1 files changed, 88 insertions, 81 deletions

diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index 7370f763346f..1f60883d78c6 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -110,11 +110,13 @@
 #define SQE_VALID_FLAGS	(SQE_COMMON_FLAGS | IOSQE_BUFFER_SELECT | \
 			IOSQE_IO_DRAIN | IOSQE_CQE_SKIP_SUCCESS)
 
+#define IO_REQ_LINK_FLAGS (REQ_F_LINK | REQ_F_HARDLINK)
+
 #define IO_REQ_CLEAN_FLAGS (REQ_F_BUFFER_SELECTED | REQ_F_NEED_CLEANUP | \
 				REQ_F_POLLED | REQ_F_INFLIGHT | REQ_F_CREDS | \
 				REQ_F_ASYNC_DATA)
 
-#define IO_REQ_CLEAN_SLOW_FLAGS (REQ_F_REFCOUNT | REQ_F_LINK | REQ_F_HARDLINK |\
+#define IO_REQ_CLEAN_SLOW_FLAGS (REQ_F_REFCOUNT | IO_REQ_LINK_FLAGS | \
 				 REQ_F_REISSUE | IO_REQ_CLEAN_FLAGS)
 
 #define IO_TCTX_REFS_CACHE_NR	(1U << 10)
@@ -131,7 +133,6 @@ struct io_defer_entry {
 
 /* requests with any of those set should undergo io_disarm_next() */
 #define IO_DISARM_MASK (REQ_F_ARM_LTIMEOUT | REQ_F_LINK_TIMEOUT | REQ_F_FAIL)
-#define IO_REQ_LINK_FLAGS (REQ_F_LINK | REQ_F_HARDLINK)
 
 /*
  * No waiters. It's larger than any valid value of the tw counter
@@ -443,24 +444,6 @@ static struct io_kiocb *__io_prep_linked_timeout(struct io_kiocb *req)
 	return req->link;
 }
 
-static inline struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req)
-{
-	if (likely(!(req->flags & REQ_F_ARM_LTIMEOUT)))
-		return NULL;
-	return __io_prep_linked_timeout(req);
-}
-
-static noinline void __io_arm_ltimeout(struct io_kiocb *req)
-{
-	io_queue_linked_timeout(__io_prep_linked_timeout(req));
-}
-
-static inline void io_arm_ltimeout(struct io_kiocb *req)
-{
-	if (unlikely(req->flags & REQ_F_ARM_LTIMEOUT))
-		__io_arm_ltimeout(req);
-}
-
 static void io_prep_async_work(struct io_kiocb *req)
 {
 	const struct io_issue_def *def = &io_issue_defs[req->opcode];
@@ -513,7 +496,6 @@ static void io_prep_async_link(struct io_kiocb *req)
 
 static void io_queue_iowq(struct io_kiocb *req)
 {
-	struct io_kiocb *link = io_prep_linked_timeout(req);
 	struct io_uring_task *tctx = req->tctx;
 
 	BUG_ON(!tctx);
@@ -538,8 +520,6 @@ static void io_queue_iowq(struct io_kiocb *req)
 
 	trace_io_uring_queue_async_work(req, io_wq_is_hashed(&req->work));
 	io_wq_enqueue(tctx->io_wq, &req->work);
-	if (link)
-		io_queue_linked_timeout(link);
 }
 
 static void io_req_queue_iowq_tw(struct io_kiocb *req, struct io_tw_state *ts)
@@ -652,6 +632,7 @@ static void __io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool dying)
 		 * to care for a non-real case.
 		 */
 		if (need_resched()) {
+			ctx->cqe_sentinel = ctx->cqe_cached;
 			io_cq_unlock_post(ctx);
 			mutex_unlock(&ctx->uring_lock);
 			cond_resched();
@@ -874,13 +855,26 @@ bool io_req_post_cqe(struct io_kiocb *req, s32 res, u32 cflags)
 	struct io_ring_ctx *ctx = req->ctx;
 	bool posted;
 
+	/*
+	 * If multishot has already posted deferred completions, ensure that
+	 * those are flushed first before posting this one. If not, CQEs
+	 * could get reordered.
+	 */
+	if (!wq_list_empty(&ctx->submit_state.compl_reqs))
+		__io_submit_flush_completions(ctx);
+
 	lockdep_assert(!io_wq_current_is_worker());
 	lockdep_assert_held(&ctx->uring_lock);
 
-	__io_cq_lock(ctx);
-	posted = io_fill_cqe_aux(ctx, req->cqe.user_data, res, cflags);
+	if (!ctx->lockless_cq) {
+		spin_lock(&ctx->completion_lock);
+		posted = io_fill_cqe_aux(ctx, req->cqe.user_data, res, cflags);
+		spin_unlock(&ctx->completion_lock);
+	} else {
+		posted = io_fill_cqe_aux(ctx, req->cqe.user_data, res, cflags);
+	}
+
 	ctx->submit_state.cq_flush = true;
-	__io_cq_unlock_post(ctx);
 	return posted;
 }
 
@@ -1080,21 +1074,22 @@ static __cold void __io_fallback_tw(struct llist_node *node, bool sync)
 	while (node) {
 		req = container_of(node, struct io_kiocb, io_task_work.node);
 		node = node->next;
-		if (sync && last_ctx != req->ctx) {
+		if (last_ctx != req->ctx) {
 			if (last_ctx) {
-				flush_delayed_work(&last_ctx->fallback_work);
+				if (sync)
+					flush_delayed_work(&last_ctx->fallback_work);
 				percpu_ref_put(&last_ctx->refs);
 			}
 			last_ctx = req->ctx;
 			percpu_ref_get(&last_ctx->refs);
 		}
-		if (llist_add(&req->io_task_work.node,
-			      &req->ctx->fallback_llist))
-			schedule_delayed_work(&req->ctx->fallback_work, 1);
+		if (llist_add(&req->io_task_work.node, &last_ctx->fallback_llist))
+			schedule_delayed_work(&last_ctx->fallback_work, 1);
 	}
 
 	if (last_ctx) {
-		flush_delayed_work(&last_ctx->fallback_work);
+		if (sync)
+			flush_delayed_work(&last_ctx->fallback_work);
 		percpu_ref_put(&last_ctx->refs);
 	}
 }
@@ -1157,7 +1152,7 @@ static inline void io_req_local_work_add(struct io_kiocb *req,
 	 * We don't know how many reuqests is there in the link and whether
 	 * they can even be queued lazily, fall back to non-lazy.
 	 */
-	if (req->flags & (REQ_F_LINK | REQ_F_HARDLINK))
+	if (req->flags & IO_REQ_LINK_FLAGS)
 		flags &= ~IOU_F_TWQ_LAZY_WAKE;
 
 	guard(rcu)();
@@ -1719,17 +1714,24 @@ static bool io_assign_file(struct io_kiocb *req, const struct io_issue_def *def,
 	return !!req->file;
 }
 
+#define REQ_ISSUE_SLOW_FLAGS	(REQ_F_CREDS | REQ_F_ARM_LTIMEOUT)
+
 static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
 {
 	const struct io_issue_def *def = &io_issue_defs[req->opcode];
 	const struct cred *creds = NULL;
+	struct io_kiocb *link = NULL;
 	int ret;
 
 	if (unlikely(!io_assign_file(req, def, issue_flags)))
 		return -EBADF;
 
-	if (unlikely((req->flags & REQ_F_CREDS) && req->creds != current_cred()))
-		creds = override_creds(req->creds);
+	if (unlikely(req->flags & REQ_ISSUE_SLOW_FLAGS)) {
+		if ((req->flags & REQ_F_CREDS) && req->creds != current_cred())
+			creds = override_creds(req->creds);
+		if (req->flags & REQ_F_ARM_LTIMEOUT)
+			link = __io_prep_linked_timeout(req);
+	}
 
 	if (!def->audit_skip)
 		audit_uring_entry(req->opcode);
@@ -1739,8 +1741,12 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
 	if (!def->audit_skip)
 		audit_uring_exit(!ret, ret);
 
-	if (creds)
-		revert_creds(creds);
+	if (unlikely(creds || link)) {
+		if (creds)
+			revert_creds(creds);
+		if (link)
+			io_queue_linked_timeout(link);
+	}
 
 	if (ret == IOU_OK) {
 		if (issue_flags & IO_URING_F_COMPLETE_DEFER)
@@ -1753,7 +1759,6 @@ static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags)
 
 	if (ret == IOU_ISSUE_SKIP_COMPLETE) {
 		ret = 0;
-		io_arm_ltimeout(req);
 
 		/* If the op doesn't have a file, we're not polling for it */
 		if ((req->ctx->flags & IORING_SETUP_IOPOLL) && def->iopoll_queue)
@@ -1774,7 +1779,7 @@ struct io_wq_work *io_wq_free_work(struct io_wq_work *work)
 	struct io_kiocb *req = container_of(work, struct io_kiocb, work);
 	struct io_kiocb *nxt = NULL;
 
-	if (req_ref_put_and_test(req)) {
+	if (req_ref_put_and_test_atomic(req)) {
 		if (req->flags & IO_REQ_LINK_FLAGS)
 			nxt = io_req_find_next(req);
 		io_free_req(req);
@@ -1796,8 +1801,6 @@ void io_wq_submit_work(struct io_wq_work *work)
 	else
 		req_ref_get(req);
 
-	io_arm_ltimeout(req);
-
 	/* either cancelled or io-wq is dying, so don't touch tctx->iowq */
 	if (atomic_read(&work->flags) & IO_WQ_WORK_CANCEL) {
 fail:
@@ -1913,15 +1916,11 @@ struct file *io_file_get_normal(struct io_kiocb *req, int fd)
 static void io_queue_async(struct io_kiocb *req, int ret)
 	__must_hold(&req->ctx->uring_lock)
 {
-	struct io_kiocb *linked_timeout;
-
 	if (ret != -EAGAIN || (req->flags & REQ_F_NOWAIT)) {
 		io_req_defer_failed(req, ret);
 		return;
 	}
 
-	linked_timeout = io_prep_linked_timeout(req);
-
 	switch (io_arm_poll_handler(req, 0)) {
 	case IO_APOLL_READY:
 		io_kbuf_recycle(req, 0);
@@ -1934,9 +1933,6 @@ static void io_queue_async(struct io_kiocb *req, int ret)
 	case IO_APOLL_OK:
 		break;
 	}
-
-	if (linked_timeout)
-		io_queue_linked_timeout(linked_timeout);
 }
 
 static inline void io_queue_sqe(struct io_kiocb *req)
@@ -3537,6 +3533,44 @@ static struct file *io_uring_get_file(struct io_ring_ctx *ctx)
 					 O_RDWR | O_CLOEXEC, NULL);
 }
 
+static int io_uring_sanitise_params(struct io_uring_params *p)
+{
+	unsigned flags = p->flags;
+
+	/* There is no way to mmap rings without a real fd */
+	if ((flags & IORING_SETUP_REGISTERED_FD_ONLY) &&
+	    !(flags & IORING_SETUP_NO_MMAP))
+		return -EINVAL;
+
+	if (flags & IORING_SETUP_SQPOLL) {
+		/* IPI related flags don't make sense with SQPOLL */
+		if (flags & (IORING_SETUP_COOP_TASKRUN |
+			     IORING_SETUP_TASKRUN_FLAG |
+			     IORING_SETUP_DEFER_TASKRUN))
+			return -EINVAL;
+	}
+
+	if (flags & IORING_SETUP_TASKRUN_FLAG) {
+		if (!(flags & (IORING_SETUP_COOP_TASKRUN |
+			       IORING_SETUP_DEFER_TASKRUN)))
+			return -EINVAL;
+	}
+
+	/* HYBRID_IOPOLL only valid with IOPOLL */
+	if ((flags & IORING_SETUP_HYBRID_IOPOLL) && !(flags & IORING_SETUP_IOPOLL))
+		return -EINVAL;
+
+	/*
+	 * For DEFER_TASKRUN we require the completion task to be the same as
+	 * the submission task. This implies that there is only one submitter.
+	 */
+	if ((flags & IORING_SETUP_DEFER_TASKRUN) &&
+	    !(flags & IORING_SETUP_SINGLE_ISSUER))
+		return -EINVAL;
+
+	return 0;
+}
+
 int io_uring_fill_params(unsigned entries, struct io_uring_params *p)
 {
 	if (!entries)
@@ -3547,10 +3581,6 @@ int io_uring_fill_params(unsigned entries, struct io_uring_params *p)
 		entries = IORING_MAX_ENTRIES;
 	}
 
-	if ((p->flags & IORING_SETUP_REGISTERED_FD_ONLY)
-	    && !(p->flags & IORING_SETUP_NO_MMAP))
-		return -EINVAL;
-
 	/*
 	 * Use twice as many entries for the CQ ring. It's possible for the
 	 * application to drive a higher depth than the size of the SQ ring,
@@ -3612,6 +3642,10 @@ static __cold int io_uring_create(unsigned entries, struct io_uring_params *p,
 	struct file *file;
 	int ret;
 
+	ret = io_uring_sanitise_params(p);
+	if (ret)
+		return ret;
+
 	ret = io_uring_fill_params(entries, p);
 	if (unlikely(ret))
 		return ret;
@@ -3659,37 +3693,10 @@ static __cold int io_uring_create(unsigned entries, struct io_uring_params *p,
 	 * For SQPOLL, we just need a wakeup, always. For !SQPOLL, if
 	 * COOP_TASKRUN is set, then IPIs are never needed by the app.
 	 */
-	ret = -EINVAL;
-	if (ctx->flags & IORING_SETUP_SQPOLL) {
-		/* IPI related flags don't make sense with SQPOLL */
-		if (ctx->flags & (IORING_SETUP_COOP_TASKRUN |
-				  IORING_SETUP_TASKRUN_FLAG |
-				  IORING_SETUP_DEFER_TASKRUN))
-			goto err;
+	if (ctx->flags & (IORING_SETUP_SQPOLL|IORING_SETUP_COOP_TASKRUN))
 		ctx->notify_method = TWA_SIGNAL_NO_IPI;
-	} else if (ctx->flags & IORING_SETUP_COOP_TASKRUN) {
-		ctx->notify_method = TWA_SIGNAL_NO_IPI;
-	} else {
-		if (ctx->flags & IORING_SETUP_TASKRUN_FLAG &&
-		    !(ctx->flags & IORING_SETUP_DEFER_TASKRUN))
-			goto err;
+	else
 		ctx->notify_method = TWA_SIGNAL;
-	}
-
-	/* HYBRID_IOPOLL only valid with IOPOLL */
-	if ((ctx->flags & (IORING_SETUP_IOPOLL|IORING_SETUP_HYBRID_IOPOLL)) ==
-			IORING_SETUP_HYBRID_IOPOLL)
-		goto err;
-
-	/*
-	 * For DEFER_TASKRUN we require the completion task to be the same as the
-	 * submission task. This implies that there is only one submitter, so enforce
-	 * that.
-	 */
-	if (ctx->flags & IORING_SETUP_DEFER_TASKRUN &&
-	    !(ctx->flags & IORING_SETUP_SINGLE_ISSUER)) {
-		goto err;
-	}
 
 	/*
 	 * This is just grabbed for accounting purposes. When a process exits,