Commit c98c70ed authored by Linus Torvalds's avatar Linus Torvalds
Browse files

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
parents 6d84c258 2ec33a6c

include/linux/io_uring_types.h

+0 −5
Original line number Diff line number Diff line
@@ -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

+2 −0
Original line number Diff line number Diff line
@@ -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

+1 −1
Original line number Diff line number Diff line
@@ -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

+21 −12
Original line number Diff line number Diff line
@@ -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) {

			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

+16 −2
Original line number Diff line number Diff line
@@ -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) {