[MODERATED] Re: [PATCH v6] boot-time control

[Josh Poimboeuf <jpoimboe@redhat.com>]

On Fri, Jul 06, 2018 at 10:19:39PM +0200, speck for Jiri Kosina wrote:
> On Fri, 6 Jul 2018, speck for Josh Poimboeuf wrote:
> 
> > Will you be doing the sysfs control next?
> 
> Once this is in, I am planning to do so, but this alone has bigger 
> priority on my list, so I want to have that in first.
> 
> [ ... snip the stylistic fixes / typos ... ]
> > > +			case L1TF_MITIGATION_FULL:
> > > +				/*
> > > +				 * Warn if SMT has been runtime-enabled. We can't
> > > +				 * really warn only once here, as SMT state is
> > > +				 * not constant.
> > > +				 */
> > > +				if (cpu_smt_control == CPU_SMT_ENABLED)
> > > +					pr_warn(L1TF_MSG_SMT);
> > 
> > I think I said this sounded ok on IRC, but...
> > 
> > This repeated warning will be very annoying for those people enabling 
> > SMT on purpose.  I think a one-time warning would be better.
> 
> No problem with that.
> 
> On Fri, 6 Jul 2018, speck for Luck, Tony wrote:
> 
> > If you feel that you must have the CVE here, 
> 
> I don't; if Intel has any better suggestions that would be equally 
> alerting the sysadmin, let's change it to whatever works the best.
> 
> > then shouldn't you have CVE-2018-3646 which covers the virtualization 
> > aspects of L1TF rather than CVE-2018-3620 which just covers OS and SMM 
> > implications?
> 
> Good catch, I copied the CVE number from what we already have in speck.git 
> (via 26acfb666a473) and didn't cross-check it. So either it has to be 
> changed via this (modified(*)) patch, or it should be fixed independetly 
> in speck.git.
> 
> (*) Thomas, all the comments so far are rather cosmetic ... if nothing 
>     else pops up, do you prefer to fix that up on your side when applying, 
>     or should I send v8?

With the above changed to a one-time warning (with comment adjusted
accordingly) and typos fixed:

Reviewed-by: Josh Poimboeuf <jpoimboe@redhat.com>

-- 
Josh