Merge branch 'pm-cpuidle'

Fix an issue in the PSCI cpuidle driver introduced recently and a nasty
x86 power regression introduced in 6.15:

 - Prevent freeing an uninitialized pointer in error path of
   dt_idle_state_present() in the PSCI cpuidle driver (Dan Carpenter).

 - Revert an x86 commit that went into 6.15 and caused idle power,
   including power in suspend-to-idle, to rise rather dramatically on
   systems booting with "nosmt" in the kernel command line (Rafael Wysocki).

* pm-cpuidle:
  Revert "x86/smp: Eliminate mwait_play_dead_cpuid_hint()"
  cpuidle: psci: Fix uninitialized variable in dt_idle_state_present()

Author: rafaeljw

arch/x86/kernel/smpboot.c

@@ -1244,10 +1244,6 @@ void play_dead_common(void)
 	local_irq_disable();
 }
 
-/*
- * We need to flush the caches before going to sleep, lest we have
- * dirty data in our caches when we come back up.
- */
 void __noreturn mwait_play_dead(unsigned int eax_hint)
 {
 	struct mwait_cpu_dead *md = this_cpu_ptr(&mwait_cpu_dead);
@@ -1293,6 +1289,50 @@ void __noreturn mwait_play_dead(unsigned int eax_hint)
 	}
 }
 
+/*
+ * We need to flush the caches before going to sleep, lest we have
+ * dirty data in our caches when we come back up.
+ */
+static inline void mwait_play_dead_cpuid_hint(void)
+{
+	unsigned int eax, ebx, ecx, edx;
+	unsigned int highest_cstate = 0;
+	unsigned int highest_subcstate = 0;
+	int i;
+
+	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD ||
+	    boot_cpu_data.x86_vendor == X86_VENDOR_HYGON)
+		return;
+	if (!this_cpu_has(X86_FEATURE_MWAIT))
+		return;
+	if (!this_cpu_has(X86_FEATURE_CLFLUSH))
+		return;
+
+	eax = CPUID_LEAF_MWAIT;
+	ecx = 0;
+	native_cpuid(&eax, &ebx, &ecx, &edx);
+
+	/*
+	 * eax will be 0 if EDX enumeration is not valid.
+	 * Initialized below to cstate, sub_cstate value when EDX is valid.
+	 */
+	if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED)) {
+		eax = 0;
+	} else {
+		edx >>= MWAIT_SUBSTATE_SIZE;
+		for (i = 0; i < 7 && edx; i++, edx >>= MWAIT_SUBSTATE_SIZE) {
+			if (edx & MWAIT_SUBSTATE_MASK) {
+				highest_cstate = i;
+				highest_subcstate = edx & MWAIT_SUBSTATE_MASK;
+			}
+		}
+		eax = (highest_cstate << MWAIT_SUBSTATE_SIZE) |
+			(highest_subcstate - 1);
+	}
+
+	mwait_play_dead(eax);
+}
+
 /*
  * Kick all "offline" CPUs out of mwait on kexec(). See comment in
  * mwait_play_dead().
@@ -1343,9 +1383,9 @@ void native_play_dead(void)
 	play_dead_common();
 	tboot_shutdown(TB_SHUTDOWN_WFS);
 
-	/* Below returns only on error. */
-	cpuidle_play_dead();
-	hlt_play_dead();
+	mwait_play_dead_cpuid_hint();
+	if (cpuidle_play_dead())
+		hlt_play_dead();
 }
 
 #else /* ... !CONFIG_HOTPLUG_CPU */

drivers/cpuidle/cpuidle-psci.c

@@ -456,14 +456,13 @@ static struct faux_device_ops psci_cpuidle_ops = {
 
 static bool __init dt_idle_state_present(void)
 {
-	struct device_node *cpu_node __free(device_node);
-	struct device_node *state_node __free(device_node);
-
-	cpu_node = of_cpu_device_node_get(cpumask_first(cpu_possible_mask));
+	struct device_node *cpu_node __free(device_node) =
+			of_cpu_device_node_get(cpumask_first(cpu_possible_mask));
 	if (!cpu_node)
 		return false;
 
-	state_node = of_get_cpu_state_node(cpu_node, 0);
+	struct device_node *state_node __free(device_node) =
+			of_get_cpu_state_node(cpu_node, 0);
 	if (!state_node)
 		return false;