x86-64: fix thread tracing

The current instrumentation point for CONFIG_TRACING added in
PR zephyrproject-rtos#28512 had two problems:

- If userspace and KPTI are enabled, the tracing point is simply
  never run if we are resuming a user thread as the
  z_x86_trampoline_to_user function is jumped to and calls
  'iret' from there

- Only %rdi is being saved. However, at that location, *all*
  caller-saved registers are in use as they contain the
  resumed thread's context

Simplest solution is to move this up near where we update page
tables. The #ifdefs are used to make sure we don't push/pop
%rdi more than once. At that point in the code only %rdi
is in use among the volatile registers.

Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>

Author: Andrew Boie

arch/x86/core/intel64/locore.S

@@ -324,17 +324,23 @@ z_x86_switch:
  */
 
 __resume:
-#ifdef CONFIG_USERSPACE
-#ifndef CONFIG_X86_KPTI
+#if (!defined(CONFIG_X86_KPTI) && defined(CONFIG_USERSPACE)) \
+		|| defined(CONFIG_TRACING)
+	pushq %rdi	/* Caller-saved, stash it */
+#if !defined(CONFIG_X86_KPTI) && defined(CONFIG_USERSPACE)
 	/* If KPTI is enabled we're always on the kernel's page tables in
 	 * this context and the appropriate page table switch takes place
 	 * when trampolining back to user mode
 	 */
-	pushq %rdi	/* Caller-saved, stash it */
 	call z_x86_swap_update_page_tables
+#endif
+#ifdef CONFIG_TRACING
+	call sys_trace_thread_switched_in
+#endif
 	popq %rdi
-#endif /* CONFIG_X86_KPTI */
+#endif /* (!CONFIG_X86_KPTI && CONFIG_USERSPACE) || CONFIG_TRACING */
 
+#ifdef CONFIG_USERSPACE
 	/* Set up exception return stack frame */
 	pushq _thread_offset_to_ss(%rdi)	/* SS */
 #else
@@ -402,12 +408,6 @@ __resume:
 	/* swapgs variant of Spectre V1. Disable speculation past this point */
 	lfence
 #endif /* CONFIG_X86_BOUNDS_CHECK_BYPASS_MITIGATION */
-
-#ifdef CONFIG_TRACING
-	pushq %rdi
-	call sys_trace_thread_switched_in
-	popq %rdi
-#endif
 	iretq