1201392>

nosimplefb=1 doesn't work for me. Thanks, Martin -- Atari FTP-site: ftp://kurobox.serveftp.net:3021 Running openSUSE Tumbleweed / KDE 5.25.4

Op Sun, 14 Aug 2022 17:36:00 +0200 Stephan Hemeier <Sauerlandlinux@gmx.de> schreef:

From your Xorg.0.log:

...

[ 52.828] (==) Using config file: "/etc/X11/xorg.conf"

Do you need that file? Can you rename it and restart?

That file isn't needed but it makes no difference. I need to set nomodeset with or without xorg.conf. TTFN, Martin -- Atari FTP-site: ftp://kurobox.serveftp.net:3021 Running openSUSE Tumbleweed / KDE 5.25.4

Op Sun, 14 Aug 2022 14:16:40 -0400 Felix Miata <mrmazda@earthlink.net> schreef:

Martin /Nightowl/ Byttebier composed on 2022-08-14 18:05 (UTC+0200):

...

Op Sun, 14 Aug 2022 17:36:00 +0200 Stephan Hemeier schreef:

...

...

From your Xorg.0.log:

...

[ 52.828] (==) Using config file: "/etc/X11/xorg.conf"

...

...

Do you need that file? Can you rename it and restart?

...

That file isn't needed but it makes no difference.

...

I need to set nomodeset with or without xorg.conf.

Is suse-prime installed? Have you ever tried to use it? Dual graphics with NVidia users normally need it, but I doubt it can work right with nomodeset blocking KMS.

Suse-prime is installed but I see no reason why I should use it. The monitor is directly hooked to a displayport of the NVIDIA graphic card. The dell XPS 8940 desktop has 3x DisplayPort 1.4a, 1x HDMI 2.1 ports on the NVIDA card The Dell has also 1 DisplayPort 1.2 and 1 HDMI 1.4. I suppose these are for the Intel graphic card.

It seems from the log as though the NVidia driver isn't causing use of (EE) where it should. All that shows up are (WW): [ 55.024] (WW) NVIDIA: No DRM device: Direct render devices found but none could be [ 55.024] (WW) NVIDIA: used. I would expect a message like that to be (EE), fatal, indicating broken NVidia driver installation and/or configuration, and/or prime not used as needed for an Optimus laptop.

Thnaks, Martin -- Atari FTP-site: ftp://kurobox.serveftp.net:3021 Running openSUSE Tumbleweed / KDE 5.25.4

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <34b96a72-4345-f2b9-eeeb-5c6a111530a5@earthlink.net> Date & Time: Sun, 14 Aug 2022 07:18:28 -0400 [FM] == Felix Miata <mrmazda@earthlink.net> has written: FM> Martin /Nightowl/ Byttebier composed on 2022-08-14 12:02 (UTC+0200): FM>>> nomodeset FM>> With nomodeset the Plymouth boot screen works like before. FM>> I wonder why I need to set this? Prior to kernel 15.18.9-2 there was no FM>> problem at all. FM> That's the time of simpledrm having been re-enabled, a major FM> change in graphics handling during boot. [...] Simpledrm has been re-enabled since 15.18.9-1, not 15.18.9-2. In Martin's case, I believe that until 15.18.9-1, he was working with nouveau, and since 15.18.9-2, nouveau and nvidia's proprietary driver coexist. In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220813230324.e038b10bf7f6f29b010382f9@telenet.be> Date & Time: Sat, 13 Aug 2022 23:03:24 +0200 [M] == Martin /Nightowl/ Byttebier <nightowl@telenet.be> has written: [...] M> cat /proc/cmdline M> BOOT_IMAGE=/boot/vmlinuz-5.19.0-1-default M> root=UUID=88da3879-fdf1-4cd5-a691-9346a9411953 pcie_aspm=off M> rd.driver.blacklist=nouveau mitigations=auto I see that rd.driver.blacklist=nouveau is set by the installer of the NVIDIA driver. I can't figure this out. If the kernel parameter nomodeset is not set, the installer of the NVIDIA driver should refuse to install the driver. But.... It has installed .... (_ _? In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <34b96a72-4345-f2b9-eeeb-5c6a111530a5@earthlink.net> Date & Time: Sun, 14 Aug 2022 07:18:28 -0400 [FM] == Felix Miata <mrmazda@earthlink.net> has written: [...] FM> Please boot without using nomodeset, and following Ctrl-Alt-F3 and logging in on FM> resulting vtty3, run the following: FM> cat /var/log/Xorg.0.log | susepaste FM> and report here [...] Certainly, Xorg.0.log is an important source of information, but I want to know the following results; 1. cat /var/log/Xorg.0.log | grep nouveau 2. cat /var/log/Xorg.0.log | grep nvidia 3. cat /var/log/Xorg.0.log | grep nv Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "It should never be said that it is OK to ignore the theoretical as long as it becomes a tool." -- T. Mori (The original is in Japanese) --

* Martin /Nightowl/ Byttebier <nightowl@telenet.be> [08-14-22 17:07]:

Op Mon, 15 Aug 2022 05:12:53 +0900 Masaru Nomiya <nomiya@galaxy.dti.ne.jp> schreef:

...

Hello,

In the Message;

Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <34b96a72-4345-f2b9-eeeb-5c6a111530a5@earthlink.net> Date & Time: Sun, 14 Aug 2022 07:18:28 -0400

[FM] == Felix Miata <mrmazda@earthlink.net> has written:

FM> Martin /Nightowl/ Byttebier composed on 2022-08-14 12:02 FM> (UTC+0200):

FM>>> nomodeset

FM>> With nomodeset the Plymouth boot screen works like before.

FM>> I wonder why I need to set this? Prior to kernel 15.18.9-2 there FM>> was no problem at all.

FM> That's the time of simpledrm having been re-enabled, a major FM> change in graphics handling during boot. [...]

Simpledrm has been re-enabled since 15.18.9-1, not 15.18.9-2.

In Martin's case, I believe that until 15.18.9-1, he was working with nouveau, and since 15.18.9-2, nouveau and nvidia's proprietary driver coexist.

As far as I know I always have used NVIDIA's driver. I bought the Dell last year in may. In the beginning with Leap 15.3 and NVIDIA (the easy way). Later on with TW and NVIDIA installed the hard way. AFAIR I couldn't install NVIDIA the easy way on TW.

I can be there are some leftover from previous attemps to install NVIDIA the easy way.

easy to remove as the "easy way" was via rpm.

...

In the Message;

Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220813230324.e038b10bf7f6f29b010382f9@telenet.be> Date & Time: Sat, 13 Aug 2022 23:03:24 +0200

[M] == Martin /Nightowl/ Byttebier <nightowl@telenet.be> has written:

[...] M> cat /proc/cmdline M> BOOT_IMAGE=/boot/vmlinuz-5.19.0-1-default M> root=UUID=88da3879-fdf1-4cd5-a691-9346a9411953 pcie_aspm=off M> rd.driver.blacklist=nouveau mitigations=auto

I see that rd.driver.blacklist=nouveau is set by the installer of the NVIDIA driver. I can't figure this out.

I've added this myself following instruction from this website

https://www.if-not-true-then-false.com/2022/opensuse-nvidia-guide/

...

If the kernel parameter nomodeset is not set, the installer of the NVIDIA driver should refuse to install the driver. But.... It has installed .... (_ _?

As said earlier before kernel 5.18.9-2 there was no need to set nomodeset.

...

In the Message;

Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <34b96a72-4345-f2b9-eeeb-5c6a111530a5@earthlink.net> Date & Time: Sun, 14 Aug 2022 07:18:28 -0400

[FM] == Felix Miata <mrmazda@earthlink.net> has written:

[...] FM> Please boot without using nomodeset, and following Ctrl-Alt-F3 FM> and logging in on resulting vtty3, run the following:

FM> cat /var/log/Xorg.0.log | susepaste

FM> and report here [...]

Certainly, Xorg.0.log is an important source of information, but I want to know the following results;

1. cat /var/log/Xorg.0.log | grep nouveau

2. cat /var/log/Xorg.0.log | grep nvidia

3. cat /var/log/Xorg.0.log | grep nv

1) martin@dell-desk:~> cat /var/log/Xorg.0.log | grep nouveau [ 58.759] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-5.19.0-1-default root=UUID=88da3879-fdf1-4cd5-a691-9346a9411953 pcie_aspm=off nomodeset rd.driver.blacklist=nouveau nvidia-drm.modeset=1 mitigations=auto [ 59.183] (==) Matched nouveau as autoconfigured driver 1 [ 59.276] (II) LoadModule: "nouveau" [ 59.276] (WW) Warning, couldn't open module nouveau [ 59.276] (EE) Failed to load module "nouveau" (module does not exist, 0)

2) martin@dell-desk:~> cat /var/log/Xorg.0.log | grep nvidia [ 58.759] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-5.19.0-1-default root=UUID=88da3879-fdf1-4cd5-a691-9346a9411953 pcie_aspm=off nomodeset rd.driver.blacklist=nouveau nvidia-drm.modeset=1 mitigations=auto [ 59.183] (==) Matched nvidia as autoconfigured driver 0 [ 59.183] (II) LoadModule: "nvidia" [ 59.184] (II) Loading /usr/lib64/xorg/modules/drivers/nvidia_drv.so [ 59.276] (II) Module nvidia: vendor="NVIDIA Corporation" [ 60.036] (II) Loading sub module "glxserver_nvidia" [ 60.036] (II) LoadModule: "glxserver_nvidia" [ 60.036] (II) Loading /usr/lib64/xorg/modules/extensions/libglxserver_nvidia.so [ 60.635] (II) Module glxserver_nvidia: vendor="NVIDIA Corporation" [ 61.894] (==) NVIDIA(0): No modes were requested; the default mode "nvidia-auto-select" [ 61.895] (II) NVIDIA(0): "DFP-1:nvidia-auto-select" [ 61.960] (II) NVIDIA(0): Setting mode "DFP-1:nvidia-auto-select" [ 62.072] (II) NVIDIA(0): [DRI2] VDPAU driver: nvidia

3) martin@dell-desk:~> cat /var/log/Xorg.0.log | grep nv [ 58.759] Kernel command line: BOOT_IMAGE=/boot/vmlinuz-5.19.0-1-default root=UUID=88da3879-fdf1-4cd5-a691-9346a9411953 pcie_aspm=off nomodeset rd.driver.blacklist=nouveau nvidia-drm.modeset=1 mitigations=auto [ 59.183] (==) Matched nvidia as autoconfigured driver 0 [ 59.183] (==) Matched nv as autoconfigured driver 2 [ 59.183] (II) LoadModule: "nvidia" [ 59.184] (II) Loading /usr/lib64/xorg/modules/drivers/nvidia_drv.so [ 59.276] (II) Module nvidia: vendor="NVIDIA Corporation" [ 59.276] (II) LoadModule: "nv" [ 59.276] (WW) Warning, couldn't open module nv [ 59.276] (EE) Failed to load module "nv" (module does not exist, 0) [ 60.036] (II) Loading sub module "glxserver_nvidia" [ 60.036] (II) LoadModule: "glxserver_nvidia" [ 60.036] (II) Loading /usr/lib64/xorg/modules/extensions/libglxserver_nvidia.so [ 60.635] (II) Module glxserver_nvidia: vendor="NVIDIA Corporation" [ 61.894] (==) NVIDIA(0): No modes were requested; the default mode "nvidia-auto-select" [ 61.895] (II) NVIDIA(0): "DFP-1:nvidia-auto-select" [ 61.960] (II) NVIDIA(0): Setting mode "DFP-1:nvidia-auto-select" [ 62.072] (II) NVIDIA(0): [DRI2] VDPAU driver: nvidia

Regards, Martin

--

Atari FTP-site: ftp://kurobox.serveftp.net:3021 Running openSUSE Tumbleweed / KDE 5.25.4

-- (paka)Patrick Shanahan Plainfield, Indiana, USA @ptilopteri http://en.opensuse.org openSUSE Community Member facebook/ptilopteri Photos: http://wahoo.no-ip.org/piwigo paka @ IRCnet oftc

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <87y1vqmn50.wl-nomiya@galaxy.dti.ne.jp> Date & Time: Mon, 15 Aug 2022 08:16:43 +0900 [MN] == 野宮　賢 / NOMIYA Masaru <nomiya@galaxy.dti.ne.jp> has written: [...] MN> > Simpledrm has been re-enabled since 15.18.9-1, not 15.18.9-2. MN> This is for sure. Oh my God! I was totally mistaken． Simpledrm has been re-enabled since 15.18.9-2. Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "Tim Cook, the C.E.O. of Apple, said earlier this year that he would not let his nephew join social networks. Bill Gates banned cellphone until his children were teenagers, and Melinda Gates wrote that she wished they had waited even longer. Steve Jobs would not let his young children near iPads." -- The New York Times --

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220815221000.899915dd2e0c8ab82bad0c7d@telenet.be> Date & Time: Mon, 15 Aug 2022 22:10:00 +0200 [M] == Martin /Nightowl/ Byttebier <nightowl@telenet.be> has written: [...] MN> > Could you tell me what the easy way and the hard way are? M> The easy way is via yast or zypper Add the right repository, install M> the right NVDIA-drivers. M> For some reason I couldn't use this on TW. M> Maybe I should try it again. This is also strange. M> The hard way is via the console (init 3) M> Download the right driver from NVIDIA, boot into runlevel 3 M> install the driver sh NVIDIA-Linux-x86_64-515.65.01.run M> answer a few questions. After install mkinitrd, then reboot. # sh ./NVIDIA-Linux-x86_64-515.65.01.run -aq suppress the questions. M> Prior to kernel 5.18..9.2 I could do this without nomodeset, but now I M> need to set nomodeset as I get a black screen. Please show the results of; 1. $ lspci | grep VGA 2. $ lsmod Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ " Today’s China is not the old China humiliated and bullied over 100 years ago. It is time for these people to wake up from their imperial dream." -- Hua Chunying’s Regular Press Conference on August 4, 2022 --

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220815235841.8153b07de142cd9a934e01b4@telenet.be> Date & Time: Mon, 15 Aug 2022 23:58:41 +0200 [M] == Martin /Nightowl/ Byttebier <nightowl@telenet.be> has written: M> Op Tue, 16 Aug 2022 05:57:05 +0900 M> Masaru Nomiya <nomiya@galaxy.dti.ne.jp> schreef: [...] M>> Prior to kernel 5.18..9.2 I could do this without nomodeset, but M>> M> now I need to set nomodeset as I get a black screen. MN> > Please show the results of; MN> > MN> > 1. $ lspci | grep VGA martin@dell-desk:~> lspci | grep VGA M> 00:02.0 VGA compatible controller: Intel Corporation RocketLake-S GT1 [UHD Graphics 750] (rev 04) M> 02:00.0 VGA compatible controller: NVIDIA Corporation GA104 [GeForce RTX 3060 Ti] (rev a1) I learned for the first time that Intel drivers and nvidia drivers can coexist. In other words, without nomodeset, there is a conflict between the intel driver and the nvidia driver, resulting in a black screen. So, I set up nomodeset, which suppresses the loading of the intel driver, so I can get to the login screen. In https://wiki.archlinux.org/title/intel_graphics, it says; The Intel kernel module should load fine automatically on system boot. If it does not happen, then: Make sure you do not have nomodeset as a kernel parameter, since Intel requires kernel mode-setting. I learned for the first time that Intel driver and nvidia driver can coexist. In other words, without nomodeset parameter, there is a conflict between the intel driver and the nvidia driver, resulting in a black screen. So, the nomodeset setteing suppresses the loading of the intel driver, then you can get to the login screen. You may be wondering why the problem started with Tumbleweed. Since GPU (=Intel Corporation RocketLake-S GT1) is so new that the corresponding kernel module has only recently been developed, so the intel driver autoloading did not occur until then. To avoid this problem, replace the nomodeset in the kernel parameter with i915.modeset=0. Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ " Today’s China is not the old China humiliated and bullied over 100 years ago. It is time for these people to wake up from their imperial dream." -- Hua Chunying’s Regular Press Conference on August 4, 2022 --

Hello, The new mail delivery notification system is working properly. Thanks. In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <4923df5a-8564-c42d-61f7-5d654e994030@earthlink.net> Date & Time: Tue, 16 Aug 2022 04:33:27 -0400 [FM] == Felix Miata <mrmazda@earthlink.net> has written: FM> Masaru Nomiya composed on 2022-08-16 08:10 (UTC+0900): [...] MN> > You may be wondering why the problem started with Tumbleweed. MN> > Since GPU (=Intel Corporation RocketLake-S GT1) is so new that the MN> > corresponding kernel module has only recently been developed, so the MN> > intel driver autoloading did not occur until then. FM> 18 months isn't new in the PC world. Rocket Lake is more than 18 FM> months old, at least one full generation prior to latest. Mine is FM> one of the newer RKLs, yet six quarters old: FM> <https://ark.intel.com/content/www/us/en/ark/products/212270/intel-core-i511400-processor-12m-cache-up-to-4-40-ghz.html> FM> <https://lists.opensuse.org/archives/list/users@lists.opensuse.org/message/QDSIV5KTTMSJMESJ2DDKSPLZZN7EL2EF/> Yours is Intel RocketLake-S GT1 [UHD Graphics 730]. But, Martin's is Intel RocketLake-S GT1 [UHD Graphics 750]. As for UHD Graphics 750, there are some trouble reports about i985 module, and this is one of them; https://gist.github.com/Postrediori/556706b28aff3b831d9e41acb47418c5 Based on the status of the trouble reports, I think that the i985 module was extended to support UHD Graphics 750 from 5.18.9-2. MN> > To avoid this problem, replace the nomodeset in the kernel parameter MN> > with i915.modeset=0. FM> A normal average desktop PC with PCIe slots and graphics outputs FM> supported by a CPU containing an IGP by default has a BIOS option FM> set to prefer enabling only the add-in GPU if present. It FM> shouldn't be necessary to disable the IGP in software, e.g using FM> i915.modeset=0, but that doesn't mean disabling in software can't FM> orshouldn't work. BIOS vary, and many are buggy. Martin's issue FM> might be affected by updating his UEFI BIOS, or simply making a FM> change to graphics settings in it. I thought that the internal GPU could be disabled by installing a graphic card in the PCI slot, but I found that Martin's model does not have such function. I thought it was not possible to disable the GPU in the BIOS since it is an integrated chip. FM> I'm guessing we're off the trail here, that i915.modeset=0 isn't FM> going to work, and that this simpledrm/NVidia combo has a FM> software bug, or a BIOS bug triggered by that FM> combination. cf. <https://gitlab.freedesktop.org/drm/intel/-/issues/4762> I can't understand what is being said? When he installed the NVIDIA driver without nomodeset setting and started it, the Intel and NVIDIA drivers are batting and he got a black screen. However, when he added nomodeset as the kernel parameter, the problem disappeared, which I believe is because the i985 module has been disabled and only the NVIDIA driver is running. Therefore, I do not think it is a software bug, but rather a problem with the kernel parameter settings. If it is a bug in softeware, it should not be possible to boot without any problem. Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "Bill! You married with Computer. Not with Me!" "No..., with money."

Hello, Sorry for late reply. I got some time. In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <2c4ccf29-6724-23ee-e934-fb1cbca535b5@earthlink.net> Date & Time: Tue, 16 Aug 2022 19:00:43 -0400 [FM] == Felix Miata <mrmazda@earthlink.net> has written: FM> Masaru Nomiya composed on 2022-08-16 19:27 (UTC+0900): MN> > Yours is Intel RocketLake-S GT1 [UHD Graphics 730]. MN> > But, Martin's is Intel RocketLake-S GT1 [UHD Graphics 750]. FM> A distinction without a difference for the current FM> context. Officially our CPUs were launched at the same time, FM> while the higher priced models, including Martin's as usual, were FM> first to actually become available. FM> Martin's: FM> <https://www.intel.com/content/www/us/en/products/sku/212279/intel-core-i711700-processor-16m-cache-up-to-4-90-ghz/specifications.html> FM> top CPU price tiers: i7 (i9): FM> i7-11700 launched Q1'21 FM> RocketLake-S GT1 [UHD Graphics 750]; device ID 8086:4c8a FM> FP32 ALUs : EUs : Subslices 256:32:4 (for faster/pricier products) FM> FP32: https://en.wikipedia.org/wiki/Single-precision_floating-point_format FM> Mine: FM> <https://ark.intel.com/content/www/us/en/ark/products/212270/intel-core-i511400-processor-12m-cache-up-to-4-40-ghz.html> FM> lesser CPU price tier: i5: FM> i5-11400 launched Q1'21 FM> RocketLake-S GT1 [UHD Graphics 730]; device ID 8086:4c8b FM> FP32 ALUs : EUs : Subslices 192:24:3 (75% of 4c8a, for slower/cheaper products) FM> 730 has simpler specs and thus (theoretically at least) is cheaper. It wouldn't FM> surprise me if they were all the same, with the lower tier(s) having the other 25% FM> of FP32s disabled. I can understand your doubts. My doubts are; 1. is it possible to reach the login screen without the parameter nvidia-drm.modeset=1? 2. even if I can reach the login screen, can I install NVIDIA's driver without the parameter nomodeset? In my Tumbleweed environment, it is impossible to install the NVIDIA driver without the nomodeset parameter, and even if it is successfully installed, it hangs with a black screen if I try to boot without the nvidia-drm.modeset=1 parameter. I tried, although the kernel are different (5.18.9-2, 5.19.1-1). The results are as follows; 1. Without the parameter nvidia-drm.modeset=1, I cannot reach the login screen, but I can enter terminal mode. 2. the parameter rd.driver.blacklist=nouveau instead of nomodeset installs the NVIDIA driver without any problem. Many Web have shown that even if there are two GPUs (Intel and NVIDIA), as long as NVIDIA is recognized, the driver can be installed and no problem will occur. Based on the above, I believe that the black screen after rebooting after installing the NVIDIA driver is the key to resolving our suspicions. What is important in this regard is the fact that the parameter nomodeset was not set until kernel 5.18-9-2. Then, it is natural to think that the Intel i915 module was improved in kernel 5.18.9-2, and Intel GPU, which reject nomodeset setteing, are also supported, which caused the conflict between drivers, isn't it? I suppose this is the reason why the nomodeset parameter has to be set, since kernel 5.18.9-2. Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "Tim Cook, the C.E.O. of Apple, said earlier this year that he would not let his nephew join social networks. Bill Gates banned cellphone until his children were teenagers, and Melinda Gates wrote that she wished they had waited even longer. Steve Jobs would not let his young children near iPads." -- The New York Times --

Am 18.08.22 um 13:28 schrieb Patrick Shanahan:

fwiw: having utilized nvidia gpus for many years, I cannot remember *ever* having to use "nvidia-drm.modeset=1" kernel command line parameter. and currently am not with a G04 and a G06.

Up to your post, I was convinced I hadn't this directive, either. But on my 15.4 box, I have two files regarding nvidia, both from the same package. The second file uses the modeset option. # cat /etc/modprobe.d/nvidia-default.conf blacklist nouveau # rpm -qf /etc/modprobe.d/nvidia-default.conf nvidia-gfxG06-kmp-default-515.65.01_k5.14.21_150400.22-lp154.15.1.x86_64 # cat /etc/modprobe.d/50-nvidia-default.conf options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 NVreg_DeviceFileMode=0660 options nvidia-drm modeset=1 install nvidia PATH=$PATH:/bin:/usr/bin; if /sbin/modprobe --ignore-install nvidia; then if /sbin/modprobe nvidia_uvm; then if [ ! -c /dev/nvidia-uvm ]; then mknod -m 660 /dev/nvidia-uvm c $(cat /proc/devices | while read major device; do if [ "$device" == "nvidia-uvm" ]; then echo $major; break; fi ; done) 0; chown :video /dev/nvidia-uvm; fi; if [ ! -c /dev/nvidia-uvm-tools ]; then mknod -m 660 /dev/nvidia-uvm-tools c $(cat /proc/devices | while read major device; do if [ "$device" == "nvidia-uvm" ]; then echo $major; break; fi ; done) 1; chown :video /dev/nvidia-uvm-tools; fi; fi; if [ ! -c /dev/nvidiactl ]; then mknod -m 660 /dev/nvidiactl c 195 255; chown :video /dev/nvidiactl; fi; devid=-1; for dev in $(ls -d /sys/bus/pci/devices/*); do vendorid=$(cat $dev/vendor); if [ "$vendorid" == "0x10de" ]; then class=$(cat $dev/class); classid=${class%%00}; if [ "$classid" == "0x0300" -o "$classid" == "0x0302" ]; then devid=$((devid+1)); if [ ! -c /dev/nvidia${devid} ]; then mknod -m 660 /dev/nvidia${devid} c 195 ${devid}; chown :video /dev/nvidia${devid}; fi; fi; fi; done; /sbin/modprobe nvidia_drm; if [ ! -c /dev/nvidia-modeset ]; then mknod -m 660 /dev/nvidia-modeset c 195 254; chown :video /dev/nvidia-modeset; fi; fi # rpm -qf /etc/modprobe.d/50-nvidia-default.conf nvidia-gfxG06-kmp-default-515.65.01_k5.14.21_150400.22-lp154.15.1.x86_64 -- Werner

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220818230202.GF22216@wahoo.no-ip.org> Date & Time: Thu, 18 Aug 2022 19:02:02 -0400 [PS] == Patrick Shanahan <paka@opensuse.org> has written: PS> * Werner Flamme <werner.flamme@email.de> [08-18-22 17:43]: WF> > Am 18.08.22 um 13:28 schrieb Patrick Shanahan: PS> > > fwiw: having utilized nvidia gpus for many years, I cannot remember *ever* PS> > > having to use "nvidia-drm.modeset=1" kernel command line parameter. PS> > > and currently am not with a G04 and a G06. WF> > Up to your post, I was convinced I hadn't this directive, either. But on WF> > my 15.4 box, I have two files regarding nvidia, both from the same WF> > package. The second file uses the modeset option. WF> > WF> > # cat /etc/modprobe.d/nvidia-default.conf WF> > blacklist nouveau WF> > # rpm -qf /etc/modprobe.d/nvidia-default.conf WF> > nvidia-gfxG06-kmp-default-515.65.01_k5.14.21_150400.22-lp154.15.1.x86_64 WF> > WF> > # cat /etc/modprobe.d/50-nvidia-default.conf WF> > options nvidia NVreg_DeviceFileUID=0 NVreg_DeviceFileGID=33 WF> > NVreg_DeviceFileMode=0660 WF> > options nvidia-drm modeset=1 WF> > install nvidia PATH=$PATH:/bin:/usr/bin; if /sbin/modprobe [...] PS> and now I see this also, but... I never had to physically add to kernel PS> boot cmd line1 The difference is whether you use the package or not. If you do not use the package, you must set the kernel parameter separately. Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "She continues to deeply divide opinion, nearly a decade after her death in 2013, due to her policies of privatisation, breaking the power of trade unions, and selling off public housing." -- World Is One News --

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220819013211.GH22216@wahoo.no-ip.org> Date & Time: Thu, 18 Aug 2022 21:32:11 -0400 [PS] == Patrick Shanahan <paka@opensuse.org> has written: PS> * Masaru Nomiya <nomiya@galaxy.dti.ne.jp> [08-18-22 20:16]: PS> > Hello, [...] PS>>> and now I see this also, but... I never had to physically add to kernel PS>>> boot cmd line1 MN>> The difference is whether you use the package or not. MN>> If you do not use the package, you must set the kernel parameter MN>> separately. PS> I still do not recall *ever* using nomodeset when I used the XX.run PS> packages (the hard way). Is there a configuration file in /etc/modprobe.d that contains any of the following? 1. blacklist nouveau 2. blacklist nouveau options nouveau modeset=0 Regards. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "It should never be said that it is OK to ignore the theoretical as long as it becomes a tool." -- T. Mori (The original is in Japanese) --

Hello, In the Message; Subject : Re: TW: boot problem kernel 5.18.9-2 and higher Message-ID : <20220819025112.GI22216@wahoo.no-ip.org> Date & Time: Thu, 18 Aug 2022 22:51:12 -0400 [PS] == Patrick Shanahan <paka@opensuse.org> has written: [...] PS> > Is there a configuration file in /etc/modprobe.d that contains any of PS> > the following? PS> > PS> > 1. blacklist nouveau PS> > PS> > 2. blacklist nouveau PS> > options nouveau modeset=0 PS> I have nowhere to look now as I am currently and for the last several PS> years using rpm to update my nvidia drivers. How about this; /usr/lib/modprobe.d/nvidia-default.conf Reagrds. --- ┏━━┓彡 Masaru Nomiya mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛ "No Windows, no gains!" ... "Why, I am wrong?" -- Bill --

On Thu, 2022-08-18 at 23:14 -0400, Patrick Shanahan wrote:

* Masaru Nomiya <nomiya@galaxy.dti.ne.jp> [08-18-22 23:11]:

...

Hello,

In the Message;

  Subject    : Re: TW: boot problem kernel 5.18.9-2 and higher   Message-ID : <20220819025112.GI22216@wahoo.no-ip.org>   Date & Time: Thu, 18 Aug 2022 22:51:12 -0400

[PS] == Patrick Shanahan <paka@opensuse.org> has written:

[...] PS>  > Is there a configuration file in /etc/modprobe.d that contains any of PS>  > the following? PS>  > PS>  > 1. blacklist nouveau PS>  > PS>  > 2. blacklist nouveau PS>  >    options nouveau modeset=0

PS>  I have nowhere to look now as I am currently and for the last several PS>  years using rpm to update my nvidia drivers.

How about this;

   /usr/lib/modprobe.d/nvidia-default.conf

the three system I currently have with nvidia all show:    > cat /usr/lib/modprobe.d/nvidia-default.conf    blacklist nouveau      

I had some derivation of G04/G05 installed EARLY on for the Dell which has an Intel iGPU and nvidia T1000. Either due to my misunderstanding or an actual problem (I didn't recognize), I had little to no success with the RPM version of nvidia drivers ~ so I ditched them for .run installs (which I also sign and have secure boot enabled since the laptop is also for use with Windows 11; at worst, sometimes I have to enter the bitlocker recovery key when I boot windows, but I don't lose sleep over it). I don't really care about using the iGPU or switching between them, my goal was to just use the T1000 via the laptop screen (so no external display plugged in as I'd generally ssh to it or use Xpra). My main issue, which had persisted for at least 6mos, was that it'd boot to a black screen (on what I assume was vt7). Using the .run always worked fine (w/o X11 running of course) and I'd `sudo systemctl restart display-manager.service` and be greeted with a lightdm login screen. But subsequent reboots always yielded a black screen. That was solved with a CTRL+ALT+F4, login, `sudo modprobe nvidia_drm` and `sudo systemctl restart display-manager.service` ; Very simple, but I wanted to know why it was always the case. I don't use any extra boot parameters and have these ACTIVE files in: /etc/modprobe.d -- $ /bin/cat nvidia-installer-disable-nouveau.conf # generated by nvidia-installer blacklist nouveau options nouveau modeset=0 $ /bin/cat 50-nvidia-smb.conf options nvidia_drm modeset=1 /usr/lib/modprobe.d -- $ /bin/cat nvidia-installer-disable-nouveau.conf # generated by nvidia-installer blacklist nouveau options nouveau modeset=0 /usr/lib/modules-load.d -- $ /bin/cat nvidia_drm.conf nvidia_drm The '/etc/modules-load.d' directory is empty. The 'prime-select' service always fails to start at boot, but `sudo prime-select get-current` returns 'nvidia': $ sudo prime-select get-current Driver configured: nvidia NVIDIA modules are loaded I also had next to no luck following what bbswitch was actually doing and at one point it'd just completely disable the NVIDIA card; so I've disabled '/usr/lib/modules-load.d/bbswitch.conf' by renaming '.conf' to 'DOTconf'. I see these nvidia related messages after boot through dmesg: $ dmesg | grep -i nvidia [ 30.338461] audit: type=1400 audit(1660917198.283:6): apparmor="STATUS" operation="profile_load" profile="unconfined" name="nvidia_modprobe" pid=792 comm="apparmor_parser" [ 30.339146] audit: type=1400 audit(1660917198.283:7): apparmor="STATUS" operation="profile_load" profile="unconfined" name="nvidia_modprobe//kmod" pid=792 comm="apparmor_parser" [ 42.359200] nvidia: loading out-of-tree module taints kernel. [ 42.359653] nvidia: module license 'NVIDIA' taints kernel. [ 42.370739] nvidia: module verification failed: signature and/or required key missing - tainting kernel [ 42.437506] nvidia-nvlink: Nvlink Core is being initialized, major device number 237 [ 42.438514] nvidia 0000:01:00.0: enabling device (0000 -> 0003) [ 42.486075] NVRM: loading NVIDIA UNIX x86_64 Kernel Module 515.65.01 Wed Jul 20 14:00:58 UTC 2022 [ 44.789252] nvidia-modeset: Loading NVIDIA Kernel Mode Setting Driver for UNIX platforms 515.65.01 Wed Jul 20 13:43:59 UTC 2022 [ 45.731149] [drm] [nvidia-drm] [GPU ID 0x00000100] Loading driver [ 46.310497] [drm] Initialized nvidia-drm 0.0.0 20160202 for 0000:01:00.0 on minor 0 And these for i915: $ dmesg | grep -i i915 [ 2.413937] i915 0000:00:02.0: [drm] VT-d active for gfx access [ 2.497787] i915 0000:00:02.0: vgaarb: deactivate vga console [ 2.497815] i915 0000:00:02.0: [drm] Transparent Hugepage mode 'huge=within_size' [ 2.519429] i915 0000:00:02.0: vgaarb: changed VGA decodes: olddecodes=io+mem,decodes=io+mem:owns=io+mem [ 2.520378] i915 0000:00:02.0: [drm] Finished loading DMC firmware i915/kbl_dmc_ver1_04.bin (v1.4) [ 2.627659] [drm] Initialized i915 1.6.0 20201103 for 0000:00:02.0 on minor 1 [ 2.645536] fbcon: i915drmfb (fb0) is primary device [ 3.800019] i915 0000:00:02.0: [drm] fb0: i915drmfb frame buffer device [ 37.275611] snd_hda_intel 0000:00:1f.3: bound 0000:00:02.0 (ops i915_audio_component_bind_ops [i915]) [ 41.460005] mei_hdcp 0000:00:16.0-b638ab7e-94e2-4ea2-a552-d1c54b627f04: bound 0000:00:02.0 (ops i915_hdcp_component_ops [i915]) Now at [re]boot I'm greeted with lightdm consistently. -- ~ Scott Bradnick |- Windows Subsystem for Linux (WSL) Developer |-- Tumbleweed: |--- Dell Precision 5540 [NVIDIA Quadro T1000] (x86_64) |--- O-DROID H2+ [UHD Graphics 600] (x86_64) |--- IceWhale ZimaBoard 832 [Intel HD Graphics 500] (x86_64) |--- 2x Raspberry Pi 4 Model B Rev 1.2 (aarch64) |--- 1x Raspberry Pi 3 Model B Rev 1.2 (aarch64) |--- WinBook TW100 (x86_64) https://keys.openpgp.org/ :: DBC5AA9A2D2BAEBC

On Sat, 2022-08-20 at 08:54 +0900, Masaru Nomiya wrote:

Hello,

In the Message;

  Subject    : Re: TW: boot problem kernel 5.18.9-2 and higher   Message-ID : <8b7827c580db80b0678a2fbd0a4291d339f2c1da.camel@suse.com>   Date & Time: Fri, 19 Aug 2022 15:30:22 +0000

[SB] == Scott Bradnick <scott.bradnick@suse.com> has written:

[...] SB>  I had some derivation of G04/G05 installed EARLY on for the Dell which has an SB>  Intel iGPU and nvidia T1000. Either due to my misunderstanding or an actual SB>  problem (I didn't recognize), I had little to no success with the RPM version of SB>  nvidia drivers ~ so I ditched them for .run installs (which I also sign and have SB>  secure boot enabled since the laptop is also for use with Windows 11; at worst, SB>  sometimes I have to enter the bitlocker recovery key when I boot windows, but I SB>  don't lose sleep over it).

SB>  I don't really care about using the iGPU or switching between them, my goal was SB>  to just use the T1000 via the laptop screen (so no external display plugged in SB>  as I'd generally ssh to it or use Xpra). [...] SB>  And these for i915:

SB>  $ dmesg | grep -i i915 SB>  [    2.413937] i915 0000:00:02.0: [drm] VT-d active for gfx access SB>  [    2.497787] i915 0000:00:02.0: vgaarb: deactivate vga console SB>  [    2.497815] i915 0000:00:02.0: [drm] Transparent Hugepage mode SB>  'huge=within_size' SB>  [    2.519429] i915 0000:00:02.0: vgaarb: changed VGA decodes: SB>  olddecodes=io+mem,decodes=io+mem:owns=io+mem SB>  [    2.520378] i915 0000:00:02.0: [drm] Finished loading DMC firmware SB>  i915/kbl_dmc_ver1_04.bin (v1.4) SB>  [    2.627659] [drm] Initialized i915 1.6.0 20201103 for 0000:00:02.0 on minor 1 SB>  [    2.645536] fbcon: i915drmfb (fb0) is primary device SB>  [    3.800019] i915 0000:00:02.0: [drm] fb0: i915drmfb frame buffer device SB>  [   37.275611] snd_hda_intel 0000:00:1f.3: bound 0000:00:02.0 (ops SB>  i915_audio_component_bind_ops [i915]) SB>  [   41.460005] mei_hdcp 0000:00:16.0-b638ab7e-94e2-4ea2-a552- d1c54b627f04: bound SB>  0000:00:02.0 (ops i915_hdcp_component_ops [i915]) [...]

How about adding this to the kernel parameter?

   i915.modeset=0

Regards.

--- ┏━━┓彡   Masaru Nomiya                mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛       "A bachelor’s degree still holds prestige as a ticket to the                middle class, but its value has received increasing scrutiny.                In the last several years, rising tuition and student loan debt                have led more Americans to reconsider an investment in                postsecondary education."

                                                        -- Washington Post --

Why would I do that (if things are working in a way I'm happy with)? To squash those early-in-the-boot-process i915 messages? -- ~ Scott Bradnick |- Windows Subsystem for Linux (WSL) Developer |-- Tumbleweed: |--- Dell Precision 5540 [NVIDIA Quadro T1000] (x86_64) |--- O-DROID H2+ [UHD Graphics 600] (x86_64) |--- IceWhale ZimaBoard 832 [Intel HD Graphics 500] (x86_64) |--- 2x Raspberry Pi 4 Model B Rev 1.2 (aarch64) |--- 1x Raspberry Pi 3 Model B Rev 1.2 (aarch64) |--- WinBook TW100 (x86_64) https://keys.openpgp.org/ :: DBC5AA9A2D2BAEBC

On Sat, 2022-08-20 at 10:04 +0900, Masaru Nomiya wrote:

Hello,

In the Message;

  Subject    : Re: TW: boot problem kernel 5.18.9-2 and higher   Message-ID : <047196af5184d55ed3998c99bea097797a94be83.camel@suse.com>   Date & Time: Sat, 20 Aug 2022 00:08:54 +0000

[SB] == Scott Bradnick <scott.bradnick@suse.com> has written:

[...] MN> > How about adding this to the kernel parameter? MN> > MN> >    i915.modeset=0 [...]

SB>  Why would I do that (if things are working in a way I'm happy SB>  with)? To squash those early-in-the-boot-process i915 messages?

In the Message;

  Subject    : Re: TW: boot problem kernel 5.18.9-2 and higher   Message-ID : <8b7827c580db80b0678a2fbd0a4291d339f2c1da.camel@suse.com>   Date & Time: Fri, 19 Aug 2022 15:30:22 +0000

[SB] == Scott Bradnick <scott.bradnick@suse.com> has written:

[...] SB>  My main issue, which had persisted for at least 6mos, was that it'd boot to a SB>  black screen (on what I assume was vt7). Using the .run always worked fine (w/o SB>  X11 running of course) and I'd `sudo systemctl restart display- manager.service` SB>  and be greeted with a lightdm login screen. But subsequent reboots always SB>  yielded a black screen. That was solved with a CTRL+ALT+F4, login, `sudo SB>  modprobe nvidia_drm` and `sudo systemctl restart display-manager.service` ; Very SB>  simple, but I wanted to know why it was always the case.

Sorry, I was distracted by this sentence.

The i915.modesete=0 is a setting that does not allow Intel drivers to be loaded.

In the Message;

  Subject    : Re: TW: boot problem kernel 5.18.9-2 and higher   Message-ID : <8b7827c580db80b0678a2fbd0a4291d339f2c1da.camel@suse.com>   Date & Time: Fri, 19 Aug 2022 15:30:22 +0000

[SB] == Scott Bradnick <scott.bradnick@suse.com> has written:

[...] Now at [re]boot I'm greeted with lightdm consistently. [...]

You seem to be using two GPUs, one Intel and one NVIDIA. Of course, this is possible.

What settings are you using?

Regards.

--- ┏━━┓彡  Masaru Nomiya                 mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛       "Bill!                You married with Computer.                Not with Me!"                "No..., with money."

I probably wasn't explicit enough. I'm not having an issue I'm looking for advice on - I'm sharing my experience with 'the hard way' (which I prefer) and [maybe wrongfully] assuming it could help here or someone else. This Dell Precision 5540 is an Optimus laptop w/ an nvidia T1000 & intel iGPU. The prime and bbswitch stuff seems to make usage harder or inconsistent w/ my expectations, so I'm not using prime in any way and have stopped bbswitch module from loading. 'dmesg' does report something around the i915 driver(s) being loaded, but the T1000 is definitely active according to `glxinfo -B` using 'nvidia_drm' according to 'lsmod'. The settings I'm using were outlined in my initial email w/ .conf files under /etc/{modprobe.d,modules-load.d} & /usr/lib/{modprobe.d,modules-load.d}; I'm not using 1-off items that'd show up under /proc/cmdline or adjusting grub config(s). Here's 'glxinfo -B' and 'inxi --graphics' output: ``` name of display: :0 display: :0 screen: 0 direct rendering: Yes Memory info (GL_NVX_gpu_memory_info): Dedicated video memory: 4096 MB Total available memory: 4096 MB Currently available dedicated video memory: 3778 MB OpenGL vendor string: NVIDIA Corporation OpenGL renderer string: Quadro T1000/PCIe/SSE2 OpenGL core profile version string: 4.6.0 NVIDIA 515.65.01 OpenGL core profile shading language version string: 4.60 NVIDIA OpenGL core profile context flags: (none) OpenGL core profile profile mask: core profile OpenGL version string: 4.6.0 NVIDIA 515.65.01 OpenGL shading language version string: 4.60 NVIDIA OpenGL context flags: (none) OpenGL profile mask: (none) OpenGL ES profile version string: OpenGL ES 3.2 NVIDIA 515.65.01 OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.20 Graphics: Device-1: Intel CoffeeLake-H GT2 [UHD Graphics 630] driver: i915 v: kernel Device-2: NVIDIA TU117GLM [Quadro T1000 Mobile] driver: nvidia v: 515.65.01 Device-3: Microdia Integrated_Webcam_HD type: USB driver: uvcvideo Display: x11 server: X.Org v: 21.1.4 with: Xwayland v: 22.1.3 driver: X: loaded: modesetting,nvidia gpu: i915 resolution: 1920x1080~60Hz OpenGL: renderer: Quadro T1000/PCIe/SSE2 v: 4.6.0 NVIDIA 515.65.01 ``` No more black screen on vt7 and manually loading nvidia_drm followed by `$ sudo systemctl restart display-manager.service` 😀️ -- ~ Scott Bradnick |- Windows Subsystem for Linux (WSL) Developer |-- Tumbleweed: |--- Dell Precision 5540 [NVIDIA Quadro T1000] (x86_64) |--- O-DROID H2+ [UHD Graphics 600] (x86_64) |--- IceWhale ZimaBoard 832 [Intel HD Graphics 500] (x86_64) |--- 2x Raspberry Pi 4 Model B Rev 1.2 (aarch64) |--- 1x Raspberry Pi 3 Model B Rev 1.2 (aarch64) |--- WinBook TW100 (x86_64) https://keys.openpgp.org/ :: DBC5AA9A2D2BAEBC

On Sun, 2022-08-21 at 07:30 +0900, Masaru Nomiya wrote:

Hello,

In the Message;

  Subject    : Re: TW: boot problem kernel 5.18.9-2 and higher   Message-ID : <557b9205ba9ca693843fa3985abc6daf5cceb254.camel@suse.com>   Date & Time: Sat, 20 Aug 2022 15:33:07 +0000

[SB] == Scott Bradnick <scott.bradnick@suse.com> has written:

[...] SB>  No more black screen on vt7 and manually loading nvidia_drm SB> followed by `$ sudo systemctl restart display-manager.service` 😀️

Please have a look;

https://forums.opensuse.org/showthread.php/521325-display-manager-service-re...

Regards.

--- ┏━━┓彡   Masaru Nomiya                  mail-to: nomiya @ galaxy.dti.ne.jp ┃＼／彡 ┗━━┛       " Today’s China is not the old China humiliated and bullied over                100 years ago. It is time for these people to wake up from their                imperial dream."

              -- Hua Chunying’s Regular Press Conference on August 4, 2022 --

SIDE NOTE: I tried 'i915.modeset=0' and while that did stop i915 from being loaded, it also stopped lightdm from loading. I'm sticking w/ NO EXTRA entries since both i915 and nvidia_drm load just fine and I'm greeted w/ a lightdm login window. Even though I've said a few times now - everything is fine for me on the Dell (I'm just trying to add some knowledge or 'this is what worked for me' info to this thread) - I looked at that URL you linked and it doesn't have anything to do w/ what I've been saying (granted, I have mentioned 'display-manager.service' but I NEVER said I was having issues with it). I used to have to restart 'display-manager.service' AFTER manually loading 'nvidia_drm' in the past because there was some mysterious issue blocking it from loading properly UNTIL I made adjustments to .conf files in these 4 (well, technically 3 - [*]one is empty) directories: /etc/modprobe.d [*]/etc/modules-load.d /usr/lib/modprobe.d /usr/lib/modules-load.d I only injected myself into this convo to _hopefully_ assist someone else who might be fighting w/ an Optimus laptop, using nvidia drivers 'the hard way' and getting a black screen. That was happening to me for ~6mos. until I happened upon the settings I['ve been trying to] relay[ed] here. Maybe that was a foolish endeavor 😋️ I'm going to bow-out of this thread as it's gone beyond what my intention was for even getting involved. Take care! -- ~ Scott Bradnick |- Windows Subsystem for Linux (WSL) Developer |-- Tumbleweed: |--- Dell Precision 5540 [NVIDIA Quadro T1000] (x86_64) |--- O-DROID H2+ [UHD Graphics 600] (x86_64) |--- IceWhale ZimaBoard 832 [Intel HD Graphics 500] (x86_64) |--- 2x Raspberry Pi 4 Model B Rev 1.2 (aarch64) |--- 1x Raspberry Pi 3 Model B Rev 1.2 (aarch64) |--- WinBook TW100 (x86_64) https://keys.openpgp.org/ :: DBC5AA9A2D2BAEBC

Masaru Nomiya composed on 2022-08-15 05:57 (UTC+0900):

Please show the results of;

1. $ lspci | grep VGA

What does that provide that wasn't already provided as requested two days ago in <https://lists.opensuse.org/archives/list/users@lists.opensuse.org/message/C4U3UG46HTREKBNLMEISNTTX3YKXOTB4/> ??? [quote] Martin /Nightowl/ Byttebier composed on 2022-08-14 15:12 (UTC+0200): ...

inxi -Gaz

Graphics: Device-1: Intel RocketLake-S GT1 [UHD Graphics 750] vendor: Dell driver: N/A alternate: i915 arch: Gen12.1 process: Intel 10nm built: 2020-21 bus-ID: 00:02.0 chip-ID: 8086:4c8a class-ID: 0300 Device-2: NVIDIA GA104 [GeForce RTX 3060 Ti] vendor: Dell driver: nvidia v: 515.65.01 alternate: nouveau,nvidia_drm non-free: 515.xx+ status: current (as of 2022-06) arch: Ampere process: TSMC n7 (7nm) built: 2020-22 pcie: gen: 1 speed: 2.5 GT/s lanes: 16 link-max: gen: 4 speed: 16 GT/s bus-ID: 02:00.0 chip-ID: 10de:2486 class-ID: 0300[/quote] It has BUS ID for each GPU. It has manufacturer for each GPU. It has model of each GPU. For further identification, it also has, among other useful device information, Device ID, which 'lspci | grep VGA' doesn't, while 'lspci -nnk | grep VGA' would have. -- Evolution as taught in public schools is, like religion, based on faith, not based on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata

On 8/14/22 03:29, Martin /Nightowl/ Byttebier wrote:

With nvidia-drm.modeset=1 set vt 7 error is gone. But the plymouth booting screen still halts after I915 0000:00:02.0: enabling device (0006 -> 0007)

I have found plymouth much more trouble than it is worth on my system. You can take the conservative approach and test with plymouth disabled by passing the kernel parameter "plymouth.enable=0". If that works, shoot the dog in the head with: # rpm -e plymouth-dracut plymouth-scripts plymouth-branding-openSUSE plymouth-plugin-script plymouth plymouth-plugin-label-ft Never had any problem since 15.0. I treat plymouth the same way I treated the dreaded-dog (Beagle) in the past -- just shoot it in the head and be done with it ;-) -- David C. Rankin, J.D.,P.E.

On 8/20/22 05:38, Dave Howorth wrote:

...

I have found plymouth much more trouble than it is worth on my system. You can take the conservative approach and test with plymouth disabled by passing the kernel parameter "plymouth.enable=0".

If that works, shoot the dog in the head with:

# rpm -e plymouth-dracut plymouth-scripts plymouth-branding-openSUSE plymouth-plugin-script plymouth plymouth-plugin-label-ft Just taboo plymouth and all the others won't be installed

That's a good point DH. But to taboo Plymouth, I'd have to remember that during install - and human frailties being what they are... it's usually an after-the-fact issue I have to deal with -- given installs are no more frequent than a biennial event for me :) -- David C. Rankin, J.D.,P.E.