Messages in this thread Patch in this message | | | From | Jane Chu <> | Subject | [PATCH v2] x86/mce: retrieve poison range from hardware whenever supported | Date | Fri, 15 Jul 2022 21:31:04 -0600 |
| |
With Commit 7917f9cdb503 ("acpi/nfit: rely on mce->misc to determine poison granularity") that changed nfit_handle_mce() callback to report badrange according to 1ULL << MCI_MISC_ADDR_LSB(mce->misc), it's been discovered that the mce->misc LSB field is 0x1000 bytes, hence injecting 2 back-to-back poisons and the driver ends up logging 8 badblocks, because 0x1000 bytes is 8 512-byte.
Dan Williams noticed that apei_mce_report_mem_error() hardcode the LSB field to PAGE_SHIFT instead of consulting the input struct cper_sec_mem_err record. So change to rely on hardware whenever support is available.
v1: https://lkml.org/lkml/2022/7/15/1040 Signed-off-by: Jane Chu <jane.chu@oracle.com> --- arch/x86/kernel/cpu/mce/apei.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/arch/x86/kernel/cpu/mce/apei.c b/arch/x86/kernel/cpu/mce/apei.c index 717192915f28..a4d5893632e0 100644 --- a/arch/x86/kernel/cpu/mce/apei.c +++ b/arch/x86/kernel/cpu/mce/apei.c @@ -29,6 +29,7 @@ void apei_mce_report_mem_error(int severity, struct cper_sec_mem_err *mem_err) { struct mce m; + int lsb; if (!(mem_err->validation_bits & CPER_MEM_VALID_PA)) return; @@ -37,7 +38,10 @@ void apei_mce_report_mem_error(int severity, struct cper_sec_mem_err *mem_err) m.bank = -1; /* Fake a memory read error with unknown channel */ m.status = MCI_STATUS_VAL | MCI_STATUS_EN | MCI_STATUS_ADDRV | MCI_STATUS_MISCV | 0x9f; - m.misc = (MCI_MISC_ADDR_PHYS << 6) | PAGE_SHIFT; + lsb = __builtin_ffs(mem_err->physical_addr_mask) - 1; + if (lsb <= 0) + lsb = PAGE_SHIFT; + m.misc = (MCI_MISC_ADDR_PHYS << 6) | lsb; if (severity >= GHES_SEV_RECOVERABLE) m.status |= MCI_STATUS_UC; -- 2.18.4
| |