Commit cb86d0f8 authored by Daniel Borkmann's avatar Daniel Borkmann
Browse files

Merge branch 'bpf-per-cpu-cgroup-storage' 



Roman Gushchin says:

====================
This patchset implements per-cpu cgroup local storage and provides
an example how per-cpu and shared cgroup local storage can be used
for efficient accounting of network traffic.

v4->v3:
  1) incorporated Alexei's feedback

v3->v2:
  1) incorporated Song's feedback
  2) rebased on top of current bpf-next

v2->v1:
  1) added a selftest implementing network counters
  2) added a missing free() in cgroup local storage selftest
====================

Signed-off-by: default avatarDaniel Borkmann <daniel@iogearbox.net>
parents 5bf7a60b 371e4fcc

include/linux/bpf-cgroup.h

+43 −12
Original line number Diff line number Diff line
@@ -2,6 +2,7 @@ 
#ifndef _BPF_CGROUP_H

#define _BPF_CGROUP_H



#include <linux/bpf.h>

#include <linux/errno.h>

#include <linux/jump_label.h>

#include <linux/percpu.h>
@@ -22,7 +23,11 @@ struct bpf_cgroup_storage; 
extern struct static_key_false cgroup_bpf_enabled_key;

#define cgroup_bpf_enabled static_branch_unlikely(&cgroup_bpf_enabled_key)



DECLARE_PER_CPU(void*, bpf_cgroup_storage);

DECLARE_PER_CPU(struct bpf_cgroup_storage*,

		bpf_cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE]);



#define for_each_cgroup_storage_type(stype) \

	for (stype = 0; stype < MAX_BPF_CGROUP_STORAGE_TYPE; stype++)



struct bpf_cgroup_storage_map;
@@ -32,7 +37,10 @@ struct bpf_storage_buffer { 
};



struct bpf_cgroup_storage {

	union {

		struct bpf_storage_buffer *buf;

		void __percpu *percpu_buf;

	};

	struct bpf_cgroup_storage_map *map;

	struct bpf_cgroup_storage_key key;

	struct list_head list;
@@ -43,7 +51,7 @@ struct bpf_cgroup_storage { 
struct bpf_prog_list {

	struct list_head node;

	struct bpf_prog *prog;

	struct bpf_cgroup_storage *storage;

	struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE];

};



struct bpf_prog_array;
@@ -101,18 +109,26 @@ int __cgroup_bpf_run_filter_sock_ops(struct sock *sk, 
int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,

				      short access, enum bpf_attach_type type);



static inline void bpf_cgroup_storage_set(struct bpf_cgroup_storage *storage)

static inline enum bpf_cgroup_storage_type cgroup_storage_type(

	struct bpf_map *map)

{

	struct bpf_storage_buffer *buf;

	if (map->map_type == BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE)

		return BPF_CGROUP_STORAGE_PERCPU;



	return BPF_CGROUP_STORAGE_SHARED;

}



	if (!storage)

		return;

static inline void bpf_cgroup_storage_set(struct bpf_cgroup_storage

					  *storage[MAX_BPF_CGROUP_STORAGE_TYPE])

{

	enum bpf_cgroup_storage_type stype;



	buf = READ_ONCE(storage->buf);

	this_cpu_write(bpf_cgroup_storage, &buf->data[0]);

	for_each_cgroup_storage_type(stype)

		this_cpu_write(bpf_cgroup_storage[stype], storage[stype]);

}



struct bpf_cgroup_storage *bpf_cgroup_storage_alloc(struct bpf_prog *prog);

struct bpf_cgroup_storage *bpf_cgroup_storage_alloc(struct bpf_prog *prog,

					enum bpf_cgroup_storage_type stype);

void bpf_cgroup_storage_free(struct bpf_cgroup_storage *storage);

void bpf_cgroup_storage_link(struct bpf_cgroup_storage *storage,

			     struct cgroup *cgroup,
@@ -121,6 +137,10 @@ void bpf_cgroup_storage_unlink(struct bpf_cgroup_storage *storage); 
int bpf_cgroup_storage_assign(struct bpf_prog *prog, struct bpf_map *map);

void bpf_cgroup_storage_release(struct bpf_prog *prog, struct bpf_map *map);



int bpf_percpu_cgroup_storage_copy(struct bpf_map *map, void *key, void *value);

int bpf_percpu_cgroup_storage_update(struct bpf_map *map, void *key,

				     void *value, u64 flags);



/* Wrappers for __cgroup_bpf_run_filter_skb() guarded by cgroup_bpf_enabled. */

#define BPF_CGROUP_RUN_PROG_INET_INGRESS(sk, skb)			      \

({									      \
@@ -265,15 +285,24 @@ static inline int cgroup_bpf_prog_query(const union bpf_attr *attr, 
	return -EINVAL;

}



static inline void bpf_cgroup_storage_set(struct bpf_cgroup_storage *storage) {}

static inline void bpf_cgroup_storage_set(

	struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE]) {}

static inline int bpf_cgroup_storage_assign(struct bpf_prog *prog,

					    struct bpf_map *map) { return 0; }

static inline void bpf_cgroup_storage_release(struct bpf_prog *prog,

					      struct bpf_map *map) {}

static inline struct bpf_cgroup_storage *bpf_cgroup_storage_alloc(

	struct bpf_prog *prog) { return 0; }

	struct bpf_prog *prog, enum bpf_cgroup_storage_type stype) { return 0; }

static inline void bpf_cgroup_storage_free(

	struct bpf_cgroup_storage *storage) {}

static inline int bpf_percpu_cgroup_storage_copy(struct bpf_map *map, void *key,

						 void *value) {

	return 0;

}

static inline int bpf_percpu_cgroup_storage_update(struct bpf_map *map,

					void *key, void *value, u64 flags) {

	return 0;

}



#define cgroup_bpf_enabled (0)

#define BPF_CGROUP_PRE_CONNECT_ENABLED(sk) (0)
@@ -293,6 +322,8 @@ static inline void bpf_cgroup_storage_free( 
#define BPF_CGROUP_RUN_PROG_SOCK_OPS(sock_ops) ({ 0; })

#define BPF_CGROUP_RUN_PROG_DEVICE_CGROUP(type,major,minor,access) ({ 0; })



#define for_each_cgroup_storage_type(stype) for (; false; )



#endif /* CONFIG_CGROUP_BPF */



#endif /* _BPF_CGROUP_H */

include/linux/bpf.h

+10 −2
Original line number Diff line number Diff line
@@ -272,6 +272,14 @@ struct bpf_prog_offload { 
	u32			jited_len;

};



enum bpf_cgroup_storage_type {

	BPF_CGROUP_STORAGE_SHARED,

	BPF_CGROUP_STORAGE_PERCPU,

	__BPF_CGROUP_STORAGE_MAX

};



#define MAX_BPF_CGROUP_STORAGE_TYPE __BPF_CGROUP_STORAGE_MAX



struct bpf_prog_aux {

	atomic_t refcnt;

	u32 used_map_cnt;
@@ -289,7 +297,7 @@ struct bpf_prog_aux { 
	struct bpf_prog *prog;

	struct user_struct *user;

	u64 load_time; /* ns since boottime */

	struct bpf_map *cgroup_storage;

	struct bpf_map *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];

	char name[BPF_OBJ_NAME_LEN];

#ifdef CONFIG_SECURITY

	void *security;
@@ -358,7 +366,7 @@ int bpf_prog_test_run_skb(struct bpf_prog *prog, const union bpf_attr *kattr, 
 */

struct bpf_prog_array_item {

	struct bpf_prog *prog;

	struct bpf_cgroup_storage *cgroup_storage;

	struct bpf_cgroup_storage *cgroup_storage[MAX_BPF_CGROUP_STORAGE_TYPE];

};



struct bpf_prog_array {

include/linux/bpf_types.h

+1 −0
Original line number Diff line number Diff line
@@ -43,6 +43,7 @@ BPF_MAP_TYPE(BPF_MAP_TYPE_CGROUP_ARRAY, cgroup_array_map_ops) 
#endif

#ifdef CONFIG_CGROUP_BPF

BPF_MAP_TYPE(BPF_MAP_TYPE_CGROUP_STORAGE, cgroup_storage_map_ops)

BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE, cgroup_storage_map_ops)

#endif

BPF_MAP_TYPE(BPF_MAP_TYPE_HASH, htab_map_ops)

BPF_MAP_TYPE(BPF_MAP_TYPE_PERCPU_HASH, htab_percpu_map_ops)

include/uapi/linux/bpf.h

+1 −0
Original line number Diff line number Diff line
@@ -127,6 +127,7 @@ enum bpf_map_type { 
	BPF_MAP_TYPE_SOCKHASH,

	BPF_MAP_TYPE_CGROUP_STORAGE,

	BPF_MAP_TYPE_REUSEPORT_SOCKARRAY,

	BPF_MAP_TYPE_PERCPU_CGROUP_STORAGE,

};



enum bpf_prog_type {

kernel/bpf/cgroup.c

+52 −22
Original line number Diff line number Diff line
@@ -25,6 +25,7 @@ EXPORT_SYMBOL(cgroup_bpf_enabled_key); 
 */

void cgroup_bpf_put(struct cgroup *cgrp)

{

	enum bpf_cgroup_storage_type stype;

	unsigned int type;



	for (type = 0; type < ARRAY_SIZE(cgrp->bpf.progs); type++) {
@@ -34,8 +35,10 @@ void cgroup_bpf_put(struct cgroup *cgrp) 
		list_for_each_entry_safe(pl, tmp, progs, node) {

			list_del(&pl->node);

			bpf_prog_put(pl->prog);

			bpf_cgroup_storage_unlink(pl->storage);

			bpf_cgroup_storage_free(pl->storage);

			for_each_cgroup_storage_type(stype) {

				bpf_cgroup_storage_unlink(pl->storage[stype]);

				bpf_cgroup_storage_free(pl->storage[stype]);

			}

			kfree(pl);

			static_branch_dec(&cgroup_bpf_enabled_key);

		}
@@ -97,6 +100,7 @@ static int compute_effective_progs(struct cgroup *cgrp, 
				   enum bpf_attach_type type,

				   struct bpf_prog_array __rcu **array)

{

	enum bpf_cgroup_storage_type stype;

	struct bpf_prog_array *progs;

	struct bpf_prog_list *pl;

	struct cgroup *p = cgrp;
@@ -125,7 +129,9 @@ static int compute_effective_progs(struct cgroup *cgrp, 
				continue;



			progs->items[cnt].prog = pl->prog;

			progs->items[cnt].cgroup_storage = pl->storage;

			for_each_cgroup_storage_type(stype)

				progs->items[cnt].cgroup_storage[stype] =

					pl->storage[stype];

			cnt++;

		}

	} while ((p = cgroup_parent(p)));
@@ -232,7 +238,9 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog, 
{

	struct list_head *progs = &cgrp->bpf.progs[type];

	struct bpf_prog *old_prog = NULL;

	struct bpf_cgroup_storage *storage, *old_storage = NULL;

	struct bpf_cgroup_storage *storage[MAX_BPF_CGROUP_STORAGE_TYPE],

		*old_storage[MAX_BPF_CGROUP_STORAGE_TYPE] = {NULL};

	enum bpf_cgroup_storage_type stype;

	struct bpf_prog_list *pl;

	bool pl_was_allocated;

	int err;
@@ -254,34 +262,44 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog, 
	if (prog_list_length(progs) >= BPF_CGROUP_MAX_PROGS)

		return -E2BIG;



	storage = bpf_cgroup_storage_alloc(prog);

	if (IS_ERR(storage))

	for_each_cgroup_storage_type(stype) {

		storage[stype] = bpf_cgroup_storage_alloc(prog, stype);

		if (IS_ERR(storage[stype])) {

			storage[stype] = NULL;

			for_each_cgroup_storage_type(stype)

				bpf_cgroup_storage_free(storage[stype]);

			return -ENOMEM;

		}

	}



	if (flags & BPF_F_ALLOW_MULTI) {

		list_for_each_entry(pl, progs, node) {

			if (pl->prog == prog) {

				/* disallow attaching the same prog twice */

				bpf_cgroup_storage_free(storage);

				for_each_cgroup_storage_type(stype)

					bpf_cgroup_storage_free(storage[stype]);

				return -EINVAL;

			}

		}



		pl = kmalloc(sizeof(*pl), GFP_KERNEL);

		if (!pl) {

			bpf_cgroup_storage_free(storage);

			for_each_cgroup_storage_type(stype)

				bpf_cgroup_storage_free(storage[stype]);

			return -ENOMEM;

		}



		pl_was_allocated = true;

		pl->prog = prog;

		pl->storage = storage;

		for_each_cgroup_storage_type(stype)

			pl->storage[stype] = storage[stype];

		list_add_tail(&pl->node, progs);

	} else {

		if (list_empty(progs)) {

			pl = kmalloc(sizeof(*pl), GFP_KERNEL);

			if (!pl) {

				bpf_cgroup_storage_free(storage);

				for_each_cgroup_storage_type(stype)

					bpf_cgroup_storage_free(storage[stype]);

				return -ENOMEM;

			}

			pl_was_allocated = true;
@@ -289,12 +307,15 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog, 
		} else {

			pl = list_first_entry(progs, typeof(*pl), node);

			old_prog = pl->prog;

			old_storage = pl->storage;

			bpf_cgroup_storage_unlink(old_storage);

			for_each_cgroup_storage_type(stype) {

				old_storage[stype] = pl->storage[stype];

				bpf_cgroup_storage_unlink(old_storage[stype]);

			}

			pl_was_allocated = false;

		}

		pl->prog = prog;

		pl->storage = storage;

		for_each_cgroup_storage_type(stype)

			pl->storage[stype] = storage[stype];

	}



	cgrp->bpf.flags[type] = flags;
@@ -304,21 +325,27 @@ int __cgroup_bpf_attach(struct cgroup *cgrp, struct bpf_prog *prog, 
		goto cleanup;



	static_branch_inc(&cgroup_bpf_enabled_key);

	if (old_storage)

		bpf_cgroup_storage_free(old_storage);

	for_each_cgroup_storage_type(stype) {

		if (!old_storage[stype])

			continue;

		bpf_cgroup_storage_free(old_storage[stype]);

	}

	if (old_prog) {

		bpf_prog_put(old_prog);

		static_branch_dec(&cgroup_bpf_enabled_key);

	}

	bpf_cgroup_storage_link(storage, cgrp, type);

	for_each_cgroup_storage_type(stype)

		bpf_cgroup_storage_link(storage[stype], cgrp, type);

	return 0;



cleanup:

	/* and cleanup the prog list */

	pl->prog = old_prog;

	bpf_cgroup_storage_free(pl->storage);

	pl->storage = old_storage;

	bpf_cgroup_storage_link(old_storage, cgrp, type);

	for_each_cgroup_storage_type(stype) {

		bpf_cgroup_storage_free(pl->storage[stype]);

		pl->storage[stype] = old_storage[stype];

		bpf_cgroup_storage_link(old_storage[stype], cgrp, type);

	}

	if (pl_was_allocated) {

		list_del(&pl->node);

		kfree(pl);
@@ -339,6 +366,7 @@ int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog, 
			enum bpf_attach_type type, u32 unused_flags)

{

	struct list_head *progs = &cgrp->bpf.progs[type];

	enum bpf_cgroup_storage_type stype;

	u32 flags = cgrp->bpf.flags[type];

	struct bpf_prog *old_prog = NULL;

	struct bpf_prog_list *pl;
@@ -385,8 +413,10 @@ int __cgroup_bpf_detach(struct cgroup *cgrp, struct bpf_prog *prog, 


	/* now can actually delete it from this cgroup list */

	list_del(&pl->node);

	bpf_cgroup_storage_unlink(pl->storage);

	bpf_cgroup_storage_free(pl->storage);

	for_each_cgroup_storage_type(stype) {

		bpf_cgroup_storage_unlink(pl->storage[stype]);

		bpf_cgroup_storage_free(pl->storage[stype]);

	}

	kfree(pl);

	if (list_empty(progs))

		/* last program was detached, reset flags to zero */

CRA Git | Maintained and supported by SUSTech CRA and CCSE