1 files changed, 170 insertions, 95 deletions

diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 514b4681a90a..afb414b26d01 100644
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -22,6 +22,45 @@
 DEFINE_STATIC_KEY_ARRAY_FALSE(cgroup_bpf_enabled_key, MAX_CGROUP_BPF_ATTACH_TYPE);
 EXPORT_SYMBOL(cgroup_bpf_enabled_key);
 
+/* __always_inline is necessary to prevent indirect call through run_prog
+ * function pointer.
+ */
+static __always_inline int
+bpf_prog_run_array_cg(const struct cgroup_bpf *cgrp,
+		      enum cgroup_bpf_attach_type atype,
+		      const void *ctx, bpf_prog_run_fn run_prog,
+		      int retval, u32 *ret_flags)
+{
+	const struct bpf_prog_array_item *item;
+	const struct bpf_prog *prog;
+	const struct bpf_prog_array *array;
+	struct bpf_run_ctx *old_run_ctx;
+	struct bpf_cg_run_ctx run_ctx;
+	u32 func_ret;
+
+	run_ctx.retval = retval;
+	migrate_disable();
+	rcu_read_lock();
+	array = rcu_dereference(cgrp->effective[atype]);
+	item = &array->items[0];
+	old_run_ctx = bpf_set_run_ctx(&run_ctx.run_ctx);
+	while ((prog = READ_ONCE(item->prog))) {
+		run_ctx.prog_item = item;
+		func_ret = run_prog(prog, ctx);
+		if (ret_flags) {
+			*(ret_flags) |= (func_ret >> 1);
+			func_ret &= 1;
+		}
+		if (!func_ret && !IS_ERR_VALUE((long)run_ctx.retval))
+			run_ctx.retval = -EPERM;
+		item++;
+	}
+	bpf_reset_run_ctx(old_run_ctx);
+	rcu_read_unlock();
+	migrate_enable();
+	return run_ctx.retval;
+}
+
 void cgroup_bpf_offline(struct cgroup *cgrp)
 {
 	cgroup_get(cgrp);
@@ -1031,7 +1070,7 @@ int cgroup_bpf_prog_query(const union bpf_attr *attr,
  * __cgroup_bpf_run_filter_skb() - Run a program for packet filtering
  * @sk: The socket sending or receiving traffic
  * @skb: The skb that is being sent or received
- * @type: The type of program to be exectuted
+ * @type: The type of program to be executed
  *
  * If no socket is passed, or the socket is not of type INET or INET6,
  * this function does nothing and returns 0.
@@ -1044,7 +1083,7 @@ int cgroup_bpf_prog_query(const union bpf_attr *attr,
  *   NET_XMIT_DROP       (1)	- drop packet and notify TCP to call cwr
  *   NET_XMIT_CN         (2)	- continue with packet output and notify TCP
  *				  to call cwr
- *   -EPERM			- drop packet
+ *   -err			- drop packet
  *
  * For ingress packets, this function will return -EPERM if any
  * attached program was found and if it returned != 1 during execution.
@@ -1075,12 +1114,40 @@ int __cgroup_bpf_run_filter_skb(struct sock *sk,
 	bpf_compute_and_save_data_end(skb, &saved_data_end);
 
 	if (atype == CGROUP_INET_EGRESS) {
-		ret = BPF_PROG_CGROUP_INET_EGRESS_RUN_ARRAY(
-			cgrp->bpf.effective[atype], skb, __bpf_prog_run_save_cb);
+		u32 flags = 0;
+		bool cn;
+
+		ret = bpf_prog_run_array_cg(&cgrp->bpf, atype, skb,
+					    __bpf_prog_run_save_cb, 0, &flags);
+
+		/* Return values of CGROUP EGRESS BPF programs are:
+		 *   0: drop packet
+		 *   1: keep packet
+		 *   2: drop packet and cn
+		 *   3: keep packet and cn
+		 *
+		 * The returned value is then converted to one of the NET_XMIT
+		 * or an error code that is then interpreted as drop packet
+		 * (and no cn):
+		 *   0: NET_XMIT_SUCCESS  skb should be transmitted
+		 *   1: NET_XMIT_DROP     skb should be dropped and cn
+		 *   2: NET_XMIT_CN       skb should be transmitted and cn
+		 *   3: -err              skb should be dropped
+		 */
+
+		cn = flags & BPF_RET_SET_CN;
+		if (ret && !IS_ERR_VALUE((long)ret))
+			ret = -EFAULT;
+		if (!ret)
+			ret = (cn ? NET_XMIT_CN : NET_XMIT_SUCCESS);
+		else
+			ret = (cn ? NET_XMIT_DROP : ret);
 	} else {
-		ret = BPF_PROG_RUN_ARRAY_CG(cgrp->bpf.effective[atype], skb,
-					    __bpf_prog_run_save_cb);
-		ret = (ret == 1 ? 0 : -EPERM);
+		ret = bpf_prog_run_array_cg(&cgrp->bpf, atype,
+					    skb, __bpf_prog_run_save_cb, 0,
+					    NULL);
+		if (ret && !IS_ERR_VALUE((long)ret))
+			ret = -EFAULT;
 	}
 	bpf_restore_data_end(skb, saved_data_end);
 	__skb_pull(skb, offset);
@@ -1093,7 +1160,7 @@ EXPORT_SYMBOL(__cgroup_bpf_run_filter_skb);
 /**
  * __cgroup_bpf_run_filter_sk() - Run a program on a sock
  * @sk: sock structure to manipulate
- * @type: The type of program to be exectuted
+ * @type: The type of program to be executed
  *
  * socket is passed is expected to be of type INET or INET6.
  *
@@ -1107,10 +1174,9 @@ int __cgroup_bpf_run_filter_sk(struct sock *sk,
 			       enum cgroup_bpf_attach_type atype)
 {
 	struct cgroup *cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
-	int ret;
 
-	ret = BPF_PROG_RUN_ARRAY_CG(cgrp->bpf.effective[atype], sk, bpf_prog_run);
-	return ret == 1 ? 0 : -EPERM;
+	return bpf_prog_run_array_cg(&cgrp->bpf, atype, sk, bpf_prog_run, 0,
+				     NULL);
 }
 EXPORT_SYMBOL(__cgroup_bpf_run_filter_sk);
 
@@ -1119,7 +1185,7 @@ EXPORT_SYMBOL(__cgroup_bpf_run_filter_sk);
  *                                       provided by user sockaddr
  * @sk: sock struct that will use sockaddr
  * @uaddr: sockaddr struct provided by user
- * @type: The type of program to be exectuted
+ * @type: The type of program to be executed
  * @t_ctx: Pointer to attach type specific context
  * @flags: Pointer to u32 which contains higher bits of BPF program
  *         return value (OR'ed together).
@@ -1142,7 +1208,6 @@ int __cgroup_bpf_run_filter_sock_addr(struct sock *sk,
 	};
 	struct sockaddr_storage unspec;
 	struct cgroup *cgrp;
-	int ret;
 
 	/* Check socket family since not all sockets represent network
 	 * endpoint (e.g. AF_UNIX).
@@ -1156,10 +1221,8 @@ int __cgroup_bpf_run_filter_sock_addr(struct sock *sk,
 	}
 
 	cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
-	ret = BPF_PROG_RUN_ARRAY_CG_FLAGS(cgrp->bpf.effective[atype], &ctx,
-				          bpf_prog_run, flags);
-
-	return ret == 1 ? 0 : -EPERM;
+	return bpf_prog_run_array_cg(&cgrp->bpf, atype, &ctx, bpf_prog_run,
+				     0, flags);
 }
 EXPORT_SYMBOL(__cgroup_bpf_run_filter_sock_addr);
 
@@ -1169,7 +1232,7 @@ EXPORT_SYMBOL(__cgroup_bpf_run_filter_sock_addr);
  * @sock_ops: bpf_sock_ops_kern struct to pass to program. Contains
  * sk with connection information (IP addresses, etc.) May not contain
  * cgroup info if it is a req sock.
- * @type: The type of program to be exectuted
+ * @type: The type of program to be executed
  *
  * socket passed is expected to be of type INET or INET6.
  *
@@ -1184,11 +1247,9 @@ int __cgroup_bpf_run_filter_sock_ops(struct sock *sk,
 				     enum cgroup_bpf_attach_type atype)
 {
 	struct cgroup *cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data);
-	int ret;
 
-	ret = BPF_PROG_RUN_ARRAY_CG(cgrp->bpf.effective[atype], sock_ops,
-				    bpf_prog_run);
-	return ret == 1 ? 0 : -EPERM;
+	return bpf_prog_run_array_cg(&cgrp->bpf, atype, sock_ops, bpf_prog_run,
+				     0, NULL);
 }
 EXPORT_SYMBOL(__cgroup_bpf_run_filter_sock_ops);
 
@@ -1201,17 +1262,47 @@ int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,
 		.major = major,
 		.minor = minor,
 	};
-	int allow;
+	int ret;
 
 	rcu_read_lock();
 	cgrp = task_dfl_cgroup(current);
-	allow = BPF_PROG_RUN_ARRAY_CG(cgrp->bpf.effective[atype], &ctx,
-				      bpf_prog_run);
+	ret = bpf_prog_run_array_cg(&cgrp->bpf, atype, &ctx, bpf_prog_run, 0,
+				    NULL);
 	rcu_read_unlock();
 
-	return !allow;
+	return ret;
 }
 
+BPF_CALL_0(bpf_get_retval)
+{
+	struct bpf_cg_run_ctx *ctx =
+		container_of(current->bpf_ctx, struct bpf_cg_run_ctx, run_ctx);
+
+	return ctx->retval;
+}
+
+static const struct bpf_func_proto bpf_get_retval_proto = {
+	.func		= bpf_get_retval,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+};
+
+BPF_CALL_1(bpf_set_retval, int, retval)
+{
+	struct bpf_cg_run_ctx *ctx =
+		container_of(current->bpf_ctx, struct bpf_cg_run_ctx, run_ctx);
+
+	ctx->retval = retval;
+	return 0;
+}
+
+static const struct bpf_func_proto bpf_set_retval_proto = {
+	.func		= bpf_set_retval,
+	.gpl_only	= false,
+	.ret_type	= RET_INTEGER,
+	.arg1_type	= ARG_ANYTHING,
+};
+
 static const struct bpf_func_proto *
 cgroup_base_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 {
@@ -1224,6 +1315,10 @@ cgroup_base_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
 		return &bpf_get_current_cgroup_id_proto;
 	case BPF_FUNC_perf_event_output:
 		return &bpf_event_output_data_proto;
+	case BPF_FUNC_get_retval:
+		return &bpf_get_retval_proto;
+	case BPF_FUNC_set_retval:
+		return &bpf_set_retval_proto;
 	default:
 		return bpf_base_func_proto(func_id);
 	}
@@ -1337,7 +1432,8 @@ int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
 
 	rcu_read_lock();
 	cgrp = task_dfl_cgroup(current);
-	ret = BPF_PROG_RUN_ARRAY_CG(cgrp->bpf.effective[atype], &ctx, bpf_prog_run);
+	ret = bpf_prog_run_array_cg(&cgrp->bpf, atype, &ctx, bpf_prog_run, 0,
+				    NULL);
 	rcu_read_unlock();
 
 	kfree(ctx.cur_val);
@@ -1350,24 +1446,10 @@ int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
 		kfree(ctx.new_val);
 	}
 
-	return ret == 1 ? 0 : -EPERM;
+	return ret;
 }
 
 #ifdef CONFIG_NET
-static bool __cgroup_bpf_prog_array_is_empty(struct cgroup *cgrp,
-					     enum cgroup_bpf_attach_type attach_type)
-{
-	struct bpf_prog_array *prog_array;
-	bool empty;
-
-	rcu_read_lock();
-	prog_array = rcu_dereference(cgrp->bpf.effective[attach_type]);
-	empty = bpf_prog_array_is_empty(prog_array);
-	rcu_read_unlock();
-
-	return empty;
-}
-
 static int sockopt_alloc_buf(struct bpf_sockopt_kern *ctx, int max_optlen,
 			     struct bpf_sockopt_buf *buf)
 {
@@ -1426,19 +1508,11 @@ int __cgroup_bpf_run_filter_setsockopt(struct sock *sk, int *level,
 	};
 	int ret, max_optlen;
 
-	/* Opportunistic check to see whether we have any BPF program
-	 * attached to the hook so we don't waste time allocating
-	 * memory and locking the socket.
-	 */
-	if (__cgroup_bpf_prog_array_is_empty(cgrp, CGROUP_SETSOCKOPT))
-		return 0;
-
 	/* Allocate a bit more than the initial user buffer for
 	 * BPF program. The canonical use case is overriding
 	 * TCP_CONGESTION(nv) to TCP_CONGESTION(cubic).
 	 */
 	max_optlen = max_t(int, 16, *optlen);
-
 	max_optlen = sockopt_alloc_buf(&ctx, max_optlen, &buf);
 	if (max_optlen < 0)
 		return max_optlen;
@@ -1451,14 +1525,12 @@ int __cgroup_bpf_run_filter_setsockopt(struct sock *sk, int *level,
 	}
 
 	lock_sock(sk);
-	ret = BPF_PROG_RUN_ARRAY_CG(cgrp->bpf.effective[CGROUP_SETSOCKOPT],
-				    &ctx, bpf_prog_run);
+	ret = bpf_prog_run_array_cg(&cgrp->bpf, CGROUP_SETSOCKOPT,
+				    &ctx, bpf_prog_run, 0, NULL);
 	release_sock(sk);
 
-	if (!ret) {
-		ret = -EPERM;
+	if (ret)
 		goto out;
-	}
 
 	if (ctx.optlen == -1) {
 		/* optlen set to -1, bypass kernel */
@@ -1518,19 +1590,11 @@ int __cgroup_bpf_run_filter_getsockopt(struct sock *sk, int level,
 		.sk = sk,
 		.level = level,
 		.optname = optname,
-		.retval = retval,
+		.current_task = current,
 	};
 	int ret;
 
-	/* Opportunistic check to see whether we have any BPF program
-	 * attached to the hook so we don't waste time allocating
-	 * memory and locking the socket.
-	 */
-	if (__cgroup_bpf_prog_array_is_empty(cgrp, CGROUP_GETSOCKOPT))
-		return retval;
-
 	ctx.optlen = max_optlen;
-
 	max_optlen = sockopt_alloc_buf(&ctx, max_optlen, &buf);
 	if (max_optlen < 0)
 		return max_optlen;
@@ -1561,28 +1625,18 @@ int __cgroup_bpf_run_filter_getsockopt(struct sock *sk, int level,
 	}
 
 	lock_sock(sk);
-	ret = BPF_PROG_RUN_ARRAY_CG(cgrp->bpf.effective[CGROUP_GETSOCKOPT],
-				    &ctx, bpf_prog_run);
+	ret = bpf_prog_run_array_cg(&cgrp->bpf, CGROUP_GETSOCKOPT,
+				    &ctx, bpf_prog_run, retval, NULL);
 	release_sock(sk);
 
-	if (!ret) {
-		ret = -EPERM;
+	if (ret < 0)
 		goto out;
-	}
 
 	if (ctx.optlen > max_optlen || ctx.optlen < 0) {
 		ret = -EFAULT;
 		goto out;
 	}
 
-	/* BPF programs only allowed to set retval to 0, not some
-	 * arbitrary value.
-	 */
-	if (ctx.retval != 0 && ctx.retval != retval) {
-		ret = -EFAULT;
-		goto out;
-	}
-
 	if (ctx.optlen != 0) {
 		if (copy_to_user(optval, ctx.optval, ctx.optlen) ||
 		    put_user(ctx.optlen, optlen)) {
@@ -1591,8 +1645,6 @@ int __cgroup_bpf_run_filter_getsockopt(struct sock *sk, int level,
 		}
 	}
 
-	ret = ctx.retval;
-
 out:
 	sockopt_free_buf(&ctx, &buf);
 	return ret;
@@ -1607,10 +1659,10 @@ int __cgroup_bpf_run_filter_getsockopt_kern(struct sock *sk, int level,
 		.sk = sk,
 		.level = level,
 		.optname = optname,
-		.retval = retval,
 		.optlen = *optlen,
 		.optval = optval,
 		.optval_end = optval + *optlen,
+		.current_task = current,
 	};
 	int ret;
 
@@ -1622,26 +1674,20 @@ int __cgroup_bpf_run_filter_getsockopt_kern(struct sock *sk, int level,
 	 * be called if that data shouldn't be "exported".
 	 */
 
-	ret = BPF_PROG_RUN_ARRAY_CG(cgrp->bpf.effective[CGROUP_GETSOCKOPT],
-				    &ctx, bpf_prog_run);
-	if (!ret)
-		return -EPERM;
+	ret = bpf_prog_run_array_cg(&cgrp->bpf, CGROUP_GETSOCKOPT,
+				    &ctx, bpf_prog_run, retval, NULL);
+	if (ret < 0)
+		return ret;
 
 	if (ctx.optlen > *optlen)
 		return -EFAULT;
 
-	/* BPF programs only allowed to set retval to 0, not some
-	 * arbitrary value.
-	 */
-	if (ctx.retval != 0 && ctx.retval != retval)
-		return -EFAULT;
-
 	/* BPF programs can shrink the buffer, export the modifications.
 	 */
 	if (ctx.optlen != 0)
 		*optlen = ctx.optlen;
 
-	return ctx.retval;
+	return ret;
 }
 #endif
 
@@ -2057,10 +2103,39 @@ static u32 cg_sockopt_convert_ctx_access(enum bpf_access_type type,
 			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, optlen);
 		break;
 	case offsetof(struct bpf_sockopt, retval):
-		if (type == BPF_WRITE)
-			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_STX_MEM, retval);
-		else
-			*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, retval);
+		BUILD_BUG_ON(offsetof(struct bpf_cg_run_ctx, run_ctx) != 0);
+
+		if (type == BPF_WRITE) {
+			int treg = BPF_REG_9;
+
+			if (si->src_reg == treg || si->dst_reg == treg)
+				--treg;
+			if (si->src_reg == treg || si->dst_reg == treg)
+				--treg;
+			*insn++ = BPF_STX_MEM(BPF_DW, si->dst_reg, treg,
+					      offsetof(struct bpf_sockopt_kern, tmp_reg));
+			*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct bpf_sockopt_kern, current_task),
+					      treg, si->dst_reg,
+					      offsetof(struct bpf_sockopt_kern, current_task));
+			*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct task_struct, bpf_ctx),
+					      treg, treg,
+					      offsetof(struct task_struct, bpf_ctx));
+			*insn++ = BPF_STX_MEM(BPF_FIELD_SIZEOF(struct bpf_cg_run_ctx, retval),
+					      treg, si->src_reg,
+					      offsetof(struct bpf_cg_run_ctx, retval));
+			*insn++ = BPF_LDX_MEM(BPF_DW, treg, si->dst_reg,
+					      offsetof(struct bpf_sockopt_kern, tmp_reg));
+		} else {
+			*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct bpf_sockopt_kern, current_task),
+					      si->dst_reg, si->src_reg,
+					      offsetof(struct bpf_sockopt_kern, current_task));
+			*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct task_struct, bpf_ctx),
+					      si->dst_reg, si->dst_reg,
+					      offsetof(struct task_struct, bpf_ctx));
+			*insn++ = BPF_LDX_MEM(BPF_FIELD_SIZEOF(struct bpf_cg_run_ctx, retval),
+					      si->dst_reg, si->dst_reg,
+					      offsetof(struct bpf_cg_run_ctx, retval));
+		}
 		break;
 	case offsetof(struct bpf_sockopt, optval):
 		*insn++ = CG_SOCKOPT_ACCESS_FIELD(BPF_LDX_MEM, optval);