How to build the kernel faster?

[kirelagin]

I’d like to do some kernel patch testing and bisecting, in other words, I want to rebuild the kernel 20 times in a row.

My CPU is AMD Ryzen 7 7840U, so not the worst out there, but the build takes like half an hour, which, I think, is kinda insane, and I am not really sure why. Am I right in thinking that the default NixOS config is just really heavy and builds way too many modules? I was expecting it to take a couple of minutes.

Is there some best-practice / guide on making the kernel build fast? I think I am inclined to simply start from a defconfig and tweak it to make sure it runs on my hardware.

[7c6f434c]

@kirelagin

builds way too many modules

it answers “m” if possible

So yes.

But indeed you can pass your own configfile

[rnhmjoj]

See this:

Kernel build uses more than 16 GB of space

The default kernel configuration in NixOS is insanely bloated, don’t even try to compile it unless you have some server as a dedicated build farm. It can easily take hours to build on a desktop, even if you manage to get enough disk and memory to do it. So, you can’t simply add your patch to boot.kernelPatches and call it a day, but there is a way around. You can generate a kernel configuration that enables only the modules you really need by looking at what is currently loaded: you have to run…

[twoolie]

You can use CCache using an overlay, and it will cache partial build products between runs. This may speed up kernel bisects.

There’s also this thread discussing how to get it working for kernel builds.