Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next

Daniel Borkmann says:

====================
pull-request: bpf-next 2020-03-13

The following pull-request contains BPF updates for your *net-next* tree.

We've added 86 non-merge commits during the last 12 day(s) which contain
a total of 107 files changed, 5771 insertions(+), 1700 deletions(-).

The main changes are:

1) Add modify_return attach type which allows to attach to a function via
   BPF trampoline and is run after the fentry and before the fexit programs
   and can pass a return code to the original caller, from KP Singh.

2) Generalize BPF's kallsyms handling and add BPF trampoline and dispatcher
   objects to be visible in /proc/kallsyms so they can be annotated in
   stack traces, from Jiri Olsa.

3) Extend BPF sockmap to allow for UDP next to existing TCP support in order
   in order to enable this for BPF based socket dispatch, from Lorenz Bauer.

4) Introduce a new bpftool 'prog profile' command which attaches to existing
   BPF programs via fentry and fexit hooks and reads out hardware counters
   during that period, from Song Liu. Example usage:

   bpftool prog profile id 337 duration 3 cycles instructions llc_misses

        4228 run_cnt
     3403698 cycles                                              (84.08%)
     3525294 instructions   #  1.04 insn per cycle               (84.05%)
          13 llc_misses     #  3.69 LLC misses per million isns  (83.50%)

5) Batch of improvements to libbpf, bpftool and BPF selftests. Also addition
   of a new bpf_link abstraction to keep in particular BPF tracing programs
   attached even when the applicaion owning them exits, from Andrii Nakryiko.

6) New bpf_get_current_pid_tgid() helper for tracing to perform PID filtering
   and which returns the PID as seen by the init namespace, from Carlos Neira.

7) Refactor of RISC-V JIT code to move out common pieces and addition of a
   new RV32G BPF JIT compiler, from Luke Nelson.

8) Add gso_size context member to __sk_buff in order to be able to know whether
   a given skb is GSO or not, from Willem de Bruijn.

9) Add a new bpf_xdp_output() helper which reuses XDP's existing perf RB output
   implementation but can be called from tracepoint programs, from Eelco Chaudron.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

12 files changed, 542 insertions, 219 deletions

diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
index c498f0fffb40..ca5cc8cdb6eb 100644
--- a/kernel/bpf/bpf_struct_ops.c
+++ b/kernel/bpf/bpf_struct_ops.c
@@ -320,6 +320,7 @@ static int bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 	struct bpf_struct_ops_value *uvalue, *kvalue;
 	const struct btf_member *member;
 	const struct btf_type *t = st_ops->type;
+	struct bpf_tramp_progs *tprogs = NULL;
 	void *udata, *kdata;
 	int prog_fd, err = 0;
 	void *image;
@@ -343,6 +344,10 @@ static int bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 	if (uvalue->state || refcount_read(&uvalue->refcnt))
 		return -EINVAL;
 
+	tprogs = kcalloc(BPF_TRAMP_MAX, sizeof(*tprogs), GFP_KERNEL);
+	if (!tprogs)
+		return -ENOMEM;
+
 	uvalue = (struct bpf_struct_ops_value *)st_map->uvalue;
 	kvalue = (struct bpf_struct_ops_value *)&st_map->kvalue;
 
@@ -425,10 +430,12 @@ static int bpf_struct_ops_map_update_elem(struct bpf_map *map, void *key,
 			goto reset_unlock;
 		}
 
+		tprogs[BPF_TRAMP_FENTRY].progs[0] = prog;
+		tprogs[BPF_TRAMP_FENTRY].nr_progs = 1;
 		err = arch_prepare_bpf_trampoline(image,
 						  st_map->image + PAGE_SIZE,
 						  &st_ops->func_models[i], 0,
-						  &prog, 1, NULL, 0, NULL);
+						  tprogs, NULL);
 		if (err < 0)
 			goto reset_unlock;
 
@@ -469,6 +476,7 @@ reset_unlock:
 	memset(uvalue, 0, map->value_size);
 	memset(kvalue, 0, map->value_size);
 unlock:
+	kfree(tprogs);
 	mutex_unlock(&st_map->lock);
 	return err;
 }
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 787140095e58..50080add2ab9 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -3710,13 +3710,26 @@ bool btf_ctx_access(int off, int size, enum bpf_access_type type,
 		nr_args--;
 	}
 
-	if (prog->expected_attach_type == BPF_TRACE_FEXIT &&
-	    arg == nr_args) {
-		if (!t)
-			/* Default prog with 5 args. 6th arg is retval. */
-			return true;
-		/* function return type */
-		t = btf_type_by_id(btf, t->type);
+	if (arg == nr_args) {
+		if (prog->expected_attach_type == BPF_TRACE_FEXIT) {
+			if (!t)
+				return true;
+			t = btf_type_by_id(btf, t->type);
+		} else if (prog->expected_attach_type == BPF_MODIFY_RETURN) {
+			/* For now the BPF_MODIFY_RETURN can only be attached to
+			 * functions that return an int.
+			 */
+			if (!t)
+				return false;
+
+			t = btf_type_skip_modifiers(btf, t->type, NULL);
+			if (!btf_type_is_int(t)) {
+				bpf_log(log,
+					"ret type %s not allowed for fmod_ret\n",
+					btf_kind_str[BTF_INFO_KIND(t->info)]);
+				return false;
+			}
+		}
 	} else if (arg >= nr_args) {
 		bpf_log(log, "func '%s' doesn't have %d-th argument\n",
 			tname, arg + 1);
diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c
index 973a20d49749..914f3463aa41 100644
--- a/kernel/bpf/core.c
+++ b/kernel/bpf/core.c
@@ -97,7 +97,7 @@ struct bpf_prog *bpf_prog_alloc_no_stats(unsigned int size, gfp_t gfp_extra_flag
 	fp->aux->prog = fp;
 	fp->jit_requested = ebpf_jit_enabled();
 
-	INIT_LIST_HEAD_RCU(&fp->aux->ksym_lnode);
+	INIT_LIST_HEAD_RCU(&fp->aux->ksym.lnode);
 
 	return fp;
 }
@@ -523,22 +523,22 @@ int bpf_jit_kallsyms __read_mostly = IS_BUILTIN(CONFIG_BPF_JIT_DEFAULT_ON);
 int bpf_jit_harden   __read_mostly;
 long bpf_jit_limit   __read_mostly;
 
-static __always_inline void
-bpf_get_prog_addr_region(const struct bpf_prog *prog,
-			 unsigned long *symbol_start,
-			 unsigned long *symbol_end)
+static void
+bpf_prog_ksym_set_addr(struct bpf_prog *prog)
 {
 	const struct bpf_binary_header *hdr = bpf_jit_binary_hdr(prog);
 	unsigned long addr = (unsigned long)hdr;
 
 	WARN_ON_ONCE(!bpf_prog_ebpf_jited(prog));
 
-	*symbol_start = addr;
-	*symbol_end   = addr + hdr->pages * PAGE_SIZE;
+	prog->aux->ksym.start = (unsigned long) prog->bpf_func;
+	prog->aux->ksym.end   = addr + hdr->pages * PAGE_SIZE;
 }
 
-void bpf_get_prog_name(const struct bpf_prog *prog, char *sym)
+static void
+bpf_prog_ksym_set_name(struct bpf_prog *prog)
 {
+	char *sym = prog->aux->ksym.name;
 	const char *end = sym + KSYM_NAME_LEN;
 	const struct btf_type *type;
 	const char *func_name;
@@ -572,36 +572,27 @@ void bpf_get_prog_name(const struct bpf_prog *prog, char *sym)
 		*sym = 0;
 }
 
-static __always_inline unsigned long
-bpf_get_prog_addr_start(struct latch_tree_node *n)
+static unsigned long bpf_get_ksym_start(struct latch_tree_node *n)
 {
-	unsigned long symbol_start, symbol_end;
-	const struct bpf_prog_aux *aux;
-
-	aux = container_of(n, struct bpf_prog_aux, ksym_tnode);
-	bpf_get_prog_addr_region(aux->prog, &symbol_start, &symbol_end);
-
-	return symbol_start;
+	return container_of(n, struct bpf_ksym, tnode)->start;
 }
 
 static __always_inline bool bpf_tree_less(struct latch_tree_node *a,
 					  struct latch_tree_node *b)
 {
-	return bpf_get_prog_addr_start(a) < bpf_get_prog_addr_start(b);
+	return bpf_get_ksym_start(a) < bpf_get_ksym_start(b);
 }
 
 static __always_inline int bpf_tree_comp(void *key, struct latch_tree_node *n)
 {
 	unsigned long val = (unsigned long)key;
-	unsigned long symbol_start, symbol_end;
-	const struct bpf_prog_aux *aux;
+	const struct bpf_ksym *ksym;
 
-	aux = container_of(n, struct bpf_prog_aux, ksym_tnode);
-	bpf_get_prog_addr_region(aux->prog, &symbol_start, &symbol_end);
+	ksym = container_of(n, struct bpf_ksym, tnode);
 
-	if (val < symbol_start)
+	if (val < ksym->start)
 		return -1;
-	if (val >= symbol_end)
+	if (val >= ksym->end)
 		return  1;
 
 	return 0;
@@ -616,20 +607,29 @@ static DEFINE_SPINLOCK(bpf_lock);
 static LIST_HEAD(bpf_kallsyms);
 static struct latch_tree_root bpf_tree __cacheline_aligned;
 
-static void bpf_prog_ksym_node_add(struct bpf_prog_aux *aux)
+void bpf_ksym_add(struct bpf_ksym *ksym)
 {
-	WARN_ON_ONCE(!list_empty(&aux->ksym_lnode));
-	list_add_tail_rcu(&aux->ksym_lnode, &bpf_kallsyms);
-	latch_tree_insert(&aux->ksym_tnode, &bpf_tree, &bpf_tree_ops);
+	spin_lock_bh(&bpf_lock);
+	WARN_ON_ONCE(!list_empty(&ksym->lnode));
+	list_add_tail_rcu(&ksym->lnode, &bpf_kallsyms);
+	latch_tree_insert(&ksym->tnode, &bpf_tree, &bpf_tree_ops);
+	spin_unlock_bh(&bpf_lock);
 }
 
-static void bpf_prog_ksym_node_del(struct bpf_prog_aux *aux)
+static void __bpf_ksym_del(struct bpf_ksym *ksym)
 {
-	if (list_empty(&aux->ksym_lnode))
+	if (list_empty(&ksym->lnode))
 		return;
 
-	latch_tree_erase(&aux->ksym_tnode, &bpf_tree, &bpf_tree_ops);
-	list_del_rcu(&aux->ksym_lnode);
+	latch_tree_erase(&ksym->tnode, &bpf_tree, &bpf_tree_ops);
+	list_del_rcu(&ksym->lnode);
+}
+
+void bpf_ksym_del(struct bpf_ksym *ksym)
+{
+	spin_lock_bh(&bpf_lock);
+	__bpf_ksym_del(ksym);
+	spin_unlock_bh(&bpf_lock);
 }
 
 static bool bpf_prog_kallsyms_candidate(const struct bpf_prog *fp)
@@ -639,8 +639,8 @@ static bool bpf_prog_kallsyms_candidate(const struct bpf_prog *fp)
 
 static bool bpf_prog_kallsyms_verify_off(const struct bpf_prog *fp)
 {
-	return list_empty(&fp->aux->ksym_lnode) ||
-	       fp->aux->ksym_lnode.prev == LIST_POISON2;
+	return list_empty(&fp->aux->ksym.lnode) ||
+	       fp->aux->ksym.lnode.prev == LIST_POISON2;
 }
 
 void bpf_prog_kallsyms_add(struct bpf_prog *fp)
@@ -649,9 +649,11 @@ void bpf_prog_kallsyms_add(struct bpf_prog *fp)
 	    !capable(CAP_SYS_ADMIN))
 		return;
 
-	spin_lock_bh(&bpf_lock);
-	bpf_prog_ksym_node_add(fp->aux);
-	spin_unlock_bh(&bpf_lock);
+	bpf_prog_ksym_set_addr(fp);
+	bpf_prog_ksym_set_name(fp);
+	fp->aux->ksym.prog = true;
+
+	bpf_ksym_add(&fp->aux->ksym);
 }
 
 void bpf_prog_kallsyms_del(struct bpf_prog *fp)
@@ -659,33 +661,30 @@ void bpf_prog_kallsyms_del(struct bpf_prog *fp)
 	if (!bpf_prog_kallsyms_candidate(fp))
 		return;
 
-	spin_lock_bh(&bpf_lock);
-	bpf_prog_ksym_node_del(fp->aux);
-	spin_unlock_bh(&bpf_lock);
+	bpf_ksym_del(&fp->aux->ksym);
 }
 
-static struct bpf_prog *bpf_prog_kallsyms_find(unsigned long addr)
+static struct bpf_ksym *bpf_ksym_find(unsigned long addr)
 {
 	struct latch_tree_node *n;
 
 	n = latch_tree_find((void *)addr, &bpf_tree, &bpf_tree_ops);
-	return n ?
-	       container_of(n, struct bpf_prog_aux, ksym_tnode)->prog :
-	       NULL;
+	return n ? container_of(n, struct bpf_ksym, tnode) : NULL;
 }
 
 const char *__bpf_address_lookup(unsigned long addr, unsigned long *size,
 				 unsigned long *off, char *sym)
 {
-	unsigned long symbol_start, symbol_end;
-	struct bpf_prog *prog;
+	struct bpf_ksym *ksym;
 	char *ret = NULL;
 
 	rcu_read_lock();
-	prog = bpf_prog_kallsyms_find(addr);
-	if (prog) {
-		bpf_get_prog_addr_region(prog, &symbol_start, &symbol_end);
-		bpf_get_prog_name(prog, sym);
+	ksym = bpf_ksym_find(addr);
+	if (ksym) {
+		unsigned long symbol_start = ksym->start;
+		unsigned long symbol_end = ksym->end;
+
+		strncpy(sym, ksym->name, KSYM_NAME_LEN);
 
 		ret = sym;
 		if (size)
@@ -703,19 +702,28 @@ bool is_bpf_text_address(unsigned long addr)
 	bool ret;
 
 	rcu_read_lock();
-	ret = bpf_prog_kallsyms_find(addr) != NULL;
+	ret = bpf_ksym_find(addr) != NULL;
 	rcu_read_unlock();
 
 	return ret;
 }
 
+static struct bpf_prog *bpf_prog_ksym_find(unsigned long addr)
+{
+	struct bpf_ksym *ksym = bpf_ksym_find(addr);
+
+	return ksym && ksym->prog ?
+	       container_of(ksym, struct bpf_prog_aux, ksym)->prog :
+	       NULL;
+}
+
 const struct exception_table_entry *search_bpf_extables(unsigned long addr)
 {
 	const struct exception_table_entry *e = NULL;
 	struct bpf_prog *prog;
 
 	rcu_read_lock();
-	prog = bpf_prog_kallsyms_find(addr);
+	prog = bpf_prog_ksym_find(addr);
 	if (!prog)
 		goto out;
 	if (!prog->aux->num_exentries)
@@ -730,7 +738,7 @@ out:
 int bpf_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
 		    char *sym)
 {
-	struct bpf_prog_aux *aux;
+	struct bpf_ksym *ksym;
 	unsigned int it = 0;
 	int ret = -ERANGE;
 
@@ -738,13 +746,13 @@ int bpf_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
 		return ret;
 
 	rcu_read_lock();
-	list_for_each_entry_rcu(aux, &bpf_kallsyms, ksym_lnode) {
+	list_for_each_entry_rcu(ksym, &bpf_kallsyms, lnode) {
 		if (it++ != symnum)
 			continue;
 
-		bpf_get_prog_name(aux->prog, sym);
+		strncpy(sym, ksym->name, KSYM_NAME_LEN);
 
-		*value = (unsigned long)aux->prog->bpf_func;
+		*value = ksym->start;
 		*type  = BPF_SYM_ELF_TYPE;
 
 		ret = 0;
@@ -2149,6 +2157,7 @@ const struct bpf_func_proto bpf_get_current_uid_gid_proto __weak;
 const struct bpf_func_proto bpf_get_current_comm_proto __weak;
 const struct bpf_func_proto bpf_get_current_cgroup_id_proto __weak;
 const struct bpf_func_proto bpf_get_local_storage_proto __weak;
+const struct bpf_func_proto bpf_get_ns_current_pid_tgid_proto __weak;
 
 const struct bpf_func_proto * __weak bpf_get_trace_printk_proto(void)
 {
diff --git a/kernel/bpf/dispatcher.c b/kernel/bpf/dispatcher.c
index b3e5b214fed8..2444bd15cc2d 100644
--- a/kernel/bpf/dispatcher.c
+++ b/kernel/bpf/dispatcher.c
@@ -113,7 +113,7 @@ static void bpf_dispatcher_update(struct bpf_dispatcher *d, int prev_num_progs)
 		noff = 0;
 	} else {
 		old = d->image + d->image_off;
-		noff = d->image_off ^ (BPF_IMAGE_SIZE / 2);
+		noff = d->image_off ^ (PAGE_SIZE / 2);
 	}
 
 	new = d->num_progs ? d->image + noff : NULL;
@@ -140,9 +140,10 @@ void bpf_dispatcher_change_prog(struct bpf_dispatcher *d, struct bpf_prog *from,
 
 	mutex_lock(&d->mutex);
 	if (!d->image) {
-		d->image = bpf_image_alloc();
+		d->image = bpf_jit_alloc_exec_page();
 		if (!d->image)
 			goto out;
+		bpf_image_ksym_add(d->image, &d->ksym);
 	}
 
 	prev_num_progs = d->num_progs;
diff --git a/kernel/bpf/helpers.c b/kernel/bpf/helpers.c
index d8b7b110a1c5..01878db15eaf 100644
--- a/kernel/bpf/helpers.c
+++ b/kernel/bpf/helpers.c
@@ -12,6 +12,8 @@
 #include <linux/filter.h>
 #include <linux/ctype.h>
 #include <linux/jiffies.h>
+#include <linux/pid_namespace.h>
+#include <linux/proc_ns.h>
 
 #include "../../lib/kstrtox.h"
 
@@ -499,3 +501,46 @@ const struct bpf_func_proto bpf_strtoul_proto = {
 	.arg4_type	= ARG_PTR_TO_LONG,
 };
 #endif
+
+BPF_CALL_4(bpf_get_ns_current_pid_tgid, u64, dev, u64, ino,
+	   struct bpf_pidns_info *, nsdata, u32, size)
+{
+	struct task_struct *task = current;
+	struct pid_namespace *pidns;
+	int err = -EINVAL;
+
+	if (unlikely(size != sizeof(struct bpf_pidns_info)))
+		goto clear;
+
+	if (unlikely((u64)(dev_t)dev != dev))
+		goto clear;
+
+	if (unlikely(!task))
+		goto clear;
+
+	pidns = task_active_pid_ns(task);
+	if (unlikely(!pidns)) {
+		err = -ENOENT;
+		goto clear;
+	}
+
+	if (!ns_match(&pidns->ns, (dev_t)dev, ino))
+		goto clear;
+
+	nsdata->pid = task_pid_nr_ns(task, pidns);
+	nsdata->tgid = task_tgid_nr_ns(task, pidns);
+	return 0;
+clear:
+	memset((void *)nsdata, 0, (size_t) size);
+	return err;
+}
+
+const struct bpf_func_proto bpf_get_ns_current_pid_tgid_proto = {
+	.func		= bpf_get_ns_current_pid_tgid,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_ANYTHING,
+	.arg2_type	= ARG_ANYTHING,
+	.arg3_type      = ARG_PTR_TO_UNINIT_MEM,
+	.arg4_type      = ARG_CONST_SIZE,
+};
diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c
index 5e40e7fccc21..95087d9f4ed3 100644
--- a/kernel/bpf/inode.c
+++ b/kernel/bpf/inode.c
@@ -25,6 +25,7 @@ enum bpf_type {
 	BPF_TYPE_UNSPEC	= 0,
 	BPF_TYPE_PROG,
 	BPF_TYPE_MAP,
+	BPF_TYPE_LINK,
 };
 
 static void *bpf_any_get(void *raw, enum bpf_type type)
@@ -36,6 +37,9 @@ static void *bpf_any_get(void *raw, enum bpf_type type)
 	case BPF_TYPE_MAP:
 		bpf_map_inc_with_uref(raw);
 		break;
+	case BPF_TYPE_LINK:
+		bpf_link_inc(raw);
+		break;
 	default:
 		WARN_ON_ONCE(1);
 		break;
@@ -53,6 +57,9 @@ static void bpf_any_put(void *raw, enum bpf_type type)
 	case BPF_TYPE_MAP:
 		bpf_map_put_with_uref(raw);
 		break;
+	case BPF_TYPE_LINK:
+		bpf_link_put(raw);
+		break;
 	default:
 		WARN_ON_ONCE(1);
 		break;
@@ -63,20 +70,32 @@ static void *bpf_fd_probe_obj(u32 ufd, enum bpf_type *type)
 {
 	void *raw;
 
-	*type = BPF_TYPE_MAP;
 	raw = bpf_map_get_with_uref(ufd);
-	if (IS_ERR(raw)) {
+	if (!IS_ERR(raw)) {
+		*type = BPF_TYPE_MAP;
+		return raw;
+	}
+
+	raw = bpf_prog_get(ufd);
+	if (!IS_ERR(raw)) {
 		*type = BPF_TYPE_PROG;
-		raw = bpf_prog_get(ufd);
+		return raw;
 	}
 
-	return raw;
+	raw = bpf_link_get_from_fd(ufd);
+	if (!IS_ERR(raw)) {
+		*type = BPF_TYPE_LINK;
+		return raw;
+	}
+
+	return ERR_PTR(-EINVAL);
 }
 
 static const struct inode_operations bpf_dir_iops;
 
 static const struct inode_operations bpf_prog_iops = { };
 static const struct inode_operations bpf_map_iops  = { };
+static const struct inode_operations bpf_link_iops  = { };
 
 static struct inode *bpf_get_inode(struct super_block *sb,
 				   const struct inode *dir,
@@ -114,6 +133,8 @@ static int bpf_inode_type(const struct inode *inode, enum bpf_type *type)
 		*type = BPF_TYPE_PROG;
 	else if (inode->i_op == &bpf_map_iops)
 		*type = BPF_TYPE_MAP;
+	else if (inode->i_op == &bpf_link_iops)
+		*type = BPF_TYPE_LINK;
 	else
 		return -EACCES;
 
@@ -335,6 +356,12 @@ static int bpf_mkmap(struct dentry *dentry, umode_t mode, void *arg)
 			     &bpffs_map_fops : &bpffs_obj_fops);
 }
 
+static int bpf_mklink(struct dentry *dentry, umode_t mode, void *arg)
+{
+	return bpf_mkobj_ops(dentry, mode, arg, &bpf_link_iops,
+			     &bpffs_obj_fops);
+}
+
 static struct dentry *
 bpf_lookup(struct inode *dir, struct dentry *dentry, unsigned flags)
 {
@@ -411,6 +438,9 @@ static int bpf_obj_do_pin(const char __user *pathname, void *raw,
 	case BPF_TYPE_MAP:
 		ret = vfs_mkobj(dentry, mode, bpf_mkmap, raw);
 		break;
+	case BPF_TYPE_LINK:
+		ret = vfs_mkobj(dentry, mode, bpf_mklink, raw);
+		break;
 	default:
 		ret = -EPERM;
 	}
@@ -487,6 +517,8 @@ int bpf_obj_get_user(const char __user *pathname, int flags)
 		ret = bpf_prog_new_fd(raw);
 	else if (type == BPF_TYPE_MAP)
 		ret = bpf_map_new_fd(raw, f_flags);
+	else if (type == BPF_TYPE_LINK)
+		ret = bpf_link_new_fd(raw);
 	else
 		return -ENOENT;
 
@@ -504,6 +536,8 @@ static struct bpf_prog *__get_prog_inode(struct inode *inode, enum bpf_prog_type
 
 	if (inode->i_op == &bpf_map_iops)
 		return ERR_PTR(-EINVAL);
+	if (inode->i_op == &bpf_link_iops)
+		return ERR_PTR(-EINVAL);
 	if (inode->i_op != &bpf_prog_iops)
 		return ERR_PTR(-EACCES);
 
diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index c536c65256ad..85567a6ea5f9 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -2173,84 +2173,274 @@ static int bpf_obj_get(const union bpf_attr *attr)
 				attr->file_flags);
 }
 
-static int bpf_tracing_prog_release(struct inode *inode, struct file *filp)
+struct bpf_link {
+	atomic64_t refcnt;
+	const struct bpf_link_ops *ops;
+	struct bpf_prog *prog;
+	struct work_struct work;
+};
+
+void bpf_link_init(struct bpf_link *link, const struct bpf_link_ops *ops,
+		   struct bpf_prog *prog)
 {
-	struct bpf_prog *prog = filp->private_data;
+	atomic64_set(&link->refcnt, 1);
+	link->ops = ops;
+	link->prog = prog;
+}
 
-	WARN_ON_ONCE(bpf_trampoline_unlink_prog(prog));
-	bpf_prog_put(prog);
+/* Clean up bpf_link and corresponding anon_inode file and FD. After
+ * anon_inode is created, bpf_link can't be just kfree()'d due to deferred
+ * anon_inode's release() call. This helper manages marking bpf_link as
+ * defunct, releases anon_inode file and puts reserved FD.
+ */
+static void bpf_link_cleanup(struct bpf_link *link, struct file *link_file,
+			     int link_fd)
+{
+	link->prog = NULL;
+	fput(link_file);
+	put_unused_fd(link_fd);
+}
+
+void bpf_link_inc(struct bpf_link *link)
+{
+	atomic64_inc(&link->refcnt);
+}
+
+/* bpf_link_free is guaranteed to be called from process context */
+static void bpf_link_free(struct bpf_link *link)
+{
+	if (link->prog) {
+		/* detach BPF program, clean up used resources */
+		link->ops->release(link);
+		bpf_prog_put(link->prog);
+	}
+	/* free bpf_link and its containing memory */
+	link->ops->dealloc(link);
+}
+
+static void bpf_link_put_deferred(struct work_struct *work)
+{
+	struct bpf_link *link = container_of(work, struct bpf_link, work);
+
+	bpf_link_free(link);
+}
+
+/* bpf_link_put can be called from atomic context, but ensures that resources
+ * are freed from process context
+ */
+void bpf_link_put(struct bpf_link *link)
+{
+	if (!atomic64_dec_and_test(&link->refcnt))
+		return;
+
+	if (in_atomic()) {
+		INIT_WORK(&link->work, bpf_link_put_deferred);
+		schedule_work(&link->work);
+	} else {
+		bpf_link_free(link);
+	}
+}
+
+static int bpf_link_release(struct inode *inode, struct file *filp)
+{
+	struct bpf_link *link = filp->private_data;
+
+	bpf_link_put(link);
 	return 0;
 }
 
-static const struct file_operations bpf_tracing_prog_fops = {
-	.release	= bpf_tracing_prog_release,
+#ifdef CONFIG_PROC_FS
+static const struct bpf_link_ops bpf_raw_tp_lops;
+static const struct bpf_link_ops bpf_tracing_link_lops;
+static const struct bpf_link_ops bpf_xdp_link_lops;
+
+static void bpf_link_show_fdinfo(struct seq_file *m, struct file *filp)
+{
+	const struct bpf_link *link = filp->private_data;
+	const struct bpf_prog *prog = link->prog;
+	char prog_tag[sizeof(prog->tag) * 2 + 1] = { };
+	const char *link_type;
+
+	if (link->ops == &bpf_raw_tp_lops)
+		link_type = "raw_tracepoint";
+	else if (link->ops == &bpf_tracing_link_lops)
+		link_type = "tracing";
+	else
+		link_type = "unknown";
+
+	bin2hex(prog_tag, prog->tag, sizeof(prog->tag));
+	seq_printf(m,
+		   "link_type:\t%s\n"
+		   "prog_tag:\t%s\n"
+		   "prog_id:\t%u\n",
+		   link_type,
+		   prog_tag,
+		   prog->aux->id);
+}
+#endif
+
+const struct file_operations bpf_link_fops = {
+#ifdef CONFIG_PROC_FS
+	.show_fdinfo	= bpf_link_show_fdinfo,
+#endif
+	.release	= bpf_link_release,
 	.read		= bpf_dummy_read,
 	.write		= bpf_dummy_write,
 };
 
+int bpf_link_new_fd(struct bpf_link *link)
+{
+	return anon_inode_getfd("bpf-link", &bpf_link_fops, link, O_CLOEXEC);
+}
+
+/* Similar to bpf_link_new_fd, create anon_inode for given bpf_link, but
+ * instead of immediately installing fd in fdtable, just reserve it and
+ * return. Caller then need to either install it with fd_install(fd, file) or
+ * release with put_unused_fd(fd).
+ * This is useful for cases when bpf_link attachment/detachment are
+ * complicated and expensive operations and should be delayed until all the fd
+ * reservation and anon_inode creation succeeds.
+ */
+struct file *bpf_link_new_file(struct bpf_link *link, int *reserved_fd)
+{
+	struct file *file;
+	int fd;
+
+	fd = get_unused_fd_flags(O_CLOEXEC);
+	if (fd < 0)
+		return ERR_PTR(fd);
+
+	file = anon_inode_getfile("bpf_link", &bpf_link_fops, link, O_CLOEXEC);
+	if (IS_ERR(file)) {
+		put_unused_fd(fd);
+		return file;
+	}
+
+	*reserved_fd = fd;
+	return file;
+}
+
+struct bpf_link *bpf_link_get_from_fd(u32 ufd)
+{
+	struct fd f = fdget(ufd);
+	struct bpf_link *link;
+
+	if (!f.file)
+		return ERR_PTR(-EBADF);
+	if (f.file->f_op != &bpf_link_fops) {
+		fdput(f);
+		return ERR_PTR(-EINVAL);
+	}
+
+	link = f.file->private_data;
+	bpf_link_inc(link);
+	fdput(f);
+
+	return link;
+}
+
+struct bpf_tracing_link {
+	struct bpf_link link;
+};
+
+static void bpf_tracing_link_release(struct bpf_link *link)
+{
+	WARN_ON_ONCE(bpf_trampoline_unlink_prog(link->prog));
+}
+
+static void bpf_tracing_link_dealloc(struct bpf_link *link)
+{
+	struct bpf_tracing_link *tr_link =
+		container_of(link, struct bpf_tracing_link, link);
+
+	kfree(tr_link);
+}
+
+static const struct bpf_link_ops bpf_tracing_link_lops = {
+	.release = bpf_tracing_link_release,
+	.dealloc = bpf_tracing_link_dealloc,
+};
+
 static int bpf_tracing_prog_attach(struct bpf_prog *prog)
 {
-	int tr_fd, err;
+	struct bpf_tracing_link *link;
+	struct file *link_file;
+	int link_fd, err;
 
 	if (prog->expected_attach_type != BPF_TRACE_FENTRY &&
 	    prog->expected_attach_type != BPF_TRACE_FEXIT &&
+	    prog->expected_attach_type != BPF_MODIFY_RETURN &&
 	    prog->type != BPF_PROG_TYPE_EXT) {
 		err = -EINVAL;
 		goto out_put_prog;
 	}
 
-	err = bpf_trampoline_link_prog(prog);
-	if (err)
+	link = kzalloc(sizeof(*link), GFP_USER);
+	if (!link) {
+		err = -ENOMEM;
+		goto out_put_prog;
+	}
+	bpf_link_init(&link->link, &bpf_tracing_link_lops, prog);
+
+	link_file = bpf_link_new_file(&link->link, &link_fd);
+	if (IS_ERR(link_file)) {
+		kfree(link);
+		err = PTR_ERR(link_file);
 		goto out_put_prog;
+	}
 
-	tr_fd = anon_inode_getfd("bpf-tracing-prog", &bpf_tracing_prog_fops,
-				 prog, O_CLOEXEC);
-	if (tr_fd < 0) {
-		WARN_ON_ONCE(bpf_trampoline_unlink_prog(prog));
-		err = tr_fd;
+	err = bpf_trampoline_link_prog(prog);
+	if (err) {
+		bpf_link_cleanup(&link->link, link_file, link_fd);
 		goto out_put_prog;
 	}
-	return tr_fd;
+
+	fd_install(link_fd, link_file);
+	return link_fd;
 
 out_put_prog:
 	bpf_prog_put(prog);
 	return err;
 }
 
-struct bpf_raw_tracepoint {
+struct bpf_raw_tp_link {
+	struct bpf_link link;
 	struct bpf_raw_event_map *btp;
-	struct bpf_prog *prog;
 };
 
-static int bpf_raw_tracepoint_release(struct inode *inode, struct file *filp)
+static void bpf_raw_tp_link_release(struct bpf_link *link)
 {
-	struct bpf_raw_tracepoint *raw_tp = filp->private_data;
+	struct bpf_raw_tp_link *raw_tp =
+		container_of(link, struct bpf_raw_tp_link, link);
 
-	if (raw_tp->prog) {
-		bpf_probe_unregister(raw_tp->btp, raw_tp->prog);
-		bpf_prog_put(raw_tp->prog);
-	}
+	bpf_probe_unregister(raw_tp->btp, raw_tp->link.prog);
 	bpf_put_raw_tracepoint(raw_tp->btp);
+}
+
+static void bpf_raw_tp_link_dealloc(struct bpf_link *link)
+{
+	struct bpf_raw_tp_link *raw_tp =
+		container_of(link, struct bpf_raw_tp_link, link);
+
 	kfree(raw_tp);
-	return 0;
 }
 
-static const struct file_operations bpf_raw_tp_fops = {
-	.release	= bpf_raw_tracepoint_release,
-	.read		= bpf_dummy_read,
-	.write		= bpf_dummy_write,
+static const struct bpf_link_ops bpf_raw_tp_lops = {
+	.release = bpf_raw_tp_link_release,
+	.dealloc = bpf_raw_tp_link_dealloc,
 };
 
 #define BPF_RAW_TRACEPOINT_OPEN_LAST_FIELD raw_tracepoint.prog_fd
 
 static int bpf_raw_tracepoint_open(const union bpf_attr *attr)
 {
-	struct bpf_raw_tracepoint *raw_tp;
+	struct bpf_raw_tp_link *link;
 	struct bpf_raw_event_map *btp;
+	struct file *link_file;
 	struct bpf_prog *prog;
 	const char *tp_name;
 	char buf[128];
-	int tp_fd, err;
+	int link_fd, err;
 
 	if (CHECK_ATTR(BPF_RAW_TRACEPOINT_OPEN))
 		return -EINVAL;
@@ -2297,29 +2487,30 @@ static int bpf_raw_tracepoint_open(const union bpf_attr *attr)
 		goto out_put_prog;
 	}
 
-	raw_tp = kzalloc(sizeof(*raw_tp), GFP_USER);
-	if (!raw_tp) {
+	link = kzalloc(sizeof(*link), GFP_USER);
+	if (!link) {
 		err = -ENOMEM;
 		goto out_put_btp;
 	}
-	raw_tp->btp = btp;
-	raw_tp->prog = prog;
+	bpf_link_init(&link->link, &bpf_raw_tp_lops, prog);
+	link->btp = btp;
 
-	err = bpf_probe_register(raw_tp->btp, prog);
-	if (err)
-		goto out_free_tp;
+	link_file = bpf_link_new_file(&link->link, &link_fd);
+	if (IS_ERR(link_file)) {
+		kfree(link);
+		err = PTR_ERR(link_file);
+		goto out_put_btp;
+	}
 
-	tp_fd = anon_inode_getfd("bpf-raw-tracepoint", &bpf_raw_tp_fops, raw_tp,
-				 O_CLOEXEC);
-	if (tp_fd < 0) {
-		bpf_probe_unregister(raw_tp->btp, prog);
-		err = tp_fd;
-		goto out_free_tp;
+	err = bpf_probe_register(link->btp, prog);
+	if (err) {
+		bpf_link_cleanup(&link->link, link_file, link_fd);
+		goto out_put_btp;
 	}
-	return tp_fd;
 
-out_free_tp:
-	kfree(raw_tp);
+	fd_install(link_fd, link_file);
+	return link_fd;
+
 out_put_btp:
 	bpf_put_raw_tracepoint(btp);
 out_put_prog:
@@ -3266,15 +3457,21 @@ static int bpf_task_fd_query(const union bpf_attr *attr,
 	if (err)
 		goto out;
 
-	if (file->f_op == &bpf_raw_tp_fops) {
-		struct bpf_raw_tracepoint *raw_tp = file->private_data;
-		struct bpf_raw_event_map *btp = raw_tp->btp;
+	if (file->f_op == &bpf_link_fops) {
+		struct bpf_link *link = file->private_data;
 
-		err = bpf_task_fd_query_copy(attr, uattr,
-					     raw_tp->prog->aux->id,
-					     BPF_FD_TYPE_RAW_TRACEPOINT,
-					     btp->tp->name, 0, 0);
-		goto put_file;
+		if (link->ops == &bpf_raw_tp_lops) {
+			struct bpf_raw_tp_link *raw_tp =
+				container_of(link, struct bpf_raw_tp_link, link);
+			struct bpf_raw_event_map *btp = raw_tp->btp;
+
+			err = bpf_task_fd_query_copy(attr, uattr,
+						     raw_tp->link.prog->aux->id,
+						     BPF_FD_TYPE_RAW_TRACEPOINT,
+						     btp->tp->name, 0, 0);
+			goto put_file;
+		}
+		goto out_not_supp;
 	}
 
 	event = perf_get_event(file);
@@ -3294,6 +3491,7 @@ static int bpf_task_fd_query(const union bpf_attr *attr,
 		goto put_file;
 	}
 
+out_not_supp:
 	err = -ENOTSUPP;
 put_file:
 	fput(file);
diff --git a/kernel/bpf/trampoline.c b/kernel/bpf/trampoline.c
index 704fa787fec0..f30bca2a4d01 100644
--- a/kernel/bpf/trampoline.c
+++ b/kernel/bpf/trampoline.c
@@ -5,6 +5,7 @@
 #include <linux/filter.h>
 #include <linux/ftrace.h>
 #include <linux/rbtree_latch.h>
+#include <linux/perf_event.h>
 
 /* dummy _ops. The verifier will operate on target program's ops. */
 const struct bpf_verifier_ops bpf_extension_verifier_ops = {
@@ -17,12 +18,11 @@ const struct bpf_prog_ops bpf_extension_prog_ops = {
 #define TRAMPOLINE_TABLE_SIZE (1 << TRAMPOLINE_HASH_BITS)
 
 static struct hlist_head trampoline_table[TRAMPOLINE_TABLE_SIZE];
-static struct latch_tree_root image_tree __cach