Arch Linux User Repository

@graysky This is a forum post I made on the issue. It looks like folders under the work directory are not able to be identified as mount points. If you install archiso, and just run sudo cp -r /usr/share/archiso/configs/releng/ ~/archlive; cd ~/archlive; sudo ./build.sh -v The issue should reproduce itself.

I just happened to decide to try the Arch kernel to see if that resolved the build failure I was receiving, and everything just built successfully. Looks like others have identified this issue in the past, but did not report it. I have a feeling it has something to do with mkarchiso.

https://bbs.archlinux.org/viewtopic.php?id=248473

https://unix.stackexchange.com/questions/460043/how-can-i-successfully-build-an-archiso-image-airootfs-is-not-a-mountpoint

You have provided nothing to go on...

I'm having issues getting archiso to successfully build an image when running the CK kernel. Everything builds fine with the default kernel. Is anyone else having this issue? Could it be from disabling initramfs support, other than LZ4? Is this a CK issue or archiso issue?

@charveey thx but fix pkgbuild please ;D

Will the AMD Llano processors family be added to the kernel?

@rzax666 if you're interested you can just install my package, linux-bootsplash which provides the kernel with framebuffer bootsplash support. I personally run the 5.2.5 ck kernel with the patch you want. I got a hardware issue on my computer so I can't test the latest version (5.2.6)

please add this patch - https://github.com/philmmanjaro/linux-bootsplash

https://gitlab.manjaro.org/packages/core/linux52

Yeah, I think you shouldn't use native in non-interactive mode. See the comments:

Note - the march=native option is unavailable by this method, use the and manually select it.

Since switching from _subarch=20 (MIVYBRIDGE) to _subarch=28 (MNATIVE) in the PKGBUILD, it won't build:

  24. Intel Skylake X (MSKYLAKEX) (NEW)  
  25. Intel Cannon Lake (MCANNONLAKE) (NEW)  
  26. Intel Ice Lake (MICELAKE) (NEW)  
> 27. Generic-x86-64 (GENERIC_CPU)  
  28. Native optimizations autodetected by GCC (MNATIVE) (NEW)  
choice[1-28?]: 28  
Support for P6_NOPs on Intel chips (X86_P6_NOP) [N/y/?] (NEW) 28  
Support for P6_NOPs on Intel chips (X86_P6_NOP) [N/y/?] (NEW) 28  
Support for P6_NOPs on Intel chips (X86_P6_NOP) [N/y/?] (NEW) 28  
Support for P6_NOPs on Intel chips (X86_P6_NOP) [N/y/?] (NEW) 28  
Support for P6_NOPs on Intel chips (X86_P6_NOP) [N/y/?] (NEW) 28  
Support for P6_NOPs on Intel chips (X86_P6_NOP) [N/y/?] (NEW) 28  

The last message keeps looping forever.

When I switch back to the default _subarch= with no value, the process/interaction seems suboptimal:
- it asks me choice[1-28?]:
- I choose 28
- it asks me Support for P6_NOPs on Intel chips (X86_P6_NOP) [N/y/?] (NEW)
- I choose y
- it does the modprobed-db recall (as configured)
- it asks me BT8XX GPIO abuser (GPIO_BT8XX) [N/m/y/?] (NEW)
- I choose n (why does this ask me every time?)
- it asks me Studio Evolution SE6X (SND_SE6X) [N/m/?] (NEW)
- I choose n (again, why does this ask me every time?)
- ==> Sources are ready.

@graysky - Thanks, and understood.

@zerophase - Thanks, what you said made sense, and got me thinking about my chroot. As I had been using makechrootpkg of late to build linux-ck, I completely forgot that command neither cleans, nor updates a chroot.

As soon as I built linux-ck with my typical extra-x86_64-build my chroot got cleaned and updated, which resolved all the DKMS incompatible gcc versions installation issues.