Merge branch 'ras-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull x86 RAS updates from Ingo Molnar: "The main changes in this cycle were: - AMD MCE support/decoding improvements (Yazen Ghannam) - general MCE header cleanups and reorganization (Borislav Petkov)" * 'ras-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: Revert "x86/mce/AMD: Collect error info even if valid bits are not set" x86/MCE: Cleanup and complete struct mce fields definitions x86/mce/AMD: Carve out SMCA get_block_address() code x86/mce/AMD: Get address from already initialized block x86/mce/AMD, EDAC/mce_amd: Enumerate Reserved SMCA bank type x86/mce/AMD: Pass the bank number to smca_get_bank_type() x86/mce/AMD: Collect error info even if valid bits are not set x86/mce: Issue the 'mcelog --ascii' message only on !AMD x86/mce: Convert 'struct mca_config' bools to a bitfield x86/mce: Put private structures and definitions into the internal header
2018-04-02 11:47:07 -07:00
--- a/arch/x86/kernel/cpu/mcheck/mce-internal.h
+++ b/arch/x86/kernel/cpu/mcheck/mce-internal.h
@@ -113,8 +113,6 @@ static inline void mce_register_injector_chain(struct notifier_block *nb)	{ }
 static inline void mce_unregister_injector_chain(struct notifier_block *nb)	{ }
 #endif

-extern struct mca_config mca_cfg;
-
 #ifndef CONFIG_X86_64
 /*
 * On 32-bit systems it would be difficult to safely unmap a poison page
@@ -130,4 +128,61 @@ static inline void mce_unmap_kpfn(unsigned long pfn) {}
 #define mce_unmap_kpfn mce_unmap_kpfn
 #endif

+struct mca_config {
+	bool dont_log_ce;
+	bool cmci_disabled;
+	bool ignore_ce;
+
+	__u64 lmce_disabled		: 1,
+	      disabled			: 1,
+	      ser			: 1,
+	      recovery			: 1,
+	      bios_cmci_threshold	: 1,
+	      __reserved		: 59;
+
+	u8 banks;
+	s8 bootlog;
+	int tolerant;
+	int monarch_timeout;
+	int panic_timeout;
+	u32 rip_msr;
+};
+
+extern struct mca_config mca_cfg;
+
+struct mce_vendor_flags {
+	/*
+	 * Indicates that overflow conditions are not fatal, when set.
+	 */
+	__u64 overflow_recov	: 1,
+
+	/*
+	 * (AMD) SUCCOR stands for S/W UnCorrectable error COntainment and
+	 * Recovery. It indicates support for data poisoning in HW and deferred
+	 * error interrupts.
+	 */
+	      succor		: 1,
+
+	/*
+	 * (AMD) SMCA: This bit indicates support for Scalable MCA which expands
+	 * the register space for each MCA bank and also increases number of
+	 * banks. Also, to accommodate the new banks and registers, the MCA
+	 * register space is moved to a new MSR range.
+	 */
+	      smca		: 1,
+
+	      __reserved_0	: 61;
+};
+
+extern struct mce_vendor_flags mce_flags;
+
+struct mca_msr_regs {
+	u32 (*ctl)	(int bank);
+	u32 (*status)	(int bank);
+	u32 (*addr)	(int bank);
+	u32 (*misc)	(int bank);
+};
+
+extern struct mca_msr_regs msr_ops;
+
 #endif /* __X86_MCE_INTERNAL_H__ */
--- a/arch/x86/kernel/cpu/mcheck/mce.c
+++ b/arch/x86/kernel/cpu/mcheck/mce.c
@@ -273,7 +273,9 @@ static void __print_mce(struct mce *m)
 static void print_mce(struct mce *m)
 {
 	__print_mce(m);
-	pr_emerg_ratelimited(HW_ERR "Run the above through 'mcelog --ascii'\n");
+
+	if (m->cpuvendor != X86_VENDOR_AMD)
+		pr_emerg_ratelimited(HW_ERR "Run the above through 'mcelog --ascii'\n");
 }

 #define PANIC_TIMEOUT 5 /* 5 seconds */
@@ -1516,7 +1518,7 @@ static int __mcheck_cpu_cap_init(void)
 		mca_cfg.rip_msr = MSR_IA32_MCG_EIP;

 	if (cap & MCG_SER_P)
-		mca_cfg.ser = true;
+		mca_cfg.ser = 1;

 	return 0;
 }
@@ -1824,12 +1826,12 @@ void mcheck_cpu_init(struct cpuinfo_x86 *c)
 		return;

 	if (__mcheck_cpu_cap_init() < 0 || __mcheck_cpu_apply_quirks(c) < 0) {
-		mca_cfg.disabled = true;
+		mca_cfg.disabled = 1;
 		return;
 	}

 	if (mce_gen_pool_init()) {
-		mca_cfg.disabled = true;
+		mca_cfg.disabled = 1;
 		pr_emerg("Couldn't allocate MCE records pool!\n");
 		return;
 	}
@@ -1907,11 +1909,11 @@ static int __init mcheck_enable(char *str)
 	if (*str == '=')
 		str++;
 	if (!strcmp(str, "off"))
-		cfg->disabled = true;
+		cfg->disabled = 1;
 	else if (!strcmp(str, "no_cmci"))
 		cfg->cmci_disabled = true;
 	else if (!strcmp(str, "no_lmce"))
-		cfg->lmce_disabled = true;
+		cfg->lmce_disabled = 1;
 	else if (!strcmp(str, "dont_log_ce"))
 		cfg->dont_log_ce = true;
 	else if (!strcmp(str, "ignore_ce"))
@@ -1919,9 +1921,9 @@ static int __init mcheck_enable(char *str)
 	else if (!strcmp(str, "bootlog") || !strcmp(str, "nobootlog"))
 		cfg->bootlog = (str[0] == 'b');
 	else if (!strcmp(str, "bios_cmci_threshold"))
-		cfg->bios_cmci_threshold = true;
+		cfg->bios_cmci_threshold = 1;
 	else if (!strcmp(str, "recovery"))
-		cfg->recovery = true;
+		cfg->recovery = 1;
 	else if (isdigit(str[0])) {
 		if (get_option(&str, &cfg->tolerant) == 2)
 			get_option(&str, &(cfg->monarch_timeout));
@@ -2403,7 +2405,7 @@ device_initcall_sync(mcheck_init_device);
 */
 static int __init mcheck_disable(char *str)
 {
-	mca_cfg.disabled = true;
+	mca_cfg.disabled = 1;
 	return 1;
 }
 __setup("nomce", mcheck_disable);
--- a/arch/x86/kernel/cpu/mcheck/mce_amd.c
+++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c
@@ -82,6 +82,7 @@ static struct smca_bank_name smca_names[] = {
 	[SMCA_IF]	= { "insn_fetch",	"Instruction Fetch Unit" },
 	[SMCA_L2_CACHE]	= { "l2_cache",		"L2 Cache" },
 	[SMCA_DE]	= { "decode_unit",	"Decode Unit" },
+	[SMCA_RESERVED]	= { "reserved",		"Reserved" },
 	[SMCA_EX]	= { "execution_unit",	"Execution Unit" },
 	[SMCA_FP]	= { "floating_point",	"Floating Point Unit" },
 	[SMCA_L3_CACHE]	= { "l3_cache",		"L3 Cache" },
@@ -110,14 +111,14 @@ const char *smca_get_long_name(enum smca_bank_types t)
 }
 EXPORT_SYMBOL_GPL(smca_get_long_name);

-static enum smca_bank_types smca_get_bank_type(struct mce *m)
+static enum smca_bank_types smca_get_bank_type(unsigned int bank)
 {
 	struct smca_bank *b;

-	if (m->bank >= N_SMCA_BANK_TYPES)
+	if (bank >= MAX_NR_BANKS)
 		return N_SMCA_BANK_TYPES;

-	b = &smca_banks[m->bank];
+	b = &smca_banks[bank];
 	if (!b->hwid)
 		return N_SMCA_BANK_TYPES;

@@ -127,6 +128,9 @@ static enum smca_bank_types smca_get_bank_type(struct mce *m)
 static struct smca_hwid smca_hwid_mcatypes[] = {
 	/* { bank_type, hwid_mcatype, xec_bitmap } */

+	/* Reserved type */
+	{ SMCA_RESERVED, HWID_MCATYPE(0x00, 0x0), 0x0 },
+
 	/* ZN Core (HWID=0xB0) MCA types */
 	{ SMCA_LS,	 HWID_MCATYPE(0xB0, 0x0), 0x1FFFEF },
 	{ SMCA_IF,	 HWID_MCATYPE(0xB0, 0x1), 0x3FFF },
@@ -427,35 +431,58 @@ static void deferred_error_interrupt_enable(struct cpuinfo_x86 *c)
 	wrmsr(MSR_CU_DEF_ERR, low, high);
 }

+static u32 smca_get_block_address(unsigned int cpu, unsigned int bank,
+				  unsigned int block)
+{
+	u32 low, high;
+	u32 addr = 0;
+
+	if (smca_get_bank_type(bank) == SMCA_RESERVED)
+		return addr;
+
+	if (!block)
+		return MSR_AMD64_SMCA_MCx_MISC(bank);
+
+	/*
+	 * For SMCA enabled processors, BLKPTR field of the first MISC register
+	 * (MCx_MISC0) indicates presence of additional MISC regs set (MISC1-4).
+	 */
+	if (rdmsr_safe_on_cpu(cpu, MSR_AMD64_SMCA_MCx_CONFIG(bank), &low, &high))
+		return addr;
+
+	if (!(low & MCI_CONFIG_MCAX))
+		return addr;
+
+	if (!rdmsr_safe_on_cpu(cpu, MSR_AMD64_SMCA_MCx_MISC(bank), &low, &high) &&
+	    (low & MASK_BLKPTR_LO))
+		return MSR_AMD64_SMCA_MCx_MISCy(bank, block - 1);
+
+	return addr;
+}
+
 static u32 get_block_address(unsigned int cpu, u32 current_addr, u32 low, u32 high,
 			     unsigned int bank, unsigned int block)
 {
 	u32 addr = 0, offset = 0;

-	if (mce_flags.smca) {
-		if (!block) {
-			addr = MSR_AMD64_SMCA_MCx_MISC(bank);
-		} else {
-			/*
-			 * For SMCA enabled processors, BLKPTR field of the
-			 * first MISC register (MCx_MISC0) indicates presence of
-			 * additional MISC register set (MISC1-4).
-			 */
-			u32 low, high;
-
-			if (rdmsr_safe_on_cpu(cpu, MSR_AMD64_SMCA_MCx_CONFIG(bank), &low, &high))
-				return addr;
-
-			if (!(low & MCI_CONFIG_MCAX))
-				return addr;
-
-			if (!rdmsr_safe_on_cpu(cpu, MSR_AMD64_SMCA_MCx_MISC(bank), &low, &high) &&
-			    (low & MASK_BLKPTR_LO))
-				addr = MSR_AMD64_SMCA_MCx_MISCy(bank, block - 1);
-		}
+	if ((bank >= mca_cfg.banks) || (block >= NR_BLOCKS))
 		return addr;
+
+	/* Get address from already initialized block. */
+	if (per_cpu(threshold_banks, cpu)) {
+		struct threshold_bank *bankp = per_cpu(threshold_banks, cpu)[bank];
+
+		if (bankp && bankp->blocks) {
+			struct threshold_block *blockp = &bankp->blocks[block];
+
+			if (blockp)
+				return blockp->address;
+		}
 	}

+	if (mce_flags.smca)
+		return smca_get_block_address(cpu, bank, block);
+
 	/* Fall back to method we used for older processors: */
 	switch (block) {
 	case 0:
@@ -760,7 +787,7 @@ bool amd_mce_is_memory_error(struct mce *m)
 	u8 xec = (m->status >> 16) & 0x1f;

 	if (mce_flags.smca)
-		return smca_get_bank_type(m) == SMCA_UMC && xec == 0x0;
+		return smca_get_bank_type(m->bank) == SMCA_UMC && xec == 0x0;

 	return m->bank == 4 && xec == 0x8;
 }
@@ -1063,7 +1090,7 @@ static struct kobj_type threshold_ktype = {

 static const char *get_name(unsigned int bank, struct threshold_block *b)
 {
-	unsigned int bank_type;
+	enum smca_bank_types bank_type;

 	if (!mce_flags.smca) {
 		if (b && bank == 4)
@@ -1072,11 +1099,10 @@ static const char *get_name(unsigned int bank, struct threshold_block *b)
 		return th_names[bank];
 	}

-	if (!smca_banks[bank].hwid)
+	bank_type = smca_get_bank_type(bank);
+	if (bank_type >= N_SMCA_BANK_TYPES)
 		return NULL;

-	bank_type = smca_banks[bank].hwid->bank_type;
-
 	if (b && bank_type == SMCA_UMC) {
 		if (b->block < ARRAY_SIZE(smca_umc_block_names))
 			return smca_umc_block_names[b->block];