Arch Linux User Repository

==> dkms install --no-depmod vboxhost/7.0.8_OSE -k 6.3.1-1-ck-generic-v3
Error! Bad return status for module build on kernel: 6.3.1-1-ck-generic-v3 (x86_64)
Consult /var/lib/dkms/vboxhost/7.0.8_OSE/build/make.log for more information.

Anyone else getting errors when trying to build 6.3.1-1?

btf_encoder__encode: btf__dedup failed!
Failed to encode BTF
  LD      .tmp_vmlinux.kallsyms1
  NM      .tmp_vmlinux.kallsyms1.syms
  KSYMS   .tmp_vmlinux.kallsyms1.S
  AS      .tmp_vmlinux.kallsyms1.S
  LD      .tmp_vmlinux.kallsyms2
  NM      .tmp_vmlinux.kallsyms2.syms
  KSYMS   .tmp_vmlinux.kallsyms2.S
  AS      .tmp_vmlinux.kallsyms2.S
  LD      vmlinux
  BTFIDS  vmlinux
libbpf: failed to find '.BTF' ELF section in vmlinux
FAILED: load BTF from vmlinux: No data available
make[1]: *** [scripts/Makefile.vmlinux:35: vmlinux] Error 255
make[1]: *** Deleting file 'vmlinux'
make: *** [Makefile:1249: vmlinux] Error 2

Fails on both clang and gcc.

I have the same one: 57c562c3cd2753f232549cab05c8ad770ed848ae86401619c7581bdffaeea4fe

I just checked the given signature for the file with this hash:

sal ➜  Downloads  xz -cd linux-6.2.10.tar.xz | gpg --verify linux-6.2.10.tar.sign -                                    
gpg: Signature made Thu 06 Apr 2023 05:13:42 AM CDT
gpg:                using RSA key 647F28654894E3BD457199BE38DBBDC86092693E
gpg: Good signature from "Greg Kroah-Hartman <gregkh@linuxfoundation.org>" [unknown]
gpg:                 aka "Greg Kroah-Hartman <gregkh@kernel.org>" [unknown]
gpg:                 aka "Greg Kroah-Hartman (Linux kernel stable release signing key) <greg@kroah.com>" [unknown]
gpg: WARNING: This key is not certified with a trusted signature!
gpg:          There is no indication that the signature belongs to the owner.
Primary key fingerprint: 647F 2865 4894 E3BD 4571  99BE 38DB BDC8 6092 693E

I agree/I get the same checksum

something is strange.

On my notebook I could build the update on release day. on my Desktop that I can only update now, I get a sha256sum mismatch when trying to build. No matter how often I download https://www.kernel.org/pub/linux/kernel/v6.x/linux-6.2.10.tar.xz the sha256sum I get is 57c562c3cd2753f232549cab05c8ad770ed848ae86401619c7581bdffaeea4fe not 7de856a284e57e23b8744b2708fe49d49aebe3510c3eee2dd6342e3b5e46ecf2 like expected in the PKGBUILD - is upstream broken/infected or is something off on my end?

Edit: I did now download the file on my phone using mobile internet, and the sha256sum is 57c562c3cd2753f232549cab05c8ad770ed848ae86401619c7581bdffaeea4fe too - so something definitely changed upstream after the release ...

Thanks @glitsj16 - I will update the repo with 6.2.8-2. Others please comment here if you notice any performance regressions.

Building 6.2.8-2 in progress. But after your latest rebase the ck-hrtimer patches are exactly the same as what I used in my 6.2.7-1 / 6.2.8-1. Can confirm both build fine. Didn't notice any regressions or performance drops compared to 6.1.12-1 (the latest available in your repo). All in all I think the 6.2.y is looking good now.

Ah, I missed that one. See here: https://github.com/graysky2/linux-patches/commit/a8be5531ff07e300525c20bdbbe38cc6a665a0c4

Does building 6.2.8-2 work for you? Any performance issues?

@graysky I do notice a difference between our 0004-hrtimer-Replace-all-schedule-timeout-1-with-schedule.patch files. Here's mine:

https://gist.github.com/glitsj16/b082da47c46ef9f3cb444798b32b35f7

@graysky I'm doing a rebuild of 6.2.8 using your rebased patches right now. My earlier build didn't have the 0002-bpf-x86-Fix-IP-after-emitting-call-depth-accounting.patch, so I included that too. Will post the results here asap. Thanks for your work!