Arch Linux User Repository

Hi!, for me PGP signatures are not matching :c

gotcha, thanks! @ANDRoid7890

@JoeSchr I don't plan to add any patches to this package, I only keep it up-to-date with all changes from linux-xanmod and manjaro linux, so you can ask linux-xanmod maintainer to add this. If they add this - I will do the same

he, I love this kernel but I had some weird connectivity issues on wired network since using it.

Is this patch in: https://bugzilla.kernel.org/show_bug.cgi?id=205047#c31

It seems the issue can be traced to that. Thanks

@deemon i don't use yay, with Pamac you can edit before compiling, so do it at least once, there's no need for the maintainer to change the pkgbuild for you.

Regarding the error in the dkms, this is happening with any 5.11 kernel, it is not a problem specifically with this package.

Greetings.

  YACC    scripts/kconfig/parser.tab.[ch]
/bin/sh: line 1: bison: command not found
make[1]: *** [scripts/Makefile.host:17: scripts/kconfig/parser.tab.h] Error 127
make: *** [Makefile:602: olddefconfig] Error 2
==> ERROR: A failure occurred in prepare().
    Aborting...
error making: linux-manjaro-xanmod (linux-manjaro-xanmod linux-manjaro-xanmod-headers)

Installed bison manually and then it was able to make it -- missing makedep?.

then a bit later (did not fail for those modules for official 5.10.15, 5.9.16 and 5.4.95 kernels. Only for xanmod):

==> dkms install --no-depmod -m vboxhost -v 6.1.18_OSE -k 5.10.14-xanmod1-MANJARO
Error! Bad return status for module build on kernel: 5.10.14-xanmod1-MANJARO (x86_64)
Consult /var/lib/dkms/vboxhost/6.1.18_OSE/build/make.log for more information.

about the pkgbuild thing... wait what? I run yay and it does download and compile and install the stuff. there is no inbetween time for me to go and change pkgbuild :D

Can this optimisation setting 99 be made default instead of 0? :)

@ANDRoid7890 the script only does something if the user chooses in pkgbuild for which processor the kernel should be compiled, if it leaves the default it will build generically, if it chooses 99 the script will choose the native option.

There is nothing wrong with that, the user just needs to read pkgbuild to know what to do.

@deemon you use 0 or 99, there is no specific number for Zen 3 yet.

Greetings.

@deemon you'd better ask linux-xanmod maintainer, for some reason we have to look in that script and set architecture number in PKGBUILD manually

I just forked from his AUR package and I don't know why this script does nothing

The following file hasn't been updated against the main xanmod kernel.

The generic optimisation is setting 99 now.

choose-gcc-optimization.sh

Pardon my n00bism on this matter, but ... while installing/building the new version with yay I noticed something:

==> Detected CPU architecture: znver2

    Available CPU microarchitectures:

    1) AMD K6/K6-II/K6-III
    2) AMD Athlon/Duron/K7
<snip>
   13) AMD Family 17h (Zen)
   14) AMD Family 17h (Zen 2)
<snip>
   42) Native optimizations autodetected by GCC

    0) Generic (default)

==> WARNING: According to PKGBUILD variable _microarchitecture, your choice is 0
==> Building this package for microarchitecture: CONFIG_GENERIC_CPU

when it detected znver2, why it did pick the choice 0 and built it for generic cpu? :D how can I make it choose option 14 ??? (do I have to like... not use YAY but do it somehow semi manually? any hints please? :-) )

And what special sauce would be in the specific CPU version kernel compared to generic cpu? Also while on the subject, the real question is how to compile it for Zen 3, which I actually have, not znver2.