Re: [GIT PULL] x86 fixes

From: Ingo Molnar

Date: Sun Dec 14 2025 - 02:51:57 EST



* Borislav Petkov <bp@xxxxxxxxx> wrote:

> On Sat, Dec 13, 2025 at 11:00:11AM +0100, Ingo Molnar wrote:
> > Yeah, but that's not really how we apply -stable tags. If it's good for
> > -stable, it's good for Linus's tree, full stop.
>
> Full stop, schmulstop... I'm not going to send a not as much as possibly
> tested fix to Linus for something which was a nasty pain to get right
> a couple of times because of a bunch of SNAFUs by BIOS and Qemu and
> whatnot, just to break other machines and then hurry and scramble to fix
> it again.

It's Linus's preference: a couple of years ago, when I did something
similar to what you did here Linus requested that fixes with -stable
tags not live in -next indefinitely, but be sent to his tree.
-next should not be a dumping ground for long-term testing.

> Rather, I'd let it cook for a whole cycle so that we're sure.

If a fix is important enough to get a -stable tag, it should go
upstream sooner rather than later.

Thanks,

Ingo