From: Huang Ying ying.huang@intel.com

mainline inclusion from mainline-v6.7-rc1 commit 07a8bdd4120ced3490ef9adf51b8086af0aaa8e7 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I90U82

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...

--------------------------------

Patch series "memory tiering: calculate abstract distance based on ACPI HMAT", v4.

We have the explicit memory tiers framework to manage systems with multiple types of memory, e.g., DRAM in DIMM slots and CXL memory devices. Where, same kind of memory devices will be grouped into memory types, then put into memory tiers. To describe the performance of a memory type, abstract distance is defined. Which is in direct proportion to the memory latency and inversely proportional to the memory bandwidth. To keep the code as simple as possible, fixed abstract distance is used in dax/kmem to describe slow memory such as Optane DCPMM.

To support more memory types, in this series, we added the abstract distance calculation algorithm management mechanism, provided a algorithm implementation based on ACPI HMAT, and used the general abstract distance calculation interface in dax/kmem driver. So, dax/kmem can support HBM (high bandwidth memory) in addition to the original Optane DCPMM.

This patch (of 4):

The abstract distance may be calculated by various drivers, such as ACPI HMAT, CXL CDAT, etc. While it may be used by various code which hot-add memory node, such as dax/kmem etc. To decouple the algorithm users and the providers, the abstract distance calculation algorithms management mechanism is implemented in this patch. It provides interface for the providers to register the implementation, and interface for the users.

Multiple algorithm implementations can cooperate via calculating abstract distance for different memory nodes. The preference of algorithm implementations can be specified via priority (notifier_block.priority).

Link: https://lkml.kernel.org/r/20230926060628.265989-1-ying.huang@intel.com Link: https://lkml.kernel.org/r/20230926060628.265989-2-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Tested-by: Bharata B Rao bharata@amd.com Reviewed-by: Alistair Popple apopple@nvidia.com Reviewed-by: Dave Jiang dave.jiang@intel.com Cc: Aneesh Kumar K.V aneesh.kumar@linux.ibm.com Cc: Wei Xu weixugc@google.com Cc: Dan Williams dan.j.williams@intel.com Cc: Dave Hansen dave.hansen@intel.com Cc: Davidlohr Bueso dave@stgolabs.net Cc: Johannes Weiner hannes@cmpxchg.org Cc: Jonathan Cameron Jonathan.Cameron@huawei.com Cc: Michal Hocko mhocko@kernel.org Cc: Yang Shi shy828301@gmail.com Cc: Rafael J Wysocki rafael.j.wysocki@intel.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Tong Tiangen tongtiangen@huawei.com --- include/linux/memory-tiers.h | 19 ++++++++++++ mm/memory-tiers.c | 59 ++++++++++++++++++++++++++++++++++++ 2 files changed, 78 insertions(+)

diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h index 437441cdf78f..c8382220cced 100644 --- a/include/linux/memory-tiers.h +++ b/include/linux/memory-tiers.h @@ -6,6 +6,7 @@ #include <linux/nodemask.h> #include <linux/kref.h> #include <linux/mmzone.h> +#include <linux/notifier.h> /* * Each tier cover a abstrace distance chunk size of 128 */ @@ -36,6 +37,9 @@ struct memory_dev_type *alloc_memory_type(int adistance); void put_memory_type(struct memory_dev_type *memtype); void init_node_memory_type(int node, struct memory_dev_type *default_type); void clear_node_memory_type(int node, struct memory_dev_type *memtype); +int register_mt_adistance_algorithm(struct notifier_block *nb); +int unregister_mt_adistance_algorithm(struct notifier_block *nb); +int mt_calc_adistance(int node, int *adist); #ifdef CONFIG_MIGRATION int next_demotion_node(int node); void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets); @@ -97,5 +101,20 @@ static inline bool node_is_toptier(int node) { return true; } + +static inline int register_mt_adistance_algorithm(struct notifier_block *nb) +{ + return 0; +} + +static inline int unregister_mt_adistance_algorithm(struct notifier_block *nb) +{ + return 0; +} + +static inline int mt_calc_adistance(int node, int *adist) +{ + return NOTIFY_DONE; +} #endif /* CONFIG_NUMA */ #endif /* _LINUX_MEMORY_TIERS_H */ diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c index 37a4f59d9585..76c0ad47a5ad 100644 --- a/mm/memory-tiers.c +++ b/mm/memory-tiers.c @@ -5,6 +5,7 @@ #include <linux/kobject.h> #include <linux/memory.h> #include <linux/memory-tiers.h> +#include <linux/notifier.h>

#include "internal.h"

@@ -105,6 +106,8 @@ static int top_tier_adistance; static struct demotion_nodes *node_demotion __read_mostly; #endif /* CONFIG_MIGRATION */

+static BLOCKING_NOTIFIER_HEAD(mt_adistance_algorithms); + static inline struct memory_tier *to_memory_tier(struct device *device) { return container_of(device, struct memory_tier, dev); @@ -592,6 +595,62 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype) } EXPORT_SYMBOL_GPL(clear_node_memory_type);

+/** + * register_mt_adistance_algorithm() - Register memory tiering abstract distance algorithm + * @nb: The notifier block which describe the algorithm + * + * Return: 0 on success, errno on error. + * + * Every memory tiering abstract distance algorithm provider needs to + * register the algorithm with register_mt_adistance_algorithm(). To + * calculate the abstract distance for a specified memory node, the + * notifier function will be called unless some high priority + * algorithm has provided result. The prototype of the notifier + * function is as follows, + * + * int (*algorithm_notifier)(struct notifier_block *nb, + * unsigned long nid, void *data); + * + * Where "nid" specifies the memory node, "data" is the pointer to the + * returned abstract distance (that is, "int *adist"). If the + * algorithm provides the result, NOTIFY_STOP should be returned. + * Otherwise, return_value & %NOTIFY_STOP_MASK == 0 to allow the next + * algorithm in the chain to provide the result. + */ +int register_mt_adistance_algorithm(struct notifier_block *nb) +{ + return blocking_notifier_chain_register(&mt_adistance_algorithms, nb); +} +EXPORT_SYMBOL_GPL(register_mt_adistance_algorithm); + +/** + * unregister_mt_adistance_algorithm() - Unregister memory tiering abstract distance algorithm + * @nb: the notifier block which describe the algorithm + * + * Return: 0 on success, errno on error. + */ +int unregister_mt_adistance_algorithm(struct notifier_block *nb) +{ + return blocking_notifier_chain_unregister(&mt_adistance_algorithms, nb); +} +EXPORT_SYMBOL_GPL(unregister_mt_adistance_algorithm); + +/** + * mt_calc_adistance() - Calculate abstract distance with registered algorithms + * @node: the node to calculate abstract distance for + * @adist: the returned abstract distance + * + * Return: if return_value & %NOTIFY_STOP_MASK != 0, then some + * abstract distance algorithm provides the result, and return it via + * @adist. Otherwise, no algorithm can provide the result and @adist + * will be kept as it is. + */ +int mt_calc_adistance(int node, int *adist) +{ + return blocking_notifier_call_chain(&mt_adistance_algorithms, node, adist); +} +EXPORT_SYMBOL_GPL(mt_calc_adistance); + static int __meminit memtier_hotplug_callback(struct notifier_block *self, unsigned long action, void *_arg) {

From: Huang Ying ying.huang@intel.com

mainline inclusion from mainline-v6.7-rc1 commit 3718c02dbd4c88d47b5af003acdb3d1112604ea3 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I90U82

Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i...

--------------------------------

A memory tiering abstract distance calculation algorithm based on ACPI HMAT is implemented. The basic idea is as follows.

The performance attributes of system default DRAM nodes are recorded as the base line. Whose abstract distance is MEMTIER_ADISTANCE_DRAM. Then, the ratio of the abstract distance of a memory node (target) to MEMTIER_ADISTANCE_DRAM is scaled based on the ratio of the performance attributes of the node to that of the default DRAM nodes.

The functions to record the read/write latency/bandwidth of the default DRAM nodes and calculate abstract distance according to read/write latency/bandwidth ratio will be used by CXL CDAT (Coherent Device Attribute Table) and other memory device drivers. So, they are put in memory-tiers.c.

Link: https://lkml.kernel.org/r/20230926060628.265989-4-ying.huang@intel.com Signed-off-by: "Huang, Ying" ying.huang@intel.com Tested-by: Bharata B Rao bharata@amd.com Reviewed-by: Dave Jiang dave.jiang@intel.com Reviewed-by: Alistair Popple apopple@nvidia.com Cc: Aneesh Kumar K.V aneesh.kumar@linux.ibm.com Cc: Wei Xu weixugc@google.com Cc: Dan Williams dan.j.williams@intel.com Cc: Dave Hansen dave.hansen@intel.com Cc: Davidlohr Bueso dave@stgolabs.net Cc: Johannes Weiner hannes@cmpxchg.org Cc: Jonathan Cameron Jonathan.Cameron@huawei.com Cc: Michal Hocko mhocko@kernel.org Cc: Yang Shi shy828301@gmail.com Cc: Rafael J Wysocki rafael.j.wysocki@intel.com Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Tong Tiangen tongtiangen@huawei.com --- drivers/acpi/numa/hmat.c | 65 +++++++++++++++++++++- include/linux/memory-tiers.h | 18 ++++++ mm/memory-tiers.c | 103 ++++++++++++++++++++++++++++++++++- 3 files changed, 183 insertions(+), 3 deletions(-)

diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c index 2dee0098f1a9..9ef5f1bdcfdb 100644 --- a/drivers/acpi/numa/hmat.c +++ b/drivers/acpi/numa/hmat.c @@ -24,6 +24,7 @@ #include <linux/node.h> #include <linux/sysfs.h> #include <linux/dax.h> +#include <linux/memory-tiers.h>

static u8 hmat_revision; static int hmat_disable __initdata; @@ -759,6 +760,61 @@ static int hmat_callback(struct notifier_block *self, return NOTIFY_OK; }

+static int hmat_set_default_dram_perf(void) +{ + int rc; + int nid, pxm; + struct memory_target *target; + struct node_hmem_attrs *attrs; + + if (!default_dram_type) + return -EIO; + + for_each_node_mask(nid, default_dram_type->nodes) { + pxm = node_to_pxm(nid); + target = find_mem_target(pxm); + if (!target) + continue; + attrs = &target->hmem_attrs[1]; + rc = mt_set_default_dram_perf(nid, attrs, "ACPI HMAT"); + if (rc) + return rc; + } + + return 0; +} + +static int hmat_calculate_adistance(struct notifier_block *self, + unsigned long nid, void *data) +{ + static DECLARE_BITMAP(p_nodes, MAX_NUMNODES); + struct memory_target *target; + struct node_hmem_attrs *perf; + int *adist = data; + int pxm; + + pxm = node_to_pxm(nid); + target = find_mem_target(pxm); + if (!target) + return NOTIFY_OK; + + mutex_lock(&target_lock); + hmat_update_target_attrs(target, p_nodes, 1); + mutex_unlock(&target_lock); + + perf = &target->hmem_attrs[1]; + + if (mt_perf_to_adistance(perf, adist)) + return NOTIFY_OK; + + return NOTIFY_STOP; +} + +static struct notifier_block hmat_adist_nb __meminitdata = { + .notifier_call = hmat_calculate_adistance, + .priority = 100, +}; + static __init void hmat_free_structures(void) { struct memory_target *target, *tnext; @@ -841,8 +897,13 @@ static __init int hmat_init(void) hmat_register_targets();

/* Keep the table and structures if the notifier may use them */ - if (!hotplug_memory_notifier(hmat_callback, HMAT_CALLBACK_PRI)) - return 0; + if (hotplug_memory_notifier(hmat_callback, HMAT_CALLBACK_PRI)) + goto out_put; + + if (!hmat_set_default_dram_perf()) + register_mt_adistance_algorithm(&hmat_adist_nb); + + return 0; out_put: hmat_free_structures(); acpi_put_table(tbl); diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h index c8382220cced..9d27ca3b143e 100644 --- a/include/linux/memory-tiers.h +++ b/include/linux/memory-tiers.h @@ -31,8 +31,11 @@ struct memory_dev_type { struct kref kref; };

+struct node_hmem_attrs; + #ifdef CONFIG_NUMA extern bool numa_demotion_enabled; +extern struct memory_dev_type *default_dram_type; struct memory_dev_type *alloc_memory_type(int adistance); void put_memory_type(struct memory_dev_type *memtype); void init_node_memory_type(int node, struct memory_dev_type *default_type); @@ -40,6 +43,9 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype); int register_mt_adistance_algorithm(struct notifier_block *nb); int unregister_mt_adistance_algorithm(struct notifier_block *nb); int mt_calc_adistance(int node, int *adist); +int mt_set_default_dram_perf(int nid, struct node_hmem_attrs *perf, + const char *source); +int mt_perf_to_adistance(struct node_hmem_attrs *perf, int *adist); #ifdef CONFIG_MIGRATION int next_demotion_node(int node); void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets); @@ -64,6 +70,7 @@ static inline bool node_is_toptier(int node) #else

#define numa_demotion_enabled false +#define default_dram_type NULL /* * CONFIG_NUMA implementation returns non NULL error. */ @@ -116,5 +123,16 @@ static inline int mt_calc_adistance(int node, int *adist) { return NOTIFY_DONE; } + +static inline int mt_set_default_dram_perf(int nid, struct node_hmem_attrs *perf, + const char *source) +{ + return -EIO; +} + +static inline int mt_perf_to_adistance(struct node_hmem_attrs *perf, int *adist) +{ + return -EIO; +} #endif /* CONFIG_NUMA */ #endif /* _LINUX_MEMORY_TIERS_H */ diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c index 76c0ad47a5ad..fa1a8b418f9a 100644 --- a/mm/memory-tiers.c +++ b/mm/memory-tiers.c @@ -37,7 +37,7 @@ struct node_memory_type_map { static DEFINE_MUTEX(memory_tier_lock); static LIST_HEAD(memory_tiers); static struct node_memory_type_map node_memory_types[MAX_NUMNODES]; -static struct memory_dev_type *default_dram_type; +struct memory_dev_type *default_dram_type;

static struct bus_type memory_tier_subsys = { .name = "memory_tiering", @@ -108,6 +108,11 @@ static struct demotion_nodes *node_demotion __read_mostly;

static BLOCKING_NOTIFIER_HEAD(mt_adistance_algorithms);

+static bool default_dram_perf_error; +static struct node_hmem_attrs default_dram_perf; +static int default_dram_perf_ref_nid = NUMA_NO_NODE; +static const char *default_dram_perf_ref_source; + static inline struct memory_tier *to_memory_tier(struct device *device) { return container_of(device, struct memory_tier, dev); @@ -595,6 +600,102 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype) } EXPORT_SYMBOL_GPL(clear_node_memory_type);

+static void dump_hmem_attrs(struct node_hmem_attrs *attrs, const char *prefix) +{ + pr_info( +"%sread_latency: %u, write_latency: %u, read_bandwidth: %u, write_bandwidth: %u\n", + prefix, attrs->read_latency, attrs->write_latency, + attrs->read_bandwidth, attrs->write_bandwidth); +} + +int mt_set_default_dram_perf(int nid, struct node_hmem_attrs *perf, + const char *source) +{ + int rc = 0; + + mutex_lock(&memory_tier_lock); + if (default_dram_perf_error) { + rc = -EIO; + goto out; + } + + if (perf->read_latency + perf->write_latency == 0 || + perf->read_bandwidth + perf->write_bandwidth == 0) { + rc = -EINVAL; + goto out; + } + + if (default_dram_perf_ref_nid == NUMA_NO_NODE) { + default_dram_perf = *perf; + default_dram_perf_ref_nid = nid; + default_dram_perf_ref_source = kstrdup(source, GFP_KERNEL); + goto out; + } + + /* + * The performance of all default DRAM nodes is expected to be + * same (that is, the variation is less than 10%). And it + * will be used as base to calculate the abstract distance of + * other memory nodes. + */ + if (abs(perf->read_latency - default_dram_perf.read_latency) * 10 > + default_dram_perf.read_latency || + abs(perf->write_latency - default_dram_perf.write_latency) * 10 > + default_dram_perf.write_latency || + abs(perf->read_bandwidth - default_dram_perf.read_bandwidth) * 10 > + default_dram_perf.read_bandwidth || + abs(perf->write_bandwidth - default_dram_perf.write_bandwidth) * 10 > + default_dram_perf.write_bandwidth) { + pr_info( +"memory-tiers: the performance of DRAM node %d mismatches that of the reference\n" +"DRAM node %d.\n", nid, default_dram_perf_ref_nid); + pr_info(" performance of reference DRAM node %d:\n", + default_dram_perf_ref_nid); + dump_hmem_attrs(&default_dram_perf, " "); + pr_info(" performance of DRAM node %d:\n", nid); + dump_hmem_attrs(perf, " "); + pr_info( +" disable default DRAM node performance based abstract distance algorithm.\n"); + default_dram_perf_error = true; + rc = -EINVAL; + } + +out: + mutex_unlock(&memory_tier_lock); + return rc; +} + +int mt_perf_to_adistance(struct node_hmem_attrs *perf, int *adist) +{ + if (default_dram_perf_error) + return -EIO; + + if (default_dram_perf_ref_nid == NUMA_NO_NODE) + return -ENOENT; + + if (perf->read_latency + perf->write_latency == 0 || + perf->read_bandwidth + perf->write_bandwidth == 0) + return -EINVAL; + + mutex_lock(&memory_tier_lock); + /* + * The abstract distance of a memory node is in direct proportion to + * its memory latency (read + write) and inversely proportional to its + * memory bandwidth (read + write). The abstract distance, memory + * latency, and memory bandwidth of the default DRAM nodes are used as + * the base. + */ + *adist = MEMTIER_ADISTANCE_DRAM * + (perf->read_latency + perf->write_latency) / + (default_dram_perf.read_latency + default_dram_perf.write_latency) * + (default_dram_perf.read_bandwidth + default_dram_perf.write_bandwidth) / + (perf->read_bandwidth + perf->write_bandwidth); + mutex_unlock(&memory_tier_lock); + + return 0; +} +EXPORT_SYMBOL_GPL(mt_perf_to_adistance); + /** * register_mt_adistance_algorithm() - Register memory tiering abstract distance algorithm * @nb: The notifier block which describe the algorithm