Merge tag 'io_uring-5.11-2021-01-24' of git://git.kernel.dk/linux-block

Pull io_uring fixes from Jens Axboe:
 "Still need a final cancelation fix that isn't quite done done,
  expected in the next day or two. That said, this contains:

   - Wakeup fix for IOPOLL requests

   - SQPOLL split close op handling fix

   - Ensure that any use of io_uring fd itself is marked as inflight

   - Short non-regular file read fix (Pavel)

   - Fix up bad false positive warning (Pavel)

   - SQPOLL fixes (Pavel)

   - In-flight removal fix (Pavel)"

* tag 'io_uring-5.11-2021-01-24' of git://git.kernel.dk/linux-block:
  io_uring: account io_uring internal files as REQ_F_INFLIGHT
  io_uring: fix sleeping under spin in __io_clean_op
  io_uring: fix short read retries for non-reg files
  io_uring: fix SQPOLL IORING_OP_CLOSE cancelation state
  io_uring: fix skipping disabling sqo on exec
  io_uring: fix uring_flush in exit_files() warning
  io_uring: fix false positive sqo warning on flush
  io_uring: iopoll requests should also wake task ->in_idle state

fs/io_uring.c

@@ -1025,6 +1025,7 @@ static ssize_t io_import_iovec(int rw, struct io_kiocb *req,
 static int io_setup_async_rw(struct io_kiocb *req, const struct iovec *iovec,
 			     const struct iovec *fast_iov,
 			     struct iov_iter *iter, bool force);
+static void io_req_drop_files(struct io_kiocb *req);
 
 static struct kmem_cache *req_cachep;
 
@@ -1048,8 +1049,7 @@ EXPORT_SYMBOL(io_uring_get_socket);
 
 static inline void io_clean_op(struct io_kiocb *req)
 {
-	if (req->flags & (REQ_F_NEED_CLEANUP | REQ_F_BUFFER_SELECTED |
-			  REQ_F_INFLIGHT))
+	if (req->flags & (REQ_F_NEED_CLEANUP | REQ_F_BUFFER_SELECTED))
 		__io_clean_op(req);
 }
 
@@ -1075,8 +1075,11 @@ static bool io_match_task(struct io_kiocb *head,
 		return true;
 
 	io_for_each_link(req, head) {
-		if ((req->flags & REQ_F_WORK_INITIALIZED) &&
-		    (req->work.flags & IO_WQ_WORK_FILES) &&
+		if (!(req->flags & REQ_F_WORK_INITIALIZED))
+			continue;
+		if (req->file && req->file->f_op == &io_uring_fops)
+			return true;
+		if ((req->work.flags & IO_WQ_WORK_FILES) &&
 		    req->work.identity->files == files)
 			return true;
 	}
@@ -1394,6 +1397,8 @@ static void io_req_clean_work(struct io_kiocb *req)
 			free_fs_struct(fs);
 		req->work.flags &= ~IO_WQ_WORK_FS;
 	}
+	if (req->flags & REQ_F_INFLIGHT)
+		io_req_drop_files(req);
 
 	io_put_identity(req->task->io_uring, req);
 }
@@ -1503,11 +1508,14 @@ static bool io_grab_identity(struct io_kiocb *req)
 			return false;
 		atomic_inc(&id->files->count);
 		get_nsproxy(id->nsproxy);
-		req->flags |= REQ_F_INFLIGHT;
 
-		spin_lock_irq(&ctx->inflight_lock);
-		list_add(&req->inflight_entry, &ctx->inflight_list);
-		spin_unlock_irq(&ctx->inflight_lock);
+		if (!(req->flags & REQ_F_INFLIGHT)) {
+			req->flags |= REQ_F_INFLIGHT;
+
+			spin_lock_irq(&ctx->inflight_lock);
+			list_add(&req->inflight_entry, &ctx->inflight_list);
+			spin_unlock_irq(&ctx->inflight_lock);
+		}
 		req->work.flags |= IO_WQ_WORK_FILES;
 	}
 	if (!(req->work.flags & IO_WQ_WORK_MM) &&
@@ -2270,6 +2278,8 @@ static void io_req_free_batch_finish(struct io_ring_ctx *ctx,
 		struct io_uring_task *tctx = rb->task->io_uring;
 
 		percpu_counter_sub(&tctx->inflight, rb->task_refs);
+		if (atomic_read(&tctx->in_idle))
+			wake_up(&tctx->wait);
 		put_task_struct_many(rb->task, rb->task_refs);
 		rb->task = NULL;
 	}
@@ -2288,6 +2298,8 @@ static void io_req_free_batch(struct req_batch *rb, struct io_kiocb *req)
 			struct io_uring_task *tctx = rb->task->io_uring;
 
 			percpu_counter_sub(&tctx->inflight, rb->task_refs);
+			if (atomic_read(&tctx->in_idle))
+				wake_up(&tctx->wait);
 			put_task_struct_many(rb->task, rb->task_refs);
 		}
 		rb->task = req->task;
@@ -3548,7 +3560,7 @@ static int io_read(struct io_kiocb *req, bool force_nonblock,
 
 	/* read it all, or we did blocking attempt. no retry. */
 	if (!iov_iter_count(iter) || !force_nonblock ||
-	    (req->file->f_flags & O_NONBLOCK))
+	    (req->file->f_flags & O_NONBLOCK) || !(req->flags & REQ_F_ISREG))
 		goto done;
 
 	io_size -= ret;
@@ -4468,7 +4480,6 @@ static int io_close_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 	 * io_wq_work.flags, so initialize io_wq_work firstly.
 	 */
 	io_req_init_async(req);
-	req->work.flags |= IO_WQ_WORK_NO_CANCEL;
 
 	if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL))
 		return -EINVAL;
@@ -4501,6 +4512,8 @@ static int io_close(struct io_kiocb *req, bool force_nonblock,
 
 	/* if the file has a flush method, be safe and punt to async */
 	if (close->put_file->f_op->flush && force_nonblock) {
+		/* not safe to cancel at this point */
+		req->work.flags |= IO_WQ_WORK_NO_CANCEL;
 		/* was never set, but play safe */
 		req->flags &= ~REQ_F_NOWAIT;
 		/* avoid grabbing files - we don't need the files */
@@ -6157,8 +6170,10 @@ static void io_req_drop_files(struct io_kiocb *req)
 	struct io_uring_task *tctx = req->task->io_uring;
 	unsigned long flags;
 
-	put_files_struct(req->work.identity->files);
-	put_nsproxy(req->work.identity->nsproxy);
+	if (req->work.flags & IO_WQ_WORK_FILES) {
+		put_files_struct(req->work.identity->files);
+		put_nsproxy(req->work.identity->nsproxy);
+	}
 	spin_lock_irqsave(&ctx->inflight_lock, flags);
 	list_del(&req->inflight_entry);
 	spin_unlock_irqrestore(&ctx->inflight_lock, flags);
@@ -6225,9 +6240,6 @@ static void __io_clean_op(struct io_kiocb *req)
 		}
 		req->flags &= ~REQ_F_NEED_CLEANUP;
 	}
-
-	if (req->flags & REQ_F_INFLIGHT)
-		io_req_drop_files(req);
 }
 
 static int io_issue_sqe(struct io_kiocb *req, bool force_nonblock,
@@ -6446,6 +6458,15 @@ static struct file *io_file_get(struct io_submit_state *state,
 		file = __io_file_get(state, fd);
 	}
 
+	if (file && file->f_op == &io_uring_fops) {
+		io_req_init_async(req);
+		req->flags |= REQ_F_INFLIGHT;
+
+		spin_lock_irq(&ctx->inflight_lock);
+		list_add(&req->inflight_entry, &ctx->inflight_list);
+		spin_unlock_irq(&ctx->inflight_lock);
+	}
+
 	return file;
 }
 
@@ -8856,8 +8877,7 @@ static void io_uring_cancel_files(struct io_ring_ctx *ctx,
 
 		spin_lock_irq(&ctx->inflight_lock);
 		list_for_each_entry(req, &ctx->inflight_list, inflight_entry) {
-			if (req->task != task ||
-			    req->work.identity->files != files)
+			if (!io_match_task(req, task, files))
 				continue;
 			found = true;
 			break;
@@ -8874,6 +8894,7 @@ static void io_uring_cancel_files(struct io_ring_ctx *ctx,
 		io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, &cancel, true);
 		io_poll_remove_all(ctx, task, files);
 		io_kill_timeouts(ctx, task, files);
+		io_cqring_overflow_flush(ctx, true, task, files);
 		/* cancellations _may_ trigger task work */
 		io_run_task_work();
 		schedule();
@@ -8914,8 +8935,6 @@ static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx,
 
 static void io_disable_sqo_submit(struct io_ring_ctx *ctx)
 {
-	WARN_ON_ONCE(ctx->sqo_task != current);
-
 	mutex_lock(&ctx->uring_lock);
 	ctx->sqo_dead = 1;
 	mutex_unlock(&ctx->uring_lock);
@@ -8937,6 +8956,7 @@ static void io_uring_cancel_task_requests(struct io_ring_ctx *ctx,
 
 	if ((ctx->flags & IORING_SETUP_SQPOLL) && ctx->sq_data) {
 		/* for SQPOLL only sqo_task has task notes */
+		WARN_ON_ONCE(ctx->sqo_task != current);
 		io_disable_sqo_submit(ctx);
 		task = ctx->sq_data->thread;
 		atomic_inc(&task->io_uring->in_idle);
@@ -9082,6 +9102,10 @@ void __io_uring_task_cancel(void)
 	/* make sure overflow events are dropped */
 	atomic_inc(&tctx->in_idle);
 
+	/* trigger io_disable_sqo_submit() */
+	if (tctx->sqpoll)
+		__io_uring_files_cancel(NULL);
+
 	do {
 		/* read completions before cancelations */
 		inflight = tctx_inflight(tctx);
@@ -9128,7 +9152,10 @@ static int io_uring_flush(struct file *file, void *data)
 
 	if (ctx->flags & IORING_SETUP_SQPOLL) {
 		/* there is only one file note, which is owned by sqo_task */
-		WARN_ON_ONCE((ctx->sqo_task == current) ==
+		WARN_ON_ONCE(ctx->sqo_task != current &&
+			     xa_load(&tctx->xa, (unsigned long)file));
+		/* sqo_dead check is for when this happens after cancellation */
+		WARN_ON_ONCE(ctx->sqo_task == current && !ctx->sqo_dead &&
 			     !xa_load(&tctx->xa, (unsigned long)file));
 
 		io_disable_sqo_submit(ctx);