Genkernel error with drivers

[caughtintheweb]

Hi, another day, another error, that is the motto for my Gentoo installation process.

Anyway, after fixing a simple error with emerging certain packages, I went to simply do genkernel all, and it was going smoothly, when an error happened.

After going to the log (and scrolling down a few thousand lines lmao) I found the error/s, where the first one was along the lines of "null characters ignored in a #define". Btw, both of these errors were in the same file, a GPU driver in drivers/drm/amd/amndgpu etc, etc.
The weird part about that, is that I don't even HAVE an amd gpu, I have an Nvidia RTX 2070 super.
And the second error I somewhat understood, but there is an aspect of weirdness. It's along the lines of "invalid preprocessing directive #defin_".
Now obviously, it's supposed to be #define, but the weird part is that I haven't touched any files like that, so something else must have messed it up.

If anybody has a fix, or needs me to give any form of output logs or anything like that, simply ask and you shall recieve.

-caughtintheweb

[Hu]

You configured the kernel to build with AMDGPU support, or at least genkernel did so for you. Are you sure about that second error? e is 0x65, and _ is 0x5f, so a single bit flip cannot explain that corruption.

Beyond that, there is not much we can do with the information you have provided. You should get in the habit of providing the standard logs without being prompted. If you wait for someone to prompt you, then your post becomes "answered" (has responses) before you get a solution. Some users run a canned search for threads with zero replies as a way of finding posts to answer. Thus, they would no longer see your thread on the list of unanswered threads.

[caughtintheweb]

[caughtintheweb]

Still not sure what to do about this one...

[caughtintheweb]

Hi, this is effectively the same post as before, but I'm gonna include logs from the start so yall will be able to understand it easier.
After going to the log (and scrolling down a few thousand lines lmao) I found the error/s, where the first one was along the lines of "null characters ignored in a #define". Btw, both of these errors were in the same file, a GPU driver in drivers/drm/amd/amndgpu etc, etc.
The weird part about that, is that I don't even HAVE an amd gpu, I have an Nvidia RTX 2070 super.
And the second error I somewhat understood, but there is an aspect of weirdness. It's along the lines of "invalid preprocessing directive #defin_".
Now obviously, it's supposed to be #define, but the weird part is that I haven't touched any files like that, so something else must have messed it up.

I figured out later that the first one wasn't an error, but in fact just a warning, so it's probably nothing to worry about.
Here's the logs:
In file included from drivers/gpu/drm/amd/amdgpu/nbio_2_3.c:29:
./drivers/gpu/drm/amd/amdgpu/../include/asic_reg/nbio/nbio_2_3_sh.h:2957:2: error: invalid preprocessing directive #defin_
2957 | #defin_EPF0_VF12_0_BASE_CLASS 0x000b
| ^~~~~~~~~~~~~~~~~~~~~
make[4]: *** [scripts/Makefile.build:266: drivers/gpu/drm/amd/amdgpu/nbio_2_3.o] Error 1
make[3]: *** [scripts/Makefile.build:500: drivers/gpu/drm/amd/amdgpu] Error 2
make[2]: *** [scripts/Makefile.build:500: drivers/gpu/drm] Error 2
make[1]: *** [scripts/Makefile.build:500: drivers/gpu] Error 2
make: *** [Makefile:1706: drivers] Error 2

* ERROR: compile_modules(): compile_generic() failed to compile the "modules" target!

Lmk if you guys have a fix asap.

[caughtintheweb]

Because there's been nearly 200 views and zero responses since the first one saying that I didn't add logs, I'm still not sure what to do next.
Google didn't give me any good advice, (and i checked other links from there as well), and I didn't see anybody with a similar issue.
I did see on the Arch wiki that you could omit installing a kernel when installing in a container, is it similar when installing in a virtualbox vm?
If not, then somebody PLEASE give me some advice, because I have zero idea of how to continue from here, and I've had my VM window open for several days now.

[Hu]

Since you don't need amdgpu support, what about bypassing the problem by configuring the kernel not to build amdgpu? If it were me, I would want to know why the source file was installed/generated incorrectly. That #define should not be ill-formed like that.

No, if you install in a virtual machine, you need a kernel for that virtual machine. Containers reuse the host kernel. Virtual machines do not.

[caughtintheweb]

Alright, I guess I'm going with manual configuration on this one. Will make a post if there's any issues.

[Hu]

I think genkernel can let you edit an existing configuration. I have never used genkernel, and do not recall the option to make it do that.

[caughtintheweb]

I'll look into it, and if I can't find anything, I'll go back to fixing manual configuration, as I had a problem installing pciutils.
The errors never stop, do they?

[caughtintheweb]

Alright, apparently genkernel CAN let you edit configurations, by using the makeconfig option.
So I excluded the amdgpu module, and the ati radeon module, hoping that would fix it.
Apparently not, because after I made the config and did the genkernel all command, I found not only the old error, but a new one.
Now, there are hundreds in a row of "error: stray (a number between 200 and 400) in program", and every new error the number is completely different, sometimes with a random slash or white box character next to it.
There was also some times where I got that error, then below it is just a long string of random garbled junk, that doesn't make any sense.
At the end, there was also an error for "missing terminating ' character", not quite sure where the character would be, it had below it a #define PSWUSCFG0_0_PCIE_VC1_RESOURCE_CNTL__LOAD_PORT_ARB_TABLE_MASK
Most of the stray in program errors had the #define i just listed btw.
My suspect is some sort of data corruption or something, if anybody knows how to fix that kinda stuff on vbox lmk.

[x90e]

I'm assuming you're already chroot'd into the liveCD install environment, if you're trying select text with your mouse to copy and paste, you'd need to first emerge gpmd and then start gpmd /etc/init.d/gpmd start to get a mouse working in the console, then you can highlight the text which should automaticaly copy it or you can do ctrl+shift+c, then switch to a second terminal with ctrl + alt + f2 {ctrl + alt +f1 to switch back, f2, f3, f4, f5, f6 are all available} and say, open up a nano session and right click or middle click (you'll find that whichever of these two you try first, if it doesnt work, you'll have to go back and reselect / recopy the text before you try the other, like if you try to right click first and it just selects the entire line, you'd have to go back and reselect the text on the first console before you tried to middle click) / ctrl+shift+v to paste the output when you can then upload via pastebin or something OR you can just pipe the output of the genkernel command to a text file. command > text_file or command | tee text_file and then wait for the command to finish.

you can configure your kernel yourself (removing amdgpu / ati radeon modules and selecting nouveau or nvidia) by cd /usr/src/linux and then make menuconfig, selecting support you need from lspci after which it will be saved in (default) .config which you can provide to genkernel either by cp .config /usr/share/genkernel/defaults/kernel-generic-config( which I wouldn't recommend,) or you can use the command line option genkernel --kernel-config=/usr/src/linux/.config (before you do this, make a backup of the .config file because genkernel default will overwrite the .config file. It should make a copy into /etc/kernels/ but I would do this myself before running genkernel. you might find you need to select some other options like --no-mrproper or --install / --no-install. check the manpage for genkernel and the gentoo handbook