Re: [PATCH RESEND] x86/speculation: Fix user-mode spectre-v2 protection with KERNEL_IBRS

[KP Singh]

On Mon, Feb 20, 2023 at 9:59 AM Josh Poimboeuf <jpoimboe@xxxxxxxxxx> wrote:
>
> On Mon, Feb 20, 2023 at 06:46:04PM +0100, Borislav Petkov wrote:
> > On Mon, Feb 20, 2023 at 08:34:42AM -0800, Josh Poimboeuf wrote:
> > > We will never enable IBRS in user space.  We tried that years ago and it
> > > was very slow.
> >
> > Then I don't know what this is trying to fix.
> >
> > We'd need a proper bug statement in the commit message what the problem
> > is. As folks have established, the hw vuln mess is a whole universe of
> > crazy. So without proper documenting, this spaghetti madness will be
> > completely unreadable.
>
> Agreed, and there seems to be a lot of confusion around this patch.  I
> think I understand the issue, let me write up a new patch which is
> hopefully clearer.

Feel free to write a patch, but I don't get the confusion.

Well, we disable IBRS userspace (this is KENREL_IBRS), because it is
slow. Now if a user space process wants to protect itself from cross
thread training, it should be able to do it, either by turning STIBP
always on or using a prctl to enable. With the current logic, it's
unable to do so. All of this is mentioned in the commit message.

>
> --
> Josh