From: Rakie Kim <rakie.kim@sk.com> mainline inclusion from mainline-v6.9-rc1 commit dce41f5ae2539d1c20ae8de4e039630aec3c3f3c category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I9OHHN Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?i... -------------------------------- Patch series "mm/mempolicy: weighted interleave mempolicy and sysfs extension", v5. Weighted interleave is a new interleave policy intended to make use of heterogeneous memory environments appearing with CXL. The existing interleave mechanism does an even round-robin distribution of memory across all nodes in a nodemask, while weighted interleave distributes memory across nodes according to a provided weight. (Weight = Weighted interleave is intended to reduce average latency when bandwidth is pressured - therefore increasing total throughput. In other words: It allows greater use of the total available bandwidth in a heterogeneous hardware environment (different hardware provides different bandwidth capacity). As bandwidth is pressured, latency increases - first linearly and then exponentially. By keeping bandwidth usage distributed according to available bandwidth, we therefore can reduce the average latency of a cacheline fetch. A good explanation of the bandwidth vs latency response curve: https://mahmoudhatem.wordpress.com/2017/11/07/memory-bandwidth-vs-latency-re... From the article: ``` Constant region: The latency response is fairly constant for the first 40% of the sustained bandwidth. Linear region: In between 40% to 80% of the sustained bandwidth, the latency response increases almost linearly with the bandwidth demand of the system due to contention overhead by numerous memory requests. Exponential region: Between 80% to 100% of the sustained bandwidth, the memory latency is dominated by the contention latency which can be as much as twice the idle latency or more. Maximum sustained bandwidth : Is 65% to 75% of the theoretical maximum bandwidth. ``` As a general rule of thumb: * If bandwidth usage is low, latency does not increase. It is optimal to place data in the nearest (lowest latency) device. * If bandwidth usage is high, latency increases. It is optimal to place data such that bandwidth use is optimized per-device. This is the top line goal: Provide a user a mechanism to target using the "maximum sustained bandwidth" of each hardware component in a heterogenous memory system. For example, the stream benchmark demonstrates that 1:1 (default) interleave is actively harmful, while weighted interleave can be beneficial. Default interleave distributes data such that too much pressure is placed on devices with lower available bandwidth. Stream Benchmark (vs DRAM, 1 Socket + 1 CXL Device) Default interleave : -78% (slower than DRAM) Global weighting : -6% to +4% (workload dependant) Targeted weights : +2.5% to +4% (consistently better than DRAM) Global means the task-policy was set (set_mempolicy), while targeted means VMA policies were set (mbind2). We see weighted interleave is not always beneficial when applied globally, but is always beneficial when applied to bandwidth-driving memory regions. There are 4 patches in this set: 1) Implement system-global interleave weights as sysfs extension in mm/mempolicy.c. These weights are RCU protected, and a default weight set is provided (all weights are 1 by default). In future work, we intend to expose an interface for HMAT/CDAT code to set reasonable default values based on the memory configuration of the system discovered at boot/hotplug. 2) A mild refactor of some interleave-logic for re-use in the new weighted interleave logic. 3) MPOL_WEIGHTED_INTERLEAVE extension for set_mempolicy/mbind 4) Protect interleave logic (weighted and normal) with the mems_allowed seq cookie. If the nodemask changes while accessing it during a rebind, just retry the access. Included below are some performance and LTP test information, and a sample numactl branch which can be used for testing. = Performance summary = (tests may have different configurations, see extended info below) 1) MLC (W2) : +38% over DRAM. +264% over default interleave. MLC (W5) : +40% over DRAM. +226% over default interleave. 2) Stream : -6% to +4% over DRAM, +430% over default interleave. 3) XSBench : +19% over DRAM. +47% over default interleave. = LTP Testing Summary = existing mempolicy & mbind tests: pass mempolicy & mbind + weighted interleave (global weights): pass = version history v5: - style fixes - mems_allowed cookie protection to detect rebind issues, prevents spurious allocation failures and/or mis-allocations - sparse warning fixes related to __rcu on local variables ===================================================================== Performance tests - MLC From - Ravi Jonnalagadda <ravis.opensrc@micron.com> Hardware: Single-socket, multiple CXL memory expanders. Workload: W2 Data Signature: 2:1 read:write DRAM only bandwidth (GBps): 298.8 DRAM + CXL (default interleave) (GBps): 113.04 DRAM + CXL (weighted interleave)(GBps): 412.5 Gain over DRAM only: 1.38x Gain over default interleave: 2.64x Workload: W5 Data Signature: 1:1 read:write DRAM only bandwidth (GBps): 273.2 DRAM + CXL (default interleave) (GBps): 117.23 DRAM + CXL (weighted interleave)(GBps): 382.7 Gain over DRAM only: 1.4x Gain over default interleave: 2.26x ===================================================================== Performance test - Stream From - Gregory Price <gregory.price@memverge.com> Hardware: Single socket, single CXL expander numactl extension: https://github.com/gmprice/numactl/tree/weighted_interleave_master Summary: 64 threads, ~18GB workload, 3GB per array, executed 100 times Default interleave : -78% (slower than DRAM) Global weighting : -6% to +4% (workload dependant) mbind2 weights : +2.5% to +4% (consistently better than DRAM) dram only: numactl --cpunodebind=1 --membind=1 ./stream_c.exe --ntimes 100 --array-size 400M --malloc Function Direction BestRateMBs AvgTime MinTime MaxTime Copy: 0->0 200923.2 0.032662 0.031853 0.033301 Scale: 0->0 202123.0 0.032526 0.031664 0.032970 Add: 0->0 208873.2 0.047322 0.045961 0.047884 Triad: 0->0 208523.8 0.047262 0.046038 0.048414 CXL-only: numactl --cpunodebind=1 -w --membind=2 ./stream_c.exe --ntimes 100 --array-size 400M --malloc Copy: 0->0 22209.7 0.288661 0.288162 0.289342 Scale: 0->0 22288.2 0.287549 0.287147 0.288291 Add: 0->0 24419.1 0.393372 0.393135 0.393735 Triad: 0->0 24484.6 0.392337 0.392083 0.394331 Based on the above, the optimal weights are ~9:1 echo 9 > /sys/kernel/mm/mempolicy/weighted_interleave/node1 echo 1 > /sys/kernel/mm/mempolicy/weighted_interleave/node2 default interleave: numactl --cpunodebind=1 --interleave=1,2 ./stream_c.exe --ntimes 100 --array-size 400M --malloc Copy: 0->0 44666.2 0.143671 0.143285 0.144174 Scale: 0->0 44781.6 0.143256 0.142916 0.143713 Add: 0->0 48600.7 0.197719 0.197528 0.197858 Triad: 0->0 48727.5 0.197204 0.197014 0.197439 global weighted interleave: numactl --cpunodebind=1 -w --interleave=1,2 ./stream_c.exe --ntimes 100 --array-size 400M --malloc Copy: 0->0 190085.9 0.034289 0.033669 0.034645 Scale: 0->0 207677.4 0.031909 0.030817 0.033061 Add: 0->0 202036.8 0.048737 0.047516 0.053409 Triad: 0->0 217671.5 0.045819 0.044103 0.046755 targted regions w/ global weights (modified stream to mbind2 malloc'd regions)) numactl --cpunodebind=1 --membind=1 ./stream_c.exe -b --ntimes 100 --array-size 400M --malloc Copy: 0->0 205827.0 0.031445 0.031094 0.031984 Scale: 0->0 208171.8 0.031320 0.030744 0.032505 Add: 0->0 217352.0 0.045087 0.044168 0.046515 Triad: 0->0 216884.8 0.045062 0.044263 0.046982 ===================================================================== Performance tests - XSBench From - Hyeongtak Ji <hyeongtak.ji@sk.com> Hardware: Single socket, Single CXL memory Expander NUMA node 0: 56 logical cores, 128 GB memory NUMA node 2: 96 GB CXL memory Threads: 56 Lookups: 170,000,000 Summary: +19% over DRAM. +47% over default interleave. Performance tests - XSBench 1. dram only $ numactl -m 0 ./XSBench -s XL –p 5000000 Runtime: 36.235 seconds Lookups/s: 4,691,618 2. default interleave $ numactl –i 0,2 ./XSBench –s XL –p 5000000 Runtime: 55.243 seconds Lookups/s: 3,077,293 3. weighted interleave numactl –w –i 0,2 ./XSBench –s XL –p 5000000 Runtime: 29.262 seconds Lookups/s: 5,809,513 ===================================================================== LTP Tests: https://github.com/gmprice/ltp/tree/mempolicy2 = Existing tests set_mempolicy, get_mempolicy, mbind MPOL_WEIGHTED_INTERLEAVE added manually to test basic functionality but did not adjust tests for weighting. Basically the weights were set to 1, which is the default, and it should behave the same as MPOL_INTERLEAVE if logic is correct. == set_mempolicy01 : passed 18, failed 0 == set_mempolicy02 : passed 10, failed 0 == set_mempolicy03 : passed 64, failed 0 == set_mempolicy04 : passed 32, failed 0 == set_mempolicy05 - n/a on non-x86 == set_mempolicy06 : passed 10, failed 0 this is set_mempolicy02 + MPOL_WEIGHTED_INTERLEAVE == set_mempolicy07 : passed 32, failed 0 set_mempolicy04 + MPOL_WEIGHTED_INTERLEAVE == get_mempolicy01 : passed 12, failed 0 change: added MPOL_WEIGHTED_INTERLEAVE == get_mempolicy02 : passed 2, failed 0 == mbind01 : passed 15, failed 0 added MPOL_WEIGHTED_INTERLEAVE == mbind02 : passed 4, failed 0 added MPOL_WEIGHTED_INTERLEAVE == mbind03 : passed 16, failed 0 added MPOL_WEIGHTED_INTERLEAVE == mbind04 : passed 48, failed 0 added MPOL_WEIGHTED_INTERLEAVE ===================================================================== numactl (set_mempolicy) w/ global weighting test numactl fork: https://github.com/gmprice/numactl/tree/weighted_interleave_master command: numactl -w --interleave=0,1 ./eatmem result (weights 1:1): 0176a000 weighted interleave:0-1 heap anon=65793 dirty=65793 active=0 N0=32897 N1=32896 kernelpagesize_kB=4 7fceeb9ff000 weighted interleave:0-1 anon=65537 dirty=65537 active=0 N0=32768 N1=32769 kernelpagesize_kB=4 50% distribution is correct result (weights 5:1): 01b14000 weighted interleave:0-1 heap anon=65793 dirty=65793 active=0 N0=54828 N1=10965 kernelpagesize_kB=4 7f47a1dff000 weighted interleave:0-1 anon=65537 dirty=65537 active=0 N0=54614 N1=10923 kernelpagesize_kB=4 16.666% distribution is correct result (weights 1:5): 01f07000 weighted interleave:0-1 heap anon=65793 dirty=65793 active=0 N0=10966 N1=54827 kernelpagesize_kB=4 7f17b1dff000 weighted interleave:0-1 anon=65537 dirty=65537 active=0 N0=10923 N1=54614 kernelpagesize_kB=4 16.666% distribution is correct int main (void) { char* mem = malloc(1024*1024*256); memset(mem, 1, 1024*1024*256); for (int i = 0; i < ((1024*1024*256)/4096); i++) { mem = malloc(4096); mem[0] = 1; } printf("done\n"); getchar(); return 0; } This patch (of 4): This patch provides a way to set interleave weight information under sysfs at /sys/kernel/mm/mempolicy/weighted_interleave/nodeN The sysfs structure is designed as follows. $ tree /sys/kernel/mm/mempolicy/ /sys/kernel/mm/mempolicy/ [1] └── weighted_interleave [2] ├── node0 [3] └── node1 Each file above can be explained as follows. [1] mm/mempolicy: configuration interface for mempolicy subsystem [2] weighted_interleave/: config interface for weighted interleave policy [3] weighted_interleave/nodeN: weight for nodeN If a node value is set to `0`, the system-default value will be used. As of this patch, the system-default for all nodes is always 1. Link: https://lkml.kernel.org/r/20240202170238.90004-1-gregory.price@memverge.com Link: https://lkml.kernel.org/r/20240202170238.90004-2-gregory.price@memverge.com Suggested-by: "Huang, Ying" <ying.huang@intel.com> Signed-off-by: Rakie Kim <rakie.kim@sk.com> Signed-off-by: Honggyu Kim <honggyu.kim@sk.com> Co-developed-by: Gregory Price <gregory.price@memverge.com> Signed-off-by: Gregory Price <gregory.price@memverge.com> Co-developed-by: Hyeongtak Ji <hyeongtak.ji@sk.com> Signed-off-by: Hyeongtak Ji <hyeongtak.ji@sk.com> Reviewed-by: "Huang, Ying" <ying.huang@intel.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Gregory Price <gourry.memverge@gmail.com> Cc: Hasan Al Maruf <Hasan.Maruf@amd.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Jonathan Corbet <corbet@lwn.net> Cc: Michal Hocko <mhocko@kernel.org> Cc: Srinivasulu Thanneeru <sthanneeru.opensrc@micron.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Ze Zuo <zuoze1@huawei.com> --- .../ABI/testing/sysfs-kernel-mm-mempolicy | 4 + ...fs-kernel-mm-mempolicy-weighted-interleave | 25 ++ mm/mempolicy.c | 223 ++++++++++++++++++ 3 files changed, 252 insertions(+) create mode 100644 Documentation/ABI/testing/sysfs-kernel-mm-mempolicy create mode 100644 Documentation/ABI/testing/sysfs-kernel-mm-mempolicy-weighted-interleave diff --git a/Documentation/ABI/testing/sysfs-kernel-mm-mempolicy b/Documentation/ABI/testing/sysfs-kernel-mm-mempolicy new file mode 100644 index 000000000000..8ac327fd7fb6 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-kernel-mm-mempolicy @@ -0,0 +1,4 @@ +What: /sys/kernel/mm/mempolicy/ +Date: January 2024 +Contact: Linux memory management mailing list <linux-mm@kvack.org> +Description: Interface for Mempolicy diff --git a/Documentation/ABI/testing/sysfs-kernel-mm-mempolicy-weighted-interleave b/Documentation/ABI/testing/sysfs-kernel-mm-mempolicy-weighted-interleave new file mode 100644 index 000000000000..0b7972de04e9 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-kernel-mm-mempolicy-weighted-interleave @@ -0,0 +1,25 @@ +What: /sys/kernel/mm/mempolicy/weighted_interleave/ +Date: January 2024 +Contact: Linux memory management mailing list <linux-mm@kvack.org> +Description: Configuration Interface for the Weighted Interleave policy + +What: /sys/kernel/mm/mempolicy/weighted_interleave/nodeN +Date: January 2024 +Contact: Linux memory management mailing list <linux-mm@kvack.org> +Description: Weight configuration interface for nodeN + + The interleave weight for a memory node (N). These weights are + utilized by tasks which have set their mempolicy to + MPOL_WEIGHTED_INTERLEAVE. + + These weights only affect new allocations, and changes at runtime + will not cause migrations on already allocated pages. + + The minimum weight for a node is always 1. + + Minimum weight: 1 + Maximum weight: 255 + + Writing an empty string or `0` will reset the weight to the + system default. The system default may be set by the kernel + or drivers at boot or during hotplug events. diff --git a/mm/mempolicy.c b/mm/mempolicy.c index f06c45998885..bdf22ee37c72 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -133,6 +133,32 @@ static struct mempolicy default_policy = { static struct mempolicy preferred_node_policy[MAX_NUMNODES]; +/* + * iw_table is the sysfs-set interleave weight table, a value of 0 denotes + * system-default value should be used. A NULL iw_table also denotes that + * system-default values should be used. Until the system-default table + * is implemented, the system-default is always 1. + * + * iw_table is RCU protected + */ +static u8 __rcu *iw_table; +static DEFINE_MUTEX(iw_table_lock); + +static u8 get_il_weight(int node) +{ + u8 *table; + u8 weight; + + rcu_read_lock(); + table = rcu_dereference(iw_table); + /* if no iw_table, use system default */ + weight = table ? table[node] : 1; + /* if value in iw_table is 0, use system default */ + weight = weight ? weight : 1; + rcu_read_unlock(); + return weight; +} + /** * numa_nearest_node - Find nearest node by state * @node: Node id to start the search @@ -3276,3 +3302,200 @@ void mpol_to_str(char *buffer, int maxlen, struct mempolicy *pol) p += scnprintf(p, buffer + maxlen - p, ":%*pbl", nodemask_pr_args(&nodes)); } + +#ifdef CONFIG_SYSFS +struct iw_node_attr { + struct kobj_attribute kobj_attr; + int nid; +}; + +static ssize_t node_show(struct kobject *kobj, struct kobj_attribute *attr, + char *buf) +{ + struct iw_node_attr *node_attr; + u8 weight; + + node_attr = container_of(attr, struct iw_node_attr, kobj_attr); + weight = get_il_weight(node_attr->nid); + return sysfs_emit(buf, "%d\n", weight); +} + +static ssize_t node_store(struct kobject *kobj, struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct iw_node_attr *node_attr; + u8 *new; + u8 *old; + u8 weight = 0; + + node_attr = container_of(attr, struct iw_node_attr, kobj_attr); + if (count == 0 || sysfs_streq(buf, "")) + weight = 0; + else if (kstrtou8(buf, 0, &weight)) + return -EINVAL; + + new = kzalloc(nr_node_ids, GFP_KERNEL); + if (!new) + return -ENOMEM; + + mutex_lock(&iw_table_lock); + old = rcu_dereference_protected(iw_table, + lockdep_is_held(&iw_table_lock)); + if (old) + memcpy(new, old, nr_node_ids); + new[node_attr->nid] = weight; + rcu_assign_pointer(iw_table, new); + mutex_unlock(&iw_table_lock); + synchronize_rcu(); + kfree(old); + return count; +} + +static struct iw_node_attr **node_attrs; + +static void sysfs_wi_node_release(struct iw_node_attr *node_attr, + struct kobject *parent) +{ + if (!node_attr) + return; + sysfs_remove_file(parent, &node_attr->kobj_attr.attr); + kfree(node_attr->kobj_attr.attr.name); + kfree(node_attr); +} + +static void sysfs_wi_release(struct kobject *wi_kobj) +{ + int i; + + for (i = 0; i < nr_node_ids; i++) + sysfs_wi_node_release(node_attrs[i], wi_kobj); + kobject_put(wi_kobj); +} + +static const struct kobj_type wi_ktype = { + .sysfs_ops = &kobj_sysfs_ops, + .release = sysfs_wi_release, +}; + +static int add_weight_node(int nid, struct kobject *wi_kobj) +{ + struct iw_node_attr *node_attr; + char *name; + + node_attr = kzalloc(sizeof(*node_attr), GFP_KERNEL); + if (!node_attr) + return -ENOMEM; + + name = kasprintf(GFP_KERNEL, "node%d", nid); + if (!name) { + kfree(node_attr); + return -ENOMEM; + } + + sysfs_attr_init(&node_attr->kobj_attr.attr); + node_attr->kobj_attr.attr.name = name; + node_attr->kobj_attr.attr.mode = 0644; + node_attr->kobj_attr.show = node_show; + node_attr->kobj_attr.store = node_store; + node_attr->nid = nid; + + if (sysfs_create_file(wi_kobj, &node_attr->kobj_attr.attr)) { + kfree(node_attr->kobj_attr.attr.name); + kfree(node_attr); + pr_err("failed to add attribute to weighted_interleave\n"); + return -ENOMEM; + } + + node_attrs[nid] = node_attr; + return 0; +} + +static int add_weighted_interleave_group(struct kobject *root_kobj) +{ + struct kobject *wi_kobj; + int nid, err; + + wi_kobj = kzalloc(sizeof(struct kobject), GFP_KERNEL); + if (!wi_kobj) + return -ENOMEM; + + err = kobject_init_and_add(wi_kobj, &wi_ktype, root_kobj, + "weighted_interleave"); + if (err) { + kfree(wi_kobj); + return err; + } + + for_each_node_state(nid, N_POSSIBLE) { + err = add_weight_node(nid, wi_kobj); + if (err) { + pr_err("failed to add sysfs [node%d]\n", nid); + break; + } + } + if (err) + kobject_put(wi_kobj); + return 0; +} + +static void mempolicy_kobj_release(struct kobject *kobj) +{ + u8 *old; + + mutex_lock(&iw_table_lock); + old = rcu_dereference_protected(iw_table, + lockdep_is_held(&iw_table_lock)); + rcu_assign_pointer(iw_table, NULL); + mutex_unlock(&iw_table_lock); + synchronize_rcu(); + kfree(old); + kfree(node_attrs); + kfree(kobj); +} + +static const struct kobj_type mempolicy_ktype = { + .release = mempolicy_kobj_release +}; + +static int __init mempolicy_sysfs_init(void) +{ + int err; + static struct kobject *mempolicy_kobj; + + mempolicy_kobj = kzalloc(sizeof(*mempolicy_kobj), GFP_KERNEL); + if (!mempolicy_kobj) { + err = -ENOMEM; + goto err_out; + } + + node_attrs = kcalloc(nr_node_ids, sizeof(struct iw_node_attr *), + GFP_KERNEL); + if (!node_attrs) { + err = -ENOMEM; + goto mempol_out; + } + + err = kobject_init_and_add(mempolicy_kobj, &mempolicy_ktype, mm_kobj, + "mempolicy"); + if (err) + goto node_out; + + err = add_weighted_interleave_group(mempolicy_kobj); + if (err) { + pr_err("mempolicy sysfs structure failed to initialize\n"); + kobject_put(mempolicy_kobj); + return err; + } + + return err; +node_out: + kfree(node_attrs); +mempol_out: + kfree(mempolicy_kobj); +err_out: + pr_err("failed to add mempolicy kobject to the system\n"); + return err; +} + +late_initcall(mempolicy_sysfs_init); +#endif /* CONFIG_SYSFS */ -- 2.25.1