From c3d1fb567a634dcdff4c6f6095b2053260988336 Mon Sep 17 00:00:00 2001 From: "Naveen N. Rao" Date: Mon, 1 Jul 2013 21:08:47 +0530 Subject: mce: acpi/apei: Honour Firmware First for MCA banks listed in APEI HEST CMC The Corrected Machine Check structure (CMC) in HEST has a flag which can be set by the firmware to indicate to the OS that it prefers to process the corrected error events first. In this scenario, the OS is expected to not monitor for corrected errors (through CMCI/polling). Instead, the firmware notifies the OS on corrected error events through GHES. Linux already has support for GHES. This patch adds support for parsing CMC structure and to disable CMCI/polling if the firmware first flag is set. Further, the list of machine check bank structures at the end of CMC is used to determine which MCA banks function in FF mode, so that we continue to monitor error events on the other banks. Signed-off-by: Naveen N. Rao Acked-by: Borislav Petkov Signed-off-by: Tony Luck --- arch/x86/include/asm/mce.h | 3 +++ arch/x86/kernel/cpu/mcheck/mce-internal.h | 3 +++ arch/x86/kernel/cpu/mcheck/mce.c | 28 +++++++++++++++++++++ arch/x86/kernel/cpu/mcheck/mce_intel.c | 42 +++++++++++++++++++++++-------- 4 files changed, 66 insertions(+), 10 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index fa5f71e021d5..9c91683ab5e6 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -188,6 +188,9 @@ extern void register_mce_write_callback(ssize_t (*)(struct file *filp, const char __user *ubuf, size_t usize, loff_t *off)); +/* Disable CMCI/polling for MCA bank claimed by firmware */ +extern void mce_disable_bank(int bank); + /* * Exception handler */ diff --git a/arch/x86/kernel/cpu/mcheck/mce-internal.h b/arch/x86/kernel/cpu/mcheck/mce-internal.h index 5b7d4fa5d3b7..09edd0b65fef 100644 --- a/arch/x86/kernel/cpu/mcheck/mce-internal.h +++ b/arch/x86/kernel/cpu/mcheck/mce-internal.h @@ -25,15 +25,18 @@ int mce_severity(struct mce *a, int tolerant, char **msg); struct dentry *mce_get_debugfs_dir(void); extern struct mce_bank *mce_banks; +extern mce_banks_t mce_banks_ce_disabled; #ifdef CONFIG_X86_MCE_INTEL unsigned long mce_intel_adjust_timer(unsigned long interval); void mce_intel_cmci_poll(void); void mce_intel_hcpu_update(unsigned long cpu); +void cmci_disable_bank(int bank); #else # define mce_intel_adjust_timer mce_adjust_timer_default static inline void mce_intel_cmci_poll(void) { } static inline void mce_intel_hcpu_update(unsigned long cpu) { } +static inline void cmci_disable_bank(int bank) { } #endif void mce_timer_kick(unsigned long interval); diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 9239504b41cb..5bf32c70a9c0 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -94,6 +94,15 @@ DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = { [0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL }; +/* + * MCA banks controlled through firmware first for corrected errors. + * This is a global list of banks for which we won't enable CMCI and we + * won't poll. Firmware controls these banks and is responsible for + * reporting corrected errors through GHES. Uncorrected/recoverable + * errors are still notified through a machine check. + */ +mce_banks_t mce_banks_ce_disabled; + static DEFINE_PER_CPU(struct work_struct, mce_work); static void (*quirk_no_way_out)(int bank, struct mce *m, struct pt_regs *regs); @@ -1932,6 +1941,25 @@ static struct miscdevice mce_chrdev_device = { &mce_chrdev_ops, }; +static void __mce_disable_bank(void *arg) +{ + int bank = *((int *)arg); + __clear_bit(bank, __get_cpu_var(mce_poll_banks)); + cmci_disable_bank(bank); +} + +void mce_disable_bank(int bank) +{ + if (bank >= mca_cfg.banks) { + pr_warn(FW_BUG + "Ignoring request to disable invalid MCA bank %d.\n", + bank); + return; + } + set_bit(bank, mce_banks_ce_disabled); + on_each_cpu(__mce_disable_bank, &bank, 1); +} + /* * mce=off Disables machine check * mce=no_cmci Disables CMCI diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c index ae1697c2afe3..488eae3ec3e2 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c @@ -191,6 +191,10 @@ static void cmci_discover(int banks) if (test_bit(i, owned)) continue; + /* Skip banks in firmware first mode */ + if (test_bit(i, mce_banks_ce_disabled)) + continue; + rdmsrl(MSR_IA32_MCx_CTL2(i), val); /* Already owned by someone else? */ @@ -259,6 +263,19 @@ void cmci_recheck(void) local_irq_restore(flags); } +/* Caller must hold the lock on cmci_discover_lock */ +static void __cmci_disable_bank(int bank) +{ + u64 val; + + if (!test_bit(bank, __get_cpu_var(mce_banks_owned))) + return; + rdmsrl(MSR_IA32_MCx_CTL2(bank), val); + val &= ~MCI_CTL2_CMCI_EN; + wrmsrl(MSR_IA32_MCx_CTL2(bank), val); + __clear_bit(bank, __get_cpu_var(mce_banks_owned)); +} + /* * Disable CMCI on this CPU for all banks it owns when it goes down. * This allows other CPUs to claim the banks on rediscovery. @@ -268,20 +285,12 @@ void cmci_clear(void) unsigned long flags; int i; int banks; - u64 val; if (!cmci_supported(&banks)) return; raw_spin_lock_irqsave(&cmci_discover_lock, flags); - for (i = 0; i < banks; i++) { - if (!test_bit(i, __get_cpu_var(mce_banks_owned))) - continue; - /* Disable CMCI */ - rdmsrl(MSR_IA32_MCx_CTL2(i), val); - val &= ~MCI_CTL2_CMCI_EN; - wrmsrl(MSR_IA32_MCx_CTL2(i), val); - __clear_bit(i, __get_cpu_var(mce_banks_owned)); - } + for (i = 0; i < banks; i++) + __cmci_disable_bank(i); raw_spin_unlock_irqrestore(&cmci_discover_lock, flags); } @@ -315,6 +324,19 @@ void cmci_reenable(void) cmci_discover(banks); } +void cmci_disable_bank(int bank) +{ + int banks; + unsigned long flags; + + if (!cmci_supported(&banks)) + return; + + raw_spin_lock_irqsave(&cmci_discover_lock, flags); + __cmci_disable_bank(bank); + raw_spin_unlock_irqrestore(&cmci_discover_lock, flags); +} + static void intel_init_cmci(void) { int banks; -- cgit v1.2.3-70-g09d2 From 9ad95879cd1b22ed016c804f8d686ff83a41a9d4 Mon Sep 17 00:00:00 2001 From: "Naveen N. Rao" Date: Mon, 1 Jul 2013 21:08:54 +0530 Subject: mce: acpi/apei: Add a boot option to disable ff mode for corrected errors Add a boot option to disable firmware first mode for corrected errors. Signed-off-by: Naveen N. Rao Acked-by: Borislav Petkov Signed-off-by: Tony Luck --- Documentation/x86/x86_64/boot-options.txt | 5 +++++ arch/x86/include/asm/acpi.h | 2 ++ arch/x86/kernel/acpi/boot.c | 5 +++++ drivers/acpi/apei/hest.c | 3 ++- 4 files changed, 14 insertions(+), 1 deletion(-) (limited to 'arch') diff --git a/Documentation/x86/x86_64/boot-options.txt b/Documentation/x86/x86_64/boot-options.txt index e9e8ddbbf376..1228b22e142b 100644 --- a/Documentation/x86/x86_64/boot-options.txt +++ b/Documentation/x86/x86_64/boot-options.txt @@ -176,6 +176,11 @@ ACPI acpi=noirq Don't route interrupts + acpi=nocmcff Disable firmware first mode for corrected errors. This + disables parsing the HEST CMC error source to check if + firmware has set the FF flag. This may result in + duplicate corrected error reports. + PCI pci=off Don't use PCI diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index b31bf97775fc..42db2b88953b 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -86,6 +86,7 @@ extern int acpi_pci_disabled; extern int acpi_skip_timer_override; extern int acpi_use_timer_override; extern int acpi_fix_pin2_polarity; +extern int acpi_disable_cmcff; extern u8 acpi_sci_flags; extern int acpi_sci_override_gsi; @@ -168,6 +169,7 @@ static inline void arch_acpi_set_pdc_bits(u32 *buf) #define acpi_lapic 0 #define acpi_ioapic 0 +#define acpi_disable_cmcff 0 static inline void acpi_noirq_set(void) { } static inline void acpi_disable_pci(void) { } static inline void disable_acpi(void) { } diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 230c8ea878e5..d1998d580080 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -66,6 +66,7 @@ EXPORT_SYMBOL(acpi_pci_disabled); int acpi_lapic; int acpi_ioapic; int acpi_strict; +int acpi_disable_cmcff; u8 acpi_sci_flags __initdata; int acpi_sci_override_gsi __initdata; @@ -1619,6 +1620,10 @@ static int __init parse_acpi(char *arg) /* "acpi=copy_dsdt" copys DSDT */ else if (strcmp(arg, "copy_dsdt") == 0) { acpi_gbl_copy_dsdt_locally = 1; + } + /* "acpi=nocmcff" disables FF mode for corrected errors */ + else if (strcmp(arg, "nocmcff") == 0) { + acpi_disable_cmcff = 1; } else { /* Core will printk when we return error. */ return -EINVAL; diff --git a/drivers/acpi/apei/hest.c b/drivers/acpi/apei/hest.c index b108b116bf28..502024502b13 100644 --- a/drivers/acpi/apei/hest.c +++ b/drivers/acpi/apei/hest.c @@ -262,7 +262,8 @@ void __init acpi_hest_init(void) goto err; } - apei_hest_parse(hest_parse_cmc, NULL); + if (!acpi_disable_cmcff) + apei_hest_parse(hest_parse_cmc, NULL); if (!ghes_disable) { rc = apei_hest_parse(hest_parse_ghes_count, &ghes_count); -- cgit v1.2.3-70-g09d2 From 0ca06c0857aee11911f91621db14498496f2c2cd Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Wed, 24 Jul 2013 13:54:20 -0700 Subject: x86/mce: Pay no attention to 'F' bit in MCACOD when parsing 'UC' errors The 0x1000 bit of the MCACOD field of machine check MCi_STATUS registers is only defined for corrected errors (where it means that hardware may be filtering errors see SDM section 15.9.2.1). For uncorrected errors it may, or may not be set - so we should mask it out when checking for the architecturaly defined recoverable error signatures (see SDM 15.9.3.1 and 15.9.3.2) Acked-by: Naveen N. Rao Signed-off-by: Tony Luck --- arch/x86/include/asm/mce.h | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 29e3093bbd21..aa97342e7ca0 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -32,11 +32,20 @@ #define MCI_STATUS_PCC (1ULL<<57) /* processor context corrupt */ #define MCI_STATUS_S (1ULL<<56) /* Signaled machine check */ #define MCI_STATUS_AR (1ULL<<55) /* Action required */ -#define MCACOD 0xffff /* MCA Error Code */ + +/* + * Note that the full MCACOD field of IA32_MCi_STATUS MSR is + * bits 15:0. But bit 12 is the 'F' bit, defined for corrected + * errors to indicate that errors are being filtered by hardware. + * We should mask out bit 12 when looking for specific signatures + * of uncorrected errors - so the F bit is deliberately skipped + * in this #define. + */ +#define MCACOD 0xefff /* MCA Error Code */ /* Architecturally defined codes from SDM Vol. 3B Chapter 15 */ #define MCACOD_SCRUB 0x00C0 /* 0xC0-0xCF Memory Scrubbing */ -#define MCACOD_SCRUBMSK 0xfff0 +#define MCACOD_SCRUBMSK 0xeff0 /* Skip bit 12 ('F' bit) */ #define MCACOD_L3WB 0x017A /* L3 Explicit Writeback */ #define MCACOD_DATA 0x0134 /* Data Load */ #define MCACOD_INSTR 0x0150 /* Instruction Fetch */ -- cgit v1.2.3-70-g09d2 From 7d64ac6422092adbbdaa279ab32f9d4c90a84558 Mon Sep 17 00:00:00 2001 From: Aravind Gopalakrishnan Date: Fri, 2 Aug 2013 17:43:03 -0500 Subject: x86, amd_nb: Clarify F15h, model 30h GART and L3 support F15h, models 0x30 and later don't have a GART. Note that. Also check CPUID leaf 0x80000006 for L3 prescence because there are models which don't sport an L3 cache. Signed-off-by: Aravind Gopalakrishnan [ Boris: rewrite commit message, cleanup comments. ] Signed-off-by: Borislav Petkov --- arch/x86/kernel/amd_nb.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/x86/kernel/amd_nb.c b/arch/x86/kernel/amd_nb.c index 3048ded1b598..59554dca96ec 100644 --- a/arch/x86/kernel/amd_nb.c +++ b/arch/x86/kernel/amd_nb.c @@ -20,6 +20,7 @@ const struct pci_device_id amd_nb_misc_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F3) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_M10H_F3) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_M30H_NB_F3) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_16H_NB_F3) }, {} }; @@ -27,6 +28,7 @@ EXPORT_SYMBOL(amd_nb_misc_ids); static const struct pci_device_id amd_nb_link_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_F4) }, + { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_M30H_NB_F4) }, { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_16H_NB_F4) }, {} }; @@ -81,12 +83,19 @@ int amd_cache_northbridges(void) next_northbridge(misc, amd_nb_misc_ids); node_to_amd_nb(i)->link = link = next_northbridge(link, amd_nb_link_ids); - } + } + /* GART present only on Fam15h upto model 0fh */ if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10 || - boot_cpu_data.x86 == 0x15) + (boot_cpu_data.x86 == 0x15 && boot_cpu_data.x86_model < 0x10)) amd_northbridges.flags |= AMD_NB_GART; + /* + * Check for L3 cache presence. + */ + if (!cpuid_edx(0x80000006)) + return 0; + /* * Some CPU families support L3 Cache Index Disable. There are some * limitations because of E382 and E388 on family 0x10. -- cgit v1.2.3-70-g09d2