io_uring: import 5.15-stable io_uring

No upstream commit exists.

This imports the io_uring codebase from 5.15.85, wholesale. Changes
from that code base:

- Drop IOCB_ALLOC_CACHE, we don't have that in 5.10.
- Drop MKDIRAT/SYMLINKAT/LINKAT. Would require further VFS backports,
  and we don't support these in 5.10 to begin with.
- sock_from_file() old style calling convention.
- Use compat_get_bitmap() only for CONFIG_COMPAT=y

Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Makefile

@@ -1128,7 +1128,7 @@ export MODORDER := $(extmod-prefix)modules.order
 export MODULES_NSDEPS := $(extmod-prefix)modules.nsdeps
 
 ifeq ($(KBUILD_EXTMOD),)
-core-y		+= kernel/ certs/ mm/ fs/ ipc/ security/ crypto/ block/
+core-y		+= kernel/ certs/ mm/ fs/ ipc/ security/ crypto/ block/ io_uring/
 
 vmlinux-dirs	:= $(patsubst %/,%,$(filter %/, \
 		     $(core-y) $(core-m) $(drivers-y) $(drivers-m) \

fs/Makefile

@@ -32,8 +32,6 @@ obj-$(CONFIG_TIMERFD)		+= timerfd.o
 obj-$(CONFIG_EVENTFD)		+= eventfd.o
 obj-$(CONFIG_USERFAULTFD)	+= userfaultfd.o
 obj-$(CONFIG_AIO)               += aio.o
-obj-$(CONFIG_IO_URING)		+= io_uring.o
-obj-$(CONFIG_IO_WQ)		+= io-wq.o
 obj-$(CONFIG_FS_DAX)		+= dax.o
 obj-$(CONFIG_FS_ENCRYPTION)	+= crypto/
 obj-$(CONFIG_FS_VERITY)		+= verity/

include/linux/io_uring.h

@@ -5,50 +5,20 @@
 #include <linux/sched.h>
 #include <linux/xarray.h>
 
-struct io_identity {
-	struct files_struct		*files;
-	struct mm_struct		*mm;
-#ifdef CONFIG_BLK_CGROUP
-	struct cgroup_subsys_state	*blkcg_css;
-#endif
-	const struct cred		*creds;
-	struct nsproxy			*nsproxy;
-	struct fs_struct		*fs;
-	unsigned long			fsize;
-#ifdef CONFIG_AUDIT
-	kuid_t				loginuid;
-	unsigned int			sessionid;
-#endif
-	refcount_t			count;
-};
-
-struct io_uring_task {
-	/* submission side */
-	struct xarray		xa;
-	struct wait_queue_head	wait;
-	struct file		*last;
-	struct percpu_counter	inflight;
-	struct io_identity	__identity;
-	struct io_identity	*identity;
-	atomic_t		in_idle;
-	bool			sqpoll;
-};
-
 #if defined(CONFIG_IO_URING)
 struct sock *io_uring_get_socket(struct file *file);
-void __io_uring_task_cancel(void);
+void __io_uring_cancel(bool cancel_all);
-void __io_uring_files_cancel(struct files_struct *files);
 void __io_uring_free(struct task_struct *tsk);
 
+static inline void io_uring_files_cancel(void)
+{
+	if (current->io_uring)
+		__io_uring_cancel(false);
+}
 static inline void io_uring_task_cancel(void)
 {
-	if (current->io_uring && !xa_empty(&current->io_uring->xa))
+	if (current->io_uring)
-		__io_uring_task_cancel();
+		__io_uring_cancel(true);
-}
-static inline void io_uring_files_cancel(struct files_struct *files)
-{
-	if (current->io_uring && !xa_empty(&current->io_uring->xa))
-		__io_uring_files_cancel(files);
 }
 static inline void io_uring_free(struct task_struct *tsk)
 {
@@ -63,7 +33,7 @@ static inline struct sock *io_uring_get_socket(struct file *file)
 static inline void io_uring_task_cancel(void)
 {
 }
-static inline void io_uring_files_cancel(struct files_struct *files)
+static inline void io_uring_files_cancel(void)
 {
 }
 static inline void io_uring_free(struct task_struct *tsk)

include/linux/sched.h

@@ -885,6 +885,9 @@ struct task_struct {
 	/* CLONE_CHILD_CLEARTID: */
 	int __user			*clear_child_tid;
 
+	/* PF_IO_WORKER */
+	void				*pf_io_worker;
+
 	u64				utime;
 	u64				stime;
 #ifdef CONFIG_ARCH_HAS_SCALED_CPUTIME

include/linux/syscalls.h

@@ -341,7 +341,7 @@ asmlinkage long sys_io_uring_setup(u32 entries,
 				struct io_uring_params __user *p);
 asmlinkage long sys_io_uring_enter(unsigned int fd, u32 to_submit,
 				u32 min_complete, u32 flags,
-				const sigset_t __user *sig, size_t sigsz);
+				const void __user *argp, size_t argsz);
 asmlinkage long sys_io_uring_register(unsigned int fd, unsigned int op,
 				void __user *arg, unsigned int nr_args);
 

include/trace/events/io_uring.h

@@ -12,11 +12,11 @@ struct io_wq_work;
 /**
  * io_uring_create - called after a new io_uring context was prepared
  *
- * @fd:			corresponding file descriptor
+ * @fd:		corresponding file descriptor
- * @ctx:		pointer to a ring context structure
+ * @ctx:	pointer to a ring context structure
  * @sq_entries:	actual SQ size
  * @cq_entries:	actual CQ size
- * @flags:		SQ ring flags, provided to io_uring_setup(2)
+ * @flags:	SQ ring flags, provided to io_uring_setup(2)
  *
  * Allows to trace io_uring creation and provide pointer to a context, that can
  * be used later to find correlated events.
@@ -49,15 +49,15 @@ TRACE_EVENT(io_uring_create,
 );
 
 /**
- * io_uring_register - called after a buffer/file/eventfd was succesfully
+ * io_uring_register - called after a buffer/file/eventfd was successfully
  * 					   registered for a ring
  *
- * @ctx:			pointer to a ring context structure
+ * @ctx:		pointer to a ring context structure
- * @opcode:			describes which operation to perform
+ * @opcode:		describes which operation to perform
  * @nr_user_files:	number of registered files
  * @nr_user_bufs:	number of registered buffers
  * @cq_ev_fd:		whether eventfs registered or not
- * @ret:			return code
+ * @ret:		return code
  *
  * Allows to trace fixed files/buffers/eventfds, that could be registered to
  * avoid an overhead of getting references to them for every operation. This
@@ -142,16 +142,16 @@ TRACE_EVENT(io_uring_queue_async_work,
 	TP_ARGS(ctx, rw, req, work, flags),
 
 	TP_STRUCT__entry (
-		__field(  void *,				ctx		)
+		__field(  void *,			ctx	)
-		__field(  int,					rw		)
+		__field(  int,				rw	)
-		__field(  void *,				req		)
+		__field(  void *,			req	)
 		__field(  struct io_wq_work *,		work	)
 		__field(  unsigned int,			flags	)
 	),
 
 	TP_fast_assign(
 		__entry->ctx	= ctx;
-		__entry->rw		= rw;
+		__entry->rw	= rw;
 		__entry->req	= req;
 		__entry->work	= work;
 		__entry->flags	= flags;
@@ -196,10 +196,10 @@ TRACE_EVENT(io_uring_defer,
 
 /**
  * io_uring_link - called before the io_uring request added into link_list of
- * 				   another request
+ * 		   another request
  *
- * @ctx:			pointer to a ring context structure
+ * @ctx:		pointer to a ring context structure
- * @req:			pointer to a linked request
+ * @req:		pointer to a linked request
  * @target_req:		pointer to a previous request, that would contain @req
  *
  * Allows to track linked requests, to understand dependencies between requests
@@ -212,8 +212,8 @@ TRACE_EVENT(io_uring_link,
 	TP_ARGS(ctx, req, target_req),
 
 	TP_STRUCT__entry (
-		__field(  void *,	ctx			)
+		__field(  void *,	ctx		)
-		__field(  void *,	req			)
+		__field(  void *,	req		)
 		__field(  void *,	target_req	)
 	),
 
@@ -244,7 +244,7 @@ TRACE_EVENT(io_uring_cqring_wait,
 	TP_ARGS(ctx, min_events),
 
 	TP_STRUCT__entry (
-		__field(  void *,	ctx			)
+		__field(  void *,	ctx		)
 		__field(  int,		min_events	)
 	),
 
@@ -272,7 +272,7 @@ TRACE_EVENT(io_uring_fail_link,
 	TP_ARGS(req, link),
 
 	TP_STRUCT__entry (
-		__field(  void *,	req		)
+		__field(  void *,	req	)
 		__field(  void *,	link	)
 	),
 
@@ -290,38 +290,42 @@ TRACE_EVENT(io_uring_fail_link,
  * @ctx:		pointer to a ring context structure
  * @user_data:		user data associated with the request
  * @res:		result of the request
+ * @cflags:		completion flags
  *
  */
 TRACE_EVENT(io_uring_complete,
 
-	TP_PROTO(void *ctx, u64 user_data, long res),
+	TP_PROTO(void *ctx, u64 user_data, int res, unsigned cflags),
 
-	TP_ARGS(ctx, user_data, res),
+	TP_ARGS(ctx, user_data, res, cflags),
 
 	TP_STRUCT__entry (
 		__field(  void *,	ctx		)
 		__field(  u64,		user_data	)
-		__field(  long,		res		)
+		__field(  int,		res		)
+		__field(  unsigned,	cflags		)
 	),
 
 	TP_fast_assign(
 		__entry->ctx		= ctx;
 		__entry->user_data	= user_data;
 		__entry->res		= res;
+		__entry->cflags		= cflags;
 	),
 
-	TP_printk("ring %p, user_data 0x%llx, result %ld",
+	TP_printk("ring %p, user_data 0x%llx, result %d, cflags %x",
 			  __entry->ctx, (unsigned long long)__entry->user_data,
-			  __entry->res)
+			  __entry->res, __entry->cflags)
 );
 
-
 /**
  * io_uring_submit_sqe - called before submitting one SQE
  *
  * @ctx:		pointer to a ring context structure
+ * @req:		pointer to a submitted request
  * @opcode:		opcode of request
  * @user_data:		user data associated with the request
+ * @flags		request flags
  * @force_nonblock:	whether a context blocking or not
  * @sq_thread:		true if sq_thread has submitted this SQE
  *
@@ -330,41 +334,60 @@ TRACE_EVENT(io_uring_complete,
  */
 TRACE_EVENT(io_uring_submit_sqe,
 
-	TP_PROTO(void *ctx, u8 opcode, u64 user_data, bool force_nonblock,
+	TP_PROTO(void *ctx, void *req, u8 opcode, u64 user_data, u32 flags,
-		 bool sq_thread),
+		 bool force_nonblock, bool sq_thread),
 
-	TP_ARGS(ctx, opcode, user_data, force_nonblock, sq_thread),
+	TP_ARGS(ctx, req, opcode, user_data, flags, force_nonblock, sq_thread),
 
 	TP_STRUCT__entry (
 		__field(  void *,	ctx		)
+		__field(  void *,	req		)
 		__field(  u8,		opcode		)
 		__field(  u64,		user_data	)
+		__field(  u32,		flags		)
 		__field(  bool,		force_nonblock	)
 		__field(  bool,		sq_thread	)
 	),
 
 	TP_fast_assign(
 		__entry->ctx		= ctx;
+		__entry->req		= req;
 		__entry->opcode		= opcode;
 		__entry->user_data	= user_data;
+		__entry->flags		= flags;
 		__entry->force_nonblock	= force_nonblock;
 		__entry->sq_thread	= sq_thread;
 	),
 
-	TP_printk("ring %p, op %d, data 0x%llx, non block %d, sq_thread %d",
+	TP_printk("ring %p, req %p, op %d, data 0x%llx, flags %u, "
-			  __entry->ctx, __entry->opcode,
+		  "non block %d, sq_thread %d", __entry->ctx, __entry->req,
-			  (unsigned long long) __entry->user_data,
+		  __entry->opcode, (unsigned long long)__entry->user_data,
-			  __entry->force_nonblock, __entry->sq_thread)
+		  __entry->flags, __entry->force_nonblock, __entry->sq_thread)
 );
 
+/*
+ * io_uring_poll_arm - called after arming a poll wait if successful
+ *
+ * @ctx:		pointer to a ring context structure
+ * @req:		pointer to the armed request
+ * @opcode:		opcode of request
+ * @user_data:		user data associated with the request
+ * @mask:		request poll events mask
+ * @events:		registered events of interest
+ *
+ * Allows to track which fds are waiting for and what are the events of
+ * interest.
+ */
 TRACE_EVENT(io_uring_poll_arm,
 
-	TP_PROTO(void *ctx, u8 opcode, u64 user_data, int mask, int events),
+	TP_PROTO(void *ctx, void *req, u8 opcode, u64 user_data,
+		 int mask, int events),
 
-	TP_ARGS(ctx, opcode, user_data, mask, events),
+	TP_ARGS(ctx, req, opcode, user_data, mask, events),
 
 	TP_STRUCT__entry (
 		__field(  void *,	ctx		)
+		__field(  void *,	req		)
 		__field(  u8,		opcode		)
 		__field(  u64,		user_data	)
 		__field(  int,		mask		)
@@ -373,16 +396,17 @@ TRACE_EVENT(io_uring_poll_arm,
 
 	TP_fast_assign(
 		__entry->ctx		= ctx;
+		__entry->req		= req;
 		__entry->opcode		= opcode;
 		__entry->user_data	= user_data;
 		__entry->mask		= mask;
 		__entry->events		= events;
 	),
 
-	TP_printk("ring %p, op %d, data 0x%llx, mask 0x%x, events 0x%x",
+	TP_printk("ring %p, req %p, op %d, data 0x%llx, mask 0x%x, events 0x%x",
-			  __entry->ctx, __entry->opcode,
+		  __entry->ctx, __entry->req, __entry->opcode,
-			  (unsigned long long) __entry->user_data,
+		  (unsigned long long) __entry->user_data,
-			  __entry->mask, __entry->events)
+		  __entry->mask, __entry->events)
 );
 
 TRACE_EVENT(io_uring_poll_wake,
@@ -437,27 +461,40 @@ TRACE_EVENT(io_uring_task_add,
 			  __entry->mask)
 );
 
+/*
+ * io_uring_task_run - called when task_work_run() executes the poll events
+ *                     notification callbacks
+ *
+ * @ctx:		pointer to a ring context structure
+ * @req:		pointer to the armed request
+ * @opcode:		opcode of request
+ * @user_data:		user data associated with the request
+ *
+ * Allows to track when notified poll events are processed
+ */
 TRACE_EVENT(io_uring_task_run,
 
-	TP_PROTO(void *ctx, u8 opcode, u64 user_data),
+	TP_PROTO(void *ctx, void *req, u8 opcode, u64 user_data),
 
-	TP_ARGS(ctx, opcode, user_data),
+	TP_ARGS(ctx, req, opcode, user_data),
 
 	TP_STRUCT__entry (
 		__field(  void *,	ctx		)
+		__field(  void *,	req		)
 		__field(  u8,		opcode		)
 		__field(  u64,		user_data	)
 	),
 
 	TP_fast_assign(
 		__entry->ctx		= ctx;
+		__entry->req		= req;
 		__entry->opcode		= opcode;
 		__entry->user_data	= user_data;
 	),
 
-	TP_printk("ring %p, op %d, data 0x%llx",
+	TP_printk("ring %p, req %p, op %d, data 0x%llx",
-			  __entry->ctx, __entry->opcode,
+		  __entry->ctx, __entry->req, __entry->opcode,
-			  (unsigned long long) __entry->user_data)
+		  (unsigned long long) __entry->user_data)
 );
 
 #endif /* _TRACE_IO_URING_H */

include/uapi/linux/io_uring.h

@@ -42,23 +42,25 @@ struct io_uring_sqe {
 		__u32		statx_flags;
 		__u32		fadvise_advice;
 		__u32		splice_flags;
+		__u32		rename_flags;
+		__u32		unlink_flags;
+		__u32		hardlink_flags;
 	};
 	__u64	user_data;	/* data to be passed back at completion time */
+	/* pack this to avoid bogus arm OABI complaints */
 	union {
-		struct {
+		/* index into fixed buffers, if used */
-			/* pack this to avoid bogus arm OABI complaints */
+		__u16	buf_index;
-			union {
+		/* for grouped buffer selection */
-				/* index into fixed buffers, if used */
+		__u16	buf_group;
-				__u16	buf_index;
+	} __attribute__((packed));
-				/* for grouped buffer selection */
+	/* personality to use, if used */
-				__u16	buf_group;
+	__u16	personality;
-			} __attribute__((packed));
+	union {
-			/* personality to use, if used */
+		__s32	splice_fd_in;
-			__u16	personality;
+		__u32	file_index;
-			__s32	splice_fd_in;
-		};
-		__u64	__pad2[3];
 	};
+	__u64	__pad2[2];
 };
 
 enum {
@@ -132,6 +134,9 @@ enum {
 	IORING_OP_PROVIDE_BUFFERS,
 	IORING_OP_REMOVE_BUFFERS,
 	IORING_OP_TEE,
+	IORING_OP_SHUTDOWN,
+	IORING_OP_RENAMEAT,
+	IORING_OP_UNLINKAT,
 
 	/* this goes last, obviously */
 	IORING_OP_LAST,
@@ -145,14 +150,34 @@ enum {
 /*
  * sqe->timeout_flags
  */
-#define IORING_TIMEOUT_ABS	(1U << 0)
+#define IORING_TIMEOUT_ABS		(1U << 0)
-
+#define IORING_TIMEOUT_UPDATE		(1U << 1)
+#define IORING_TIMEOUT_BOOTTIME		(1U << 2)
+#define IORING_TIMEOUT_REALTIME		(1U << 3)
+#define IORING_LINK_TIMEOUT_UPDATE	(1U << 4)
+#define IORING_TIMEOUT_CLOCK_MASK	(IORING_TIMEOUT_BOOTTIME | IORING_TIMEOUT_REALTIME)
+#define IORING_TIMEOUT_UPDATE_MASK	(IORING_TIMEOUT_UPDATE | IORING_LINK_TIMEOUT_UPDATE)
 /*
  * sqe->splice_flags
  * extends splice(2) flags
  */
 #define SPLICE_F_FD_IN_FIXED	(1U << 31) /* the last bit of __u32 */
 
+/*
+ * POLL_ADD flags. Note that since sqe->poll_events is the flag space, the
+ * command flags for POLL_ADD are stored in sqe->len.
+ *
+ * IORING_POLL_ADD_MULTI	Multishot poll. Sets IORING_CQE_F_MORE if
+ *				the poll handler will continue to report
+ *				CQEs on behalf of the same SQE.
+ *
+ * IORING_POLL_UPDATE		Update existing poll request, matching
+ *				sqe->addr as the old user_data field.
+ */
+#define IORING_POLL_ADD_MULTI	(1U << 0)
+#define IORING_POLL_UPDATE_EVENTS	(1U << 1)
+#define IORING_POLL_UPDATE_USER_DATA	(1U << 2)
+
 /*
  * IO completion data structure (Completion Queue Entry)
  */
@@ -166,8 +191,10 @@ struct io_uring_cqe {
  * cqe->flags
  *
  * IORING_CQE_F_BUFFER	If set, the upper 16 bits are the buffer ID
+ * IORING_CQE_F_MORE	If set, parent SQE will generate more CQE entries
  */
 #define IORING_CQE_F_BUFFER		(1U << 0)
+#define IORING_CQE_F_MORE		(1U << 1)
 
 enum {
 	IORING_CQE_BUFFER_SHIFT		= 16,
@@ -226,6 +253,7 @@ struct io_cqring_offsets {
 #define IORING_ENTER_GETEVENTS	(1U << 0)
 #define IORING_ENTER_SQ_WAKEUP	(1U << 1)
 #define IORING_ENTER_SQ_WAIT	(1U << 2)
+#define IORING_ENTER_EXT_ARG	(1U << 3)
 
 /*
  * Passed in for io_uring_setup(2). Copied back with updated info on success
@@ -253,6 +281,10 @@ struct io_uring_params {
 #define IORING_FEAT_CUR_PERSONALITY	(1U << 4)
 #define IORING_FEAT_FAST_POLL		(1U << 5)
 #define IORING_FEAT_POLL_32BITS 	(1U << 6)
+#define IORING_FEAT_SQPOLL_NONFIXED	(1U << 7)
+#define IORING_FEAT_EXT_ARG		(1U << 8)
+#define IORING_FEAT_NATIVE_WORKERS	(1U << 9)
+#define IORING_FEAT_RSRC_TAGS		(1U << 10)
 
 /*
  * io_uring_register(2) opcodes and arguments
@@ -272,16 +304,62 @@ enum {
 	IORING_REGISTER_RESTRICTIONS		= 11,
 	IORING_REGISTER_ENABLE_RINGS		= 12,
 
+	/* extended with tagging */
+	IORING_REGISTER_FILES2			= 13,
+	IORING_REGISTER_FILES_UPDATE2		= 14,
+	IORING_REGISTER_BUFFERS2		= 15,
+	IORING_REGISTER_BUFFERS_UPDATE		= 16,
+
+	/* set/clear io-wq thread affinities */
+	IORING_REGISTER_IOWQ_AFF		= 17,
+	IORING_UNREGISTER_IOWQ_AFF		= 18,
+
+	/* set/get max number of io-wq workers */
+	IORING_REGISTER_IOWQ_MAX_WORKERS	= 19,
+
 	/* this goes last */
 	IORING_REGISTER_LAST
 };
 
+/* io-wq worker categories */
+enum {
+	IO_WQ_BOUND,
+	IO_WQ_UNBOUND,
+};
+
+/* deprecated, see struct io_uring_rsrc_update */
 struct io_uring_files_update {
 	__u32 offset;
 	__u32 resv;
 	__aligned_u64 /* __s32 * */ fds;
 };
 
+struct io_uring_rsrc_register {
+	__u32 nr;
+	__u32 resv;
+	__u64 resv2;
+	__aligned_u64 data;
+	__aligned_u64 tags;
+};
+
+struct io_uring_rsrc_update {
+	__u32 offset;
+	__u32 resv;
+	__aligned_u64 data;
+};
+
+struct io_uring_rsrc_update2 {
+	__u32 offset;
+	__u32 resv;
+	__aligned_u64 data;
+	__aligned_u64 tags;
+	__u32 nr;
+	__u32 resv2;
+};
+
+/* Skip updating fd indexes set to this value in the fd table */
+#define IORING_REGISTER_FILES_SKIP	(-2)
+
 #define IO_URING_OP_SUPPORTED	(1U << 0)
 
 struct io_uring_probe_op {
@@ -329,4 +407,11 @@ enum {
 	IORING_RESTRICTION_LAST
 };
 
+struct io_uring_getevents_arg {
+	__u64	sigmask;
+	__u32	sigmask_sz;
+	__u32	pad;
+	__u64	ts;
+};
+
 #endif

io_uring/Makefile

@@ -0,0 +1,6 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for io_uring
+
+obj-$(CONFIG_IO_URING)		+= io_uring.o
+obj-$(CONFIG_IO_WQ)		+= io-wq.o

io_uring/io-wq.h

@@ -1,7 +1,7 @@
 #ifndef INTERNAL_IO_WQ_H
 #define INTERNAL_IO_WQ_H
 
-#include <linux/io_uring.h>
+#include <linux/refcount.h>
 
 struct io_wq;
 
@@ -9,16 +9,8 @@ enum {
 	IO_WQ_WORK_CANCEL	= 1,
 	IO_WQ_WORK_HASHED	= 2,
 	IO_WQ_WORK_UNBOUND	= 4,
-	IO_WQ_WORK_NO_CANCEL	= 8,
 	IO_WQ_WORK_CONCURRENT	= 16,
 
-	IO_WQ_WORK_FILES	= 32,
-	IO_WQ_WORK_FS		= 64,
-	IO_WQ_WORK_MM		= 128,
-	IO_WQ_WORK_CREDS	= 256,
-	IO_WQ_WORK_BLKCG	= 512,
-	IO_WQ_WORK_FSIZE	= 1024,
-
 	IO_WQ_HASH_SHIFT	= 24,	/* upper 8 bits are used for hash key */
 };
 
@@ -52,6 +44,7 @@ static inline void wq_list_add_after(struct io_wq_work_node *node,
 static inline void wq_list_add_tail(struct io_wq_work_node *node,
 				    struct io_wq_work_list *list)
 {
+	node->next = NULL;
 	if (!list->first) {
 		list->last = node;
 		WRITE_ONCE(list->first, node);
@@ -59,7 +52,6 @@ static inline void wq_list_add_tail(struct io_wq_work_node *node,
 		list->last->next = node;
 		list->last = node;
 	}
-	node->next = NULL;
 }
 
 static inline void wq_list_cut(struct io_wq_work_list *list,
@@ -95,7 +87,6 @@ static inline void wq_list_del(struct io_wq_work_list *list,
 
 struct io_wq_work {
 	struct io_wq_work_node list;
-	struct io_identity *identity;
 	unsigned flags;
 };
 
@@ -107,37 +98,48 @@ static inline struct io_wq_work *wq_next_work(struct io_wq_work *work)
 	return container_of(work->list.next, struct io_wq_work, list);
 }
 
-typedef void (free_work_fn)(struct io_wq_work *);
+typedef struct io_wq_work *(free_work_fn)(struct io_wq_work *);
-typedef struct io_wq_work *(io_wq_work_fn)(struct io_wq_work *);
+typedef void (io_wq_work_fn)(struct io_wq_work *);
+
+struct io_wq_hash {
+	refcount_t refs;
+	unsigned long map;
+	struct wait_queue_head wait;
+};
+
+static inline void io_wq_put_hash(struct io_wq_hash *hash)
+{
+	if (refcount_dec_and_test(&hash->refs))
+		kfree(hash);
+}
 
 struct io_wq_data {
-	struct user_struct *user;
+	struct io_wq_hash *hash;
-
+	struct task_struct *task;
 	io_wq_work_fn *do_work;
 	free_work_fn *free_work;
 };
 
 struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data);
-bool io_wq_get(struct io_wq *wq, struct io_wq_data *data);
+void io_wq_exit_start(struct io_wq *wq);
-void io_wq_destroy(struct io_wq *wq);
+void io_wq_put_and_exit(struct io_wq *wq);
 
 void io_wq_enqueue(struct io_wq *wq, struct io_wq_work *work);
 void io_wq_hash_work(struct io_wq_work *work, void *val);
 
+int io_wq_cpu_affinity(struct io_wq *wq, cpumask_var_t mask);
+int io_wq_max_workers(struct io_wq *wq, int *new_count);
+
 static inline bool io_wq_is_hashed(struct io_wq_work *work)
 {
 	return work->flags & IO_WQ_WORK_HASHED;
 }
 
-void io_wq_cancel_all(struct io_wq *wq);
-
 typedef bool (work_cancel_fn)(struct io_wq_work *, void *);
 
 enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel,
 					void *data, bool cancel_all);
 
-struct task_struct *io_wq_get_task(struct io_wq *wq);
-
 #if defined(CONFIG_IO_WQ)
 extern void io_wq_worker_sleeping(struct task_struct *);
 extern void io_wq_worker_running(struct task_struct *);
@@ -152,6 +154,7 @@ static inline void io_wq_worker_running(struct task_struct *tsk)
 
 static inline bool io_wq_current_is_worker(void)
 {
-	return in_task() && (current->flags & PF_IO_WORKER);
+	return in_task() && (current->flags & PF_IO_WORKER) &&
+		current->pf_io_worker;
 }
 #endif

kernel/exit.c

@@ -763,7 +763,7 @@ void __noreturn do_exit(long code)
 		schedule();
 	}
 
-	io_uring_files_cancel(tsk->files);
+	io_uring_files_cancel();
 	exit_signals(tsk);  /* sets PF_EXITING */
 
 	/* sync mm's RSS info before statistics gathering */

kernel/fork.c

@@ -926,6 +926,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node)
 	tsk->splice_pipe = NULL;
 	tsk->task_frag.page = NULL;
 	tsk->wake_q.next = NULL;
+	tsk->pf_io_worker = NULL;
 
 	account_kernel_stack(tsk, 1);
 

kernel/sched/core.c

@@ -21,7 +21,7 @@
 #include <asm/tlb.h>
 
 #include "../workqueue_internal.h"
-#include "../../fs/io-wq.h"
+#include "../../io_uring/io-wq.h"
 #include "../smpboot.h"
 
 #include "pelt.h"