1 files changed, 67 insertions, 53 deletions

diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c
index cf348c33f485..c170a2b8d2cf 100644
--- a/io_uring/io_uring.c
+++ b/io_uring/io_uring.c
@@ -147,6 +147,7 @@ static bool io_uring_try_cancel_requests(struct io_ring_ctx *ctx,
 static void io_queue_sqe(struct io_kiocb *req);
 
 struct kmem_cache *req_cachep;
+static struct workqueue_struct *iou_wq __ro_after_init;
 
 static int __read_mostly sysctl_io_uring_disabled;
 static int __read_mostly sysctl_io_uring_group = -1;
@@ -350,7 +351,6 @@ static __cold struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p)
 err:
 	kfree(ctx->cancel_table.hbs);
 	kfree(ctx->cancel_table_locked.hbs);
-	kfree(ctx->io_bl);
 	xa_destroy(&ctx->io_bl_xa);
 	kfree(ctx);
 	return NULL;
@@ -1982,10 +1982,15 @@ fail:
 		err = -EBADFD;
 		if (!io_file_can_poll(req))
 			goto fail;
-		err = -ECANCELED;
-		if (io_arm_poll_handler(req, issue_flags) != IO_APOLL_OK)
-			goto fail;
-		return;
+		if (req->file->f_flags & O_NONBLOCK ||
+		    req->file->f_mode & FMODE_NOWAIT) {
+			err = -ECANCELED;
+			if (io_arm_poll_handler(req, issue_flags) != IO_APOLL_OK)
+				goto fail;
+			return;
+		} else {
+			req->flags &= ~REQ_F_APOLL_MULTISHOT;
+		}
 	}
 
 	if (req->flags & REQ_F_FORCE_ASYNC) {
@@ -2181,6 +2186,13 @@ static void io_init_req_drain(struct io_kiocb *req)
 	}
 }
 
+static __cold int io_init_fail_req(struct io_kiocb *req, int err)
+{
+	/* ensure per-opcode data is cleared if we fail before prep */
+	memset(&req->cmd.data, 0, sizeof(req->cmd.data));
+	return err;
+}
+
 static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 		       const struct io_uring_sqe *sqe)
 	__must_hold(&ctx->uring_lock)
@@ -2202,29 +2214,29 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 
 	if (unlikely(opcode >= IORING_OP_LAST)) {
 		req->opcode = 0;
-		return -EINVAL;
+		return io_init_fail_req(req, -EINVAL);
 	}
 	def = &io_issue_defs[opcode];
 	if (unlikely(sqe_flags & ~SQE_COMMON_FLAGS)) {
 		/* enforce forwards compatibility on users */
 		if (sqe_flags & ~SQE_VALID_FLAGS)
-			return -EINVAL;
+			return io_init_fail_req(req, -EINVAL);
 		if (sqe_flags & IOSQE_BUFFER_SELECT) {
 			if (!def->buffer_select)
-				return -EOPNOTSUPP;
+				return io_init_fail_req(req, -EOPNOTSUPP);
 			req->buf_index = READ_ONCE(sqe->buf_group);
 		}
 		if (sqe_flags & IOSQE_CQE_SKIP_SUCCESS)
 			ctx->drain_disabled = true;
 		if (sqe_flags & IOSQE_IO_DRAIN) {
 			if (ctx->drain_disabled)
-				return -EOPNOTSUPP;
+				return io_init_fail_req(req, -EOPNOTSUPP);
 			io_init_req_drain(req);
 		}
 	}
 	if (unlikely(ctx->restricted || ctx->drain_active || ctx->drain_next)) {
 		if (ctx->restricted && !io_check_restriction(ctx, req, sqe_flags))
-			return -EACCES;
+			return io_init_fail_req(req, -EACCES);
 		/* knock it to the slow queue path, will be drained there */
 		if (ctx->drain_active)
 			req->flags |= REQ_F_FORCE_ASYNC;
@@ -2237,9 +2249,9 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 	}
 
 	if (!def->ioprio && sqe->ioprio)
-		return -EINVAL;
+		return io_init_fail_req(req, -EINVAL);
 	if (!def->iopoll && (ctx->flags & IORING_SETUP_IOPOLL))
-		return -EINVAL;
+		return io_init_fail_req(req, -EINVAL);
 
 	if (def->needs_file) {
 		struct io_submit_state *state = &ctx->submit_state;
@@ -2263,12 +2275,12 @@ static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 
 		req->creds = xa_load(&ctx->personalities, personality);
 		if (!req->creds)
-			return -EINVAL;
+			return io_init_fail_req(req, -EINVAL);
 		get_cred(req->creds);
 		ret = security_uring_override_creds(req->creds);
 		if (ret) {
 			put_cred(req->creds);
-			return ret;
+			return io_init_fail_req(req, ret);
 		}
 		req->flags |= REQ_F_CREDS;
 	}
@@ -2539,7 +2551,7 @@ static bool current_pending_io(void)
 static inline int io_cqring_wait_schedule(struct io_ring_ctx *ctx,
 					  struct io_wait_queue *iowq)
 {
-	int io_wait, ret;
+	int ret;
 
 	if (unlikely(READ_ONCE(ctx->check_cq)))
 		return 1;
@@ -2557,7 +2569,6 @@ static inline int io_cqring_wait_schedule(struct io_ring_ctx *ctx,
 	 * can take into account that the task is waiting for IO - turns out
 	 * to be important for low QD IO.
 	 */
-	io_wait = current->in_iowait;
 	if (current_pending_io())
 		current->in_iowait = 1;
 	ret = 0;
@@ -2565,7 +2576,7 @@ static inline int io_cqring_wait_schedule(struct io_ring_ctx *ctx,
 		schedule();
 	else if (!schedule_hrtimeout(&iowq->timeout, HRTIMER_MODE_ABS))
 		ret = -ETIME;
-	current->in_iowait = io_wait;
+	current->in_iowait = 0;
 	return ret;
 }
 
@@ -2591,19 +2602,6 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
 	if (__io_cqring_events_user(ctx) >= min_events)
 		return 0;
 
-	if (sig) {
-#ifdef CONFIG_COMPAT
-		if (in_compat_syscall())
-			ret = set_compat_user_sigmask((const compat_sigset_t __user *)sig,
-						      sigsz);
-		else
-#endif
-			ret = set_user_sigmask(sig, sigsz);
-
-		if (ret)
-			return ret;
-	}
-
 	init_waitqueue_func_entry(&iowq.wq, io_wake_function);
 	iowq.wq.private = current;
 	INIT_LIST_HEAD(&iowq.wq.entry);
@@ -2622,6 +2620,19 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
 		io_napi_adjust_timeout(ctx, &iowq, &ts);
 	}
 
+	if (sig) {
+#ifdef CONFIG_COMPAT
+		if (in_compat_syscall())
+			ret = set_compat_user_sigmask((const compat_sigset_t __user *)sig,
+						      sigsz);
+		else
+#endif
+			ret = set_user_sigmask(sig, sigsz);
+
+		if (ret)
+			return ret;
+	}
+
 	io_napi_busy_loop(ctx, &iowq);
 
 	trace_io_uring_cqring_wait(ctx, min_events);
@@ -2697,13 +2708,9 @@ void io_mem_free(void *ptr)
 
 static void io_pages_free(struct page ***pages, int npages)
 {
-	struct page **page_array;
+	struct page **page_array = *pages;
 	int i;
 
-	if (!pages)
-		return;
-
-	page_array = *pages;
 	if (!page_array)
 		return;
 
@@ -2719,7 +2726,7 @@ static void *__io_uaddr_map(struct page ***pages, unsigned short *npages,
 	struct page **page_array;
 	unsigned int nr_pages;
 	void *page_addr;
-	int ret, i;
+	int ret, i, pinned;
 
 	*npages = 0;
 
@@ -2733,12 +2740,12 @@ static void *__io_uaddr_map(struct page ***pages, unsigned short *npages,
 	if (!page_array)
 		return ERR_PTR(-ENOMEM);
 
-	ret = pin_user_pages_fast(uaddr, nr_pages, FOLL_WRITE | FOLL_LONGTERM,
-					page_array);
-	if (ret != nr_pages) {
-err:
-		io_pages_free(&page_array, ret > 0 ? ret : 0);
-		return ret < 0 ? ERR_PTR(ret) : ERR_PTR(-EFAULT);
+
+	pinned = pin_user_pages_fast(uaddr, nr_pages, FOLL_WRITE | FOLL_LONGTERM,
+				     page_array);
+	if (pinned != nr_pages) {
+		ret = (pinned < 0) ? pinned : -EFAULT;
+		goto free_pages;
 	}
 
 	page_addr = page_address(page_array[0]);
@@ -2752,7 +2759,7 @@ err:
 		 * didn't support this feature.
 		 */
 		if (PageHighMem(page_array[i]))
-			goto err;
+			goto free_pages;
 
 		/*
 		 * No support for discontig pages for now, should either be a
@@ -2761,13 +2768,17 @@ err:
 		 * just fail them with EINVAL.
 		 */
 		if (page_address(page_array[i]) != page_addr)
-			goto err;
+			goto free_pages;
 		page_addr += PAGE_SIZE;
 	}
 
 	*pages = page_array;
 	*npages = nr_pages;
 	return page_to_virt(page_array[0]);
+
+free_pages:
+	io_pages_free(&page_array, pinned > 0 ? pinned : 0);
+	return ERR_PTR(ret);
 }
 
 static void *io_rings_map(struct io_ring_ctx *ctx, unsigned long uaddr,
@@ -2789,14 +2800,15 @@ static void io_rings_free(struct io_ring_ctx *ctx)
 	if (!(ctx->flags & IORING_SETUP_NO_MMAP)) {
 		io_mem_free(ctx->rings);
 		io_mem_free(ctx->sq_sqes);
-		ctx->rings = NULL;
-		ctx->sq_sqes = NULL;
 	} else {
 		io_pages_free(&ctx->ring_pages, ctx->n_ring_pages);
 		ctx->n_ring_pages = 0;
 		io_pages_free(&ctx->sqe_pages, ctx->n_sqe_pages);
 		ctx->n_sqe_pages = 0;
 	}
+
+	ctx->rings = NULL;
+	ctx->sq_sqes = NULL;
 }
 
 void *io_mem_alloc(size_t size)
@@ -2919,7 +2931,6 @@ static __cold void io_ring_ctx_free(struct io_ring_ctx *ctx)
 	io_napi_free(ctx);
 	kfree(ctx->cancel_table.hbs);
 	kfree(ctx->cancel_table_locked.hbs);
-	kfree(ctx->io_bl);
 	xa_destroy(&ctx->io_bl_xa);
 	kfree(ctx);
 }
@@ -3154,7 +3165,7 @@ static __cold void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)
 	 * noise and overhead, there's no discernable change in runtime
 	 * over using system_wq.
 	 */
-	queue_work(system_unbound_wq, &ctx->exit_work);
+	queue_work(iou_wq, &ctx->exit_work);
 }
 
 static int io_uring_release(struct inode *inode, struct file *file)
@@ -3436,14 +3447,15 @@ static void *io_uring_validate_mmap_request(struct file *file,
 		ptr = ctx->sq_sqes;
 		break;
 	case IORING_OFF_PBUF_RING: {
+		struct io_buffer_list *bl;
 		unsigned int bgid;
 
 		bgid = (offset & ~IORING_OFF_MMAP_MASK) >> IORING_OFF_PBUF_SHIFT;
-		rcu_read_lock();
-		ptr = io_pbuf_get_address(ctx, bgid);
-		rcu_read_unlock();
-		if (!ptr)
-			return ERR_PTR(-EINVAL);
+		bl = io_pbuf_get_bl(ctx, bgid);
+		if (IS_ERR(bl))
+			return bl;
+		ptr = bl->buf_ring;
+		io_put_bl(ctx, bl);
 		break;
 		}
 	default:
@@ -4178,6 +4190,8 @@ static int __init io_uring_init(void)
 	io_buf_cachep = KMEM_CACHE(io_buffer,
 					  SLAB_HWCACHE_ALIGN | SLAB_PANIC | SLAB_ACCOUNT);
 
+	iou_wq = alloc_workqueue("iou_exit", WQ_UNBOUND, 64);
+
 #ifdef CONFIG_SYSCTL
 	register_sysctl_init("kernel", kernel_io_uring_disabled_table);
 #endif