i387: ptrace breaks the lazy-fpu-restore logic
authorOleg Nesterov <oleg@redhat.com>
Mon, 16 Apr 2012 20:48:15 +0000 (22:48 +0200)
committerH. Peter Anvin <hpa@linux.intel.com>
Mon, 16 Apr 2012 23:23:59 +0000 (16:23 -0700)
commit089f9fba56faf33cc6dd2a6442b7ac92c58b8209
treefa86fb7efe7253f344fca3f0395aee2d92d13fa1
parent68894632afb2729a1d8785c877840953894c7283
i387: ptrace breaks the lazy-fpu-restore logic

Starting from 7e16838d "i387: support lazy restore of FPU state"
we assume that fpu_owner_task doesn't need restore_fpu_checking()
on the context switch, its FPU state should match what we already
have in the FPU on this CPU.

However, debugger can change the tracee's FPU state, in this case
we should reset fpu.last_cpu to ensure fpu_lazy_restore() can't
return true.

Change init_fpu() to do this, it is called by user_regset->set()
methods.

Reported-by: Jan Kratochvil <jan.kratochvil@redhat.com>
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Link: http://lkml.kernel.org/r/20120416204815.GB24884@redhat.com
Cc: <stable@vger.kernel.org> v3.3
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
arch/x86/kernel/i387.c