81e42fc1 - EDAC: Update memory repair control interface for memory sparing feature

Mon, 24 Feb 2025 11:13:40 +0000

EDAC: Update memory repair control interface for memory sparing featureUpdate memory repair control interface for memory sparing feature.CXL memory devices can support soft and hard memory sparing at cacheline,row, bank and rank granularities. Memory sparing is defined as a repairfunction that replaces a portion of memory with a portion of functionalmemory at that same granularity.When a CXL device detects an error in memory, it will report to the hostthat there's need for a repair maintenance operation by using an eventrecord where the "maintenance needed" flag is set.The event records contain the device physical address (DPA) and otherattributes of the memory to repair such as bank group, bank, rank, row,column, channel etc.The kernel will report the corresponding CXL general media or DRAM traceevent to userspace, and userspace tools (e.g. rasdaemon) will initiatea repair operation in response to the device request via the sysfsrepair control. [ bp: Massage. ]Signed-off-by: Shiju Jose Signed-off-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/r/20250212143654.1893-15-shiju.jose@huawei.com List of files: /linux-6.15/Documentation/ABI/testing/sysfs-edac-memory-repair

699ea521 - EDAC: Add a memory repair control feature

Wed, 12 Feb 2025 14:36:42 +0000

EDAC: Add a memory repair control featureAdd a generic EDAC memory repair control driver to manage memory repairs inthe system, such as CXL Post Package Repair (PPR) and other soft and hard PPRfeatures.For example, a CXL device with DRAM components that support PPR features mayimplement PPR maintenance operations. DRAM components may support two types ofPPR: - hard PPR, for a permanent row repair, and - soft PPR, for a temporary row repair.Soft PPR is much faster than hard PPR, but the repair is lost with a powercycle.When a CXL device detects an error in a memory, it may report the need fora repair maintenance operation by using an event record where the "maintenanceneeded" flag is set. The event records contain the device physicaladdress (DPA) and other optional attributes of the memory to repair.The kernel will report the corresponding CXL general media or DRAM trace eventto userspace, and userspace tools (e.g. rasdaemon) will initiate a repairoperation in response to the device request via the sysfs repair control.Device with memory repair features registers with EDAC device driver, whichretrieves a memory repair descriptor from EDAC memory repair driver and exposesthe sysfs repair control attributes to userspace in /sys/bus/edac/devices//mem_repairX/.The common memory repair control interface abstracts the control of arbitrarymemory repair functionality into a standardized set of functions. The sysfsmemory repair attribute nodes are only available if the client driver hasimplemented the corresponding attribute callback function and providedoperations to the EDAC device driver during registration. [ bp: Massage, fixup edac_dev_register() retvals, merge write_overflow fix to mem_repair_create_desc() ]Signed-off-by: Shiju Jose Signed-off-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/r/20250212143654.1893-5-shiju.jose@huawei.com List of files: /linux-6.15/Documentation/ABI/testing/sysfs-edac-memory-repair

Changes in sysfs-edac-memory-repair

81e42fc1 - EDAC: Update memory repair control interface for memory sparing feature

699ea521 - EDAC: Add a memory repair control feature