Re: [PATCH v3 3/4] x86/alternative: Rewrite optimize_nops() some

[Peter Zijlstra]

On Wed, Feb 08, 2023 at 09:44:04PM +0100, Peter Zijlstra wrote:

> Works as advertised. Next up are 8 12 byte nopes and 8 13 byte nops,
> after that we're left with:

Yeah, we can ignore those, those are the retpoline thunks being patched.
They should be totally unused. These NOPs are after an indirect jmp and
as such pure speculation food.

Arguably I should just merge that patch that turns them into UD2 --
funneling and all that.

But there was some i386 issues... oh well.