Re: [PATCH] x86/bugs: Allow STIBP with IBRS
From: KP Singh
Date: Mon Feb 20 2023 - 14:40:13 EST
On Mon, Feb 20, 2023 at 11:35 AM Josh Poimboeuf <jpoimboe@xxxxxxxxxx> wrote:
>
> On Mon, Feb 20, 2023 at 11:16:45AM -0800, KP Singh wrote:
> > > As far as I can tell, that document was never updated to describe
> > > spectre_v2=ibrs in the first place. That would be a whole 'nother patch
> > > which I'm not volunteering for. Nice try ;-)
> >
> > This should at least be documented in the code.
> >
> > Now it seems like it is not okay to work with people on the list and
> > just send revisions bypassing them. This is not something we do in the
> > kernel area I come from (an x86 favorite ;)). Please feel free to go
> > with Josh's version (or its future revisions). If you want me to
> > re-spin with some comments, happy to. If not, please do at least give
> > me Reported-by here.
>
> It's a common practice even outside of x86. I'd recommend not taking it
> personally. In the end it's all about what produces better code.
>
> And please don't take this the wrong way, but sometimes it takes a bad
> patch to inspire a better one. I mean that with no disrespect, I myself
> have sent many bad patches.
Appreciate the clarification! Thank you so much.
>
> And if you don't like my patch, fine, send a v2...
I can also take a stab at the Documentation piece, this will make
Boris happy too. Will send a v2 later today.
>
> --
> Josh