io_uring: simplify big_cqe handling

Don't keep big_cqe bits of req in a union with hash_node, find a
separate space for it. It's bit safer, but also if we keep it always
initialised, we can get rid of ugly REQ_F_CQE32_INIT handling.

Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Link: https://lore.kernel.org/r/447aa1b2968978c99e655ba88db536e903df0fe9.1692916914.git.asml.silence@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>

include/linux/io_uring_types.h

@@ -409,7 +409,6 @@ enum {
 	REQ_F_SINGLE_POLL_BIT,
 	REQ_F_DOUBLE_POLL_BIT,
 	REQ_F_PARTIAL_IO_BIT,
-	REQ_F_CQE32_INIT_BIT,
 	REQ_F_APOLL_MULTISHOT_BIT,
 	REQ_F_CLEAR_POLLIN_BIT,
 	REQ_F_HASH_LOCKED_BIT,
@@ -479,8 +478,6 @@ enum {
 	REQ_F_PARTIAL_IO	= BIT(REQ_F_PARTIAL_IO_BIT),
 	/* fast poll multishot mode */
 	REQ_F_APOLL_MULTISHOT	= BIT(REQ_F_APOLL_MULTISHOT_BIT),
-	/* ->extra1 and ->extra2 are initialised */
-	REQ_F_CQE32_INIT	= BIT(REQ_F_CQE32_INIT_BIT),
 	/* recvmsg special flag, clear EPOLLIN */
 	REQ_F_CLEAR_POLLIN	= BIT(REQ_F_CLEAR_POLLIN_BIT),
 	/* hashed into ->cancel_hash_locked, protected by ->uring_lock */
@@ -579,13 +576,7 @@ struct io_kiocb {
 	struct io_task_work		io_task_work;
 	unsigned			nr_tw;
 	/* for polled requests, i.e. IORING_OP_POLL_ADD and async armed poll */
-	union {
-		struct hlist_node	hash_node;
-		struct {
-			u64		extra1;
-			u64		extra2;
-		};
-	};
+	struct hlist_node		hash_node;
 	/* internal polling, see IORING_FEAT_FAST_POLL */
 	struct async_poll		*apoll;
 	/* opcode allocated if it needs to store data for async defer */
@@ -595,6 +586,11 @@ struct io_kiocb {
 	/* custom credentials, valid IFF REQ_F_CREDS is set */
 	const struct cred		*creds;
 	struct io_wq_work		work;
+
+	struct {
+		u64			extra1;
+		u64			extra2;
+	} big_cqe;
 };
 
 struct io_overflow_cqe {

io_uring/io_uring.c

@@ -807,13 +807,10 @@ static bool io_cqring_event_overflow(struct io_ring_ctx *ctx, u64 user_data,
 
 void io_req_cqe_overflow(struct io_kiocb *req)
 {
-	if (!(req->flags & REQ_F_CQE32_INIT)) {
-		req->extra1 = 0;
-		req->extra2 = 0;
-	}
 	io_cqring_event_overflow(req->ctx, req->cqe.user_data,
 				req->cqe.res, req->cqe.flags,
-				req->extra1, req->extra2);
+				req->big_cqe.extra1, req->big_cqe.extra2);
+	memset(&req->big_cqe, 0, sizeof(req->big_cqe));
 }
 
 /*
@@ -1057,6 +1054,7 @@ static void io_preinit_req(struct io_kiocb *req, struct io_ring_ctx *ctx)
 	req->async_data = NULL;
 	/* not necessary, but safer to zero */
 	memset(&req->cqe, 0, sizeof(req->cqe));
+	memset(&req->big_cqe, 0, sizeof(req->big_cqe));
 }
 
 static void io_flush_cached_locked_reqs(struct io_ring_ctx *ctx,

io_uring/io_uring.h

@@ -148,21 +148,12 @@ static inline bool io_fill_cqe_req(struct io_ring_ctx *ctx, struct io_kiocb *req
 	if (trace_io_uring_complete_enabled())
 		trace_io_uring_complete(req->ctx, req, req->cqe.user_data,
 					req->cqe.res, req->cqe.flags,
-					(req->flags & REQ_F_CQE32_INIT) ? req->extra1 : 0,
-					(req->flags & REQ_F_CQE32_INIT) ? req->extra2 : 0);
+					req->big_cqe.extra1, req->big_cqe.extra2);
 
 	memcpy(cqe, &req->cqe, sizeof(*cqe));
-
 	if (ctx->flags & IORING_SETUP_CQE32) {
-		u64 extra1 = 0, extra2 = 0;
-
-		if (req->flags & REQ_F_CQE32_INIT) {
-			extra1 = req->extra1;
-			extra2 = req->extra2;
-		}
-
-		WRITE_ONCE(cqe->big_cqe[0], extra1);
-		WRITE_ONCE(cqe->big_cqe[1], extra2);
+		memcpy(cqe->big_cqe, &req->big_cqe, sizeof(*cqe));
+		memset(&req->big_cqe, 0, sizeof(req->big_cqe));
 	}
 	return true;
 }

io_uring/uring_cmd.c

@@ -43,9 +43,8 @@ EXPORT_SYMBOL_GPL(io_uring_cmd_do_in_task_lazy);
 static inline void io_req_set_cqe32_extra(struct io_kiocb *req,
 					  u64 extra1, u64 extra2)
 {
-	req->extra1 = extra1;
-	req->extra2 = extra2;
-	req->flags |= REQ_F_CQE32_INIT;
+	req->big_cqe.extra1 = extra1;
+	req->big_cqe.extra2 = extra2;
 }
 
 /*