The current map creation API does not allow to provide the numa-node preference. The memory usually comes from where the map-creation-process is running. The performance is not ideal if the bpf_prog is known to always run in a numa node different from the map-creation-process. One of the use case is sharding on CPU to different LRU maps (i.e. an array of LRU maps). Here is the test result of map_perf_test on the INNER_LRU_HASH_PREALLOC test if we force the lru map used by CPU0 to be allocated from a remote numa node: [ The machine has 20 cores. CPU0-9 at node 0. CPU10-19 at node 1 ] ># taskset -c 10 ./map_perf_test 512 8 1260000 8000000 5:inner_lru_hash_map_perf pre-alloc 1628380 events per sec 4:inner_lru_hash_map_perf pre-alloc 1626396 events per sec 3:inner_lru_hash_map_perf pre-alloc 1626144 events per sec 6:inner_lru_hash_map_perf pre-alloc 1621657 events per sec 2:inner_lru_hash_map_perf pre-alloc 1621534 events per sec 1:inner_lru_hash_map_perf pre-alloc 1620292 events per sec 7:inner_lru_hash_map_perf pre-alloc 1613305 events per sec 0:inner_lru_hash_map_perf pre-alloc 1239150 events per sec #<<< After specifying numa node: ># taskset -c 10 ./map_perf_test 512 8 1260000 8000000 5:inner_lru_hash_map_perf pre-alloc 1629627 events per sec 3:inner_lru_hash_map_perf pre-alloc 1628057 events per sec 1:inner_lru_hash_map_perf pre-alloc 1623054 events per sec 6:inner_lru_hash_map_perf pre-alloc 1616033 events per sec 2:inner_lru_hash_map_perf pre-alloc 1614630 events per sec 4:inner_lru_hash_map_perf pre-alloc 1612651 events per sec 7:inner_lru_hash_map_perf pre-alloc 1609337 events per sec 0:inner_lru_hash_map_perf pre-alloc 1619340 events per sec #<<< This patch adds one field, numa_node, to the bpf_attr. Since numa node 0 is a valid node, a new flag BPF_F_NUMA_NODE is also added. The numa_node field is honored if and only if the BPF_F_NUMA_NODE flag is set. Numa node selection is not supported for percpu map. This patch does not change all the kmalloc. F.e. 'htab = kzalloc()' is not changed since the object is small enough to stay in the cache.
Showing the changes to include/uapi/linux (1 of 9 files). Show all
Unified · Split
| @@ -165,6 +165,7 @@ enum bpf_attach_type { · 164 unmodified lines | ||
| 165 | 165 | #define BPF_NOEXIST 1 /* create new element if it didn't exist */ |
| 166 | 166 | #define BPF_EXIST 2 /* update existing element */ |
| 167 | 167 | |
| 168 | +/* flags for BPF_MAP_CREATE command */ | |
| 168 | 169 | #define BPF_F_NO_PREALLOC (1U << 0) |
| 169 | 170 | /* Instead of having one common LRU list in the |
| 170 | 171 | * BPF_MAP_TYPE_LRU_[PERCPU_]HASH map, use a percpu LRU list |
| @@ -173,6 +174,8 @@ enum bpf_attach_type { · 2 unmodified lines | ||
| 173 | 174 | * across different LRU lists. |
| 174 | 175 | */ |
| 175 | 176 | #define BPF_F_NO_COMMON_LRU (1U << 1) |
| 177 | +/* Specify numa node during map creation */ | |
| 178 | +#define BPF_F_NUMA_NODE (1U << 2) | |
| 176 | 179 | |
| 177 | 180 | union bpf_attr { |
| 178 | 181 | struct { /* anonymous struct used by BPF_MAP_CREATE command */ |
| @@ -180,8 +183,13 @@ union bpf_attr { · 1 unmodified line | ||
| 180 | 183 | __u32 key_size; /* size of key in bytes */ |
| 181 | 184 | __u32 value_size; /* size of value in bytes */ |
| 182 | 185 | __u32 max_entries; /* max number of entries in a map */ |
| 183 | − __u32 map_flags; /* prealloc or not */ | |
| 186 | + __u32 map_flags; /* BPF_MAP_CREATE related | |
| 187 | + * flags defined above. | |
| 188 | + */ | |
| 184 | 189 | __u32 inner_map_fd; /* fd pointing to the inner map */ |
| 190 | + __u32 numa_node; /* numa node (effective only if | |
| 191 | + * BPF_F_NUMA_NODE is set). | |
| 192 | + */ | |
| 185 | 193 | }; |
| 186 | 194 | |
| 187 | 195 | struct { /* anonymous struct used by BPF_MAP_*_ELEM commands */ |