Re: [PATCH 2/2] i387: support lazy restore of FPU state

[Ingo Molnar]

* Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:

> +DECLARE_PER_CPU(struct task_struct *, fpu_owner_task);

> +static inline int fpu_lazy_restore(struct task_struct *new, unsigned int cpu)
> +{
> +	return new == percpu_read_stable(fpu_owner_task) &&
> +		cpu == new->thread.fpu.last_cpu;
> +}

I guess the CPU hotplug case deserves a comment in the code: CPU 
hotplug + replug of the same (but meanwhile reset) CPU is safe 
because fpu_owner_task[cpu] gets reset to NULL.

Likewise, if we hot-unplug two CPUs and then insert the second 
one, it's the same CPU index but not the same FPU state anymore. 
There too the CPU hotplug code resetting fpu_owner_task to NULL 
makes this optimization safe.

Thanks,

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/