[PATCH 31/33] x86/fpu: Simplify and speed up fpu__copy()

From: Ingo Molnar
Date: Sat Sep 23 2017 - 09:02:00 EST

fpu__copy() has a preempt_disable()/enable() pair, which it had to do to
be able to atomically unlazy the current task when doing an FNSAVE.

But we don't unlazy tasks anymore, we always do direct saves/restores of
FPU context.

So remove both the unnecessary critical section, and update the comments.

Cc: Andy Lutomirski <luto@xxxxxxxxxxxxxx>
Cc: Borislav Petkov <bp@xxxxxxxxx>
Cc: Eric Biggers <ebiggers3@xxxxxxxxx>
Cc: Fenghua Yu <fenghua.yu@xxxxxxxxx>
Cc: H. Peter Anvin <hpa@xxxxxxxxx>
Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
Cc: Oleg Nesterov <oleg@xxxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxxxxx>
arch/x86/kernel/fpu/core.c | 15 +++------------
1 file changed, 3 insertions(+), 12 deletions(-)

diff --git a/arch/x86/kernel/fpu/core.c b/arch/x86/kernel/fpu/core.c
index 4acfc0ebc160..e935f8faf287 100644
--- a/arch/x86/kernel/fpu/core.c
+++ b/arch/x86/kernel/fpu/core.c
@@ -206,22 +206,13 @@ int fpu__copy(struct fpu *dst_fpu, struct fpu *src_fpu)
* Save current FPU registers directly into the child
* FPU context, without any memory-to-memory copying.
- * We have to do all this with preemption disabled,
- * mostly because of the FNSAVE case, because in that
- * case we must not allow preemption in the window
- * between the FNSAVE and us marking the context lazy.
- *
- * It shouldn't be an issue as even FNSAVE is plenty
- * fast in terms of critical section length.
+ * ( The function 'fails' in the FNSAVE case, which destroys
+ * register contents so we have to copy them back. )
- preempt_disable();
if (!copy_fpregs_to_fpstate(dst_fpu)) {
- memcpy(&src_fpu->state, &dst_fpu->state,
- fpu_kernel_xstate_size);
+ memcpy(&src_fpu->state, &dst_fpu->state, fpu_kernel_xstate_size);
- preempt_enable();