Merge branch 'x86/ras' into x86/core, to fix conflicts

Conflicts:
arch/x86/include/asm/irq_vectors.h

Signed-off-by: Ingo Molnar <[e-mail]>

Documentation/x86/x86_64/boot-options.txt

···
 		(e.g. BIOS or hardware monitoring applications), conflicting
 		with OS's error handling, and you cannot deactivate the agent,
 		then this option will be a help.
+   mce=no_lmce
+		Do not opt-in to Local MCE delivery. Use legacy method
+		to broadcast MCEs.
    mce=bootlog
 		Enable logging of machine checks left over from booting.
 		Disabled by default on AMD because some BIOS leave bogus ones.

arch/x86/include/asm/entry_arch.h

···
 BUILD_INTERRUPT(threshold_interrupt,THRESHOLD_APIC_VECTOR)
 #endif
 
+#ifdef CONFIG_X86_MCE_AMD
+BUILD_INTERRUPT(deferred_error_interrupt, DEFERRED_ERROR_VECTOR)
+#endif
 #endif

arch/x86/include/asm/hardirq.h

···
 #ifdef CONFIG_X86_MCE_THRESHOLD
 	unsigned int irq_threshold_count;
 #endif
+#ifdef CONFIG_X86_MCE_AMD
+	unsigned int irq_deferred_error_count;
+#endif
 #if IS_ENABLED(CONFIG_HYPERV) || defined(CONFIG_XEN)
 	unsigned int irq_hv_callback_count;
 #endif

arch/x86/include/asm/hw_irq.h

···
 extern asmlinkage void irq_move_cleanup_interrupt(void);
 extern asmlinkage void reboot_interrupt(void);
 extern asmlinkage void threshold_interrupt(void);
+extern asmlinkage void deferred_error_interrupt(void);
 
 extern asmlinkage void call_function_interrupt(void);
 extern asmlinkage void call_function_single_interrupt(void);
···
 extern void trace_thermal_interrupt(void);
 extern void trace_reschedule_interrupt(void);
 extern void trace_threshold_interrupt(void);
+extern void trace_deferred_error_interrupt(void);
 extern void trace_call_function_interrupt(void);
 extern void trace_call_function_single_interrupt(void);
 #define trace_irq_move_cleanup_interrupt  irq_move_cleanup_interrupt

arch/x86/include/asm/irq_vectors.h

···
  */
 #define X86_PLATFORM_IPI_VECTOR		0xf7
 
-/* Vector for KVM to deliver posted interrupt IPI */
-#ifdef CONFIG_HAVE_KVM
-#define POSTED_INTR_VECTOR		0xf2
 #define POSTED_INTR_WAKEUP_VECTOR	0xf1
-#endif
-
 /*
  * IRQ work vector:
  */
 #define IRQ_WORK_VECTOR			0xf6
 
 #define UV_BAU_MESSAGE			0xf5
+#define DEFERRED_ERROR_VECTOR		0xf4
 
 /* Vector on which hypervisor callbacks will be delivered */
 #define HYPERVISOR_CALLBACK_VECTOR	0xf3
+
+/* Vector for KVM to deliver posted interrupt IPI */
+#ifdef CONFIG_HAVE_KVM
+#define POSTED_INTR_VECTOR		0xf2
+#endif
 
 /*
  * Local APIC timer IRQ vector is on a different priority level,

arch/x86/include/asm/mce.h

···
 #define MCG_EXT_CNT(c)		(((c) & MCG_EXT_CNT_MASK) >> MCG_EXT_CNT_SHIFT)
 #define MCG_SER_P		(1ULL<<24)   /* MCA recovery/new status bits */
 #define MCG_ELOG_P		(1ULL<<26)   /* Extended error log supported */
+#define MCG_LMCE_P		(1ULL<<27)   /* Local machine check supported */
 
 /* MCG_STATUS register defines */
 #define MCG_STATUS_RIPV  (1ULL<<0)   /* restart ip valid */
 #define MCG_STATUS_EIPV  (1ULL<<1)   /* ip points to correct instruction */
 #define MCG_STATUS_MCIP  (1ULL<<2)   /* machine check in progress */
+#define MCG_STATUS_LMCES (1ULL<<3)   /* LMCE signaled */
+
+/* MCG_EXT_CTL register defines */
+#define MCG_EXT_CTL_LMCE_EN (1ULL<<0) /* Enable LMCE */
 
 /* MCi_STATUS register defines */
 #define MCI_STATUS_VAL   (1ULL<<63)  /* valid error */
···
 struct mca_config {
 	bool dont_log_ce;
 	bool cmci_disabled;
+	bool lmce_disabled;
 	bool ignore_ce;
 	bool disabled;
 	bool ser;
···
 };
 
 struct mce_vendor_flags {
-	__u64		overflow_recov	: 1, /* cpuid_ebx(80000007) */
-			__reserved_0	: 63;
+			/*
+			 * overflow recovery cpuid bit indicates that overflow
+			 * conditions are not fatal
+			 */
+	__u64		overflow_recov	: 1,
+
+			/*
+			 * SUCCOR stands for S/W UnCorrectable error COntainment
+			 * and Recovery. It indicates support for data poisoning
+			 * in HW and deferred error interrupts.
+			 */
+			succor		: 1,
+			__reserved_0	: 62;
 };
 extern struct mce_vendor_flags mce_flags;
 
···
 void cmci_reenable(void);
 void cmci_rediscover(void);
 void cmci_recheck(void);
+void lmce_clear(void);
+void lmce_enable(void);
 #else
 static inline void mce_intel_feature_init(struct cpuinfo_x86 *c) { }
 static inline void cmci_clear(void) {}
 static inline void cmci_reenable(void) {}
 static inline void cmci_rediscover(void) {}
 static inline void cmci_recheck(void) {}
+static inline void lmce_clear(void) {}
+static inline void lmce_enable(void) {}
 #endif
 
 #ifdef CONFIG_X86_MCE_AMD
···
 
 extern void (*mce_threshold_vector)(void);
 extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu);
+
+/* Deferred error interrupt handler */
+extern void (*deferred_error_int_vector)(void);
 
 /*
  * Thermal handler

arch/x86/include/asm/trace/irq_vectors.h

···
 DEFINE_IRQ_VECTOR_EVENT(threshold_apic);
 
 /*
+ * deferred_error_apic - called when entering/exiting a deferred apic interrupt
+ * vector handler
+ */
+DEFINE_IRQ_VECTOR_EVENT(deferred_error_apic);
+
+/*
  * thermal_apic - called when entering/exiting a thermal apic interrupt
  * vector handler
  */

arch/x86/include/asm/traps.h

···
 void math_emulate(struct math_emu_info *);
 #ifndef CONFIG_X86_32
 asmlinkage void smp_thermal_interrupt(void);
-asmlinkage void mce_threshold_interrupt(void);
+asmlinkage void smp_threshold_interrupt(void);
+asmlinkage void smp_deferred_error_interrupt(void);
 #endif
 
 extern enum ctx_state ist_enter(struct pt_regs *regs);

arch/x86/include/uapi/asm/msr-index.h

···
 #define MSR_IA32_MCG_CAP		0x00000179
 #define MSR_IA32_MCG_STATUS		0x0000017a
 #define MSR_IA32_MCG_CTL		0x0000017b
+#define MSR_IA32_MCG_EXT_CTL		0x000004d0
 
 #define MSR_OFFCORE_RSP_0		0x000001a6
 #define MSR_OFFCORE_RSP_1		0x000001a7
···
 #define FEATURE_CONTROL_LOCKED				(1<<0)
 #define FEATURE_CONTROL_VMXON_ENABLED_INSIDE_SMX	(1<<1)
 #define FEATURE_CONTROL_VMXON_ENABLED_OUTSIDE_SMX	(1<<2)
+#define FEATURE_CONTROL_LMCE				(1<<20)
 
 #define MSR_IA32_APICBASE		0x0000001b
 #define MSR_IA32_APICBASE_BSP		(1<<8)

arch/x86/kernel/cpu/mcheck/mce.c

···
 	char *msg = "Unknown";
 	u64 recover_paddr = ~0ull;
 	int flags = MF_ACTION_REQUIRED;
+	int lmce = 0;
 
 	prev_state = ist_enter(regs);
 
···
 		kill_it = 1;
 
 	/*
-	 * Go through all the banks in exclusion of the other CPUs.
-	 * This way we don't report duplicated events on shared banks
-	 * because the first one to see it will clear it.
+	 * Check if this MCE is signaled to only this logical processor
 	 */
-	order = mce_start(&no_way_out);
+	if (m.mcgstatus & MCG_STATUS_LMCES)
+		lmce = 1;
+	else {
+		/*
+		 * Go through all the banks in exclusion of the other CPUs.
+		 * This way we don't report duplicated events on shared banks
+		 * because the first one to see it will clear it.
+		 * If this is a Local MCE, then no need to perform rendezvous.
+		 */
+		order = mce_start(&no_way_out);
+	}
+
 	for (i = 0; i < cfg->banks; i++) {
 		__clear_bit(i, toclear);
 		if (!test_bit(i, valid_banks))
···
 	 * Do most of the synchronization with other CPUs.
 	 * When there's any problem use only local no_way_out state.
 	 */
-	if (mce_end(order) < 0)
-		no_way_out = worst >= MCE_PANIC_SEVERITY;
+	if (!lmce) {
+		if (mce_end(order) < 0)
+			no_way_out = worst >= MCE_PANIC_SEVERITY;
+	} else {
+		/*
+		 * Local MCE skipped calling mce_reign()
+		 * If we found a fatal error, we need to panic here.
+		 */
+		 if (worst >= MCE_PANIC_SEVERITY && mca_cfg.tolerant < 3)
+			mce_panic("Machine check from unknown source",
+				NULL, NULL);
+	}
 
 	/*
 	 * At insane "tolerant" levels we take no action. Otherwise
···
 		mce_intel_feature_init(c);
 		mce_adjust_timer = cmci_intel_adjust_timer;
 		break;
-	case X86_VENDOR_AMD:
+
+	case X86_VENDOR_AMD: {
+		u32 ebx = cpuid_ebx(0x80000007);
+
 		mce_amd_feature_init(c);
-		mce_flags.overflow_recov = cpuid_ebx(0x80000007) & 0x1;
+		mce_flags.overflow_recov = !!(ebx & BIT(0));
+		mce_flags.succor	 = !!(ebx & BIT(1));
 		break;
+		}
+
 	default:
 		break;
 	}
···
 /*
  * mce=off Disables machine check
  * mce=no_cmci Disables CMCI
+ * mce=no_lmce Disables LMCE
  * mce=dont_log_ce Clears corrected events silently, no log created for CEs.
  * mce=ignore_ce Disables polling and CMCI, corrected events are not cleared.
  * mce=TOLERANCELEVEL[,monarchtimeout] (number, see above)
···
 		cfg->disabled = true;
 	else if (!strcmp(str, "no_cmci"))
 		cfg->cmci_disabled = true;
+	else if (!strcmp(str, "no_lmce"))
+		cfg->lmce_disabled = true;
 	else if (!strcmp(str, "dont_log_ce"))
 		cfg->dont_log_ce = true;
 	else if (!strcmp(str, "ignore_ce"))
···
 	else if (!strcmp(str, "bios_cmci_threshold"))
 		cfg->bios_cmci_threshold = true;
 	else if (isdigit(str[0])) {
-		get_option(&str, &(cfg->tolerant));
-		if (*str == ',') {
-			++str;
+		if (get_option(&str, &cfg->tolerant) == 2)
 			get_option(&str, &(cfg->monarch_timeout));
-		}
 	} else {
 		pr_info("mce argument %s ignored. Please use /sys\n", str);
 		return 0;

arch/x86/kernel/cpu/mcheck/mce_amd.c

···
 /*
- *  (c) 2005-2012 Advanced Micro Devices, Inc.
+ *  (c) 2005-2015 Advanced Micro Devices, Inc.
  *  Your use of this code is subject to the terms and conditions of the
  *  GNU general public license version 2. See "COPYING" or
  *  http://www.gnu.org/licenses/gpl.html
  *
  *  Written by Jacob Shin - AMD, Inc.
- *
  *  Maintained by: Borislav Petkov <[e-mail]>
  *
- *  April 2006
- *     - added support for AMD Family 0x10 processors
- *  May 2012
- *     - major scrubbing
- *
- *  All MC4_MISCi registers are shared between multi-cores
+ *  All MC4_MISCi registers are shared between cores on a node.
  */
 #include <linux/interrupt.h>
 #include <linux/notifier.h>
···
 #include <asm/idle.h>
 #include <asm/mce.h>
 #include <asm/msr.h>
+#include <asm/trace/irq_vectors.h>
 
 #define NR_BLOCKS         9
 #define THRESHOLD_MAX     0xFFF
···
 #define MASK_BLKPTR_LO    0xFF000000
 #define MCG_XBLK_ADDR     0xC0000400
 
+/* Deferred error settings */
+#define MSR_CU_DEF_ERR		0xC0000410
+#define MASK_DEF_LVTOFF		0x000000F0
+#define MASK_DEF_INT_TYPE	0x00000006
+#define DEF_LVT_OFF		0x2
+#define DEF_INT_TYPE_APIC	0x2
+
 static const char * const th_names[] = {
 	"load_store",
 	"insn_fetch",
···
 static DEFINE_PER_CPU(unsigned char, bank_map);	/* see which banks are on */
 
 static void amd_threshold_interrupt(void);
+static void amd_deferred_error_interrupt(void);
+
+static void default_deferred_error_interrupt(void)
+{
+	pr_err("Unexpected deferred interrupt at vector %x\n", DEFERRED_ERROR_VECTOR);
+}
+void (*deferred_error_int_vector)(void) = default_deferred_error_interrupt;
 
 /*
  * CPU Initialization
···
 	threshold_restart_bank(&tr);
 };
 
-static int setup_APIC_mce(int reserved, int new)
+static int setup_APIC_mce_threshold(int reserved, int new)
 {
 	if (reserved < 0 && !setup_APIC_eilvt(new, THRESHOLD_APIC_VECTOR,
 					      APIC_EILVT_MSG_FIX, 0))
 		return new;
 
 	return reserved;
+}
+
+static int setup_APIC_deferred_error(int reserved, int new)
+{
+	if (reserved < 0 && !setup_APIC_eilvt(new, DEFERRED_ERROR_VECTOR,
+					      APIC_EILVT_MSG_FIX, 0))
+		return new;
+
+	return reserved;
+}
+
+static void deferred_error_interrupt_enable(struct cpuinfo_x86 *c)
+{
+	u32 low = 0, high = 0;
+	int def_offset = -1, def_new;
+
+	if (rdmsr_safe(MSR_CU_DEF_ERR, &low, &high))
+		return;
+
+	def_new = (low & MASK_DEF_LVTOFF) >> 4;
+	if (!(low & MASK_DEF_LVTOFF)) {
+		pr_err(FW_BUG "Your BIOS is not setting up LVT offset 0x2 for deferred error IRQs correctly.\n");
+		def_new = DEF_LVT_OFF;
+		low = (low & ~MASK_DEF_LVTOFF) | (DEF_LVT_OFF << 4);
+	}
+
+	def_offset = setup_APIC_deferred_error(def_offset, def_new);
+	if ((def_offset == def_new) &&
+	    (deferred_error_int_vector != amd_deferred_error_interrupt))
+		deferred_error_int_vector = amd_deferred_error_interrupt;
+
+	low = (low & ~MASK_DEF_INT_TYPE) | DEF_INT_TYPE_APIC;
+	wrmsr(MSR_CU_DEF_ERR, low, high);
 }
 
 /* cpu init entry point, called from mce.c with preempt off */
···
 
 			b.interrupt_enable = 1;
 			new	= (high & MASK_LVTOFF_HI) >> 20;
-			offset  = setup_APIC_mce(offset, new);
+			offset  = setup_APIC_mce_threshold(offset, new);
 
 			if ((offset == new) &&
 			    (mce_threshold_vector != amd_threshold_interrupt))
···
 init:
 			mce_threshold_block_init(&b, offset);
 		}
+	}
+
+	if (mce_flags.succor)
+		deferred_error_interrupt_enable(c);
+}
+
+static void __log_error(unsigned int bank, bool threshold_err, u64 misc)
+{
+	struct mce m;
+	u64 status;
+
+	rdmsrl(MSR_IA32_MCx_STATUS(bank), status);
+	if (!(status & MCI_STATUS_VAL))
+		return;
+
+	mce_setup(&m);
+
+	m.status = status;
+	m.bank = bank;
+
+	if (threshold_err)
+		m.misc = misc;
+
+	if (m.status & MCI_STATUS_ADDRV)
+		rdmsrl(MSR_IA32_MCx_ADDR(bank), m.addr);
+
+	mce_log(&m);
+	wrmsrl(MSR_IA32_MCx_STATUS(bank), 0);
+}
+
+static inline void __smp_deferred_error_interrupt(void)
+{
+	inc_irq_stat(irq_deferred_error_count);
+	deferred_error_int_vector();
+}
+
+asmlinkage __visible void smp_deferred_error_interrupt(void)
+{
+	entering_irq();
+	__smp_deferred_error_interrupt();
+	exiting_ack_irq();
+}
+
+asmlinkage __visible void smp_trace_deferred_error_interrupt(void)
+{
+	entering_irq();
+	trace_deferred_error_apic_entry(DEFERRED_ERROR_VECTOR);
+	__smp_deferred_error_interrupt();
+	trace_deferred_error_apic_exit(DEFERRED_ERROR_VECTOR);
+	exiting_ack_irq();
+}
+
+/* APIC interrupt handler for deferred errors */
+static void amd_deferred_error_interrupt(void)
+{
+	u64 status;
+	unsigned int bank;
+
+	for (bank = 0; bank < mca_cfg.banks; ++bank) {
+		rdmsrl(MSR_IA32_MCx_STATUS(bank), status);
+
+		if (!(status & MCI_STATUS_VAL) ||
+		    !(status & MCI_STATUS_DEFERRED))
+			continue;
+
+		__log_error(bank, false, 0);
+		break;
 	}
 }
 
···
  * the interrupt goes off when error_count reaches threshold_limit.
  * the handler will simply log mcelog w/ software defined bank number.
  */
+
 static void amd_threshold_interrupt(void)
 {
 	u32 low = 0, high = 0, address = 0;
 	int cpu = smp_processor_id();
 	unsigned int bank, block;
-	struct mce m;
 
 	/* assume first bank caused it */
 	for (bank = 0; bank < mca_cfg.banks; ++bank) {
···
 	return;
 
 log:
-	mce_setup(&m);
-	rdmsrl(MSR_IA32_MCx_STATUS(bank), m.status);
-	if (!(m.status & MCI_STATUS_VAL))
-		return;
-	m.misc = ((u64)high << 32) | low;
-	m.bank = bank;
-	mce_log(&m);
-
-	wrmsrl(MSR_IA32_MCx_STATUS(bank), 0);
+	__log_error(bank, true, ((u64)high << 32) | low);
 }
 
 /*

arch/x86/kernel/cpu/mcheck/mce_intel.c

···
 	return !!(cap & MCG_CMCI_P);
 }
 
+static bool lmce_supported(void)
+{
+	u64 tmp;
+
+	if (mca_cfg.lmce_disabled)
+		return false;
+
+	rdmsrl(MSR_IA32_MCG_CAP, tmp);
+
+	/*
+	 * LMCE depends on recovery support in the processor. Hence both
+	 * MCG_SER_P and MCG_LMCE_P should be present in MCG_CAP.
+	 */
+	if ((tmp & (MCG_SER_P | MCG_LMCE_P)) !=
+		   (MCG_SER_P | MCG_LMCE_P))
+		return false;
+
+	/*
+	 * BIOS should indicate support for LMCE by setting bit 20 in
+	 * IA32_FEATURE_CONTROL without which touching MCG_EXT_CTL will
+	 * generate a #GP fault.
+	 */
+	rdmsrl(MSR_IA32_FEATURE_CONTROL, tmp);
+	if ((tmp & (FEATURE_CONTROL_LOCKED | FEATURE_CONTROL_LMCE)) ==
+		   (FEATURE_CONTROL_LOCKED | FEATURE_CONTROL_LMCE))
+		return true;
+
+	return false;
+}
+
 bool mce_intel_cmci_poll(void)
 {
 	if (__this_cpu_read(cmci_storm_state) == CMCI_STORM_NONE)
···
 	cmci_recheck();
 }
 
+void intel_init_lmce(void)
+{
+	u64 val;
+
+	if (!lmce_supported())
+		return;
+
+	rdmsrl(MSR_IA32_MCG_EXT_CTL, val);
+
+	if (!(val & MCG_EXT_CTL_LMCE_EN))
+		wrmsrl(MSR_IA32_MCG_EXT_CTL, val | MCG_EXT_CTL_LMCE_EN);
+}
+
 void mce_intel_feature_init(struct cpuinfo_x86 *c)
 {
 	intel_init_thermal(c);
 	intel_init_cmci();
+	intel_init_lmce();
 }

arch/x86/kernel/entry_64.S

···
 	threshold_interrupt smp_threshold_interrupt
 #endif
 
+#ifdef CONFIG_X86_MCE_AMD
+apicinterrupt DEFERRED_ERROR_VECTOR \
+	deferred_error_interrupt smp_deferred_error_interrupt
+#endif
+
 #ifdef CONFIG_X86_THERMAL_VECTOR
 apicinterrupt THERMAL_APIC_VECTOR \
 	thermal_interrupt smp_thermal_interrupt

arch/x86/kernel/irq.c

···
 		seq_printf(p, "%10u ", irq_stats(j)->irq_threshold_count);
 	seq_puts(p, "  Threshold APIC interrupts\n");
 #endif
+#ifdef CONFIG_X86_MCE_AMD
+	seq_printf(p, "%*s: ", prec, "DFR");
+	for_each_online_cpu(j)
+		seq_printf(p, "%10u ", irq_stats(j)->irq_deferred_error_count);
+	seq_puts(p, "  Deferred Error APIC interrupts\n");
+#endif
 #ifdef CONFIG_X86_MCE
 	seq_printf(p, "%*s: ", prec, "MCE");
 	for_each_online_cpu(j)

arch/x86/kernel/irqinit.c

···
 	alloc_intr_gate(THRESHOLD_APIC_VECTOR, threshold_interrupt);
 #endif
 
+#ifdef CONFIG_X86_MCE_AMD
+	alloc_intr_gate(DEFERRED_ERROR_VECTOR, deferred_error_interrupt);
+#endif
+
 #ifdef CONFIG_X86_LOCAL_APIC
 	/* self generated IPI for local APIC timer */
 	alloc_intr_gate(LOCAL_TIMER_VECTOR, apic_timer_interrupt);

arch/x86/kernel/traps.c

···
 do_spurious_interrupt_bug(struct pt_regs *regs, long error_code)
 {
 	conditional_sti(regs);
-#if 0
-	/* No need to warn about this any longer. */
-	pr_info("Ignoring P6 Local APIC Spurious Interrupt Bug...\n");
-#endif
-}
-
-asmlinkage __visible void __attribute__((weak)) smp_thermal_interrupt(void)
-{
-}
-
-asmlinkage __visible void __attribute__((weak)) smp_threshold_interrupt(void)
-{
 }
 
 /*