Re: arch_set_user_pkey_access only works on the current task_struct

[Thomas Gleixner]

On Tue, Jun 08 2021 at 11:16, liangjs wrote:
> On Mon, 2021-06-07 at 10:52 -0700, Dave Hansen wrote:
>> On 6/5/21 6:10 AM, Jiashuo Liang wrote:
>> > I am learning the kernel implementation of the x86 PKU feature. I find the
>> > arch_set_user_pkey_access function in arch/x86/kernel/fpu/xstate.c does not
>> > use its first parameter. So it is perhaps a bug?
>> 
>> I wouldn't really call it a bug.  But, yes, it is something we should
>> clean up.
>
> Should we remove the tsk parameter, or allow it to change the PKRU of tsk?
>
> By the way, we are calling write_pkru, which changes both the CPU's PKRU
> and the xsave one. Why is this necessary?

Because PKRU is xstate managed and there is the requirement to keep both
up to to date. There is work in progress to clean this up.

> If I want to change PKRU of a task_struct other than current, do I still
> need to call __write_pkru?

Of course not, but you _cannot_ safely update a different tasks PKRU
value except through ptrace which guarantees that the task is scheduled
out and stays that way until ptrace releases it again.

So tsk != current cannot work which means the function argument can just
go away.

Thanks,

        tglx