[RFC PATCH 05/11] x86,fpu: ensure FPU state is reloaded from memory if task is traced

[riel@redhat.com]

From: Rik van Riel <riel@redhat.com>

If the old task is in a state where its FPU state could be changed by
a debugger, ensure the FPU state is always restored from memory on the
next context switch.

Currently the system only skips FPU reloads when !eager_fpu_mode()
and the task's FPU state is still loaded on the CPU.

Signed-off-by: Rik van Riel <riel@redhat.com>
---
 arch/x86/include/asm/fpu-internal.h | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/arch/x86/include/asm/fpu-internal.h b/arch/x86/include/asm/fpu-internal.h
index 539b050..4db8781 100644
--- a/arch/x86/include/asm/fpu-internal.h
+++ b/arch/x86/include/asm/fpu-internal.h
@@ -412,8 +412,14 @@ static inline void switch_fpu_prepare(struct task_struct *old, struct task_struc
 	bool preload = tsk_used_math(new) && (use_eager_fpu() ||
 					     new->thread.fpu_counter > 5);
 	if (__thread_has_fpu(old)) {
-		if (!__save_init_fpu(old))
+		/*
+		 * Make sure the FPU state is restored from memory next time,
+		 * if the task has an FPU exception pending, or the task's in
+		 * memory FPU state could be changed by a debugger.
+		 */
+		if (!__save_init_fpu(old) || task_is_stopped_or_traced(old))
 			cpu = ~0;
+
 		old->thread.fpu.last_cpu = cpu;
 		old->thread.fpu.has_fpu = 0;	/* But leave fpu_owner_task! */
 
-- 
1.9.3