Merge branch 'x86-apic-for-linus' of git://git.kernel.org/pub/scm/lin…

…ux/kernel/git/tip/tip

Pull x96 apic updates from Thomas Gleixner:
 "Updates for the x86 APIC interrupt handling and APIC timer:

   - Fix a long standing issue with spurious interrupts which was caused
     by the big vector management rework a few years ago. Robert Hodaszi
     provided finally enough debug data and an excellent initial failure
     analysis which allowed to understand the underlying issues.

     This contains a change to the core interrupt management code which
     is required to handle this correctly for the APIC/IO_APIC. The core
     changes are NOOPs for most architectures except ARM64. ARM64 is not
     impacted by the change as confirmed by Marc Zyngier.

   - Newer systems allow to disable the PIT clock for power saving
     causing panic in the timer interrupt delivery check of the IO/APIC
     when the HPET timer is not enabled either. While the clock could be
     turned on this would cause an endless whack a mole game to chase
     the proper register in each affected chipset.

     These systems provide the relevant frequencies for TSC, CPU and the
     local APIC timer via CPUID and/or MSRs, which allows to avoid the
     PIT/HPET based calibration. As the calibration code is the only
     usage of the legacy timers on modern systems and is skipped anyway
     when the frequencies are known already, there is no point in
     setting up the PIT and actually checking for the interrupt delivery
     via IO/APIC.

     To achieve this on a wide variety of platforms, the CPUID/MSR based
     frequency readout has been made more robust, which also allowed to
     remove quite some workarounds which turned out to be not longer
     required. Thanks to Daniel Drake for analysis, patches and
     verification"

* 'x86-apic-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/irq: Seperate unused system vectors from spurious entry again
  x86/irq: Handle spurious interrupt after shutdown gracefully
  x86/ioapic: Implement irq_get_irqchip_state() callback
  genirq: Add optional hardware synchronization for shutdown
  genirq: Fix misleading synchronize_irq() documentation
  genirq: Delay deactivation in free_irq()
  x86/timer: Skip PIT initialization on modern chipsets
  x86/apic: Use non-atomic operations when possible
  x86/apic: Make apic_bsp_setup() static
  x86/tsc: Set LAPIC timer period to crystal clock frequency
  x86/apic: Rename 'lapic_timer_frequency' to 'lapic_timer_period'
  x86/tsc: Use CPUID.0x16 to calculate missing crystal frequency

arch/x86/entry/entry_32.S

@@ -1104,6 +1104,30 @@ ENTRY(irq_entries_start)
     .endr
 END(irq_entries_start)
 
+#ifdef CONFIG_X86_LOCAL_APIC
+	.align 8
+ENTRY(spurious_entries_start)
+    vector=FIRST_SYSTEM_VECTOR
+    .rept (NR_VECTORS - FIRST_SYSTEM_VECTOR)
+	pushl	$(~vector+0x80)			/* Note: always in signed byte range */
+    vector=vector+1
+	jmp	common_spurious
+	.align	8
+    .endr
+END(spurious_entries_start)
+
+common_spurious:
+	ASM_CLAC
+	addl	$-0x80, (%esp)			/* Adjust vector into the [-256, -1] range */
+	SAVE_ALL switch_stacks=1
+	ENCODE_FRAME_POINTER
+	TRACE_IRQS_OFF
+	movl	%esp, %eax
+	call	smp_spurious_interrupt
+	jmp	ret_from_intr
+ENDPROC(common_interrupt)
+#endif
+
 /*
  * the CPU automatically disables interrupts when executing an IRQ vector,
  * so IRQ-flags tracing has to follow that:

arch/x86/entry/entry_64.S

@@ -375,6 +375,18 @@ ENTRY(irq_entries_start)
     .endr
 END(irq_entries_start)
 
+	.align 8
+ENTRY(spurious_entries_start)
+    vector=FIRST_SYSTEM_VECTOR
+    .rept (NR_VECTORS - FIRST_SYSTEM_VECTOR)
+	UNWIND_HINT_IRET_REGS
+	pushq	$(~vector+0x80)			/* Note: always in signed byte range */
+	jmp	common_spurious
+	.align	8
+	vector=vector+1
+    .endr
+END(spurious_entries_start)
+
 .macro DEBUG_ENTRY_ASSERT_IRQS_OFF
 #ifdef CONFIG_DEBUG_ENTRY
 	pushq %rax
@@ -571,10 +583,20 @@ _ASM_NOKPROBE(interrupt_entry)
 
 /* Interrupt entry/exit. */
 
-	/*
-	 * The interrupt stubs push (~vector+0x80) onto the stack and
-	 * then jump to common_interrupt.
-	 */
+/*
+ * The interrupt stubs push (~vector+0x80) onto the stack and
+ * then jump to common_spurious/interrupt.
+ */
+common_spurious:
+	addq	$-0x80, (%rsp)			/* Adjust vector to [-256, -1] range */
+	call	interrupt_entry
+	UNWIND_HINT_REGS indirect=1
+	call	smp_spurious_interrupt		/* rdi points to pt_regs */
+	jmp	ret_from_intr
+END(common_spurious)
+_ASM_NOKPROBE(common_spurious)
+
+/* common_interrupt is a hotpath. Align it */
 	.p2align CONFIG_X86_L1_CACHE_SHIFT
 common_interrupt:
 	addq	$-0x80, (%rsp)			/* Adjust vector to [-256, -1] range */

arch/x86/include/asm/apic.h

@@ -53,7 +53,7 @@ extern unsigned int apic_verbosity;
 extern int local_apic_timer_c2_ok;
 
 extern int disable_apic;
-extern unsigned int lapic_timer_frequency;
+extern unsigned int lapic_timer_period;
 
 extern enum apic_intr_mode_id apic_intr_mode;
 enum apic_intr_mode_id {
@@ -155,7 +155,6 @@ static inline int apic_force_enable(unsigned long addr)
 extern int apic_force_enable(unsigned long addr);
 #endif
 
-extern void apic_bsp_setup(bool upmode);
 extern void apic_ap_setup(void);
 
 /*
@@ -175,6 +174,7 @@ extern void lapic_assign_system_vectors(void);
 extern void lapic_assign_legacy_vector(unsigned int isairq, bool replace);
 extern void lapic_online(void);
 extern void lapic_offline(void);
+extern bool apic_needs_pit(void);
 
 #else /* !CONFIG_X86_LOCAL_APIC */
 static inline void lapic_shutdown(void) { }
@@ -188,6 +188,7 @@ static inline void init_bsp_APIC(void) { }
 static inline void apic_intr_mode_init(void) { }
 static inline void lapic_assign_system_vectors(void) { }
 static inline void lapic_assign_legacy_vector(unsigned int i, bool r) { }
+static inline bool apic_needs_pit(void) { return true; }
 #endif /* !CONFIG_X86_LOCAL_APIC */
 
 #ifdef CONFIG_X86_X2APIC

arch/x86/include/asm/hw_irq.h

@@ -150,8 +150,11 @@ extern char irq_entries_start[];
 #define trace_irq_entries_start irq_entries_start
 #endif
 
+extern char spurious_entries_start[];
+
 #define VECTOR_UNUSED		NULL
-#define VECTOR_RETRIGGERED	((void *)~0UL)
+#define VECTOR_SHUTDOWN		((void *)~0UL)
+#define VECTOR_RETRIGGERED	((void *)~1UL)
 
 typedef struct irq_desc* vector_irq_t[NR_VECTORS];
 DECLARE_PER_CPU(vector_irq_t, vector_irq);

arch/x86/include/asm/time.h

@@ -7,6 +7,7 @@
 
 extern void hpet_time_init(void);
 extern void time_init(void);
+extern bool pit_timer_init(void);
 
 extern struct clock_event_device *global_clock_event;
 

arch/x86/kernel/apic/apic.c

@@ -195,7 +195,7 @@ static struct resource lapic_resource = {
 	.flags = IORESOURCE_MEM | IORESOURCE_BUSY,
 };
 
-unsigned int lapic_timer_frequency = 0;
+unsigned int lapic_timer_period = 0;
 
 static void apic_pm_activate(void);
 
@@ -501,7 +501,7 @@ lapic_timer_set_periodic_oneshot(struct clock_event_device *evt, bool oneshot)
 	if (evt->features & CLOCK_EVT_FEAT_DUMMY)
 		return 0;
 
-	__setup_APIC_LVTT(lapic_timer_frequency, oneshot, 1);
+	__setup_APIC_LVTT(lapic_timer_period, oneshot, 1);
 	return 0;
 }
 
@@ -805,11 +805,11 @@ calibrate_by_pmtimer(long deltapm, long *delta, long *deltatsc)
 
 static int __init lapic_init_clockevent(void)
 {
-	if (!lapic_timer_frequency)
+	if (!lapic_timer_period)
 		return -1;
 
 	/* Calculate the scaled math multiplication factor */
-	lapic_clockevent.mult = div_sc(lapic_timer_frequency/APIC_DIVISOR,
+	lapic_clockevent.mult = div_sc(lapic_timer_period/APIC_DIVISOR,
 					TICK_NSEC, lapic_clockevent.shift);
 	lapic_clockevent.max_delta_ns =
 		clockevent_delta2ns(0x7FFFFFFF, &lapic_clockevent);
@@ -821,6 +821,33 @@ static int __init lapic_init_clockevent(void)
 	return 0;
 }
 
+bool __init apic_needs_pit(void)
+{
+	/*
+	 * If the frequencies are not known, PIT is required for both TSC
+	 * and apic timer calibration.
+	 */
+	if (!tsc_khz || !cpu_khz)
+		return true;
+
+	/* Is there an APIC at all? */
+	if (!boot_cpu_has(X86_FEATURE_APIC))
+		return true;
+
+	/* Deadline timer is based on TSC so no further PIT action required */
+	if (boot_cpu_has(X86_FEATURE_TSC_DEADLINE_TIMER))
+		return false;
+
+	/* APIC timer disabled? */
+	if (disable_apic_timer)
+		return true;
+	/*
+	 * The APIC timer frequency is known already, no PIT calibration
+	 * required. If unknown, let the PIT be initialized.
+	 */
+	return lapic_timer_period == 0;
+}
+
 static int __init calibrate_APIC_clock(void)
 {
 	struct clock_event_device *levt = this_cpu_ptr(&lapic_events);
@@ -839,7 +866,7 @@ static int __init calibrate_APIC_clock(void)
 	 */
 	if (!lapic_init_clockevent()) {
 		apic_printk(APIC_VERBOSE, "lapic timer already calibrated %d\n",
-			    lapic_timer_frequency);
+			    lapic_timer_period);
 		/*
 		 * Direct calibration methods must have an always running
 		 * local APIC timer, no need for broadcast timer.
@@ -884,13 +911,13 @@ static int __init calibrate_APIC_clock(void)
 	pm_referenced = !calibrate_by_pmtimer(lapic_cal_pm2 - lapic_cal_pm1,
 					&delta, &deltatsc);
 
-	lapic_timer_frequency = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS;
+	lapic_timer_period = (delta * APIC_DIVISOR) / LAPIC_CAL_LOOPS;
 	lapic_init_clockevent();
 
 	apic_printk(APIC_VERBOSE, "..... delta %ld\n", delta);
 	apic_printk(APIC_VERBOSE, "..... mult: %u\n", lapic_clockevent.mult);
 	apic_printk(APIC_VERBOSE, "..... calibration result: %u\n",
-		    lapic_timer_frequency);
+		    lapic_timer_period);
 
 	if (boot_cpu_has(X86_FEATURE_TSC)) {
 		apic_printk(APIC_VERBOSE, "..... CPU clock speed is "
@@ -901,13 +928,13 @@ static int __init calibrate_APIC_clock(void)
 
 	apic_printk(APIC_VERBOSE, "..... host bus clock speed is "
 		    "%u.%04u MHz.\n",
-		    lapic_timer_frequency / (1000000 / HZ),
-		    lapic_timer_frequency % (1000000 / HZ));
+		    lapic_timer_period / (1000000 / HZ),
+		    lapic_timer_period % (1000000 / HZ));
 
 	/*
 	 * Do a sanity check on the APIC calibration result
 	 */
-	if (lapic_timer_frequency < (1000000 / HZ)) {
+	if (lapic_timer_period < (1000000 / HZ)) {
 		local_irq_enable();
 		pr_warning("APIC frequency too slow, disabling apic timer\n");
 		return -1;
@@ -1351,6 +1378,8 @@ void __init init_bsp_APIC(void)
 	apic_write(APIC_LVT1, value);
 }
 
+static void __init apic_bsp_setup(bool upmode);
+
 /* Init the interrupt delivery mode for the BSP */
 void __init apic_intr_mode_init(void)
 {
@@ -2041,21 +2070,32 @@ __visible void __irq_entry smp_spurious_interrupt(struct pt_regs *regs)
 	entering_irq();
 	trace_spurious_apic_entry(vector);
 
+	inc_irq_stat(irq_spurious_count);
+
+	/*
+	 * If this is a spurious interrupt then do not acknowledge
+	 */
+	if (vector == SPURIOUS_APIC_VECTOR) {
+		/* See SDM vol 3 */
+		pr_info("Spurious APIC interrupt (vector 0xFF) on CPU#%d, should never happen.\n",
+			smp_processor_id());
+		goto out;
+	}
+
 	/*
-	 * Check if this really is a spurious interrupt and ACK it
-	 * if it is a vectored one.  Just in case...
-	 * Spurious interrupts should not be ACKed.
+	 * If it is a vectored one, verify it's set in the ISR. If set,
+	 * acknowledge it.
 	 */
 	v = apic_read(APIC_ISR + ((vector & ~0x1f) >> 1));
-	if (v & (1 << (vector & 0x1f)))
+	if (v & (1 << (vector & 0x1f))) {
+		pr_info("Spurious interrupt (vector 0x%02x) on CPU#%d. Acked\n",
+			vector, smp_processor_id());
 		ack_APIC_irq();
-
-	inc_irq_stat(irq_spurious_count);
-
-	/* see sw-dev-man vol 3, chapter 7.4.13.5 */
-	pr_info("spurious APIC interrupt through vector %02x on CPU#%d, "
-		"should never happen.\n", vector, smp_processor_id());
-
+	} else {
+		pr_info("Spurious interrupt (vector 0x%02x) on CPU#%d. Not pending!\n",
+			vector, smp_processor_id());
+	}
+out:
 	trace_spurious_apic_exit(vector);
 	exiting_irq();
 }
@@ -2416,11 +2456,8 @@ static void __init apic_bsp_up_setup(void)
 /**
  * apic_bsp_setup - Setup function for local apic and io-apic
  * @upmode:		Force UP mode (for APIC_init_uniprocessor)
- *
- * Returns:
- * apic_id of BSP APIC
  */
-void __init apic_bsp_setup(bool upmode)
+static void __init apic_bsp_setup(bool upmode)
 {
 	connect_bsp_APIC();
 	if (upmode)

arch/x86/kernel/apic/apic_flat_64.c

@@ -78,7 +78,7 @@ flat_send_IPI_mask_allbutself(const struct cpumask *cpumask, int vector)
 	int cpu = smp_processor_id();
 
 	if (cpu < BITS_PER_LONG)
-		clear_bit(cpu, &mask);
+		__clear_bit(cpu, &mask);
 
 	_flat_send_IPI_mask(mask, vector);
 }
@@ -92,7 +92,7 @@ static void flat_send_IPI_allbutself(int vector)
 			unsigned long mask = cpumask_bits(cpu_online_mask)[0];
 
 			if (cpu < BITS_PER_LONG)
-				clear_bit(cpu, &mask);
+				__clear_bit(cpu, &mask);
 
 			_flat_send_IPI_mask(mask, vector);
 		}