Merge tag 'io_uring-6.2-2023-01-13' of git://git.kernel.dk/linux

Pull io_uring fixes from Jens Axboe:
 "A fix for a regression that happened last week, rest is fixes that
  will be headed to stable as well. In detail:

   - Fix for a regression added with the leak fix from last week (me)

   - In writing a test case for that leak, inadvertently discovered a
     case where we a poll request can race. So fix that up and mark it
     for stable, and also ensure that fdinfo covers both the poll tables
     that we have. The latter was an oversight when the split poll table
     were added (me)

   - Fix for a lockdep reported issue with IOPOLL (Pavel)"

* tag 'io_uring-6.2-2023-01-13' of git://git.kernel.dk/linux:
  io_uring: lock overflowing for IOPOLL
  io_uring/poll: attempt request issue after racy poll wakeup
  io_uring/fdinfo: include locked hash table in fdinfo output
  io_uring/poll: add hash if ready poll request can't complete inline
  io_uring/io-wq: only free worker if it was allocated for creation

io_uring/fdinfo.c

@@ -170,12 +170,11 @@ static __cold void __io_uring_show_fdinfo(struct io_ring_ctx *ctx,
 		xa_for_each(&ctx->personalities, index, cred)
 			io_uring_show_cred(m, index, cred);
 	}
-	if (has_lock)
-		mutex_unlock(&ctx->uring_lock);
 
 	seq_puts(m, "PollList:\n");
 	for (i = 0; i < (1U << ctx->cancel_table.hash_bits); i++) {
 		struct io_hash_bucket *hb = &ctx->cancel_table.hbs[i];
+		struct io_hash_bucket *hbl = &ctx->cancel_table_locked.hbs[i];
 		struct io_kiocb *req;
 
 		spin_lock(&hb->lock);
@@ -183,8 +182,17 @@ static __cold void __io_uring_show_fdinfo(struct io_ring_ctx *ctx,
 			seq_printf(m, "  op=%d, task_works=%d\n", req->opcode,
 					task_work_pending(req->task));
 		spin_unlock(&hb->lock);
+
+		if (!has_lock)
+			continue;
+		hlist_for_each_entry(req, &hbl->list, hash_node)
+			seq_printf(m, "  op=%d, task_works=%d\n", req->opcode,
+					task_work_pending(req->task));
 	}
 
+	if (has_lock)
+		mutex_unlock(&ctx->uring_lock);
+
 	seq_puts(m, "CqOverflowList:\n");
 	spin_lock(&ctx->completion_lock);
 	list_for_each_entry(ocqe, &ctx->cq_overflow_list, list) {

io_uring/io-wq.c

@@ -1230,7 +1230,12 @@ static void io_wq_cancel_tw_create(struct io_wq *wq)
 
 		worker = container_of(cb, struct io_worker, create_work);
 		io_worker_cancel_cb(worker);
-		kfree(worker);
+		/*
+		 * Only the worker continuation helper has worker allocated and
+		 * hence needs freeing.
+		 */
+		if (cb->func == create_worker_cont)
+			kfree(worker);
 	}
 }
 

io_uring/poll.c

@@ -223,21 +223,22 @@ enum {
 	IOU_POLL_DONE = 0,
 	IOU_POLL_NO_ACTION = 1,
 	IOU_POLL_REMOVE_POLL_USE_RES = 2,
+	IOU_POLL_REISSUE = 3,
 };
 
 /*
  * All poll tw should go through this. Checks for poll events, manages
  * references, does rewait, etc.
  *
- * Returns a negative error on failure. IOU_POLL_NO_ACTION when no action require,
- * which is either spurious wakeup or multishot CQE is served.
- * IOU_POLL_DONE when it's done with the request, then the mask is stored in req->cqe.res.
- * IOU_POLL_REMOVE_POLL_USE_RES indicates to remove multishot poll and that the result
- * is stored in req->cqe.
+ * Returns a negative error on failure. IOU_POLL_NO_ACTION when no action
+ * require, which is either spurious wakeup or multishot CQE is served.
+ * IOU_POLL_DONE when it's done with the request, then the mask is stored in
+ * req->cqe.res. IOU_POLL_REMOVE_POLL_USE_RES indicates to remove multishot
+ * poll and that the result is stored in req->cqe.
  */
 static int io_poll_check_events(struct io_kiocb *req, bool *locked)
 {
-	int v, ret;
+	int v;
 
 	/* req->task == current here, checking PF_EXITING is safe */
 	if (unlikely(req->task->flags & PF_EXITING))
@@ -276,10 +277,15 @@ static int io_poll_check_events(struct io_kiocb *req, bool *locked)
 		if (!req->cqe.res) {
 			struct poll_table_struct pt = { ._key = req->apoll_events };
 			req->cqe.res = vfs_poll(req->file, &pt) & req->apoll_events;
+			/*
+			 * We got woken with a mask, but someone else got to
+			 * it first. The above vfs_poll() doesn't add us back
+			 * to the waitqueue, so if we get nothing back, we
+			 * should be safe and attempt a reissue.
+			 */
+			if (unlikely(!req->cqe.res))
+				return IOU_POLL_REISSUE;
 		}
-
-		if ((unlikely(!req->cqe.res)))
-			continue;
 		if (req->apoll_events & EPOLLONESHOT)
 			return IOU_POLL_DONE;
 
@@ -294,7 +300,7 @@ static int io_poll_check_events(struct io_kiocb *req, bool *locked)
 				return IOU_POLL_REMOVE_POLL_USE_RES;
 			}
 		} else {
-			ret = io_poll_issue(req, locked);
+			int ret = io_poll_issue(req, locked);
 			if (ret == IOU_STOP_MULTISHOT)
 				return IOU_POLL_REMOVE_POLL_USE_RES;
 			if (ret < 0)
@@ -330,6 +336,9 @@ static void io_poll_task_func(struct io_kiocb *req, bool *locked)
 
 			poll = io_kiocb_to_cmd(req, struct io_poll);
 			req->cqe.res = mangle_poll(req->cqe.res & poll->events);
+		} else if (ret == IOU_POLL_REISSUE) {
+			io_req_task_submit(req, locked);
+			return;
 		} else if (ret != IOU_POLL_REMOVE_POLL_USE_RES) {
 			req->cqe.res = ret;
 			req_set_fail(req);
@@ -342,7 +351,7 @@ static void io_poll_task_func(struct io_kiocb *req, bool *locked)
 
 		if (ret == IOU_POLL_REMOVE_POLL_USE_RES)
 			io_req_task_complete(req, locked);
-		else if (ret == IOU_POLL_DONE)
+		else if (ret == IOU_POLL_DONE || ret == IOU_POLL_REISSUE)
 			io_req_task_submit(req, locked);
 		else
 			io_req_defer_failed(req, ret);
@@ -533,6 +542,14 @@ static bool io_poll_can_finish_inline(struct io_kiocb *req,
 	return pt->owning || io_poll_get_ownership(req);
 }
 
+static void io_poll_add_hash(struct io_kiocb *req)
+{
+	if (req->flags & REQ_F_HASH_LOCKED)
+		io_poll_req_insert_locked(req);
+	else
+		io_poll_req_insert(req);
+}
+
 /*
  * Returns 0 when it's handed over for polling. The caller owns the requests if
  * it returns non-zero, but otherwise should not touch it. Negative values
@@ -591,18 +608,17 @@ static int __io_arm_poll_handler(struct io_kiocb *req,
 
 	if (mask &&
 	   ((poll->events & (EPOLLET|EPOLLONESHOT)) == (EPOLLET|EPOLLONESHOT))) {
-		if (!io_poll_can_finish_inline(req, ipt))
+		if (!io_poll_can_finish_inline(req, ipt)) {
+			io_poll_add_hash(req);
 			return 0;
+		}
 		io_poll_remove_entries(req);
 		ipt->result_mask = mask;
 		/* no one else has access to the req, forget about the ref */
 		return 1;
 	}
 
-	if (req->flags & REQ_F_HASH_LOCKED)
-		io_poll_req_insert_locked(req);
-	else
-		io_poll_req_insert(req);
+	io_poll_add_hash(req);
 
 	if (mask && (poll->events & EPOLLET) &&
 	    io_poll_can_finish_inline(req, ipt)) {

io_uring/rw.c

@@ -1062,7 +1062,11 @@ int io_do_iopoll(struct io_ring_ctx *ctx, bool force_nonspin)
 			continue;
 
 		req->cqe.flags = io_put_kbuf(req, 0);
-		io_fill_cqe_req(req->ctx, req);
+		if (unlikely(!__io_fill_cqe_req(ctx, req))) {
+			spin_lock(&ctx->completion_lock);
+			io_req_cqe_overflow(req);
+			spin_unlock(&ctx->completion_lock);
+		}
 	}
 
 	if (unlikely(!nr_events))