io_uring: fix iovec leaks

to #26323588

commit 99bc4c38537d774e667d043c520914082da19abf upstream.

Allocated iovec is freed only in io_{read,write,send,recv)(), and just
leaves it if an error occured. There are plenty of such cases:
- cancellation of non-head requests
- fail grabbing files in __io_queue_sqe()
- set REQ_F_NOWAIT and returning in __io_queue_sqe()

Add REQ_F_NEED_CLEANUP, which will force such requests with custom
allocated resourses go through cleanup handlers on put.

Cc: stable@vger.kernel.org # 5.5
Signed-off-by: NPavel Begunkov <asml.silence@gmail.com>
Signed-off-by: NJens Axboe <axboe@kernel.dk>
Signed-off-by: NJoseph Qi <joseph.qi@linux.alibaba.com>
Acked-by: NXiaoguang Wang <xiaoguang.wang@linux.alibaba.com>

fs/io_uring.c

@@ -478,6 +478,7 @@ enum {
 	REQ_F_MUST_PUNT_BIT,
 	REQ_F_TIMEOUT_NOSEQ_BIT,
 	REQ_F_COMP_LOCKED_BIT,
+	REQ_F_NEED_CLEANUP_BIT,
 };
 
 enum {
@@ -516,6 +517,8 @@ enum {
 	REQ_F_TIMEOUT_NOSEQ	= BIT(REQ_F_TIMEOUT_NOSEQ_BIT),
 	/* completion under lock */
 	REQ_F_COMP_LOCKED	= BIT(REQ_F_COMP_LOCKED_BIT),
+	/* needs cleanup */
+	REQ_F_NEED_CLEANUP	= BIT(REQ_F_NEED_CLEANUP_BIT),
 };
 
 /*
@@ -748,6 +751,7 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx,
 				 unsigned nr_args);
 static int io_grab_files(struct io_kiocb *req);
 static void io_ring_file_ref_flush(struct fixed_file_data *data);
+static void io_cleanup_req(struct io_kiocb *req);
 
 static struct kmem_cache *req_cachep;
 
@@ -1235,6 +1239,9 @@ static void __io_free_req(struct io_kiocb *req)
 {
 	__io_req_aux_free(req);
 
+	if (req->flags & REQ_F_NEED_CLEANUP)
+		io_cleanup_req(req);
+
 	if (req->flags & REQ_F_INFLIGHT) {
 		struct io_ring_ctx *ctx = req->ctx;
 		unsigned long flags;
@@ -2125,6 +2132,8 @@ static void io_req_map_rw(struct io_kiocb *req, ssize_t io_size,
 		req->io->rw.iov = req->io->rw.fast_iov;
 		memcpy(req->io->rw.iov, fast_iov,
 			sizeof(struct iovec) * iter->nr_segs);
+	} else {
+		req->flags |= REQ_F_NEED_CLEANUP;
 	}
 }
 
@@ -2235,6 +2244,7 @@ static int io_read(struct io_kiocb *req, struct io_kiocb **nxt,
 	}
 out_free:
 	kfree(iovec);
+	req->flags &= ~REQ_F_NEED_CLEANUP;
 	return ret;
 }
 
@@ -2339,6 +2349,7 @@ static int io_write(struct io_kiocb *req, struct io_kiocb **nxt,
 		}
 	}
 out_free:
+	req->flags &= ~REQ_F_NEED_CLEANUP;
 	kfree(iovec);
 	return ret;
 }
@@ -2945,6 +2956,7 @@ static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 #if defined(CONFIG_NET)
 	struct io_sr_msg *sr = &req->sr_msg;
 	struct io_async_ctx *io = req->io;
+	int ret;
 
 	sr->msg_flags = READ_ONCE(sqe->msg_flags);
 	sr->msg = u64_to_user_ptr(READ_ONCE(sqe->addr));
@@ -2954,8 +2966,11 @@ static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 		return 0;
 
 	io->msg.iov = io->msg.fast_iov;
-	return sendmsg_copy_msghdr(&io->msg.msg, sr->msg, sr->msg_flags,
+	ret = sendmsg_copy_msghdr(&io->msg.msg, sr->msg, sr->msg_flags,
 					&io->msg.iov);
+	if (!ret)
+		req->flags |= REQ_F_NEED_CLEANUP;
+	return ret;
 #else
 	return -EOPNOTSUPP;
 #endif
@@ -3013,6 +3028,7 @@ static int io_sendmsg(struct io_kiocb *req, struct io_kiocb **nxt,
 					kfree(kmsg->iov);
 				return -ENOMEM;
 			}
+			req->flags |= REQ_F_NEED_CLEANUP;
 			memcpy(&req->io->msg, &io.msg, sizeof(io.msg));
 			return -EAGAIN;
 		}
@@ -3022,6 +3038,7 @@ static int io_sendmsg(struct io_kiocb *req, struct io_kiocb **nxt,
 
 	if (kmsg && kmsg->iov != kmsg->fast_iov)
 		kfree(kmsg->iov);
+	req->flags &= ~REQ_F_NEED_CLEANUP;
 	io_cqring_add_event(req, ret);
 	if (ret < 0)
 		req_set_fail_links(req);
@@ -3089,6 +3106,7 @@ static int io_recvmsg_prep(struct io_kiocb *req,
 #if defined(CONFIG_NET)
 	struct io_sr_msg *sr = &req->sr_msg;
 	struct io_async_ctx *io = req->io;
+	int ret;
 
 	sr->msg_flags = READ_ONCE(sqe->msg_flags);
 	sr->msg = u64_to_user_ptr(READ_ONCE(sqe->addr));
@@ -3098,8 +3116,11 @@ static int io_recvmsg_prep(struct io_kiocb *req,
 		return 0;
 
 	io->msg.iov = io->msg.fast_iov;
-	return recvmsg_copy_msghdr(&io->msg.msg, sr->msg, sr->msg_flags,
+	ret = recvmsg_copy_msghdr(&io->msg.msg, sr->msg, sr->msg_flags,
 					&io->msg.uaddr, &io->msg.iov);
+	if (!ret)
+		req->flags |= REQ_F_NEED_CLEANUP;
+	return ret;
 #else
 	return -EOPNOTSUPP;
 #endif
@@ -3160,6 +3181,7 @@ static int io_recvmsg(struct io_kiocb *req, struct io_kiocb **nxt,
 				return -ENOMEM;
 			}
 			memcpy(&req->io->msg, &io.msg, sizeof(io.msg));
+			req->flags |= REQ_F_NEED_CLEANUP;
 			return -EAGAIN;
 		}
 		if (ret == -ERESTARTSYS)
@@ -3168,6 +3190,7 @@ static int io_recvmsg(struct io_kiocb *req, struct io_kiocb **nxt,
 
 	if (kmsg && kmsg->iov != kmsg->fast_iov)
 		kfree(kmsg->iov);
+	req->flags &= ~REQ_F_NEED_CLEANUP;
 	io_cqring_add_event(req, ret);
 	if (ret < 0)
 		req_set_fail_links(req);
@@ -4178,6 +4201,30 @@ static int io_req_defer(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	return -EIOCBQUEUED;
 }
 
+static void io_cleanup_req(struct io_kiocb *req)
+{
+	struct io_async_ctx *io = req->io;
+
+	switch (req->opcode) {
+	case IORING_OP_READV:
+	case IORING_OP_READ_FIXED:
+	case IORING_OP_READ:
+	case IORING_OP_WRITEV:
+	case IORING_OP_WRITE_FIXED:
+	case IORING_OP_WRITE:
+		if (io->rw.iov != io->rw.fast_iov)
+			kfree(io->rw.iov);
+		break;
+	case IORING_OP_SENDMSG:
+	case IORING_OP_RECVMSG:
+		if (io->msg.iov != io->msg.fast_iov)
+			kfree(io->msg.iov);
+		break;
+	}
+
+	req->flags &= ~REQ_F_NEED_CLEANUP;
+}
+
 static int io_issue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 			struct io_kiocb **nxt, bool force_nonblock)
 {