The NFIT machine check handler uses the physical address from the 'mce' structure, and compares it against information in the ACPI NFIT table to determine whether that location lies on an NVDIMM. The mce->addr field however may not always be valid, and this is indicated by the MCI_STATUS_ADDRV bit in the status field.

Export mce_usable_address() which already performs validation for the address, and use it in the NFIT handler.

Reported-by: Robert Elliott elliott@hpe.com Fixes: 6839a6d96f4e ("nfit: do an ARS scrub on hitting a latent media error") Cc: stable@vger.kernel.org Cc: Dan Williams dan.j.williams@intel.com Cc: Tony Luck tony.luck@intel.com Cc: Borislav Petkov bp@alien8.de Signed-off-by: Vishal Verma vishal.l.verma@intel.com --- arch/x86/include/asm/mce.h | 1 + arch/x86/kernel/cpu/mcheck/mce.c | 3 ++- drivers/acpi/nfit/mce.c | 4 ++++ 3 files changed, 7 insertions(+), 1 deletion(-)

v3: Add this patch to address the address validation (Robert)

diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 3111b3cee2ee..eb786f90f2d3 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -217,6 +217,7 @@ static inline int umc_normaddr_to_sysaddr(u64 norm_addr, u16 nid, u8 umc, u64 *s int mce_available(struct cpuinfo_x86 *c); bool mce_is_memory_error(struct mce *m); bool mce_is_correctable(struct mce *m); +int mce_usable_address(struct mce *m);

DECLARE_PER_CPU(unsigned, mce_exception_count); DECLARE_PER_CPU(unsigned, mce_poll_count); diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 27015948bc41..cdbedeb3f3db 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -485,7 +485,7 @@ static void mce_report_event(struct pt_regs *regs) * be somewhat complicated (e.g. segment offset would require an instruction * parser). So only support physical addresses up to page granuality for now. */ -static int mce_usable_address(struct mce *m) +int mce_usable_address(struct mce *m) { if (!(m->status & MCI_STATUS_ADDRV)) return 0; @@ -505,6 +505,7 @@ static int mce_usable_address(struct mce *m)

return 1; } +EXPORT_SYMBOL_GPL(mce_usable_address);

bool mce_is_memory_error(struct mce *m) { diff --git a/drivers/acpi/nfit/mce.c b/drivers/acpi/nfit/mce.c index 7a51707f87e9..d943ed50f0b4 100644 --- a/drivers/acpi/nfit/mce.c +++ b/drivers/acpi/nfit/mce.c @@ -29,6 +29,10 @@ static int nfit_handle_mce(struct notifier_block *nb, unsigned long val, if (!mce_is_memory_error(mce) || mce_is_correctable(mce)) return NOTIFY_DONE;

+ /* Verify the address reported in the mce is valid */ + if (!mce_usable_address(mce)) + return NOTIFY_DONE; + /* * mce->addr contains the physical addr accessed that caused the * machine check. We need to walk through the list of NFITs, and see

Hi, Vishal,

Vishal Verma vishal.l.verma@intel.com writes:

The mce handler for 'nfit' devices is called for memory errors on a Non-Volatile DIMM, and adds the error location to a 'badblocks' list. This list is used by the various NVDIMM drivers to avoid consuming known poison locations during IO.

The mce handler gets called for both corrected and uncorrectable errors.

Sorry for necroposting. I thought the point of the CEC was to make sure that the other registered decoders only ever saw uncorrected errors. How do we end up getting called with a correctable error?

Thanks, Jeff

Until now, both kinds of errors have been added to the badblocks list. However, corrected memory errors indicate that the problem has already been fixed by hardware, and the resulting interrupt is merely a notification to Linux. As far as future accesses to that location are concerned, it is perfectly fine to use, and thus doesn't need to be included in the above badblocks list.

Add a check in the nfit mce handler to filter out corrected mce events, and only process uncorrectable errors.

Reported-by: Omar Avelar omar.avelar@intel.com Fixes: 6839a6d96f4e ("nfit: do an ARS scrub on hitting a latent media error") Cc: stable@vger.kernel.org Cc: Dan Williams dan.j.williams@intel.com Cc: Tony Luck tony.luck@intel.com Cc: Borislav Petkov bp@alien8.de Signed-off-by: Vishal Verma vishal.l.verma@intel.com

---

arch/x86/include/asm/mce.h | 1 + arch/x86/kernel/cpu/mcheck/mce.c | 3 ++- drivers/acpi/nfit/mce.c | 4 ++-- 3 files changed, 5 insertions(+), 3 deletions(-)

v3: Unchanged from v2

diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 3a17107594c8..3111b3cee2ee 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -216,6 +216,7 @@ static inline int umc_normaddr_to_sysaddr(u64 norm_addr, u16 nid, u8 umc, u64 *s int mce_available(struct cpuinfo_x86 *c); bool mce_is_memory_error(struct mce *m); +bool mce_is_correctable(struct mce *m); DECLARE_PER_CPU(unsigned, mce_exception_count); DECLARE_PER_CPU(unsigned, mce_poll_count); diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 953b3ce92dcc..27015948bc41 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -534,7 +534,7 @@ bool mce_is_memory_error(struct mce *m) } EXPORT_SYMBOL_GPL(mce_is_memory_error); -static bool mce_is_correctable(struct mce *m) +bool mce_is_correctable(struct mce *m) { if (m->cpuvendor == X86_VENDOR_AMD && m->status & MCI_STATUS_DEFERRED) return false; @@ -544,6 +544,7 @@ static bool mce_is_correctable(struct mce *m) return true; } +EXPORT_SYMBOL_GPL(mce_is_correctable); static bool cec_add_mce(struct mce *m) { diff --git a/drivers/acpi/nfit/mce.c b/drivers/acpi/nfit/mce.c index e9626bf6ca29..7a51707f87e9 100644 --- a/drivers/acpi/nfit/mce.c +++ b/drivers/acpi/nfit/mce.c @@ -25,8 +25,8 @@ static int nfit_handle_mce(struct notifier_block *nb, unsigned long val, struct acpi_nfit_desc *acpi_desc; struct nfit_spa *nfit_spa;

• /* We only care about memory errors */
• if (!mce_is_memory_error(mce))

• /* We only care about uncorrectable memory errors */
• if (!mce_is_memory_error(mce) || mce_is_correctable(mce)) return NOTIFY_DONE;

/*