Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch to userspace

From	Andy Lutomirski <luto@amacapital.net>
Newsgroups	linux.kernel
Subject	Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch to userspace
Date	2016-10-04 05:10 +0200
Message-ID	<soohb-5b8-5@gated-at.bofh.it> (permalink)
References	(3 earlier) <snCvT-7vS-7@gated-at.bofh.it> <soivc-1li-7@gated-at.bofh.it> <soiY9-1K3-9@gated-at.bofh.it> <soj7P-1Ne-9@gated-at.bofh.it> <sonuN-4Gn-5@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Oct 3, 2016 7:11 PM, "Rik van Riel" <riel@redhat.com> wrote:
>
> On Mon, 2016-10-03 at 14:36 -0700, Andy Lutomirski wrote:
> >
> > Anything else that tries to read task xstate from memory, i.e. MPX
> > and
> > PKRU.  (Although if we switch to eager-switched PKRU, then PKRU stops
> > mattering for this purpose.)
> >
> > Actually, I don't see any way your patches can be compatible with
> > PKRU
> > without switching to eager-switched PKRU.
>
> There is one case where the in-register PKRU state matters:
> - user space accesses to memory
>
> There are several cases where the in-memory PKRU state would
> suffice:
> - get_user_pages(_fast) to the local task (could also use registers)
> - setting VMA/PTE permission bits (could also use registers)
>
> There is one case where only in-memory PKRU state works, where
> PKRU is currently simply ignored:
> - get_user_pages to another task's memory

Also __get_user, etc.  I don't think you want to start playing with
TIF_LOAD_FPU there.  I think tracking PKRU separately and eagerly
loading it with WRPKRU may be the only decent choice here.

>
> Dave, are there major obstacles to making read_pkru and write_pkru
> work with in-memory state?
>
> Would it be better for read/write_pkru to force the FPU state
> to get loaded into registers, under the assumption that if things
> like get_user_pages_fast happens, we will likely switch back to
> userspace soon, anyway?
>
> Would that assumption be wrong with KVM? :)

>
> --
> All Rights Reversed.

Thread

[PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch to userspace riel@redhat.com - 2016-10-01 23:00 +0200
  Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch  to userspace Andy Lutomirski <luto@amacapital.net> - 2016-10-02 01:50 +0200
    Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until  switch to userspace Rik van Riel <riel@redhat.com> - 2016-10-02 02:10 +0200
      Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch  to userspace Andy Lutomirski <luto@amacapital.net> - 2016-10-03 23:00 +0200
        Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until  switch to userspace Rik van Riel <riel@redhat.com> - 2016-10-03 23:30 +0200
          Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch  to userspace Andy Lutomirski <luto@amacapital.net> - 2016-10-03 23:40 +0200
            Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until  switch to userspace Rik van Riel <riel@redhat.com> - 2016-10-04 03:30 +0200
              Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch  to userspace Andy Lutomirski <luto@amacapital.net> - 2016-10-04 04:20 +0200
                Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until  switch to userspace Rik van Riel <riel@redhat.com> - 2016-10-04 04:50 +0200
                Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch  to userspace Andy Lutomirski <luto@amacapital.net> - 2016-10-04 05:10 +0200
                Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch  to userspace Ingo Molnar <mingo@kernel.org> - 2016-10-04 08:40 +0200
                Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until  switch to userspace Rik van Riel <riel@redhat.com> - 2016-10-04 14:50 +0200
            Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until  switch to userspace Rik van Riel <riel@redhat.com> - 2016-10-04 04:20 +0200
              Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch  to userspace Andy Lutomirski <luto@amacapital.net> - 2016-10-04 05:10 +0200
    Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until  switch to userspace Rik van Riel <riel@redhat.com> - 2016-10-02 02:50 +0200
      Re: [PATCH RFC 2/5] x86,fpu: delay FPU register loading until switch  to userspace Dave Hansen <dave.hansen@linux.intel.com> - 2016-10-03 18:30 +0200

csiph-web