Tim,
On Mon, 19 Nov 2018, Tim Chen wrote:
> On 11/19/2018 06:57 AM, Peter Zijlstra wrote:
> > In particular, the SMT topology bits are set before we enable
> > interrrupts and similarly, are cleared after we disable interrupts for
> > the last time and die.
>
>
> Peter & Thomas,
>
> Any objection if I export sched_smt_present after including
> Peter's patch and use it in spec_ctrl_update_msr instead.
>
> Something like this?
>
> +#ifdef CONFIG_SCHED_SMT
> +
> +extern struct static_key_false sched_smt_present;
> +
> +static inline bool cpu_smt_present(void)
> +{
> + if (static_branch_unlikely(&sched_smt_present))
> + return true;
> + else
> + return false;
What's wrong with
return static_branch_unlikely(&sched_smt_present);
???
But that's just a stylistic nitpick. The real issue is that you prevent the
mitigation when CONFIG_SCHED_SMT=n.
Thanks,
tglx
