1 files changed, 64 insertions, 27 deletions

diff --git a/fs/io_uring.c b/fs/io_uring.c
index b07196b4511c..fb2a0cb4aaf8 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -1278,6 +1278,7 @@ static void io_refs_resurrect(struct percpu_ref *ref, struct completion *compl)
 
 static bool io_match_task(struct io_kiocb *head, struct task_struct *task,
 			  bool cancel_all)
+	__must_hold(&req->ctx->timeout_lock)
 {
 	struct io_kiocb *req;
 
@@ -1293,6 +1294,44 @@ static bool io_match_task(struct io_kiocb *head, struct task_struct *task,
 	return false;
 }
 
+static bool io_match_linked(struct io_kiocb *head)
+{
+	struct io_kiocb *req;
+
+	io_for_each_link(req, head) {
+		if (req->flags & REQ_F_INFLIGHT)
+			return true;
+	}
+	return false;
+}
+
+/*
+ * As io_match_task() but protected against racing with linked timeouts.
+ * User must not hold timeout_lock.
+ */
+static bool io_match_task_safe(struct io_kiocb *head, struct task_struct *task,
+			       bool cancel_all)
+{
+	bool matched;
+
+	if (task && head->task != task)
+		return false;
+	if (cancel_all)
+		return true;
+
+	if (head->flags & REQ_F_LINK_TIMEOUT) {
+		struct io_ring_ctx *ctx = head->ctx;
+
+		/* protect against races with linked timeouts */
+		spin_lock_irq(&ctx->timeout_lock);
+		matched = io_match_linked(head);
+		spin_unlock_irq(&ctx->timeout_lock);
+	} else {
+		matched = io_match_linked(head);
+	}
+	return matched;
+}
+
 static inline bool req_has_async_data(struct io_kiocb *req)
 {
 	return req->flags & REQ_F_ASYNC_DATA;
@@ -1502,10 +1541,10 @@ static void io_prep_async_link(struct io_kiocb *req)
 	if (req->flags & REQ_F_LINK_TIMEOUT) {
 		struct io_ring_ctx *ctx = req->ctx;
 
-		spin_lock(&ctx->completion_lock);
+		spin_lock_irq(&ctx->timeout_lock);
 		io_for_each_link(cur, req)
 			io_prep_async_work(cur);
-		spin_unlock(&ctx->completion_lock);
+		spin_unlock_irq(&ctx->timeout_lock);
 	} else {
 		io_for_each_link(cur, req)
 			io_prep_async_work(cur);
@@ -2852,9 +2891,13 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 		req->flags |= io_file_get_flags(file) << REQ_F_SUPPORT_NOWAIT_BIT;
 
 	kiocb->ki_pos = READ_ONCE(sqe->off);
-	if (kiocb->ki_pos == -1 && !(file->f_mode & FMODE_STREAM)) {
-		req->flags |= REQ_F_CUR_POS;
-		kiocb->ki_pos = file->f_pos;
+	if (kiocb->ki_pos == -1) {
+		if (!(file->f_mode & FMODE_STREAM)) {
+			req->flags |= REQ_F_CUR_POS;
+			kiocb->ki_pos = file->f_pos;
+		} else {
+			kiocb->ki_pos = 0;
+		}
 	}
 	kiocb->ki_flags = iocb_flags(file);
 	ret = kiocb_set_rw_flags(kiocb, READ_ONCE(sqe->rw_flags));
@@ -4327,6 +4370,7 @@ static int __io_remove_buffers(struct io_ring_ctx *ctx, struct io_buffer *buf,
 		kfree(nxt);
 		if (++i == nbufs)
 			return i;
+		cond_resched();
 	}
 	i++;
 	kfree(buf);
@@ -5704,7 +5748,7 @@ static __cold bool io_poll_remove_all(struct io_ring_ctx *ctx,
 
 		list = &ctx->cancel_hash[i];
 		hlist_for_each_entry_safe(req, tmp, list, hash_node) {
-			if (io_match_task(req, tsk, cancel_all))
+			if (io_match_task_safe(req, tsk, cancel_all))
 				posted += io_poll_remove_one(req);
 		}
 	}
@@ -6156,6 +6200,9 @@ static int io_timeout_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe,
 	if (get_timespec64(&data->ts, u64_to_user_ptr(sqe->addr)))
 		return -EFAULT;
 
+	if (data->ts.tv_sec < 0 || data->ts.tv_nsec < 0)
+		return -EINVAL;
+
 	data->mode = io_translate_timeout_mode(flags);
 	hrtimer_init(&data->timer, io_timeout_get_clock(data), data->mode);
 
@@ -6880,10 +6927,11 @@ static inline struct file *io_file_get(struct io_ring_ctx *ctx,
 static void io_req_task_link_timeout(struct io_kiocb *req, bool *locked)
 {
 	struct io_kiocb *prev = req->timeout.prev;
-	int ret;
+	int ret = -ENOENT;
 
 	if (prev) {
-		ret = io_try_cancel_userdata(req, prev->user_data);
+		if (!(req->task->flags & PF_EXITING))
+			ret = io_try_cancel_userdata(req, prev->user_data);
 		io_req_complete_post(req, ret ?: -ETIME, 0);
 		io_put_req(prev);
 	} else {
@@ -9255,10 +9303,8 @@ static void io_destroy_buffers(struct io_ring_ctx *ctx)
 	struct io_buffer *buf;
 	unsigned long index;
 
-	xa_for_each(&ctx->io_buffers, index, buf) {
+	xa_for_each(&ctx->io_buffers, index, buf)
 		__io_remove_buffers(ctx, buf, index, -1U);
-		cond_resched();
-	}
 }
 
 static void io_req_caches_free(struct io_ring_ctx *ctx)
@@ -9562,19 +9608,8 @@ static bool io_cancel_task_cb(struct io_wq_work *work, void *data)
 {
 	struct io_kiocb *req = container_of(work, struct io_kiocb, work);
 	struct io_task_cancel *cancel = data;
-	bool ret;
 
-	if (!cancel->all && (req->flags & REQ_F_LINK_TIMEOUT)) {
-		struct io_ring_ctx *ctx = req->ctx;
-
-		/* protect against races with linked timeouts */
-		spin_lock(&ctx->completion_lock);
-		ret = io_match_task(req, cancel->task, cancel->all);
-		spin_unlock(&ctx->completion_lock);
-	} else {
-		ret = io_match_task(req, cancel->task, cancel->all);
-	}
-	return ret;
+	return io_match_task_safe(req, cancel->task, cancel->all);
 }
 
 static __cold bool io_cancel_defer_files(struct io_ring_ctx *ctx,
@@ -9586,7 +9621,7 @@ static __cold bool io_cancel_defer_files(struct io_ring_ctx *ctx,
 
 	spin_lock(&ctx->completion_lock);
 	list_for_each_entry_reverse(de, &ctx->defer_list, list) {
-		if (io_match_task(de->req, task, cancel_all)) {
+		if (io_match_task_safe(de->req, task, cancel_all)) {
 			list_cut_position(&list, &ctx->defer_list, &de->list);
 			break;
 		}
@@ -9764,7 +9799,7 @@ static __cold void io_uring_clean_tctx(struct io_uring_task *tctx)
 	}
 	if (wq) {
 		/*
-		 * Must be after io_uring_del_task_file() (removes nodes under
+		 * Must be after io_uring_del_tctx_node() (removes nodes under
 		 * uring_lock) to avoid race with io_uring_try_cancel_iowq().
 		 */
 		io_wq_put_and_exit(wq);
@@ -9793,7 +9828,7 @@ static __cold void io_uring_drop_tctx_refs(struct task_struct *task)
 
 /*
  * Find any io_uring ctx that this task has registered or done IO on, and cancel
- * requests. @sqd should be not-null IIF it's an SQPOLL thread cancellation.
+ * requests. @sqd should be not-null IFF it's an SQPOLL thread cancellation.
  */
 static __cold void io_uring_cancel_generic(bool cancel_all,
 					   struct io_sq_data *sqd)
@@ -9835,8 +9870,10 @@ static __cold void io_uring_cancel_generic(bool cancel_all,
 							     cancel_all);
 		}
 
-		prepare_to_wait(&tctx->wait, &wait, TASK_UNINTERRUPTIBLE);
+		prepare_to_wait(&tctx->wait, &wait, TASK_INTERRUPTIBLE);
+		io_run_task_work();
 		io_uring_drop_tctx_refs(current);
+
 		/*
 		 * If we've seen completions, retry without waiting. This
 		 * avoids a race where a completion comes in before we did