Patch series "Introduce per NUMA node memory error statistics", v2.
Background
==========
In the RFC for Kernel Support of Memory Error Detection [1], one advantage
of software-based scanning over hardware patrol scrubber is the ability to
make statistics visible to system administrators. The statistics include
2 categories:
* Memory error statistics, for example, how many memory error are
encountered, how many of them are recovered by the kernel. Note these
memory errors are non-fatal to kernel: during the machine check
exception (MCE) handling kernel already classified MCE's severity to be
unnecessary to panic (but either action required or optional).
* Scanner statistics, for example how many times the scanner have fully
scanned a NUMA node, how many errors are first detected by the scanner.
The memory error statistics are useful to userspace and actually not
specific to scanner detected memory errors, and are the focus of this
patchset.
Motivation
==========
Memory error stats are important to userspace but insufficient in kernel
today. Datacenter administrators can better monitor a machine's memory
health with the visible stats. For example, while memory errors are
inevitable on servers with 10+ TB memory, starting server maintenance when
there are only 1~2 recovered memory errors could be overreacting; in cloud
production environment maintenance usually means live migrate all the
workload running on the server and this usually causes nontrivial
disruption to the customer. Providing insight into the scope of memory
errors on a system helps to determine the appropriate follow-up action.
In addition, the kernel's existing memory error stats need to be
standardized so that userspace can reliably count on their usefulness.
Today kernel provides following memory error info to userspace, but they
are not sufficient or have disadvantages:
* HardwareCorrupted in /proc/meminfo: number of bytes poisoned in total,
not per NUMA node stats though
* ras:memory_failure_event: only available after explicitly enabled
* /dev/mcelog provides many useful info about the MCEs, but doesn't
capture how memory_failure recovered memory MCEs
* kernel logs: userspace needs to process log text
Exposing memory error stats is also a good start for the in-kernel memory
error detector. Today the data source of memory error stats are either
direct memory error consumption, or hardware patrol scrubber detection
(either signaled as UCNA or SRAO). Once in-kernel memory scanner is
implemented, it will be the main source as it is usually configured to
scan memory DIMMs constantly and faster than hardware patrol scrubber.
How Implemented
===============
As Naoya pointed out [2], exposing memory error statistics to userspace is
useful independent of software or hardware scanner. Therefore we
implement the memory error statistics independent of the in-kernel memory
error detector. It exposes the following per NUMA node memory error
counters:
/sys/devices/system/node/node${X}/memory_failure/total
/sys/devices/system/node/node${X}/memory_failure/recovered
/sys/devices/system/node/node${X}/memory_failure/ignored
/sys/devices/system/node/node${X}/memory_failure/failed
/sys/devices/system/node/node${X}/memory_failure/delayed
These counters describe how many raw pages are poisoned and after the
attempted recoveries by the kernel, their resolutions: how many are
recovered, ignored, failed, or delayed respectively. This approach can be
easier to extend for future use cases than /proc/meminfo, trace event, and
log. The following math holds for the statistics:
* total = recovered + ignored + failed + delayed
These memory error stats are reset during machine boot.
The 1st commit introduces these sysfs entries. The 2nd commit populates
memory error stats every time memory_failure attempts memory error
recovery. The 3rd commit adds documentations for introduced stats.
[1] https://lore.kernel.org/linux-mm/
7E670362-C29E-4626-B546-
26530D54F937@gmail.com/T/#mc22959244f5388891c523882e61163c6e4d703af
[2] https://lore.kernel.org/linux-mm/
7E670362-C29E-4626-B546-
26530D54F937@gmail.com/T/#m52d8d7a333d8536bd7ce74253298858b1c0c0ac6
This patch (of 3):
Today kernel provides following memory error info to userspace, but each
has its own disadvantage
* HardwareCorrupted in /proc/meminfo: number of bytes poisoned in total,
not per NUMA node stats though
* ras:memory_failure_event: only available after explicitly enabled
* /dev/mcelog provides many useful info about the MCEs, but
doesn't capture how memory_failure recovered memory MCEs
* kernel logs: userspace needs to process log text
Exposes per NUMA node memory error stats as sysfs entries:
/sys/devices/system/node/node${X}/memory_failure/total
/sys/devices/system/node/node${X}/memory_failure/recovered
/sys/devices/system/node/node${X}/memory_failure/ignored
/sys/devices/system/node/node${X}/memory_failure/failed
/sys/devices/system/node/node${X}/memory_failure/delayed
These counters describe how many raw pages are poisoned and after the
attempted recoveries by the kernel, their resolutions: how many are
recovered, ignored, failed, or delayed respectively. The following math
holds for the statistics:
* total = recovered + ignored + failed + delayed
Link: https://lkml.kernel.org/r/20230120034622.2698268-1-jiaqiyan@google.com
Link: https://lkml.kernel.org/r/20230120034622.2698268-2-jiaqiyan@google.com
Signed-off-by: Jiaqi Yan <jiaqiyan@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Yang Shi <shy828301@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
#ifdef CONFIG_HAVE_ARCH_NODE_DEV_GROUP
&arch_node_dev_group,
#endif
+#ifdef CONFIG_MEMORY_FAILURE
+ &memory_failure_attr_group,
+#endif
NULL
};
MF_MSG_UNKNOWN,
};
+/*
+ * Sysfs entries for memory failure handling statistics.
+ */
+extern const struct attribute_group memory_failure_attr_group;
+
#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS)
extern void clear_huge_page(struct page *page,
unsigned long addr_hint,
};
#endif
+#ifdef CONFIG_MEMORY_FAILURE
+/*
+ * Per NUMA node memory failure handling statistics.
+ */
+struct memory_failure_stats {
+ /*
+ * Number of raw pages poisoned.
+ * Cases not accounted: memory outside kernel control, offline page,
+ * arch-specific memory_failure (SGX), hwpoison_filter() filtered
+ * error events, and unpoison actions from hwpoison_unpoison.
+ */
+ unsigned long total;
+ /*
+ * Recovery results of poisoned raw pages handled by memory_failure,
+ * in sync with mf_result.
+ * total = ignored + failed + delayed + recovered.
+ * total * PAGE_SIZE * #nodes = /proc/meminfo/HardwareCorrupted.
+ */
+ unsigned long ignored;
+ unsigned long failed;
+ unsigned long delayed;
+ unsigned long recovered;
+};
+#endif
+
/*
* On NUMA machines, each NUMA node would have a pg_data_t to describe
* it's memory layout. On UMA machines there is a single pglist_data which
#ifdef CONFIG_NUMA
struct memory_tier __rcu *memtier;
#endif
+#ifdef CONFIG_MEMORY_FAILURE
+ struct memory_failure_stats mf_stats;
+#endif
} pg_data_t;
#define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages)
memblk_nr_poison_sub(pfn, i);
}
+/**
+ * MF_ATTR_RO - Create sysfs entry for each memory failure statistics.
+ * @_name: name of the file in the per NUMA sysfs directory.
+ */
+#define MF_ATTR_RO(_name) \
+static ssize_t _name##_show(struct device *dev, \
+ struct device_attribute *attr, \
+ char *buf) \
+{ \
+ struct memory_failure_stats *mf_stats = \
+ &NODE_DATA(dev->id)->mf_stats; \
+ return sprintf(buf, "%lu\n", mf_stats->_name); \
+} \
+static DEVICE_ATTR_RO(_name)
+
+MF_ATTR_RO(total);
+MF_ATTR_RO(ignored);
+MF_ATTR_RO(failed);
+MF_ATTR_RO(delayed);
+MF_ATTR_RO(recovered);
+
+static struct attribute *memory_failure_attr[] = {
+ &dev_attr_total.attr,
+ &dev_attr_ignored.attr,
+ &dev_attr_failed.attr,
+ &dev_attr_delayed.attr,
+ &dev_attr_recovered.attr,
+ NULL,
+};
+
+const struct attribute_group memory_failure_attr_group = {
+ .name = "memory_failure",
+ .attrs = memory_failure_attr,
+};
+
/*
* Return values:
* 1: the page is dissolved (if needed) and taken off from buddy,