Re: [patch V8 13/15] SSB 13

[Thomas Gleixner <tglx@linutronix.de>]

On Tue, 1 May 2018, speck for Thomas Gleixner wrote:

> On Mon, 30 Apr 2018, speck for Kees Cook wrote:
> 
> > On Mon, Apr 30, 2018 at 05:04:36PM +0200, speck for Thomas Gleixner wrote:
> > > Subject: [patch V8 13/15] prctl: Add speculation control prctls
> > > From: Thomas Gleixner <tglx@linutronix.de>
> > > 
> > > Add two new prctls to control aspects of speculation related vulnerabilites
> > > and their mitigations to provide finer grained control over performance
> > > impacting mitigations.
> > > 
> > > PR_GET_SPECULATION_CTRL returns the state of the speculation misfeature
> > > which is selected with arg2 of prctl(2). The return value uses bit 0-2 with
> > > the following meaning:
> > > 
> > > Bit  Define           Description
> > > 0    PR_SPEC_PRCTL    Mitigation can be controlled per task by
> > >                       PR_SET_SPECULATION_CTRL
> > > 1    PR_SPEC_ENABLE   The speculation feature is enabled, mitigation is
> > >                       disabled
> > > 2    PR_SPEC_DISABLE  The speculation feature is disabled, mitigation is
> > >                       enabled
> > 
> > Err, what does it mean to have both bits 1 and 2 set?
> 
> The bits are either or. Both set are invalid. That's a leftover from
> something I experimented with earlier.
> 
> But yes, we can consolidate them into one.

Thinking about it with enough coffee applied, I know why I kept it that
way. If we just have a single bit then there is no way to check whether the
CPU is affected or not.

Bit 0  		Bit 1

0		0		PRCTL disabled, Speculation enabled
1		0		PRCTL enabled, Speculation enabled
0		1		PRCTL disabled, Speculation disabled
1		1		PRCTL enabled, Speculation disabled

with the 3 bits:

Bit 0  	Bit 1	Bit 2

0	0	0		CPU not affected
0	1	0		PRCTL disabled, Speculation enabled
1	1	0		PRCTL enaabled, Speculation enabled
0	0	1		PRCTL disabled, Speculation disabled
1	0	1		PRCTL enabled, Speculation disabled
1	1	1		Invalid (wont be returned ever)

Thanks,

	tglx