io_uring-6.1-2022-10-13

-----BEGIN PGP SIGNATURE-----
 
 iQJEBAABCAAuFiEEwPw5LcreJtl1+l5K99NY+ylx4KYFAmNIXOYQHGF4Ym9lQGtl
 cm5lbC5kawAKCRD301j7KXHgpr17D/9qA35JJ4lnZWo+ayqwRPAqpO9DRgmyt2W+
 FQAdbaWpsi2cr+Iv4AU51rIWfydP1UZCfYBPkWvLCoCzLizWUz4c6bLBc4+HvJcR
 wuiWG+js6PSi279mUA1s1XXYthYt7CiMpi3+2iatGiPBFtIIMeMqMos02ohKu/4L
 gd9qTKUfwepBcM2D61wk/jl7UXtdIb+NCh685FvkQ32Nk4kQx3L5y3BRJ1NUtCTz
 GI7HT2Lp6Im3pD6hpDPSS8H/DK7gTdyQYPiI9+rfdfIR8HW4k/3FkefYB5He554J
 8b9MsdSN6g/oIQxPhMlbbE4e24ahEPK7h5V1HYCF8i/dPbxjuZQ0HpgC+LiLG9/J
 cOVXjZVjcuzUmJM9eJ/jNZx0w3/RHGo8lstBvQ0SyuRpLYhe+pvVYlkNEkK4ptIZ
 bw/scjmQxNp3xK8PRUMaIb7sHfha/+OoIj+XHXgeiujtDIFQOc+7G14o+gCkker5
 ZIVrbXcn4cWAHKUMPlH3NWj89v6lkulfuOQ6NZPbz4CYj3tudqnG3ah810iGBTiR
 1HOhtwKPYR8WGioj1ZjUiSJZ/8aT4GUuLM2EjKsIITo09mYwx24G+pfaWz+qBnzw
 SQqg0M0XVCg+RSrfakl9LE2uX2V06YDMBg52HA014CQaRE0pJHX/0u1VKg4rdD0z
 QvdbB5peXg==
 =0dSL
 -----END PGP SIGNATURE-----

Merge tag 'io_uring-6.1-2022-10-13' of git://git.kernel.dk/linux

Pull more io_uring updates from Jens Axboe:
 "A collection of fixes that ended up either being later than the
  initial pull, or dependent on multiple branches (6.0-late being one of
  them) and hence deferred purposely. This contains:

   - Cleanup fixes for the single submitter late 6.0 change, which we
     pushed to 6.1 to keep the 6.0 changes small (Dylan, Pavel)

   - Fix for IORING_OP_CONNECT not handling -EINPROGRESS correctly (me)

   - Ensure that the zc sendmsg variant gets audited correctly (me)

   - Regression fix from this merge window where kiocb_end_write()
     doesn't always gets called, which can cause issues with fs freezing
     (me)

   - Registered files SCM handling fix (Pavel)

   - Regression fix for big sqe dumping in fdinfo (Pavel)

   - Registered buffers accounting fix (Pavel)

   - Remove leftover notification structures, we killed them off late in
     6.0 (Pavel)

   - Minor optimizations (Pavel)

   - Cosmetic variable shadowing fix (Stefan)"

* tag 'io_uring-6.1-2022-10-13' of git://git.kernel.dk/linux:
  io_uring/rw: ensure kiocb_end_write() is always called
  io_uring: fix fdinfo sqe offsets calculation
  io_uring: local variable rw shadows outer variable in io_write
  io_uring/opdef: remove 'audit_skip' from SENDMSG_ZC
  io_uring: optimise locking for local tw with submit_wait
  io_uring: remove redundant memory barrier in io_req_local_work_add
  io_uring/net: handle -EINPROGRESS correct for IORING_OP_CONNECT
  io_uring: remove notif leftovers
  io_uring: correct pinned_vm accounting
  io_uring/af_unix: defer registered files gc to io_uring release
  io_uring: limit registration w/ SINGLE_ISSUER
  io_uring: remove io_register_submitter
  io_uring: simplify __io_uring_add_tctx_node

include/linux/io_uring_types.h

@@ -34,9 +34,6 @@ struct io_file_table {
 	unsigned int alloc_hint;
 };
 
-struct io_notif;
-struct io_notif_slot;
-
 struct io_hash_bucket {
 	spinlock_t		lock;
 	struct hlist_head	list;
@@ -242,8 +239,6 @@ struct io_ring_ctx {
 		unsigned		nr_user_files;
 		unsigned		nr_user_bufs;
 		struct io_mapped_ubuf	**user_bufs;
-		struct io_notif_slot	*notif_slots;
-		unsigned		nr_notif_slots;
 
 		struct io_submit_state	submit_state;
 

include/linux/skbuff.h

@@ -803,6 +803,7 @@ typedef unsigned char *sk_buff_data_t;
  *	@csum_level: indicates the number of consecutive checksums found in
  *		the packet minus one that have been verified as
  *		CHECKSUM_UNNECESSARY (max 3)
+ *	@scm_io_uring: SKB holds io_uring registered files
  *	@dst_pending_confirm: need to confirm neighbour
  *	@decrypted: Decrypted SKB
  *	@slow_gro: state present at GRO time, slower prepare step required
@@ -982,6 +983,7 @@ struct sk_buff {
 #endif
 	__u8			slow_gro:1;
 	__u8			csum_not_inet:1;
+	__u8			scm_io_uring:1;
 
 #ifdef CONFIG_NET_SCHED
 	__u16			tc_index;	/* traffic control index */

io_uring/fdinfo.c

@@ -94,7 +94,7 @@ static __cold void __io_uring_show_fdinfo(struct io_ring_ctx *ctx,
 		sq_idx = READ_ONCE(ctx->sq_array[entry & sq_mask]);
 		if (sq_idx > sq_mask)
 			continue;
-		sqe = &ctx->sq_sqes[sq_idx << 1];
+		sqe = &ctx->sq_sqes[sq_idx << sq_shift];
 		seq_printf(m, "%5u: opcode:%s, fd:%d, flags:%x, off:%llu, "
 			      "addr:0x%llx, rw_flags:0x%x, buf_index:%d "
 			      "user_data:%llu",

io_uring/io_uring.c

@@ -1106,6 +1106,8 @@ static void io_req_local_work_add(struct io_kiocb *req)
 
 	if (!llist_add(&req->io_task_work.node, &ctx->work_llist))
 		return;
+	/* need it for the following io_cqring_wake() */
+	smp_mb__after_atomic();
 
 	if (unlikely(atomic_read(&req->task->io_uring->in_idle))) {
 		io_move_task_work_from_local(ctx);
@@ -1117,8 +1119,7 @@ static void io_req_local_work_add(struct io_kiocb *req)
 
 	if (ctx->has_evfd)
 		io_eventfd_signal(ctx);
-	io_cqring_wake(ctx);
+	__io_cqring_wake(ctx);
-
 }
 
 static inline void __io_req_task_work_add(struct io_kiocb *req, bool allow_local)
@@ -2585,12 +2586,6 @@ static void io_req_caches_free(struct io_ring_ctx *ctx)
 static __cold void io_ring_ctx_free(struct io_ring_ctx *ctx)
 {
 	io_sq_thread_finish(ctx);
-
-	if (ctx->mm_account) {
-		mmdrop(ctx->mm_account);
-		ctx->mm_account = NULL;
-	}
-
 	io_rsrc_refs_drop(ctx);
 	/* __io_rsrc_put_work() may need uring_lock to progress, wait w/o it */
 	io_wait_rsrc_data(ctx->buf_data);
@@ -2631,8 +2626,11 @@ static __cold void io_ring_ctx_free(struct io_ring_ctx *ctx)
 	}
 #endif
 	WARN_ON_ONCE(!list_empty(&ctx->ltimeout_list));
-	WARN_ON_ONCE(ctx->notif_slots || ctx->nr_notif_slots);
 
+	if (ctx->mm_account) {
+		mmdrop(ctx->mm_account);
+		ctx->mm_account = NULL;
+	}
 	io_mem_free(ctx->rings);
 	io_mem_free(ctx->sq_sqes);
 
@@ -3229,8 +3227,16 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
 			mutex_unlock(&ctx->uring_lock);
 			goto out;
 		}
-		if ((flags & IORING_ENTER_GETEVENTS) && ctx->syscall_iopoll)
+		if (flags & IORING_ENTER_GETEVENTS) {
-			goto iopoll_locked;
+			if (ctx->syscall_iopoll)
+				goto iopoll_locked;
+			/*
+			 * Ignore errors, we'll soon call io_cqring_wait() and
+			 * it should handle ownership problems if any.
+			 */
+			if (ctx->flags & IORING_SETUP_DEFER_TASKRUN)
+				(void)io_run_local_work_locked(ctx);
+		}
 		mutex_unlock(&ctx->uring_lock);
 	}
 
@@ -3355,7 +3361,7 @@ static int io_uring_install_fd(struct io_ring_ctx *ctx, struct file *file)
 	if (fd < 0)
 		return fd;
 
-	ret = __io_uring_add_tctx_node(ctx, false);
+	ret = __io_uring_add_tctx_node(ctx);
 	if (ret) {
 		put_unused_fd(fd);
 		return ret;
@@ -3890,6 +3896,9 @@ static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
 	if (WARN_ON_ONCE(percpu_ref_is_dying(&ctx->refs)))
 		return -ENXIO;
 
+	if (ctx->submitter_task && ctx->submitter_task != current)
+		return -EEXIST;
+
 	if (ctx->restricted) {
 		if (opcode >= IORING_REGISTER_LAST)
 			return -EINVAL;

io_uring/io_uring.h

@@ -203,17 +203,24 @@ static inline void io_commit_cqring(struct io_ring_ctx *ctx)
 	smp_store_release(&ctx->rings->cq.tail, ctx->cached_cq_tail);
 }
 
-static inline void io_cqring_wake(struct io_ring_ctx *ctx)
+/* requires smb_mb() prior, see wq_has_sleeper() */
+static inline void __io_cqring_wake(struct io_ring_ctx *ctx)
 {
 	/*
 	 * wake_up_all() may seem excessive, but io_wake_function() and
 	 * io_should_wake() handle the termination of the loop and only
 	 * wake as many waiters as we need to.
 	 */
-	if (wq_has_sleeper(&ctx->cq_wait))
+	if (waitqueue_active(&ctx->cq_wait))
 		wake_up_all(&ctx->cq_wait);
 }
 
+static inline void io_cqring_wake(struct io_ring_ctx *ctx)
+{
+	smp_mb();
+	__io_cqring_wake(ctx);
+}
+
 static inline bool io_sqring_full(struct io_ring_ctx *ctx)
 {
 	struct io_rings *r = ctx->rings;
@@ -268,6 +275,13 @@ static inline int io_run_task_work_ctx(struct io_ring_ctx *ctx)
 	return ret;
 }
 
+static inline int io_run_local_work_locked(struct io_ring_ctx *ctx)
+{
+	if (llist_empty(&ctx->work_llist))
+		return 0;
+	return __io_run_local_work(ctx, true);
+}
+
 static inline void io_tw_lock(struct io_ring_ctx *ctx, bool *locked)
 {
 	if (!*locked) {

io_uring/net.c

@@ -46,6 +46,7 @@ struct io_connect {
 	struct file			*file;
 	struct sockaddr __user		*addr;
 	int				addr_len;
+	bool				in_progress;
 };
 
 struct io_sr_msg {
@@ -1386,6 +1387,7 @@ int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe)
 
 	conn->addr = u64_to_user_ptr(READ_ONCE(sqe->addr));
 	conn->addr_len =  READ_ONCE(sqe->addr2);
+	conn->in_progress = false;
 	return 0;
 }
 
@@ -1397,6 +1399,16 @@ int io_connect(struct io_kiocb *req, unsigned int issue_flags)
 	int ret;
 	bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK;
 
+	if (connect->in_progress) {
+		struct socket *socket;
+
+		ret = -ENOTSOCK;
+		socket = sock_from_file(req->file);
+		if (socket)
+			ret = sock_error(socket->sk);
+		goto out;
+	}
+
 	if (req_has_async_data(req)) {
 		io = req->async_data;
 	} else {
@@ -1413,13 +1425,17 @@ int io_connect(struct io_kiocb *req, unsigned int issue_flags)
 	ret = __sys_connect_file(req->file, &io->address,
 					connect->addr_len, file_flags);
 	if ((ret == -EAGAIN || ret == -EINPROGRESS) && force_nonblock) {
-		if (req_has_async_data(req))
+		if (ret == -EINPROGRESS) {
-			return -EAGAIN;
+			connect->in_progress = true;
-		if (io_alloc_async_data(req)) {
+		} else {
-			ret = -ENOMEM;
+			if (req_has_async_data(req))
-			goto out;
+				return -EAGAIN;
+			if (io_alloc_async_data(req)) {
+				ret = -ENOMEM;
+				goto out;
+			}
+			memcpy(req->async_data, &__io, sizeof(__io));
 		}
-		memcpy(req->async_data, &__io, sizeof(__io));
 		return -EAGAIN;
 	}
 	if (ret == -ERESTARTSYS)

io_uring/opdef.c

@@ -510,7 +510,6 @@ const struct io_op_def io_op_defs[] = {
 		.needs_file		= 1,
 		.unbound_nonreg_file	= 1,
 		.pollout		= 1,
-		.audit_skip		= 1,
 		.ioprio			= 1,
 		.manual_alloc		= 1,
 #if defined(CONFIG_NET)

io_uring/rsrc.c

@@ -855,6 +855,7 @@ int __io_scm_file_account(struct io_ring_ctx *ctx, struct file *file)
 
 		UNIXCB(skb).fp = fpl;
 		skb->sk = sk;
+		skb->scm_io_uring = 1;
 		skb->destructor = unix_destruct_scm;
 		refcount_add(skb->truesize, &sk->sk_wmem_alloc);
 	}

io_uring/rw.c

@@ -234,11 +234,34 @@ static void kiocb_end_write(struct io_kiocb *req)
 	}
 }
 
+/*
+ * Trigger the notifications after having done some IO, and finish the write
+ * accounting, if any.
+ */
+static void io_req_io_end(struct io_kiocb *req)
+{
+	struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
+
+	WARN_ON(!in_task());
+
+	if (rw->kiocb.ki_flags & IOCB_WRITE) {
+		kiocb_end_write(req);
+		fsnotify_modify(req->file);
+	} else {
+		fsnotify_access(req->file);
+	}
+}
+
 static bool __io_complete_rw_common(struct io_kiocb *req, long res)
 {
 	if (unlikely(res != req->cqe.res)) {
 		if ((res == -EAGAIN || res == -EOPNOTSUPP) &&
 		    io_rw_should_reissue(req)) {
+			/*
+			 * Reissue will start accounting again, finish the
+			 * current cycle.
+			 */
+			io_req_io_end(req);
 			req->flags |= REQ_F_REISSUE | REQ_F_PARTIAL_IO;
 			return true;
 		}
@@ -264,15 +287,7 @@ static inline int io_fixup_rw_res(struct io_kiocb *req, long res)
 
 static void io_req_rw_complete(struct io_kiocb *req, bool *locked)
 {
-	struct io_rw *rw = io_kiocb_to_cmd(req, struct io_rw);
+	io_req_io_end(req);
-
-	if (rw->kiocb.ki_flags & IOCB_WRITE) {
-		kiocb_end_write(req);
-		fsnotify_modify(req->file);
-	} else {
-		fsnotify_access(req->file);
-	}
-
 	io_req_task_complete(req, locked);
 }
 
@@ -317,6 +332,11 @@ static int kiocb_done(struct io_kiocb *req, ssize_t ret,
 		req->file->f_pos = rw->kiocb.ki_pos;
 	if (ret >= 0 && (rw->kiocb.ki_complete == io_complete_rw)) {
 		if (!__io_complete_rw_common(req, ret)) {
+			/*
+			 * Safe to call io_end from here as we're inline
+			 * from the submission path.
+			 */
+			io_req_io_end(req);
 			io_req_set_res(req, final_ret,
 				       io_put_kbuf(req, issue_flags));
 			return IOU_OK;
@@ -916,7 +936,7 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags)
 			goto copy_iov;
 
 		if (ret2 != req->cqe.res && ret2 >= 0 && need_complete_io(req)) {
-			struct io_async_rw *rw;
+			struct io_async_rw *io;
 
 			trace_io_uring_short_write(req->ctx, kiocb->ki_pos - ret2,
 						req->cqe.res, ret2);
@@ -929,9 +949,9 @@ int io_write(struct io_kiocb *req, unsigned int issue_flags)
 			iov_iter_save_state(&s->iter, &s->iter_state);
 			ret = io_setup_async_rw(req, iovec, s, true);
 
-			rw = req->async_data;
+			io = req->async_data;
-			if (rw)
+			if (io)
-				rw->bytes_done += ret2;
+				io->bytes_done += ret2;
 
 			if (kiocb->ki_flags & IOCB_WRITE)
 				kiocb_end_write(req);

io_uring/tctx.c

@@ -91,32 +91,12 @@ __cold int io_uring_alloc_task_context(struct task_struct *task,
 	return 0;
 }
 
-static int io_register_submitter(struct io_ring_ctx *ctx)
+int __io_uring_add_tctx_node(struct io_ring_ctx *ctx)
-{
-	int ret = 0;
-
-	mutex_lock(&ctx->uring_lock);
-	if (!ctx->submitter_task)
-		ctx->submitter_task = get_task_struct(current);
-	else if (ctx->submitter_task != current)
-		ret = -EEXIST;
-	mutex_unlock(&ctx->uring_lock);
-
-	return ret;
-}
-
-int __io_uring_add_tctx_node(struct io_ring_ctx *ctx, bool submitter)
 {
 	struct io_uring_task *tctx = current->io_uring;
 	struct io_tctx_node *node;
 	int ret;
 
-	if ((ctx->flags & IORING_SETUP_SINGLE_ISSUER) && submitter) {
-		ret = io_register_submitter(ctx);
-		if (ret)
-			return ret;
-	}
-
 	if (unlikely(!tctx)) {
 		ret = io_uring_alloc_task_context(current, ctx);
 		if (unlikely(ret))
@@ -150,8 +130,22 @@ int __io_uring_add_tctx_node(struct io_ring_ctx *ctx, bool submitter)
 		list_add(&node->ctx_node, &ctx->tctx_list);
 		mutex_unlock(&ctx->uring_lock);
 	}
-	if (submitter)
+	return 0;
-		tctx->last = ctx;
+}
+
+int __io_uring_add_tctx_node_from_submit(struct io_ring_ctx *ctx)
+{
+	int ret;
+
+	if (ctx->flags & IORING_SETUP_SINGLE_ISSUER
+	    && ctx->submitter_task != current)
+		return -EEXIST;
+
+	ret = __io_uring_add_tctx_node(ctx);
+	if (ret)
+		return ret;
+
+	current->io_uring->last = ctx;
 	return 0;
 }
 
@@ -259,7 +253,7 @@ int io_ringfd_register(struct io_ring_ctx *ctx, void __user *__arg,
 		return -EINVAL;
 
 	mutex_unlock(&ctx->uring_lock);
-	ret = __io_uring_add_tctx_node(ctx, false);
+	ret = __io_uring_add_tctx_node(ctx);
 	mutex_lock(&ctx->uring_lock);
 	if (ret)
 		return ret;

io_uring/tctx.h

@@ -9,7 +9,8 @@ struct io_tctx_node {
 int io_uring_alloc_task_context(struct task_struct *task,
 				struct io_ring_ctx *ctx);
 void io_uring_del_tctx_node(unsigned long index);
-int __io_uring_add_tctx_node(struct io_ring_ctx *ctx, bool submitter);
+int __io_uring_add_tctx_node(struct io_ring_ctx *ctx);
+int __io_uring_add_tctx_node_from_submit(struct io_ring_ctx *ctx);
 void io_uring_clean_tctx(struct io_uring_task *tctx);
 
 void io_uring_unreg_ringfd(void);
@@ -27,5 +28,6 @@ static inline int io_uring_add_tctx_node(struct io_ring_ctx *ctx)
 
 	if (likely(tctx && tctx->last == ctx))
 		return 0;
-	return __io_uring_add_tctx_node(ctx, true);
+
+	return __io_uring_add_tctx_node_from_submit(ctx);
 }

net/unix/garbage.c

@@ -204,6 +204,7 @@ void wait_for_unix_gc(void)
 /* The external entry point: unix_gc() */
 void unix_gc(void)
 {
+	struct sk_buff *next_skb, *skb;
 	struct unix_sock *u;
 	struct unix_sock *next;
 	struct sk_buff_head hitlist;
@@ -297,11 +298,30 @@ void unix_gc(void)
 
 	spin_unlock(&unix_gc_lock);
 
+	/* We need io_uring to clean its registered files, ignore all io_uring
+	 * originated skbs. It's fine as io_uring doesn't keep references to
+	 * other io_uring instances and so killing all other files in the cycle
+	 * will put all io_uring references forcing it to go through normal
+	 * release.path eventually putting registered files.
+	 */
+	skb_queue_walk_safe(&hitlist, skb, next_skb) {
+		if (skb->scm_io_uring) {
+			__skb_unlink(skb, &hitlist);
+			skb_queue_tail(&skb->sk->sk_receive_queue, skb);
+		}
+	}
+
 	/* Here we are. Hitlist is filled. Die. */
 	__skb_queue_purge(&hitlist);
 
 	spin_lock(&unix_gc_lock);
 
+	/* There could be io_uring registered files, just push them back to
+	 * the inflight list
+	 */
+	list_for_each_entry_safe(u, next, &gc_candidates, link)
+		list_move_tail(&u->link, &gc_inflight_list);
+
 	/* All candidates should have been detached by now. */
 	BUG_ON(!list_empty(&gc_candidates));