bpf: Move getsockopt retval to struct bpf_cg_run_ctx

The retval value is moved to struct bpf_cg_run_ctx for ease of access
in different prog types with different context structs layouts. The
helper implementation (to be added in a later patch in the series) can
simply perform a container_of from current->bpf_ctx to retrieve
bpf_cg_run_ctx.

Unfortunately, there is no easy way to access the current task_struct
via the verifier BPF bytecode rewrite, aside from possibly calling a
helper, so a pointer to current task is added to struct bpf_sockopt_kern
so that the rewritten BPF bytecode can access struct bpf_cg_run_ctx with
an indirection.

For backward compatibility, if a getsockopt program rejects a syscall
by returning 0, an -EPERM will be generated, by having the
BPF_PROG_RUN_ARRAY_CG family macros automatically set the retval to
-EPERM. Unlike prior to this patch, this -EPERM will be visible to
ctx->retval for any other hooks down the line in the prog array.

Additionally, the restriction that getsockopt filters can only set
the retval to 0 is removed, considering that certain getsockopt
implementations may return optlen. Filters are now able to set the
value arbitrarily.

Signed-off-by: YiFei Zhu <zhuyifei@google.com>
Reviewed-by: Stanislav Fomichev <sdf@google.com>
Link: https://lore.kernel.org/r/73b0325f5c29912ccea7ea57ec1ed4d388fc1d37.1639619851.git.zhuyifei@google.com
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

2 files changed, 15 insertions, 10 deletions

diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 83da1764fcfe..7b0c11f414d0 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -1245,6 +1245,7 @@ struct bpf_run_ctx {};
 struct bpf_cg_run_ctx {
 	struct bpf_run_ctx run_ctx;
 	const struct bpf_prog_array_item *prog_item;
+	int retval;
 };
 
 struct bpf_trace_run_ctx {
@@ -1280,16 +1281,16 @@ typedef u32 (*bpf_prog_run_fn)(const struct bpf_prog *prog, const void *ctx);
 static __always_inline int
 BPF_PROG_RUN_ARRAY_CG_FLAGS(const struct bpf_prog_array __rcu *array_rcu,
 			    const void *ctx, bpf_prog_run_fn run_prog,
-			    u32 *ret_flags)
+			    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;
-	int ret = 0;
 	u32 func_ret;
 
+	run_ctx.retval = retval;
 	migrate_disable();
 	rcu_read_lock();
 	array = rcu_dereference(array_rcu);
@@ -1299,27 +1300,28 @@ BPF_PROG_RUN_ARRAY_CG_FLAGS(const struct bpf_prog_array __rcu *array_rcu,
 		run_ctx.prog_item = item;
 		func_ret = run_prog(prog, ctx);
 		if (!(func_ret & 1))
-			ret = -EPERM;
+			run_ctx.retval = -EPERM;
 		*(ret_flags) |= (func_ret >> 1);
 		item++;
 	}
 	bpf_reset_run_ctx(old_run_ctx);
 	rcu_read_unlock();
 	migrate_enable();
-	return ret;
+	return run_ctx.retval;
 }
 
 static __always_inline int
 BPF_PROG_RUN_ARRAY_CG(const struct bpf_prog_array __rcu *array_rcu,
-		      const void *ctx, bpf_prog_run_fn run_prog)
+		      const void *ctx, bpf_prog_run_fn run_prog,
+		      int retval)
 {
 	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;
-	int ret = 0;
 
+	run_ctx.retval = retval;
 	migrate_disable();
 	rcu_read_lock();
 	array = rcu_dereference(array_rcu);
@@ -1328,13 +1330,13 @@ BPF_PROG_RUN_ARRAY_CG(const struct bpf_prog_array __rcu *array_rcu,
 	while ((prog = READ_ONCE(item->prog))) {
 		run_ctx.prog_item = item;
 		if (!run_prog(prog, ctx))
-			ret = -EPERM;
+			run_ctx.retval = -EPERM;
 		item++;
 	}
 	bpf_reset_run_ctx(old_run_ctx);
 	rcu_read_unlock();
 	migrate_enable();
-	return ret;
+	return run_ctx.retval;
 }
 
 static __always_inline u32
@@ -1394,7 +1396,7 @@ out:
 		u32 _flags = 0;				\
 		bool _cn;				\
 		u32 _ret;				\
-		_ret = BPF_PROG_RUN_ARRAY_CG_FLAGS(array, ctx, func, &_flags); \
+		_ret = BPF_PROG_RUN_ARRAY_CG_FLAGS(array, ctx, func, 0, &_flags); \
 		_cn = _flags & BPF_RET_SET_CN;		\
 		if (!_ret)				\
 			_ret = (_cn ? NET_XMIT_CN : NET_XMIT_SUCCESS);	\
diff --git a/include/linux/filter.h b/include/linux/filter.h
index 71fa57b88bfc..d23e999dc032 100644
--- a/include/linux/filter.h
+++ b/include/linux/filter.h
@@ -1356,7 +1356,10 @@ struct bpf_sockopt_kern {
 	s32		level;
 	s32		optname;
 	s32		optlen;
-	s32		retval;
+	/* for retval in struct bpf_cg_run_ctx */
+	struct task_struct *current_task;
+	/* Temporary "register" for indirect stores to ppos. */
+	u64		tmp_reg;
 };
 
 int copy_bpf_fprog_from_user(struct sock_fprog *dst, sockptr_t src, int len);