Merge tag 'for-5.19/io_uring-passthrough-2022-05-22' of git://git.kernel.dk/linux-block

Pull io_uring NVMe command passthrough from Jens Axboe:
 "On top of everything else, this adds support for passthrough for
  io_uring.

  The initial feature for this is NVMe passthrough support, which allows
  non-filesystem based IO commands and admin commands.

  To support this, io_uring grows support for SQE and CQE members that
  are twice as big, allowing to pass in a full NVMe command without
  having to copy data around. And to complete with more than just a
  single 32-bit value as the output"

* tag 'for-5.19/io_uring-passthrough-2022-05-22' of git://git.kernel.dk/linux-block: (22 commits)
  io_uring: cleanup handling of the two task_work lists
  nvme: enable uring-passthrough for admin commands
  nvme: helper for uring-passthrough checks
  blk-mq: fix passthrough plugging
  nvme: add vectored-io support for uring-cmd
  nvme: wire-up uring-cmd support for io-passthru on char-device.
  nvme: refactor nvme_submit_user_cmd()
  block: wire-up support for passthrough plugging
  fs,io_uring: add infrastructure for uring-cmd
  io_uring: support CQE32 for nop operation
  io_uring: enable CQE32
  io_uring: support CQE32 in /proc info
  io_uring: add tracing for additional CQE32 fields
  io_uring: overflow processing for CQE32
  io_uring: flush completions for CQE32
  io_uring: modify io_get_cqe for CQE32
  io_uring: add CQE32 completion processing
  io_uring: add CQE32 setup processing
  io_uring: change ring size calculation for CQE32
  io_uring: store add. return values for CQE32
  ...

block/blk-mq.c

@@ -1169,6 +1169,62 @@ static void blk_end_sync_rq(struct request *rq, blk_status_t error)
 	complete(waiting);
 }
 
+/*
+ * Allow 2x BLK_MAX_REQUEST_COUNT requests on plug queue for multiple
+ * queues. This is important for md arrays to benefit from merging
+ * requests.
+ */
+static inline unsigned short blk_plug_max_rq_count(struct blk_plug *plug)
+{
+	if (plug->multiple_queues)
+		return BLK_MAX_REQUEST_COUNT * 2;
+	return BLK_MAX_REQUEST_COUNT;
+}
+
+static void blk_add_rq_to_plug(struct blk_plug *plug, struct request *rq)
+{
+	struct request *last = rq_list_peek(&plug->mq_list);
+
+	if (!plug->rq_count) {
+		trace_block_plug(rq->q);
+	} else if (plug->rq_count >= blk_plug_max_rq_count(plug) ||
+		   (!blk_queue_nomerges(rq->q) &&
+		    blk_rq_bytes(last) >= BLK_PLUG_FLUSH_SIZE)) {
+		blk_mq_flush_plug_list(plug, false);
+		trace_block_plug(rq->q);
+	}
+
+	if (!plug->multiple_queues && last && last->q != rq->q)
+		plug->multiple_queues = true;
+	if (!plug->has_elevator && (rq->rq_flags & RQF_ELV))
+		plug->has_elevator = true;
+	rq->rq_next = NULL;
+	rq_list_add(&plug->mq_list, rq);
+	plug->rq_count++;
+}
+
+static void __blk_execute_rq_nowait(struct request *rq, bool at_head,
+		rq_end_io_fn *done, bool use_plug)
+{
+	WARN_ON(irqs_disabled());
+	WARN_ON(!blk_rq_is_passthrough(rq));
+
+	rq->end_io = done;
+
+	blk_account_io_start(rq);
+
+	if (use_plug && current->plug) {
+		blk_add_rq_to_plug(current->plug, rq);
+		return;
+	}
+	/*
+	 * don't check dying flag for MQ because the request won't
+	 * be reused after dying flag is set
+	 */
+	blk_mq_sched_insert_request(rq, at_head, true, false);
+}
+
+
 /**
  * blk_execute_rq_nowait - insert a request to I/O scheduler for execution
  * @rq:		request to insert
@@ -1184,18 +1240,8 @@ static void blk_end_sync_rq(struct request *rq, blk_status_t error)
  */
 void blk_execute_rq_nowait(struct request *rq, bool at_head, rq_end_io_fn *done)
 {
-	WARN_ON(irqs_disabled());
-	WARN_ON(!blk_rq_is_passthrough(rq));
+	__blk_execute_rq_nowait(rq, at_head, done, true);
 
-	rq->end_io = done;
-
-	blk_account_io_start(rq);
-
-	/*
-	 * don't check dying flag for MQ because the request won't
-	 * be reused after dying flag is set
-	 */
-	blk_mq_sched_insert_request(rq, at_head, true, false);
 }
 EXPORT_SYMBOL_GPL(blk_execute_rq_nowait);
 
@@ -1233,8 +1279,13 @@ blk_status_t blk_execute_rq(struct request *rq, bool at_head)
 	DECLARE_COMPLETION_ONSTACK(wait);
 	unsigned long hang_check;
 
+	/*
+	 * iopoll requires request to be submitted to driver, so can't
+	 * use plug
+	 */
 	rq->end_io_data = &wait;
-	blk_execute_rq_nowait(rq, at_head, blk_end_sync_rq);
+	__blk_execute_rq_nowait(rq, at_head, blk_end_sync_rq,
+			!blk_rq_is_poll(rq));
 
 	/* Prevent hang_check timer from firing at us during very long I/O */
 	hang_check = sysctl_hung_task_timeout_secs;
@@ -2676,40 +2727,6 @@ void blk_mq_try_issue_list_directly(struct blk_mq_hw_ctx *hctx,
 		hctx->queue->mq_ops->commit_rqs(hctx);
 }
 
-/*
- * Allow 2x BLK_MAX_REQUEST_COUNT requests on plug queue for multiple
- * queues. This is important for md arrays to benefit from merging
- * requests.
- */
-static inline unsigned short blk_plug_max_rq_count(struct blk_plug *plug)
-{
-	if (plug->multiple_queues)
-		return BLK_MAX_REQUEST_COUNT * 2;
-	return BLK_MAX_REQUEST_COUNT;
-}
-
-static void blk_add_rq_to_plug(struct blk_plug *plug, struct request *rq)
-{
-	struct request *last = rq_list_peek(&plug->mq_list);
-
-	if (!plug->rq_count) {
-		trace_block_plug(rq->q);
-	} else if (plug->rq_count >= blk_plug_max_rq_count(plug) ||
-		   (!blk_queue_nomerges(rq->q) &&
-		    blk_rq_bytes(last) >= BLK_PLUG_FLUSH_SIZE)) {
-		blk_mq_flush_plug_list(plug, false);
-		trace_block_plug(rq->q);
-	}
-
-	if (!plug->multiple_queues && last && last->q != rq->q)
-		plug->multiple_queues = true;
-	if (!plug->has_elevator && (rq->rq_flags & RQF_ELV))
-		plug->has_elevator = true;
-	rq->rq_next = NULL;
-	rq_list_add(&plug->mq_list, rq);
-	plug->rq_count++;
-}
-
 static bool blk_mq_attempt_bio_merge(struct request_queue *q,
 				     struct bio *bio, unsigned int nr_segs)
 {

drivers/nvme/host/core.c

@@ -3146,6 +3146,7 @@ static const struct file_operations nvme_dev_fops = {
 	.release	= nvme_dev_release,
 	.unlocked_ioctl	= nvme_dev_ioctl,
 	.compat_ioctl	= compat_ptr_ioctl,
+	.uring_cmd	= nvme_dev_uring_cmd,
 };
 
 static ssize_t nvme_sysfs_reset(struct device *dev,
@@ -3699,6 +3700,7 @@ static const struct file_operations nvme_ns_chr_fops = {
 	.release	= nvme_ns_chr_release,
 	.unlocked_ioctl	= nvme_ns_chr_ioctl,
 	.compat_ioctl	= compat_ptr_ioctl,
+	.uring_cmd	= nvme_ns_chr_uring_cmd,
 };
 
 static int nvme_add_ns_cdev(struct nvme_ns *ns)

drivers/nvme/host/ioctl.c

@@ -5,6 +5,7 @@
  */
 #include <linux/ptrace.h>	/* for force_successful_syscall_return */
 #include <linux/nvme_ioctl.h>
+#include <linux/io_uring.h>
 #include "nvme.h"
 
 /*
@@ -53,10 +54,21 @@ static void *nvme_add_user_metadata(struct bio *bio, void __user *ubuf,
 	return ERR_PTR(ret);
 }
 
-static int nvme_submit_user_cmd(struct request_queue *q,
+static int nvme_finish_user_metadata(struct request *req, void __user *ubuf,
+		void *meta, unsigned len, int ret)
+{
+	if (!ret && req_op(req) == REQ_OP_DRV_IN &&
+	    copy_to_user(ubuf, meta, len))
+		ret = -EFAULT;
+	kfree(meta);
+	return ret;
+}
+
+static struct request *nvme_alloc_user_request(struct request_queue *q,
 		struct nvme_command *cmd, void __user *ubuffer,
 		unsigned bufflen, void __user *meta_buffer, unsigned meta_len,
-		u32 meta_seed, u64 *result, unsigned timeout, bool vec)
+		u32 meta_seed, void **metap, unsigned timeout, bool vec,
+		unsigned int rq_flags, blk_mq_req_flags_t blk_flags)
 {
 	bool write = nvme_is_write(cmd);
 	struct nvme_ns *ns = q->queuedata;
@@ -66,9 +78,9 @@ static int nvme_submit_user_cmd(struct request_queue *q,
 	void *meta = NULL;
 	int ret;
 
-	req = blk_mq_alloc_request(q, nvme_req_op(cmd), 0);
+	req = blk_mq_alloc_request(q, nvme_req_op(cmd) | rq_flags, blk_flags);
 	if (IS_ERR(req))
-		return PTR_ERR(req);
+		return req;
 	nvme_init_request(req, cmd);
 
 	if (timeout)
@@ -105,26 +117,50 @@ static int nvme_submit_user_cmd(struct request_queue *q,
 				goto out_unmap;
 			}
 			req->cmd_flags |= REQ_INTEGRITY;
+			*metap = meta;
 		}
 	}
 
+	return req;
+
+out_unmap:
+	if (bio)
+		blk_rq_unmap_user(bio);
+out:
+	blk_mq_free_request(req);
+	return ERR_PTR(ret);
+}
+
+static int nvme_submit_user_cmd(struct request_queue *q,
+		struct nvme_command *cmd, void __user *ubuffer,
+		unsigned bufflen, void __user *meta_buffer, unsigned meta_len,
+		u32 meta_seed, u64 *result, unsigned timeout, bool vec)
+{
+	struct request *req;
+	void *meta = NULL;
+	struct bio *bio;
+	int ret;
+
+	req = nvme_alloc_user_request(q, cmd, ubuffer, bufflen, meta_buffer,
+			meta_len, meta_seed, &meta, timeout, vec, 0, 0);
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+
+	bio = req->bio;
+
 	ret = nvme_execute_passthru_rq(req);
+
 	if (result)
 		*result = le64_to_cpu(nvme_req(req)->result.u64);
-	if (meta && !ret && !write) {
-		if (copy_to_user(meta_buffer, meta, meta_len))
-			ret = -EFAULT;
-	}
-	kfree(meta);
- out_unmap:
+	if (meta)
+		ret = nvme_finish_user_metadata(req, meta_buffer, meta,
+						meta_len, ret);
 	if (bio)
 		blk_rq_unmap_user(bio);
- out:
 	blk_mq_free_request(req);
 	return ret;
 }
 
-
 static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio)
 {
 	struct nvme_user_io io;
@@ -296,6 +332,139 @@ static int nvme_user_cmd64(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	return status;
 }
 
+struct nvme_uring_data {
+	__u64	metadata;
+	__u64	addr;
+	__u32	data_len;
+	__u32	metadata_len;
+	__u32	timeout_ms;
+};
+
+/*
+ * This overlays struct io_uring_cmd pdu.
+ * Expect build errors if this grows larger than that.
+ */
+struct nvme_uring_cmd_pdu {
+	union {
+		struct bio *bio;
+		struct request *req;
+	};
+	void *meta; /* kernel-resident buffer */
+	void __user *meta_buffer;
+	u32 meta_len;
+};
+
+static inline struct nvme_uring_cmd_pdu *nvme_uring_cmd_pdu(
+		struct io_uring_cmd *ioucmd)
+{
+	return (struct nvme_uring_cmd_pdu *)&ioucmd->pdu;
+}
+
+static void nvme_uring_task_cb(struct io_uring_cmd *ioucmd)
+{
+	struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd);
+	struct request *req = pdu->req;
+	struct bio *bio = req->bio;
+	int status;
+	u64 result;
+
+	if (nvme_req(req)->flags & NVME_REQ_CANCELLED)
+		status = -EINTR;
+	else
+		status = nvme_req(req)->status;
+
+	result = le64_to_cpu(nvme_req(req)->result.u64);
+
+	if (pdu->meta)
+		status = nvme_finish_user_metadata(req, pdu->meta_buffer,
+					pdu->meta, pdu->meta_len, status);
+	if (bio)
+		blk_rq_unmap_user(bio);
+	blk_mq_free_request(req);
+
+	io_uring_cmd_done(ioucmd, status, result);
+}
+
+static void nvme_uring_cmd_end_io(struct request *req, blk_status_t err)
+{
+	struct io_uring_cmd *ioucmd = req->end_io_data;
+	struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd);
+	/* extract bio before reusing the same field for request */
+	struct bio *bio = pdu->bio;
+
+	pdu->req = req;
+	req->bio = bio;
+	/* this takes care of moving rest of completion-work to task context */
+	io_uring_cmd_complete_in_task(ioucmd, nvme_uring_task_cb);
+}
+
+static int nvme_uring_cmd_io(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
+		struct io_uring_cmd *ioucmd, unsigned int issue_flags, bool vec)
+{
+	struct nvme_uring_cmd_pdu *pdu = nvme_uring_cmd_pdu(ioucmd);
+	const struct nvme_uring_cmd *cmd = ioucmd->cmd;
+	struct request_queue *q = ns ? ns->queue : ctrl->admin_q;
+	struct nvme_uring_data d;
+	struct nvme_command c;
+	struct request *req;
+	unsigned int rq_flags = 0;
+	blk_mq_req_flags_t blk_flags = 0;
+	void *meta = NULL;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
+	c.common.opcode = READ_ONCE(cmd->opcode);
+	c.common.flags = READ_ONCE(cmd->flags);
+	if (c.common.flags)
+		return -EINVAL;
+
+	c.common.command_id = 0;
+	c.common.nsid = cpu_to_le32(cmd->nsid);
+	if (!nvme_validate_passthru_nsid(ctrl, ns, le32_to_cpu(c.common.nsid)))
+		return -EINVAL;
+
+	c.common.cdw2[0] = cpu_to_le32(READ_ONCE(cmd->cdw2));
+	c.common.cdw2[1] = cpu_to_le32(READ_ONCE(cmd->cdw3));
+	c.common.metadata = 0;
+	c.common.dptr.prp1 = c.common.dptr.prp2 = 0;
+	c.common.cdw10 = cpu_to_le32(READ_ONCE(cmd->cdw10));
+	c.common.cdw11 = cpu_to_le32(READ_ONCE(cmd->cdw11));
+	c.common.cdw12 = cpu_to_le32(READ_ONCE(cmd->cdw12));
+	c.common.cdw13 = cpu_to_le32(READ_ONCE(cmd->cdw13));
+	c.common.cdw14 = cpu_to_le32(READ_ONCE(cmd->cdw14));
+	c.common.cdw15 = cpu_to_le32(READ_ONCE(cmd->cdw15));
+
+	d.metadata = READ_ONCE(cmd->metadata);
+	d.addr = READ_ONCE(cmd->addr);
+	d.data_len = READ_ONCE(cmd->data_len);
+	d.metadata_len = READ_ONCE(cmd->metadata_len);
+	d.timeout_ms = READ_ONCE(cmd->timeout_ms);
+
+	if (issue_flags & IO_URING_F_NONBLOCK) {
+		rq_flags = REQ_NOWAIT;
+		blk_flags = BLK_MQ_REQ_NOWAIT;
+	}
+
+	req = nvme_alloc_user_request(q, &c, nvme_to_user_ptr(d.addr),
+			d.data_len, nvme_to_user_ptr(d.metadata),
+			d.metadata_len, 0, &meta, d.timeout_ms ?
+			msecs_to_jiffies(d.timeout_ms) : 0, vec, rq_flags,
+			blk_flags);
+	if (IS_ERR(req))
+		return PTR_ERR(req);
+	req->end_io_data = ioucmd;
+
+	/* to free bio on completion, as req->bio will be null at that time */
+	pdu->bio = req->bio;
+	pdu->meta = meta;
+	pdu->meta_buffer = nvme_to_user_ptr(d.metadata);
+	pdu->meta_len = d.metadata_len;
+
+	blk_execute_rq_nowait(req, 0, nvme_uring_cmd_end_io);
+	return -EIOCBQUEUED;
+}
+
 static bool is_ctrl_ioctl(unsigned int cmd)
 {
 	if (cmd == NVME_IOCTL_ADMIN_CMD || cmd == NVME_IOCTL_ADMIN64_CMD)
@@ -387,6 +556,53 @@ long nvme_ns_chr_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	return __nvme_ioctl(ns, cmd, (void __user *)arg);
 }
 
+static int nvme_uring_cmd_checks(unsigned int issue_flags)
+{
+	/* IOPOLL not supported yet */
+	if (issue_flags & IO_URING_F_IOPOLL)
+		return -EOPNOTSUPP;
+
+	/* NVMe passthrough requires big SQE/CQE support */
+	if ((issue_flags & (IO_URING_F_SQE128|IO_URING_F_CQE32)) !=
+	    (IO_URING_F_SQE128|IO_URING_F_CQE32))
+		return -EOPNOTSUPP;
+	return 0;
+}
+
+static int nvme_ns_uring_cmd(struct nvme_ns *ns, struct io_uring_cmd *ioucmd,
+			     unsigned int issue_flags)
+{
+	struct nvme_ctrl *ctrl = ns->ctrl;
+	int ret;
+
+	BUILD_BUG_ON(sizeof(struct nvme_uring_cmd_pdu) > sizeof(ioucmd->pdu));
+
+	ret = nvme_uring_cmd_checks(issue_flags);
+	if (ret)
+		return ret;
+
+	switch (ioucmd->cmd_op) {
+	case NVME_URING_CMD_IO:
+		ret = nvme_uring_cmd_io(ctrl, ns, ioucmd, issue_flags, false);
+		break;
+	case NVME_URING_CMD_IO_VEC:
+		ret = nvme_uring_cmd_io(ctrl, ns, ioucmd, issue_flags, true);
+		break;
+	default:
+		ret = -ENOTTY;
+	}
+
+	return ret;
+}
+
+int nvme_ns_chr_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags)
+{
+	struct nvme_ns *ns = container_of(file_inode(ioucmd->file)->i_cdev,
+			struct nvme_ns, cdev);
+
+	return nvme_ns_uring_cmd(ns, ioucmd, issue_flags);
+}
+
 #ifdef CONFIG_NVME_MULTIPATH
 static int nvme_ns_head_ctrl_ioctl(struct nvme_ns *ns, unsigned int cmd,
 		void __user *argp, struct nvme_ns_head *head, int srcu_idx)
@@ -453,8 +669,46 @@ long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd,
 	srcu_read_unlock(&head->srcu, srcu_idx);
 	return ret;
 }
+
+int nvme_ns_head_chr_uring_cmd(struct io_uring_cmd *ioucmd,
+		unsigned int issue_flags)
+{
+	struct cdev *cdev = file_inode(ioucmd->file)->i_cdev;
+	struct nvme_ns_head *head = container_of(cdev, struct nvme_ns_head, cdev);
+	int srcu_idx = srcu_read_lock(&head->srcu);
+	struct nvme_ns *ns = nvme_find_path(head);
+	int ret = -EINVAL;
+
+	if (ns)
+		ret = nvme_ns_uring_cmd(ns, ioucmd, issue_flags);
+	srcu_read_unlock(&head->srcu, srcu_idx);
+	return ret;
+}
 #endif /* CONFIG_NVME_MULTIPATH */
 
+int nvme_dev_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags)
+{
+	struct nvme_ctrl *ctrl = ioucmd->file->private_data;
+	int ret;
+
+	ret = nvme_uring_cmd_checks(issue_flags);
+	if (ret)
+		return ret;
+
+	switch (ioucmd->cmd_op) {
+	case NVME_URING_CMD_ADMIN:
+		ret = nvme_uring_cmd_io(ctrl, NULL, ioucmd, issue_flags, false);
+		break;
+	case NVME_URING_CMD_ADMIN_VEC:
+		ret = nvme_uring_cmd_io(ctrl, NULL, ioucmd, issue_flags, true);
+		break;
+	default:
+		ret = -ENOTTY;
+	}
+
+	return ret;
+}
+
 static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp)
 {
 	struct nvme_ns *ns;

drivers/nvme/host/multipath.c

@@ -437,6 +437,7 @@ static const struct file_operations nvme_ns_head_chr_fops = {
 	.release	= nvme_ns_head_chr_release,
 	.unlocked_ioctl	= nvme_ns_head_chr_ioctl,
 	.compat_ioctl	= compat_ptr_ioctl,
+	.uring_cmd	= nvme_ns_head_chr_uring_cmd,
 };
 
 static int nvme_add_ns_head_cdev(struct nvme_ns_head *head)

drivers/nvme/host/nvme.h

@@ -782,7 +782,12 @@ long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd,
 		unsigned long arg);
 long nvme_dev_ioctl(struct file *file, unsigned int cmd,
 		unsigned long arg);
+int nvme_ns_chr_uring_cmd(struct io_uring_cmd *ioucmd,
+		unsigned int issue_flags);
+int nvme_ns_head_chr_uring_cmd(struct io_uring_cmd *ioucmd,
+		unsigned int issue_flags);
 int nvme_getgeo(struct block_device *bdev, struct hd_geometry *geo);
+int nvme_dev_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags);
 
 extern const struct attribute_group *nvme_ns_id_attr_groups[];
 extern const struct pr_ops nvme_pr_ops;

include/linux/fs.h

@@ -1953,6 +1953,7 @@ struct dir_context {
 #define REMAP_FILE_ADVISORY		(REMAP_FILE_CAN_SHORTEN)
 
 struct iov_iter;
+struct io_uring_cmd;
 
 struct file_operations {
 	struct module *owner;
@@ -1995,6 +1996,7 @@ struct file_operations {
 				   struct file *file_out, loff_t pos_out,
 				   loff_t len, unsigned int remap_flags);
 	int (*fadvise)(struct file *, loff_t, loff_t, int);
+	int (*uring_cmd)(struct io_uring_cmd *ioucmd, unsigned int issue_flags);
 } __randomize_layout;
 
 struct inode_operations {

include/linux/io_uring.h

@@ -5,7 +5,32 @@
 #include <linux/sched.h>
 #include <linux/xarray.h>
 
+enum io_uring_cmd_flags {
+	IO_URING_F_COMPLETE_DEFER	= 1,
+	IO_URING_F_UNLOCKED		= 2,
+	/* int's last bit, sign checks are usually faster than a bit test */
+	IO_URING_F_NONBLOCK		= INT_MIN,
+
+	/* ctx state flags, for URING_CMD */
+	IO_URING_F_SQE128		= 4,
+	IO_URING_F_CQE32		= 8,
+	IO_URING_F_IOPOLL		= 16,
+};
+
+struct io_uring_cmd {
+	struct file	*file;
+	const void	*cmd;
+	/* callback to defer completions to task context */
+	void (*task_work_cb)(struct io_uring_cmd *cmd);
+	u32		cmd_op;
+	u32		pad;
+	u8		pdu[32]; /* available inline for free use */
+};
+
 #if defined(CONFIG_IO_URING)
+void io_uring_cmd_done(struct io_uring_cmd *cmd, ssize_t ret, ssize_t res2);
+void io_uring_cmd_complete_in_task(struct io_uring_cmd *ioucmd,
+			void (*task_work_cb)(struct io_uring_cmd *));
 struct sock *io_uring_get_socket(struct file *file);
 void __io_uring_cancel(bool cancel_all);
 void __io_uring_free(struct task_struct *tsk);
@@ -30,6 +55,14 @@ static inline void io_uring_free(struct task_struct *tsk)
 		__io_uring_free(tsk);
 }
 #else
+static inline void io_uring_cmd_done(struct io_uring_cmd *cmd, ssize_t ret,
+		ssize_t ret2)
+{
+}
+static inline void io_uring_cmd_complete_in_task(struct io_uring_cmd *ioucmd,
+			void (*task_work_cb)(struct io_uring_cmd *))
+{
+}
 static inline struct sock *io_uring_get_socket(struct file *file)
 {
 	return NULL;

include/trace/events/io_uring.h

@@ -321,13 +321,16 @@ TRACE_EVENT(io_uring_fail_link,
  * @user_data:		user data associated with the request
  * @res:		result of the request
  * @cflags:		completion flags
+ * @extra1:		extra 64-bit data for CQE32
+ * @extra2:		extra 64-bit data for CQE32
  *
  */
 TRACE_EVENT(io_uring_complete,
 
-	TP_PROTO(void *ctx, void *req, u64 user_data, int res, unsigned cflags),
+	TP_PROTO(void *ctx, void *req, u64 user_data, int res, unsigned cflags,
+		 u64 extra1, u64 extra2),
 
-	TP_ARGS(ctx, req, user_data, res, cflags),
+	TP_ARGS(ctx, req, user_data, res, cflags, extra1, extra2),
 
 	TP_STRUCT__entry (
 		__field(  void *,	ctx		)
@@ -335,6 +338,8 @@ TRACE_EVENT(io_uring_complete,
 		__field(  u64,		user_data	)
 		__field(  int,		res		)
 		__field(  unsigned,	cflags		)
+		__field(  u64,		extra1		)
+		__field(  u64,		extra2		)
 	),
 
 	TP_fast_assign(
@@ -343,12 +348,17 @@ TRACE_EVENT(io_uring_complete,
 		__entry->user_data	= user_data;
 		__entry->res		= res;
 		__entry->cflags		= cflags;
+		__entry->extra1		= extra1;
+		__entry->extra2		= extra2;
 	),
 
-	TP_printk("ring %p, req %p, user_data 0x%llx, result %d, cflags 0x%x",
+	TP_printk("ring %p, req %p, user_data 0x%llx, result %d, cflags 0x%x "
+		  "extra1 %llu extra2 %llu ",
 		__entry->ctx, __entry->req,
 		__entry->user_data,
-		__entry->res, __entry->cflags)
+		__entry->res, __entry->cflags,
+		(unsigned long long) __entry->extra1,
+		(unsigned long long) __entry->extra2)
 );
 
 /**

include/uapi/linux/io_uring.h

@@ -22,6 +22,7 @@ struct io_uring_sqe {
 	union {
 		__u64	off;	/* offset into file */
 		__u64	addr2;
+		__u32	cmd_op;
 	};
 	union {
 		__u64	addr;	/* pointer to buffer or iovecs */
@@ -61,8 +62,17 @@ struct io_uring_sqe {
 		__s32	splice_fd_in;
 		__u32	file_index;
 	};
-	__u64	addr3;
-	__u64	__pad2[1];
+	union {
+		struct {
+			__u64	addr3;
+			__u64	__pad2[1];
+		};
+		/*
+		 * If the ring is initialized with IORING_SETUP_SQE128, then
+		 * this field is used for 80 bytes of arbitrary command data
+		 */
+		__u8	cmd[0];
+	};
 };
 
 /*
@@ -128,6 +138,9 @@ enum {
  */
 #define IORING_SETUP_TASKRUN_FLAG	(1U << 9)
 
+#define IORING_SETUP_SQE128		(1U << 10) /* SQEs are 128 byte */
+#define IORING_SETUP_CQE32		(1U << 11) /* CQEs are 32 byte */
+
 enum io_uring_op {
 	IORING_OP_NOP,
 	IORING_OP_READV,
@@ -175,6 +188,7 @@ enum io_uring_op {
 	IORING_OP_FGETXATTR,
 	IORING_OP_GETXATTR,
 	IORING_OP_SOCKET,
+	IORING_OP_URING_CMD,
 
 	/* this goes last, obviously */
 	IORING_OP_LAST,
@@ -251,6 +265,12 @@ struct io_uring_cqe {
 	__u64	user_data;	/* sqe->data submission passed back */
 	__s32	res;		/* result code for this event */
 	__u32	flags;
+
+	/*
+	 * If the ring is initialized with IORING_SETUP_CQE32, then this field
+	 * contains 16-bytes of padding, doubling the size of the CQE.
+	 */
+	__u64 big_cqe[];
 };
 
 /*

include/uapi/linux/nvme_ioctl.h

@@ -70,6 +70,28 @@ struct nvme_passthru_cmd64 {
 	__u64	result;
 };
 
+/* same as struct nvme_passthru_cmd64, minus the 8b result field */
+struct nvme_uring_cmd {
+	__u8	opcode;
+	__u8	flags;
+	__u16	rsvd1;
+	__u32	nsid;
+	__u32	cdw2;
+	__u32	cdw3;
+	__u64	metadata;
+	__u64	addr;
+	__u32	metadata_len;
+	__u32	data_len;
+	__u32	cdw10;
+	__u32	cdw11;
+	__u32	cdw12;
+	__u32	cdw13;
+	__u32	cdw14;
+	__u32	cdw15;
+	__u32	timeout_ms;
+	__u32   rsvd2;
+};
+
 #define nvme_admin_cmd nvme_passthru_cmd
 
 #define NVME_IOCTL_ID		_IO('N', 0x40)
@@ -83,4 +105,10 @@ struct nvme_passthru_cmd64 {
 #define NVME_IOCTL_IO64_CMD	_IOWR('N', 0x48, struct nvme_passthru_cmd64)
 #define NVME_IOCTL_IO64_CMD_VEC	_IOWR('N', 0x49, struct nvme_passthru_cmd64)
 
+/* io_uring async commands: */
+#define NVME_URING_CMD_IO	_IOWR('N', 0x80, struct nvme_uring_cmd)
+#define NVME_URING_CMD_IO_VEC	_IOWR('N', 0x81, struct nvme_uring_cmd)
+#define NVME_URING_CMD_ADMIN	_IOWR('N', 0x82, struct nvme_uring_cmd)
+#define NVME_URING_CMD_ADMIN_VEC _IOWR('N', 0x83, struct nvme_uring_cmd)
+
 #endif /* _UAPI_LINUX_NVME_IOCTL_H */