[RFC,bpf-next,13/15] bpf: Add new parameter into bpf_map_container_alloc

Message ID	20220729152316.58205-14-laoar.shao@gmail.com (mailing list archive)
State	New
Headers	show Return-Path: <owner-linux-mm@kvack.org> From: Yafang Shao <laoar.shao@gmail.com> To: ast@kernel.org, daniel@iogearbox.net, andrii@kernel.org, kafai@fb.com, songliubraving@fb.com, yhs@fb.com, john.fastabend@gmail.com, kpsingh@kernel.org, sdf@google.com, haoluo@google.com, jolsa@kernel.org, hannes@cmpxchg.org, mhocko@kernel.org, roman.gushchin@linux.dev, shakeelb@google.com, songmuchun@bytedance.com, akpm@linux-foundation.org Cc: netdev@vger.kernel.org, bpf@vger.kernel.org, linux-mm@kvack.org, Yafang Shao <laoar.shao@gmail.com> Subject: [RFC PATCH bpf-next 13/15] bpf: Add new parameter into bpf_map_container_alloc Date: Fri, 29 Jul 2022 15:23:14 +0000 Message-Id: <20220729152316.58205-14-laoar.shao@gmail.com> In-Reply-To: <20220729152316.58205-1-laoar.shao@gmail.com> References: <20220729152316.58205-1-laoar.shao@gmail.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Sender: owner-linux-mm@kvack.org Precedence: bulk
Series	bpf: Introduce selectable memcg for bpf map \| expand [RFC,bpf-next,00/15] bpf: Introduce selectable memcg for bpf map [RFC,bpf-next,01/15] bpf: Remove unneeded memset in queue_stack_map creation [RFC,bpf-next,02/15] bpf: Use bpf_map_area_free instread of kvfree [RFC,bpf-next,03/15] bpf: Make __GFP_NOWARN consistent in bpf map creation [RFC,bpf-next,04/15] bpf: Use bpf_map_area_alloc consistently on bpf map creation [RFC,bpf-next,05/15] bpf: Introduce helpers for container of struct bpf_map [RFC,bpf-next,06/15] bpf: Use bpf_map_container_alloc helpers in various bpf maps [RFC,bpf-next,07/15] bpf: Define bpf_map_get_memcg for !CONFIG_MEMCG_KMEM [RFC,bpf-next,08/15] bpf: Use scope-based charge for bpf_map_area_alloc [RFC,bpf-next,09/15] bpf: Use bpf_map_kzalloc in arraymap [RFC,bpf-next,10/15] bpf: Use bpf_map_pages_alloc in ringbuf [RFC,bpf-next,11/15] bpf: Use bpf_map_kvcalloc in bpf_local_storage [RFC,bpf-next,12/15] mm, memcg: Add new helper get_obj_cgroup_from_cgroup [RFC,bpf-next,13/15] bpf: Add new parameter into bpf_map_container_alloc [RFC,bpf-next,14/15] bpf: Add new map flag BPF_F_SELECTABLE_MEMCG [RFC,bpf-next,15/15] bpf: Introduce selectable memcg for bpf map

diff --git a/include/linux/bpf.h b/include/linux/bpf.h index 605214fedd6d..2bd941279e6c 100644 --- a/include/linux/bpf.h +++ b/include/linux/bpf.h @@ -1634,9 +1634,9 @@ void bpf_map_inc_with_uref(struct bpf_map *map); struct bpf_map * __must_check bpf_map_inc_not_zero(struct bpf_map *map); void bpf_map_put_with_uref(struct bpf_map *map); void bpf_map_put(struct bpf_map *map); -void *bpf_map_container_alloc(u64 size, int numa_node); -void *bpf_map_container_mmapable_alloc(u64 size, int numa_node, - u32 align, u32 offset); +void *bpf_map_container_alloc(union bpf_attr *attr, u64 size, int numa_node); +void *bpf_map_container_mmapable_alloc(union bpf_attr *attr, u64 size, + int numa_node, u32 align, u32 offset); void *bpf_map_area_alloc(struct bpf_map *map, u64 size, int numa_node); void *bpf_map_pages_alloc(struct bpf_map *map, struct page **pages, int nr_meta_pages, int nr_data_pages, int nid, diff --git a/kernel/bpf/arraymap.c b/kernel/bpf/arraymap.c index bfcfc5df9983..883905c6c845 100644 --- a/kernel/bpf/arraymap.c +++ b/kernel/bpf/arraymap.c @@ -131,16 +131,17 @@ static struct bpf_map *array_map_alloc(union bpf_attr *attr) void *data; /* kmalloc'ed memory can't be mmap'ed, use explicit vmalloc */ - data = bpf_map_container_mmapable_alloc(array_size, numa_node, - align, offset); - if (!data) - return ERR_PTR(-ENOMEM); + data = bpf_map_container_mmapable_alloc(attr, array_size, + numa_node, align, + offset); + if (IS_ERR(data)) + return data; array = data + align - offset; } else { - array = bpf_map_container_alloc(array_size, numa_node); + array = bpf_map_container_alloc(attr, array_size, numa_node); } - if (!array) - return ERR_PTR(-ENOMEM); + if (IS_ERR(array)) + return ERR_CAST(array); array->index_mask = index_mask; array->map.bypass_spec_v1 = bypass_spec_v1; diff --git a/kernel/bpf/bloom_filter.c b/kernel/bpf/bloom_filter.c index 8b3194903b52..9fe3c6774c40 100644 --- a/kernel/bpf/bloom_filter.c +++ b/kernel/bpf/bloom_filter.c @@ -142,10 +142,11 @@ static struct bpf_map *bloom_map_alloc(union bpf_attr *attr) } bitset_bytes = roundup(bitset_bytes, sizeof(unsigned long)); - bloom = bpf_map_container_alloc(sizeof(*bloom) + bitset_bytes, numa_node); + bloom = bpf_map_container_alloc(attr, sizeof(*bloom) + bitset_bytes, + numa_node); - if (!bloom) - return ERR_PTR(-ENOMEM); + if (IS_ERR(bloom)) + return ERR_CAST(bloom); bpf_map_init_from_attr(&bloom->map, attr); diff --git a/kernel/bpf/bpf_local_storage.c b/kernel/bpf/bpf_local_storage.c index 3440135c3612..e12891dcf2a9 100644 --- a/kernel/bpf/bpf_local_storage.c +++ b/kernel/bpf/bpf_local_storage.c @@ -610,9 +610,9 @@ struct bpf_local_storage_map *bpf_local_storage_map_alloc(union bpf_attr *attr) unsigned int i; u32 nbuckets; - smap = bpf_map_container_alloc(sizeof(*smap), NUMA_NO_NODE); - if (!smap) - return ERR_PTR(-ENOMEM); + smap = bpf_map_container_alloc(attr, sizeof(*smap), NUMA_NO_NODE); + if (IS_ERR(smap)) + return ERR_CAST(smap); bpf_map_init_from_attr(&smap->map, attr); nbuckets = roundup_pow_of_two(num_possible_cpus()); diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c index 874fda7e2b8b..51b7ce9902a8 100644 --- a/kernel/bpf/bpf_struct_ops.c +++ b/kernel/bpf/bpf_struct_ops.c @@ -618,9 +618,9 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr) */ (vt->size - sizeof(struct bpf_struct_ops_value)); - st_map = bpf_map_container_alloc(st_map_size, NUMA_NO_NODE); - if (!st_map) - return ERR_PTR(-ENOMEM); + st_map = bpf_map_container_alloc(attr, st_map_size, NUMA_NO_NODE); + if (IS_ERR(st_map)) + return ERR_CAST(st_map); st_map->st_ops = st_ops; map = &st_map->map; diff --git a/kernel/bpf/cpumap.c b/kernel/bpf/cpumap.c index 95c1642deaf6..1e915a3bd1a2 100644 --- a/kernel/bpf/cpumap.c +++ b/kernel/bpf/cpumap.c @@ -97,9 +97,9 @@ static struct bpf_map *cpu_map_alloc(union bpf_attr *attr) attr->map_flags & ~BPF_F_NUMA_NODE) return ERR_PTR(-EINVAL); - cmap = bpf_map_container_alloc(sizeof(*cmap), NUMA_NO_NODE); - if (!cmap) - return ERR_PTR(-ENOMEM); + cmap = bpf_map_container_alloc(attr, sizeof(*cmap), NUMA_NO_NODE); + if (IS_ERR(cmap)) + return ERR_CAST(cmap); bpf_map_init_from_attr(&cmap->map, attr); diff --git a/kernel/bpf/devmap.c b/kernel/bpf/devmap.c index b625d578bc93..11c7b8411b03 100644 --- a/kernel/bpf/devmap.c +++ b/kernel/bpf/devmap.c @@ -167,9 +167,9 @@ static struct bpf_map *dev_map_alloc(union bpf_attr *attr) if (!capable(CAP_NET_ADMIN)) return ERR_PTR(-EPERM); - dtab = bpf_map_container_alloc(sizeof(*dtab), NUMA_NO_NODE); - if (!dtab) - return ERR_PTR(-ENOMEM); + dtab = bpf_map_container_alloc(attr, sizeof(*dtab), NUMA_NO_NODE); + if (IS_ERR(dtab)) + return ERR_CAST(dtab); err = dev_map_init_map(dtab, attr); if (err) { diff --git a/kernel/bpf/hashtab.c b/kernel/bpf/hashtab.c index 2a34a115e14f..3cb9486eb313 100644 --- a/kernel/bpf/hashtab.c +++ b/kernel/bpf/hashtab.c @@ -496,9 +496,9 @@ static struct bpf_map *htab_map_alloc(union bpf_attr *attr) struct bpf_htab *htab; int err, i; - htab = bpf_map_container_alloc(sizeof(*htab), NUMA_NO_NODE); - if (!htab) - return ERR_PTR(-ENOMEM); + htab = bpf_map_container_alloc(attr, sizeof(*htab), NUMA_NO_NODE); + if (IS_ERR(htab)) + return ERR_CAST(htab); lockdep_register_key(&htab->lockdep_key); diff --git a/kernel/bpf/local_storage.c b/kernel/bpf/local_storage.c index 963dc82e3c50..b2bd031aba79 100644 --- a/kernel/bpf/local_storage.c +++ b/kernel/bpf/local_storage.c @@ -313,9 +313,9 @@ static struct bpf_map *cgroup_storage_map_alloc(union bpf_attr *attr) /* max_entries is not used and enforced to be 0 */ return ERR_PTR(-EINVAL); - map = bpf_map_container_alloc(sizeof(struct bpf_cgroup_storage_map), numa_node); - if (!map) - return ERR_PTR(-ENOMEM); + map = bpf_map_container_alloc(attr, sizeof(*map), numa_node); + if (IS_ERR(map)) + return ERR_CAST(map); /* copy mandatory map attributes */ bpf_map_init_from_attr(&map->map, attr); diff --git a/kernel/bpf/lpm_trie.c b/kernel/bpf/lpm_trie.c index 15ccef2f10a3..e55327ad9c3a 100644 --- a/kernel/bpf/lpm_trie.c +++ b/kernel/bpf/lpm_trie.c @@ -558,9 +558,9 @@ static struct bpf_map *trie_alloc(union bpf_attr *attr) attr->value_size > LPM_VAL_SIZE_MAX) return ERR_PTR(-EINVAL); - trie = bpf_map_container_alloc(sizeof(*trie), NUMA_NO_NODE); - if (!trie) - return ERR_PTR(-ENOMEM); + trie = bpf_map_container_alloc(attr, sizeof(*trie), NUMA_NO_NODE); + if (IS_ERR(trie)) + return ERR_CAST(trie); /* copy mandatory map attributes */ bpf_map_init_from_attr(&trie->map, attr); diff --git a/kernel/bpf/offload.c b/kernel/bpf/offload.c index 02e3f4ef19f5..34fb5261ff95 100644 --- a/kernel/bpf/offload.c +++ b/kernel/bpf/offload.c @@ -372,9 +372,9 @@ struct bpf_map *bpf_map_offload_map_alloc(union bpf_attr *attr) attr->map_type != BPF_MAP_TYPE_HASH) return ERR_PTR(-EINVAL); - offmap = bpf_map_container_alloc(sizeof(*offmap), NUMA_NO_NODE); - if (!offmap) - return ERR_PTR(-ENOMEM); + offmap = bpf_map_container_alloc(attr, sizeof(*offmap), NUMA_NO_NODE); + if (IS_ERR(offmap)) + return ERR_CAST(offmap); bpf_map_init_from_attr(&offmap->map, attr); diff --git a/kernel/bpf/queue_stack_maps.c b/kernel/bpf/queue_stack_maps.c index d4c0284c9089..9425df0695ac 100644 --- a/kernel/bpf/queue_stack_maps.c +++ b/kernel/bpf/queue_stack_maps.c @@ -74,9 +74,9 @@ static struct bpf_map *queue_stack_map_alloc(union bpf_attr *attr) size = (u64) attr->max_entries + 1; queue_size = sizeof(*qs) + size * attr->value_size; - qs = bpf_map_container_alloc(queue_size, numa_node); - if (!qs) - return ERR_PTR(-ENOMEM); + qs = bpf_map_container_alloc(attr, queue_size, numa_node); + if (IS_ERR(qs)) + return ERR_CAST(qs); bpf_map_init_from_attr(&qs->map, attr); diff --git a/kernel/bpf/reuseport_array.c b/kernel/bpf/reuseport_array.c index d36a60ca68f4..0d3091866e1d 100644 --- a/kernel/bpf/reuseport_array.c +++ b/kernel/bpf/reuseport_array.c @@ -153,14 +153,15 @@ static struct bpf_map *reuseport_array_alloc(union bpf_attr *attr) { int numa_node = bpf_map_attr_numa_node(attr); struct reuseport_array *array; + size_t size = struct_size(array, ptrs, attr->max_entries); if (!bpf_capable()) return ERR_PTR(-EPERM); /* allocate all map elements and zero-initialize them */ - array = bpf_map_container_alloc(struct_size(array, ptrs, attr->max_entries), numa_node); - if (!array) - return ERR_PTR(-ENOMEM); + array = bpf_map_container_alloc(attr, size, numa_node); + if (IS_ERR(array)) + return ERR_CAST(array); /* copy mandatory map attributes */ bpf_map_init_from_attr(&array->map, attr); diff --git a/kernel/bpf/ringbuf.c b/kernel/bpf/ringbuf.c index 25973cab251d..3be472fd55da 100644 --- a/kernel/bpf/ringbuf.c +++ b/kernel/bpf/ringbuf.c @@ -159,9 +159,9 @@ static struct bpf_map *ringbuf_map_alloc(union bpf_attr *attr) return ERR_PTR(-E2BIG); #endif - rb_map = bpf_map_container_alloc(sizeof(*rb_map), NUMA_NO_NODE); - if (!rb_map) - return ERR_PTR(-ENOMEM); + rb_map = bpf_map_container_alloc(attr, sizeof(*rb_map), NUMA_NO_NODE); + if (IS_ERR(rb_map)) + return ERR_CAST(rb_map); bpf_map_init_from_attr(&rb_map->map, attr); diff --git a/kernel/bpf/stackmap.c b/kernel/bpf/stackmap.c index c9a91ca05a03..c952c7547279 100644 --- a/kernel/bpf/stackmap.c +++ b/kernel/bpf/stackmap.c @@ -101,9 +101,9 @@ static struct bpf_map *stack_map_alloc(union bpf_attr *attr) return ERR_PTR(-E2BIG); cost = n_buckets * sizeof(struct stack_map_bucket *) + sizeof(*smap); - smap = bpf_map_container_alloc(cost, bpf_map_attr_numa_node(attr)); - if (!smap) - return ERR_PTR(-ENOMEM); + smap = bpf_map_container_alloc(attr, cost, bpf_map_attr_numa_node(attr)); + if (IS_ERR(smap)) + return ERR_CAST(smap); bpf_map_init_from_attr(&smap->map, attr); smap->n_buckets = n_buckets; diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c index f7210fa8c3be..6401cc417fa9 100644 --- a/kernel/bpf/syscall.c +++ b/kernel/bpf/syscall.c @@ -526,14 +526,14 @@ void bpf_map_area_free(void *area) * * It is used in map creation path. */ -void *bpf_map_container_alloc(u64 size, int numa_node) +void *bpf_map_container_alloc(union bpf_attr *attr, u64 size, int numa_node) { struct bpf_map *map; void *container; container = __bpf_map_area_alloc(size, numa_node, false); if (!container) - return NULL; + return ERR_PTR(-ENOMEM); map = (struct bpf_map *)container; bpf_map_save_memcg(map); @@ -541,8 +541,8 @@ void *bpf_map_container_alloc(u64 size, int numa_node) return container; } -void *bpf_map_container_mmapable_alloc(u64 size, int numa_node, u32 align, - u32 offset) +void *bpf_map_container_mmapable_alloc(union bpf_attr *attr, u64 size, + int numa_node, u32 align, u32 offset) { struct bpf_map *map; void *container; @@ -551,7 +551,7 @@ void *bpf_map_container_mmapable_alloc(u64 size, int numa_node, u32 align, /* kmalloc'ed memory can't be mmap'ed, use explicit vmalloc */ ptr = __bpf_map_area_alloc(size, numa_node, true); if (!ptr) - return NULL; + return ERR_PTR(-ENOMEM); container = ptr + align - offset; map = (struct bpf_map *)container; diff --git a/net/core/sock_map.c b/net/core/sock_map.c index 1f49dc89822c..4d9b730aa27f 100644 --- a/net/core/sock_map.c +++ b/net/core/sock_map.c @@ -41,9 +41,9 @@ static struct bpf_map *sock_map_alloc(union bpf_attr *attr) attr->map_flags & ~SOCK_CREATE_FLAG_MASK) return ERR_PTR(-EINVAL); - stab = bpf_map_container_alloc(sizeof(*stab), NUMA_NO_NODE); - if (!stab) - return ERR_PTR(-ENOMEM); + stab = bpf_map_container_alloc(attr, sizeof(*stab), NUMA_NO_NODE); + if (IS_ERR(stab)) + return ERR_CAST(stab); bpf_map_init_from_attr(&stab->map, attr); raw_spin_lock_init(&stab->lock); @@ -1077,9 +1077,9 @@ static struct bpf_map *sock_hash_alloc(union bpf_attr *attr) if (attr->key_size > MAX_BPF_STACK) return ERR_PTR(-E2BIG); - htab = bpf_map_container_alloc(sizeof(*htab), NUMA_NO_NODE); - if (!htab) - return ERR_PTR(-ENOMEM); + htab = bpf_map_container_alloc(attr, sizeof(*htab), NUMA_NO_NODE); + if (IS_ERR(htab)) + return ERR_CAST(htab); bpf_map_init_from_attr(&htab->map, attr); diff --git a/net/xdp/xskmap.c b/net/xdp/xskmap.c index 6cf98f1c621a..161fc569ed88 100644 --- a/net/xdp/xskmap.c +++ b/net/xdp/xskmap.c @@ -75,9 +75,9 @@ static struct bpf_map *xsk_map_alloc(union bpf_attr *attr) numa_node = bpf_map_attr_numa_node(attr); size = struct_size(m, xsk_map, attr->max_entries); - m = bpf_map_container_alloc(size, numa_node); - if (!m) - return ERR_PTR(-ENOMEM); + m = bpf_map_container_alloc(attr, size, numa_node); + if (IS_ERR(m)) + return ERR_CAST(m); bpf_map_init_from_attr(&m->map, attr); spin_lock_init(&m->lock);

[RFC,bpf-next,13/15] bpf: Add new parameter into bpf_map_container_alloc

Commit Message

Patch