Merge tag 'x86-urgent-2020-02-09' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 fixes from Thomas Gleixner:
 "A set of fixes for X86:

   - Ensure that the PIT is set up when the local APIC is disable or
     configured in legacy mode. This is caused by an ordering issue
     introduced in the recent changes which skip PIT initialization when
     the TSC and APIC frequencies are already known.

   - Handle malformed SRAT tables during early ACPI parsing which caused
     an infinite loop anda boot hang.

   - Fix a long standing race in the affinity setting code which affects
     PCI devices with non-maskable MSI interrupts. The problem is caused
     by the non-atomic writes of the MSI address (destination APIC id)
     and data (vector) fields which the device uses to construct the MSI
     message. The non-atomic writes are mandated by PCI.

     If both fields change and the device raises an interrupt after
     writing address and before writing data, then the MSI block
     constructs a inconsistent message which causes interrupts to be
     lost and subsequent malfunction of the device.

     The fix is to redirect the interrupt to the new vector on the
     current CPU first and then switch it over to the new target CPU.
     This allows to observe an eventually raised interrupt in the
     transitional stage (old CPU, new vector) to be observed in the APIC
     IRR and retriggered on the new target CPU and the new vector.

     The potential spurious interrupts caused by this are harmless and
     can in the worst case expose a buggy driver (all handlers have to
     be able to deal with spurious interrupts as they can and do happen
     for various reasons).

   - Add the missing suspend/resume mechanism for the HYPERV hypercall
     page which prevents resume hibernation on HYPERV guests. This
     change got lost before the merge window.

   - Mask the IOAPIC before disabling the local APIC to prevent
     potentially stale IOAPIC remote IRR bits which cause stale
     interrupt lines after resume"

* tag 'x86-urgent-2020-02-09' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/apic: Mask IOAPIC entries when disabling the local APIC
  x86/hyperv: Suspend/resume the hypercall page for hibernation
  x86/apic/msi: Plug non-maskable MSI affinity race
  x86/boot: Handle malformed SRAT tables during early ACPI parsing
  x86/timer: Don't skip PIT setup when APIC is disabled or in legacy mode

Author: torvalds

arch/x86/boot/compressed/acpi.c

@@ -393,7 +393,13 @@ int count_immovable_mem_regions(void)
 	table = table_addr + sizeof(struct acpi_table_srat);
 
 	while (table + sizeof(struct acpi_subtable_header) < table_end) {
+
 		sub_table = (struct acpi_subtable_header *)table;
+		if (!sub_table->length) {
+			debug_putstr("Invalid zero length SRAT subtable.\n");
+			return 0;
+		}
+
 		if (sub_table->type == ACPI_SRAT_TYPE_MEMORY_AFFINITY) {
 			struct acpi_srat_mem_affinity *ma;
 

arch/x86/hyperv/hv_init.c

@@ -21,11 +21,15 @@
 #include <linux/hyperv.h>
 #include <linux/slab.h>
 #include <linux/cpuhotplug.h>
+#include <linux/syscore_ops.h>
 #include <clocksource/hyperv_timer.h>
 
 void *hv_hypercall_pg;
 EXPORT_SYMBOL_GPL(hv_hypercall_pg);
 
+/* Storage to save the hypercall page temporarily for hibernation */
+static void *hv_hypercall_pg_saved;
+
 u32 *hv_vp_index;
 EXPORT_SYMBOL_GPL(hv_vp_index);
 
@@ -246,6 +250,48 @@ static int __init hv_pci_init(void)
 	return 1;
 }
 
+static int hv_suspend(void)
+{
+	union hv_x64_msr_hypercall_contents hypercall_msr;
+
+	/*
+	 * Reset the hypercall page as it is going to be invalidated
+	 * accross hibernation. Setting hv_hypercall_pg to NULL ensures
+	 * that any subsequent hypercall operation fails safely instead of
+	 * crashing due to an access of an invalid page. The hypercall page
+	 * pointer is restored on resume.
+	 */
+	hv_hypercall_pg_saved = hv_hypercall_pg;
+	hv_hypercall_pg = NULL;
+
+	/* Disable the hypercall page in the hypervisor */
+	rdmsrl(HV_X64_MSR_HYPERCALL, hypercall_msr.as_uint64);
+	hypercall_msr.enable = 0;
+	wrmsrl(HV_X64_MSR_HYPERCALL, hypercall_msr.as_uint64);
+
+	return 0;
+}
+
+static void hv_resume(void)
+{
+	union hv_x64_msr_hypercall_contents hypercall_msr;
+
+	/* Re-enable the hypercall page */
+	rdmsrl(HV_X64_MSR_HYPERCALL, hypercall_msr.as_uint64);
+	hypercall_msr.enable = 1;
+	hypercall_msr.guest_physical_address =
+		vmalloc_to_pfn(hv_hypercall_pg_saved);
+	wrmsrl(HV_X64_MSR_HYPERCALL, hypercall_msr.as_uint64);
+
+	hv_hypercall_pg = hv_hypercall_pg_saved;
+	hv_hypercall_pg_saved = NULL;
+}
+
+static struct syscore_ops hv_syscore_ops = {
+	.suspend	= hv_suspend,
+	.resume		= hv_resume,
+};
+
 /*
  * This function is to be invoked early in the boot sequence after the
  * hypervisor has been detected.
@@ -330,6 +376,8 @@ void __init hyperv_init(void)
 
 	x86_init.pci.arch_init = hv_pci_init;
 
+	register_syscore_ops(&hv_syscore_ops);
+
 	return;
 
 remove_cpuhp_state:
@@ -349,6 +397,8 @@ void hyperv_cleanup(void)
 {
 	union hv_x64_msr_hypercall_contents hypercall_msr;
 
+	unregister_syscore_ops(&hv_syscore_ops);
+
 	/* Reset our OS id */
 	wrmsrl(HV_X64_MSR_GUEST_OS_ID, 0);
 

arch/x86/include/asm/apic.h

@@ -140,6 +140,7 @@ extern void apic_soft_disable(void);
 extern void lapic_shutdown(void);
 extern void sync_Arb_IDs(void);
 extern void init_bsp_APIC(void);
+extern void apic_intr_mode_select(void);
 extern void apic_intr_mode_init(void);
 extern void init_apic_mappings(void);
 void register_lapic_address(unsigned long address);
@@ -188,6 +189,7 @@ static inline void disable_local_APIC(void) { }
 # define setup_secondary_APIC_clock x86_init_noop
 static inline void lapic_update_tsc_freq(void) { }
 static inline void init_bsp_APIC(void) { }
+static inline void apic_intr_mode_select(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) { }
@@ -452,6 +454,14 @@ static inline void ack_APIC_irq(void)
 	apic_eoi();
 }
 
+
+static inline bool lapic_vector_set_in_irr(unsigned int vector)
+{
+	u32 irr = apic_read(APIC_IRR + (vector / 32 * 0x10));
+
+	return !!(irr & (1U << (vector % 32)));
+}
+
 static inline unsigned default_get_apic_id(unsigned long x)
 {
 	unsigned int ver = GET_APIC_VERSION(apic_read(APIC_LVR));

arch/x86/include/asm/x86_init.h

@@ -51,12 +51,14 @@ struct x86_init_resources {
  *				are set up.
  * @intr_init:			interrupt init code
  * @trap_init:			platform specific trap setup
+ * @intr_mode_select:		interrupt delivery mode selection
  * @intr_mode_init:		interrupt delivery mode setup
  */
 struct x86_init_irqs {
 	void (*pre_vector_init)(void);
 	void (*intr_init)(void);
 	void (*trap_init)(void);
+	void (*intr_mode_select)(void);
 	void (*intr_mode_init)(void);
 };
 

arch/x86/kernel/apic/apic.c

@@ -830,8 +830,17 @@ bool __init apic_needs_pit(void)
 	if (!tsc_khz || !cpu_khz)
 		return true;
 
-	/* Is there an APIC at all? */
-	if (!boot_cpu_has(X86_FEATURE_APIC))
+	/* Is there an APIC at all or is it disabled? */
+	if (!boot_cpu_has(X86_FEATURE_APIC) || disable_apic)
+		return true;
+
+	/*
+	 * If interrupt delivery mode is legacy PIC or virtual wire without
+	 * configuration, the local APIC timer wont be set up. Make sure
+	 * that the PIT is initialized.
+	 */
+	if (apic_intr_mode == APIC_PIC ||
+	    apic_intr_mode == APIC_VIRTUAL_WIRE_NO_CONFIG)
 		return true;
 
 	/* Virt guests may lack ARAT, but still have DEADLINE */
@@ -1322,7 +1331,7 @@ void __init sync_Arb_IDs(void)
 
 enum apic_intr_mode_id apic_intr_mode __ro_after_init;
 
-static int __init apic_intr_mode_select(void)
+static int __init __apic_intr_mode_select(void)
 {
 	/* Check kernel option */
 	if (disable_apic) {
@@ -1384,6 +1393,12 @@ static int __init apic_intr_mode_select(void)
 	return APIC_SYMMETRIC_IO;
 }
 
+/* Select the interrupt delivery mode for the BSP */
+void __init apic_intr_mode_select(void)
+{
+	apic_intr_mode = __apic_intr_mode_select();
+}
+
 /*
  * An initial setup of the virtual wire mode.
  */
@@ -1440,8 +1455,6 @@ void __init apic_intr_mode_init(void)
 {
 	bool upmode = IS_ENABLED(CONFIG_UP_LATE_INIT);
 
-	apic_intr_mode = apic_intr_mode_select();
-
 	switch (apic_intr_mode) {
 	case APIC_PIC:
 		pr_info("APIC: Keep in PIC mode(8259)\n");
@@ -2626,6 +2639,13 @@ static int lapic_suspend(void)
 #endif
 
 	local_irq_save(flags);
+
+	/*
+	 * Mask IOAPIC before disabling the local APIC to prevent stale IRR
+	 * entries on some implementations.
+	 */
+	mask_ioapic_entries();
+
 	disable_local_APIC();
 
 	irq_remapping_disable();

arch/x86/kernel/apic/msi.c

@@ -23,10 +23,8 @@
 
 static struct irq_domain *msi_default_domain;
 
-static void irq_msi_compose_msg(struct irq_data *data, struct msi_msg *msg)
+static void __irq_msi_compose_msg(struct irq_cfg *cfg, struct msi_msg *msg)
 {
-	struct irq_cfg *cfg = irqd_cfg(data);
-
 	msg->address_hi = MSI_ADDR_BASE_HI;
 
 	if (x2apic_enabled())
@@ -47,6 +45,127 @@ static void irq_msi_compose_msg(struct irq_data *data, struct msi_msg *msg)
 		MSI_DATA_VECTOR(cfg->vector);
 }
 
+static void irq_msi_compose_msg(struct irq_data *data, struct msi_msg *msg)
+{
+	__irq_msi_compose_msg(irqd_cfg(data), msg);
+}
+
+static void irq_msi_update_msg(struct irq_data *irqd, struct irq_cfg *cfg)
+{
+	struct msi_msg msg[2] = { [1] = { }, };
+
+	__irq_msi_compose_msg(cfg, msg);
+	irq_data_get_irq_chip(irqd)->irq_write_msi_msg(irqd, msg);
+}
+
+static int
+msi_set_affinity(struct irq_data *irqd, const struct cpumask *mask, bool force)
+{
+	struct irq_cfg old_cfg, *cfg = irqd_cfg(irqd);
+	struct irq_data *parent = irqd->parent_data;
+	unsigned int cpu;
+	int ret;
+
+	/* Save the current configuration */
+	cpu = cpumask_first(irq_data_get_effective_affinity_mask(irqd));
+	old_cfg = *cfg;
+
+	/* Allocate a new target vector */
+	ret = parent->chip->irq_set_affinity(parent, mask, force);
+	if (ret < 0 || ret == IRQ_SET_MASK_OK_DONE)
+		return ret;
+
+	/*
+	 * For non-maskable and non-remapped MSI interrupts the migration
+	 * to a different destination CPU and a different vector has to be
+	 * done careful to handle the possible stray interrupt which can be
+	 * caused by the non-atomic update of the address/data pair.
+	 *
+	 * Direct update is possible when:
+	 * - The MSI is maskable (remapped MSI does not use this code path)).
+	 *   The quirk bit is not set in this case.
+	 * - The new vector is the same as the old vector
+	 * - The old vector is MANAGED_IRQ_SHUTDOWN_VECTOR (interrupt starts up)
+	 * - The new destination CPU is the same as the old destination CPU
+	 */
+	if (!irqd_msi_nomask_quirk(irqd) ||
+	    cfg->vector == old_cfg.vector ||
+	    old_cfg.vector == MANAGED_IRQ_SHUTDOWN_VECTOR ||
+	    cfg->dest_apicid == old_cfg.dest_apicid) {
+		irq_msi_update_msg(irqd, cfg);
+		return ret;
+	}
+
+	/*
+	 * Paranoia: Validate that the interrupt target is the local
+	 * CPU.
+	 */
+	if (WARN_ON_ONCE(cpu != smp_processor_id())) {
+		irq_msi_update_msg(irqd, cfg);
+		return ret;
+	}
+
+	/*
+	 * Redirect the interrupt to the new vector on the current CPU
+	 * first. This might cause a spurious interrupt on this vector if
+	 * the device raises an interrupt right between this update and the
+	 * update to the final destination CPU.
+	 *
+	 * If the vector is in use then the installed device handler will
+	 * denote it as spurious which is no harm as this is a rare event
+	 * and interrupt handlers have to cope with spurious interrupts
+	 * anyway. If the vector is unused, then it is marked so it won't
+	 * trigger the 'No irq handler for vector' warning in do_IRQ().
+	 *
+	 * This requires to hold vector lock to prevent concurrent updates to
+	 * the affected vector.
+	 */
+	lock_vector_lock();
+
+	/*
+	 * Mark the new target vector on the local CPU if it is currently
+	 * unused. Reuse the VECTOR_RETRIGGERED state which is also used in
+	 * the CPU hotplug path for a similar purpose. This cannot be
+	 * undone here as the current CPU has interrupts disabled and
+	 * cannot handle the interrupt before the whole set_affinity()
+	 * section is done. In the CPU unplug case, the current CPU is
+	 * about to vanish and will not handle any interrupts anymore. The
+	 * vector is cleaned up when the CPU comes online again.
+	 */
+	if (IS_ERR_OR_NULL(this_cpu_read(vector_irq[cfg->vector])))
+		this_cpu_write(vector_irq[cfg->vector], VECTOR_RETRIGGERED);
+
+	/* Redirect it to the new vector on the local CPU temporarily */
+	old_cfg.vector = cfg->vector;
+	irq_msi_update_msg(irqd, &old_cfg);
+
+	/* Now transition it to the target CPU */
+	irq_msi_update_msg(irqd, cfg);
+
+	/*
+	 * All interrupts after this point are now targeted at the new
+	 * vector/CPU.
+	 *
+	 * Drop vector lock before testing whether the temporary assignment
+	 * to the local CPU was hit by an interrupt raised in the device,
+	 * because the retrigger function acquires vector lock again.
+	 */
+	unlock_vector_lock();
+
+	/*
+	 * Check whether the transition raced with a device interrupt and
+	 * is pending in the local APICs IRR. It is safe to do this outside
+	 * of vector lock as the irq_desc::lock of this interrupt is still
+	 * held and interrupts are disabled: The check is not accessing the
+	 * underlying vector store. It's just checking the local APIC's
+	 * IRR.
+	 */
+	if (lapic_vector_set_in_irr(cfg->vector))
+		irq_data_get_irq_chip(irqd)->irq_retrigger(irqd);
+
+	return ret;
+}
+
 /*
  * IRQ Chip for MSI PCI/PCI-X/PCI-Express Devices,
  * which implement the MSI or MSI-X Capability Structure.
@@ -58,6 +177,7 @@ static struct irq_chip pci_msi_controller = {
 	.irq_ack		= irq_chip_ack_parent,
 	.irq_retrigger		= irq_chip_retrigger_hierarchy,
 	.irq_compose_msi_msg	= irq_msi_compose_msg,
+	.irq_set_affinity	= msi_set_affinity,
 	.flags			= IRQCHIP_SKIP_SET_WAKE,
 };
 
@@ -146,6 +266,8 @@ void __init arch_init_msi_domain(struct irq_domain *parent)
 	}
 	if (!msi_default_domain)
 		pr_warn("failed to initialize irqdomain for MSI/MSI-x.\n");
+	else
+		msi_default_domain->flags |= IRQ_DOMAIN_MSI_NOMASK_QUIRK;
 }
 
 #ifdef CONFIG_IRQ_REMAP

arch/x86/kernel/time.c

@@ -91,10 +91,18 @@ void __init hpet_time_init(void)
 
 static __init void x86_late_time_init(void)
 {
+	/*
+	 * Before PIT/HPET init, select the interrupt mode. This is required
+	 * to make the decision whether PIT should be initialized correct.
+	 */
+	x86_init.irqs.intr_mode_select();
+
+	/* Setup the legacy timers */
 	x86_init.timers.timer_init();
+
 	/*
-	 * After PIT/HPET timers init, select and setup
-	 * the final interrupt mode for delivering IRQs.
+	 * After PIT/HPET timers init, set up the final interrupt mode for
+	 * delivering IRQs.
 	 */
 	x86_init.irqs.intr_mode_init();
 	tsc_init();

arch/x86/kernel/x86_init.c

@@ -80,6 +80,7 @@ struct x86_init_ops x86_init __initdata = {
 		.pre_vector_init	= init_ISA_irqs,
 		.intr_init		= native_init_IRQ,
 		.trap_init		= x86_init_noop,
+		.intr_mode_select	= apic_intr_mode_select,
 		.intr_mode_init		= apic_intr_mode_init
 	},