Re: [RFC 1/3] x86: Add this_cpu_sp0() to read sp0 for the current cpu

From: Andy Lutomirski
Date: Wed Mar 04 2015 - 13:52:13 EST


On Wed, Mar 4, 2015 at 12:02 AM, Ingo Molnar <mingo@xxxxxxxxxx> wrote:
>
> * Andy Lutomirski <luto@xxxxxxxxxxxxxx> wrote:
>
>> We currently store references to the top of the kernel stack in
>> multiple places: kernel_stack (with an offset) and
>> init_tss.x86_tss.sp0 (no offset). The latter is defined by hardware
>> and is a clean canonical way to find the top of the stack. Add an
>
> Btw., 'per_cpu(init_tss)' is a somewhat misleading name these days, as
> there's nothing 'init' about it anymore - we load it during CPU init
> and then manually maintain its contents. A better name would be
> 'current_tss' - referring to both the current CPU and the current
> task?

Hmm. That seems a little odd to me, since we never change the TSS
pointer. It's certainly better than init_tss, though. I'll add a
followup to rename it.

>
>> This needs minor paravirt tweaks to ensure that On native, sp0
>
> nit: s/On/on/
>
>> defines the top of the kernel stack. On Xen and lguest, the
>> hypervisor tracks it, but we want to start reading sp0 in the
>> kernel. Fixing this is simple: just update our local copy of sp0 as
>> well as the hypervisor's copy on task switches.

Thanks,
Andy
--
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/