Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted

Pull networking updates from Jakub Kicinski:
 "Core:

   - Add dedicated kmem_cache for typical/small skb->head, avoid having
     to access struct page at kfree time, and improve memory use.

   - Introduce sysctl to set default RPS configuration for new netdevs.

   - Define Netlink protocol specification format which can be used to
     describe messages used by each family and auto-generate parsers.
     Add tools for generating kernel data structures and uAPI headers.

   - Expose all net/core sysctls inside netns.

   - Remove 4s sleep in netpoll if carrier is instantly detected on
     boot.

   - Add configurable limit of MDB entries per port, and port-vlan.

   - Continue populating drop reasons throughout the stack.

   - Retire a handful of legacy Qdiscs and classifiers.

  Protocols:

   - Support IPv4 big TCP (TSO frames larger than 64kB).

   - Add IP_LOCAL_PORT_RANGE socket option, to control local port range
     on socket by socket basis.

   - Track and report in procfs number of MPTCP sockets used.

   - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path
     manager.

   - IPv6: don't check net.ipv6.route.max_size and rely on garbage
     collection to free memory (similarly to IPv4).

   - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986).

   - ICMP: add per-rate limit counters.

   - Add support for user scanning requests in ieee802154.

   - Remove static WEP support.

   - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate
     reporting.

   - WiFi 7 EHT channel puncturing support (client & AP).

  BPF:

   - Add a rbtree data structure following the "next-gen data structure"
     precedent set by recently added linked list, that is, by using
     kfunc + kptr instead of adding a new BPF map type.

   - Expose XDP hints via kfuncs with initial support for RX hash and
     timestamp metadata.

   - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to
     better support decap on GRE tunnel devices not operating in collect
     metadata.

   - Improve x86 JIT's codegen for PROBE_MEM runtime error checks.

   - Remove the need for trace_printk_lock for bpf_trace_printk and
     bpf_trace_vprintk helpers.

   - Extend libbpf's bpf_tracing.h support for tracing arguments of
     kprobes/uprobes and syscall as a special case.

   - Significantly reduce the search time for module symbols by
     livepatch and BPF.

   - Enable cpumasks to be used as kptrs, which is useful for tracing
     programs tracking which tasks end up running on which CPUs in
     different time intervals.

   - Add support for BPF trampoline on s390x and riscv64.

   - Add capability to export the XDP features supported by the NIC.

   - Add __bpf_kfunc tag for marking kernel functions as kfuncs.

   - Add cgroup.memory=nobpf kernel parameter option to disable BPF
     memory accounting for container environments.

  Netfilter:

   - Remove the CLUSTERIP target. It has been marked as obsolete for
     years, and we still have WARN splats wrt races of the out-of-band
     /proc interface installed by this target.

   - Add 'destroy' commands to nf_tables. They are identical to the
     existing 'delete' commands, but do not return an error if the
     referenced object (set, chain, rule...) did not exist.

  Driver API:

   - Improve cpumask_local_spread() locality to help NICs set the right
     IRQ affinity on AMD platforms.

   - Separate C22 and C45 MDIO bus transactions more clearly.

   - Introduce new DCB table to control DSCP rewrite on egress.

   - Support configuration of Physical Layer Collision Avoidance (PLCA)
     Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of
     shared medium Ethernet.

   - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing
     preemption of low priority frames by high priority frames.

   - Add support for controlling MACSec offload using netlink SET.

   - Rework devlink instance refcounts to allow registration and
     de-registration under the instance lock. Split the code into
     multiple files, drop some of the unnecessarily granular locks and
     factor out common parts of netlink operation handling.

   - Add TX frame aggregation parameters (for USB drivers).

   - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning
     messages with notifications for debug.

   - Allow offloading of UDP NEW connections via act_ct.

   - Add support for per action HW stats in TC.

   - Support hardware miss to TC action (continue processing in SW from
     a specific point in the action chain).

   - Warn if old Wireless Extension user space interface is used with
     modern cfg80211/mac80211 drivers. Do not support Wireless
     Extensions for Wi-Fi 7 devices at all. Everyone should switch to
     using nl80211 interface instead.

   - Improve the CAN bit timing configuration. Use extack to return
     error messages directly to user space, update the SJW handling,
     including the definition of a new default value that will benefit
     CAN-FD controllers, by increasing their oscillator tolerance.

  New hardware / drivers:

   - Ethernet:
      - nVidia BlueField-3 support (control traffic driver)
      - Ethernet support for imx93 SoCs
      - Motorcomm yt8531 gigabit Ethernet PHY
      - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA)
      - Microchip LAN8841 PHY (incl. cable diagnostics and PTP)
      - Amlogic gxl MDIO mux

   - WiFi:
      - RealTek RTL8188EU (rtl8xxxu)
      - Qualcomm Wi-Fi 7 devices (ath12k)

   - CAN:
      - Renesas R-Car V4H

  Drivers:

   - Bluetooth:
      - Set Per Platform Antenna Gain (PPAG) for Intel controllers.

   - Ethernet NICs:
      - Intel (1G, igc):
         - support TSN / Qbv / packet scheduling features of i226 model
      - Intel (100G, ice):
         - use GNSS subsystem instead of TTY
         - multi-buffer XDP support
         - extend support for GPIO pins to E823 devices
      - nVidia/Mellanox:
         - update the shared buffer configuration on PFC commands
         - implement PTP adjphase function for HW offset control
         - TC support for Geneve and GRE with VF tunnel offload
         - more efficient crypto key management method
         - multi-port eswitch support
      - Netronome/Corigine:
         - add DCB IEEE support
         - support IPsec offloading for NFP3800
      - Freescale/NXP (enetc):
         - support XDP_REDIRECT for XDP non-linear buffers
         - improve reconfig, avoid link flap and waiting for idle
         - support MAC Merge layer
      - Other NICs:
         - sfc/ef100: add basic devlink support for ef100
         - ionic: rx_push mode operation (writing descriptors via MMIO)
         - bnxt: use the auxiliary bus abstraction for RDMA
         - r8169: disable ASPM and reset bus in case of tx timeout
         - cpsw: support QSGMII mode for J721e CPSW9G
         - cpts: support pulse-per-second output
         - ngbe: add an mdio bus driver
         - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing
         - r8152: handle devices with FW with NCM support
         - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation
         - virtio-net: support multi buffer XDP
         - virtio/vsock: replace virtio_vsock_pkt with sk_buff
         - tsnep: XDP support

   - Ethernet high-speed switches:
      - nVidia/Mellanox (mlxsw):
         - add support for latency TLV (in FW control messages)
      - Microchip (sparx5):
         - separate explicit and implicit traffic forwarding rules, make
           the implicit rules always active
         - add support for egress DSCP rewrite
         - IS0 VCAP support (Ingress Classification)
         - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS
           etc.)
         - ES2 VCAP support (Egress Access Control)
         - support for Per-Stream Filtering and Policing (802.1Q,
           8.6.5.1)

   - Ethernet embedded switches:
      - Marvell (mv88e6xxx):
         - add MAB (port auth) offload support
         - enable PTP receive for mv88e6390
      - NXP (ocelot):
         - support MAC Merge layer
         - support for the the vsc7512 internal copper phys
      - Microchip:
         - lan9303: convert to PHYLINK
         - lan966x: support TC flower filter statistics
         - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x
         - lan937x: support Credit Based Shaper configuration
         - ksz9477: support Energy Efficient Ethernet
      - other:
         - qca8k: convert to regmap read/write API, use bulk operations
         - rswitch: Improve TX timestamp accuracy

   - Intel WiFi (iwlwifi):
      - EHT (Wi-Fi 7) rate reporting
      - STEP equalizer support: transfer some STEP (connection to radio
        on platforms with integrated wifi) related parameters from the
        BIOS to the firmware.

   - Qualcomm 802.11ax WiFi (ath11k):
      - IPQ5018 support
      - Fine Timing Measurement (FTM) responder role support
      - channel 177 support

   - MediaTek WiFi (mt76):
      - per-PHY LED support
      - mt7996: EHT (Wi-Fi 7) support
      - Wireless Ethernet Dispatch (WED) reset support
      - switch to using page pool allocator

   - RealTek WiFi (rtw89):
      - support new version of Bluetooth co-existance

   - Mobile:
      - rmnet: support TX aggregation"

* tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits)
  page_pool: add a comment explaining the fragment counter usage
  net: ethtool: fix __ethtool_dev_mm_supported() implementation
  ethtool: pse-pd: Fix double word in comments
  xsk: add linux/vmalloc.h to xsk.c
  sefltests: netdevsim: wait for devlink instance after netns removal
  selftest: fib_tests: Always cleanup before exit
  net/mlx5e: Align IPsec ASO result memory to be as required by hardware
  net/mlx5e: TC, Set CT miss to the specific ct action instance
  net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG
  net/mlx5: Refactor tc miss handling to a single function
  net/mlx5: Kconfig: Make tc offload depend on tc skb extension
  net/sched: flower: Support hardware miss to tc action
  net/sched: flower: Move filter handle initialization earlier
  net/sched: cls_api: Support hardware miss to tc action
  net/sched: Rename user cookie and act cookie
  sfc: fix builds without CONFIG_RTC_LIB
  sfc: clean up some inconsistent indentings
  net/mlx4_en: Introduce flexible array to silence overflow warning
  net: lan966x: Fix possible deadlock inside PTP
  net/ulp: Remove redundant ->clone() test in inet_clone_ulp().
  ...

1 files changed, 990 insertions, 0 deletions

diff --git a/drivers/sbus/char/oradax.c b/drivers/sbus/char/oradax.c
new file mode 100644
index 000000000..21b7cb6e7
--- /dev/null
+++ b/drivers/sbus/char/oradax.c
@@ -0,0 +1,990 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Copyright (c) 2017, Oracle and/or its affiliates. All rights reserved.
+ */
+
+/*
+ * Oracle Data Analytics Accelerator (DAX)
+ *
+ * DAX is a coprocessor which resides on the SPARC M7 (DAX1) and M8
+ * (DAX2) processor chips, and has direct access to the CPU's L3
+ * caches as well as physical memory. It can perform several
+ * operations on data streams with various input and output formats.
+ * The driver provides a transport mechanism only and has limited
+ * knowledge of the various opcodes and data formats. A user space
+ * library provides high level services and translates these into low
+ * level commands which are then passed into the driver and
+ * subsequently the hypervisor and the coprocessor.  The library is
+ * the recommended way for applications to use the coprocessor, and
+ * the driver interface is not intended for general use.
+ *
+ * See Documentation/sparc/oradax/oracle-dax.rst for more details.
+ */
+
+#include <linux/uaccess.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/cdev.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+
+#include <asm/hypervisor.h>
+#include <asm/mdesc.h>
+#include <asm/oradax.h>
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Driver for Oracle Data Analytics Accelerator");
+
+#define	DAX_DBG_FLG_BASIC	0x01
+#define	DAX_DBG_FLG_STAT	0x02
+#define	DAX_DBG_FLG_INFO	0x04
+#define	DAX_DBG_FLG_ALL		0xff
+
+#define	dax_err(fmt, ...)      pr_err("%s: " fmt "\n", __func__, ##__VA_ARGS__)
+#define	dax_info(fmt, ...)     pr_info("%s: " fmt "\n", __func__, ##__VA_ARGS__)
+
+#define	dax_dbg(fmt, ...)	do {					\
+					if (dax_debug & DAX_DBG_FLG_BASIC)\
+						dax_info(fmt, ##__VA_ARGS__); \
+				} while (0)
+#define	dax_stat_dbg(fmt, ...)	do {					\
+					if (dax_debug & DAX_DBG_FLG_STAT) \
+						dax_info(fmt, ##__VA_ARGS__); \
+				} while (0)
+#define	dax_info_dbg(fmt, ...)	do { \
+					if (dax_debug & DAX_DBG_FLG_INFO) \
+						dax_info(fmt, ##__VA_ARGS__); \
+				} while (0)
+
+#define	DAX1_MINOR		1
+#define	DAX1_MAJOR		1
+#define	DAX2_MINOR		0
+#define	DAX2_MAJOR		2
+
+#define	DAX1_STR    "ORCL,sun4v-dax"
+#define	DAX2_STR    "ORCL,sun4v-dax2"
+
+#define	DAX_CA_ELEMS		(DAX_MMAP_LEN / sizeof(struct dax_cca))
+
+#define	DAX_CCB_USEC		100
+#define	DAX_CCB_RETRIES		10000
+
+/* stream types */
+enum {
+	OUT,
+	PRI,
+	SEC,
+	TBL,
+	NUM_STREAM_TYPES
+};
+
+/* completion status */
+#define	CCA_STAT_NOT_COMPLETED	0
+#define	CCA_STAT_COMPLETED	1
+#define	CCA_STAT_FAILED		2
+#define	CCA_STAT_KILLED		3
+#define	CCA_STAT_NOT_RUN	4
+#define	CCA_STAT_PIPE_OUT	5
+#define	CCA_STAT_PIPE_SRC	6
+#define	CCA_STAT_PIPE_DST	7
+
+/* completion err */
+#define	CCA_ERR_SUCCESS		0x0	/* no error */
+#define	CCA_ERR_OVERFLOW	0x1	/* buffer overflow */
+#define	CCA_ERR_DECODE		0x2	/* CCB decode error */
+#define	CCA_ERR_PAGE_OVERFLOW	0x3	/* page overflow */
+#define	CCA_ERR_KILLED		0x7	/* command was killed */
+#define	CCA_ERR_TIMEOUT		0x8	/* Timeout */
+#define	CCA_ERR_ADI		0x9	/* ADI error */
+#define	CCA_ERR_DATA_FMT	0xA	/* data format error */
+#define	CCA_ERR_OTHER_NO_RETRY	0xE	/* Other error, do not retry */
+#define	CCA_ERR_OTHER_RETRY	0xF	/* Other error, retry */
+#define	CCA_ERR_PARTIAL_SYMBOL	0x80	/* QP partial symbol warning */
+
+/* CCB address types */
+#define	DAX_ADDR_TYPE_NONE	0
+#define	DAX_ADDR_TYPE_VA_ALT	1	/* secondary context */
+#define	DAX_ADDR_TYPE_RA	2	/* real address */
+#define	DAX_ADDR_TYPE_VA	3	/* virtual address */
+
+/* dax_header_t opcode */
+#define	DAX_OP_SYNC_NOP		0x0
+#define	DAX_OP_EXTRACT		0x1
+#define	DAX_OP_SCAN_VALUE	0x2
+#define	DAX_OP_SCAN_RANGE	0x3
+#define	DAX_OP_TRANSLATE	0x4
+#define	DAX_OP_SELECT		0x5
+#define	DAX_OP_INVERT		0x10	/* OR with translate, scan opcodes */
+
+struct dax_header {
+	u32 ccb_version:4;	/* 31:28 CCB Version */
+				/* 27:24 Sync Flags */
+	u32 pipe:1;		/* Pipeline */
+	u32 longccb:1;		/* Longccb. Set for scan with lu2, lu3, lu4. */
+	u32 cond:1;		/* Conditional */
+	u32 serial:1;		/* Serial */
+	u32 opcode:8;		/* 23:16 Opcode */
+				/* 15:0 Address Type. */
+	u32 reserved:3;		/* 15:13 reserved */
+	u32 table_addr_type:2;	/* 12:11 Huffman Table Address Type */
+	u32 out_addr_type:3;	/* 10:8 Destination Address Type */
+	u32 sec_addr_type:3;	/* 7:5 Secondary Source Address Type */
+	u32 pri_addr_type:3;	/* 4:2 Primary Source Address Type */
+	u32 cca_addr_type:2;	/* 1:0 Completion Address Type */
+};
+
+struct dax_control {
+	u32 pri_fmt:4;		/* 31:28 Primary Input Format */
+	u32 pri_elem_size:5;	/* 27:23 Primary Input Element Size(less1) */
+	u32 pri_offset:3;	/* 22:20 Primary Input Starting Offset */
+	u32 sec_encoding:1;	/* 19    Secondary Input Encoding */
+				/*	 (must be 0 for Select) */
+	u32 sec_offset:3;	/* 18:16 Secondary Input Starting Offset */
+	u32 sec_elem_size:2;	/* 15:14 Secondary Input Element Size */
+				/*	 (must be 0 for Select) */
+	u32 out_fmt:2;		/* 13:12 Output Format */
+	u32 out_elem_size:2;	/* 11:10 Output Element Size */
+	u32 misc:10;		/* 9:0 Opcode specific info */
+};
+
+struct dax_data_access {
+	u64 flow_ctrl:2;	/* 63:62 Flow Control Type */
+	u64 pipe_target:2;	/* 61:60 Pipeline Target */
+	u64 out_buf_size:20;	/* 59:40 Output Buffer Size */
+				/*	 (cachelines less 1) */
+	u64 unused1:8;		/* 39:32 Reserved, Set to 0 */
+	u64 out_alloc:5;	/* 31:27 Output Allocation */
+	u64 unused2:1;		/* 26	 Reserved */
+	u64 pri_len_fmt:2;	/* 25:24 Input Length Format */
+	u64 pri_len:24;		/* 23:0  Input Element/Byte/Bit Count */
+				/*	 (less 1) */
+};
+
+struct dax_ccb {
+	struct dax_header hdr;	/* CCB Header */
+	struct dax_control ctrl;/* Control Word */
+	void *ca;		/* Completion Address */
+	void *pri;		/* Primary Input Address */
+	struct dax_data_access dac; /* Data Access Control */
+	void *sec;		/* Secondary Input Address */
+	u64 dword5;		/* depends on opcode */
+	void *out;		/* Output Address */
+	void *tbl;		/* Table Address or bitmap */
+};
+
+struct dax_cca {
+	u8	status;		/* user may mwait on this address */
+	u8	err;		/* user visible error notification */
+	u8	rsvd[2];	/* reserved */
+	u32	n_remaining;	/* for QP partial symbol warning */
+	u32	output_sz;	/* output in bytes */
+	u32	rsvd2;		/* reserved */
+	u64	run_cycles;	/* run time in OCND2 cycles */
+	u64	run_stats;	/* nothing reported in version 1.0 */
+	u32	n_processed;	/* number input elements */
+	u32	rsvd3[5];	/* reserved */
+	u64	retval;		/* command return value */
+	u64	rsvd4[8];	/* reserved */
+};
+
+/* per thread CCB context */
+struct dax_ctx {
+	struct dax_ccb		*ccb_buf;
+	u64			ccb_buf_ra;	/* cached RA of ccb_buf  */
+	struct dax_cca		*ca_buf;
+	u64			ca_buf_ra;	/* cached RA of ca_buf   */
+	struct page		*pages[DAX_CA_ELEMS][NUM_STREAM_TYPES];
+						/* array of locked pages */
+	struct task_struct	*owner;		/* thread that owns ctx  */
+	struct task_struct	*client;	/* requesting thread     */
+	union ccb_result	result;
+	u32			ccb_count;
+	u32			fail_count;
+};
+
+/* driver public entry points */
+static int dax_open(struct inode *inode, struct file *file);
+static ssize_t dax_read(struct file *filp, char __user *buf,
+			size_t count, loff_t *ppos);
+static ssize_t dax_write(struct file *filp, const char __user *buf,
+			 size_t count, loff_t *ppos);
+static int dax_devmap(struct file *f, struct vm_area_struct *vma);
+static int dax_close(struct inode *i, struct file *f);
+
+static const struct file_operations dax_fops = {
+	.owner	=	THIS_MODULE,
+	.open	=	dax_open,
+	.read	=	dax_read,
+	.write	=	dax_write,
+	.mmap	=	dax_devmap,
+	.release =	dax_close,
+};
+
+static int dax_ccb_exec(struct dax_ctx *ctx, const char __user *buf,
+			size_t count, loff_t *ppos);
+static int dax_ccb_info(u64 ca, struct ccb_info_result *info);
+static int dax_ccb_kill(u64 ca, u16 *kill_res);
+
+static struct cdev c_dev;
+static struct class *cl;
+static dev_t first;
+
+static int max_ccb_version;
+static int dax_debug;
+module_param(dax_debug, int, 0644);
+MODULE_PARM_DESC(dax_debug, "Debug flags");
+
+static int __init dax_attach(void)
+{
+	unsigned long dummy, hv_rv, major, minor, minor_requested, max_ccbs;
+	struct mdesc_handle *hp = mdesc_grab();
+	char *prop, *dax_name;
+	bool found = false;
+	int len, ret = 0;
+	u64 pn;
+
+	if (hp == NULL) {
+		dax_err("Unable to grab mdesc");
+		return -ENODEV;
+	}
+
+	mdesc_for_each_node_by_name(hp, pn, "virtual-device") {
+		prop = (char *)mdesc_get_property(hp, pn, "name", &len);
+		if (prop == NULL)
+			continue;
+		if (strncmp(prop, "dax", strlen("dax")))
+			continue;
+		dax_dbg("Found node 0x%llx = %s", pn, prop);
+
+		prop = (char *)mdesc_get_property(hp, pn, "compatible", &len);
+		if (prop == NULL)
+			continue;
+		dax_dbg("Found node 0x%llx = %s", pn, prop);
+		found = true;
+		break;
+	}
+
+	if (!found) {
+		dax_err("No DAX device found");
+		ret = -ENODEV;
+		goto done;
+	}
+
+	if (strncmp(prop, DAX2_STR, strlen(DAX2_STR)) == 0) {
+		dax_name = DAX_NAME "2";
+		major = DAX2_MAJOR;
+		minor_requested = DAX2_MINOR;
+		max_ccb_version = 1;
+		dax_dbg("MD indicates DAX2 coprocessor");
+	} else if (strncmp(prop, DAX1_STR, strlen(DAX1_STR)) == 0) {
+		dax_name = DAX_NAME "1";
+		major = DAX1_MAJOR;
+		minor_requested = DAX1_MINOR;
+		max_ccb_version = 0;
+		dax_dbg("MD indicates DAX1 coprocessor");
+	} else {
+		dax_err("Unknown dax type: %s", prop);
+		ret = -ENODEV;
+		goto done;
+	}
+
+	minor = minor_requested;
+	dax_dbg("Registering DAX HV api with major %ld minor %ld", major,
+		minor);
+	if (sun4v_hvapi_register(HV_GRP_DAX, major, &minor)) {
+		dax_err("hvapi_register failed");
+		ret = -ENODEV;
+		goto done;
+	} else {
+		dax_dbg("Max minor supported by HV = %ld (major %ld)", minor,
+			major);
+		minor = min(minor, minor_requested);
+		dax_dbg("registered DAX major %ld minor %ld", major, minor);
+	}
+
+	/* submit a zero length ccb array to query coprocessor queue size */
+	hv_rv = sun4v_ccb_submit(0, 0, HV_CCB_QUERY_CMD, 0, &max_ccbs, &dummy);
+	if (hv_rv != 0) {
+		dax_err("get_hwqueue_size failed with status=%ld and max_ccbs=%ld",
+			hv_rv, max_ccbs);
+		ret = -ENODEV;
+		goto done;
+	}
+
+	if (max_ccbs != DAX_MAX_CCBS) {
+		dax_err("HV reports unsupported max_ccbs=%ld", max_ccbs);
+		ret = -ENODEV;
+		goto done;
+	}
+
+	if (alloc_chrdev_region(&first, 0, 1, DAX_NAME) < 0) {
+		dax_err("alloc_chrdev_region failed");
+		ret = -ENXIO;
+		goto done;
+	}
+
+	cl = class_create(THIS_MODULE, DAX_NAME);
+	if (IS_ERR(cl)) {
+		dax_err("class_create failed");
+		ret = PTR_ERR(cl);
+		goto class_error;
+	}
+
+	if (device_create(cl, NULL, first, NULL, dax_name) == NULL) {
+		dax_err("device_create failed");
+		ret = -ENXIO;
+		goto device_error;
+	}
+
+	cdev_init(&c_dev, &dax_fops);
+	if (cdev_add(&c_dev, first, 1) == -1) {
+		dax_err("cdev_add failed");
+		ret = -ENXIO;
+		goto cdev_error;
+	}
+
+	pr_info("Attached DAX module\n");
+	goto done;
+
+cdev_error:
+	device_destroy(cl, first);
+device_error:
+	class_destroy(cl);
+class_error:
+	unregister_chrdev_region(first, 1);
+done:
+	mdesc_release(hp);
+	return ret;
+}
+module_init(dax_attach);
+
+static void __exit dax_detach(void)
+{
+	pr_info("Cleaning up DAX module\n");
+	cdev_del(&c_dev);
+	device_destroy(cl, first);
+	class_destroy(cl);
+	unregister_chrdev_region(first, 1);
+}
+module_exit(dax_detach);
+
+/* map completion area */
+static int dax_devmap(struct file *f, struct vm_area_struct *vma)
+{
+	struct dax_ctx *ctx = (struct dax_ctx *)f->private_data;
+	size_t len = vma->vm_end - vma->vm_start;
+
+	dax_dbg("len=0x%lx, flags=0x%lx", len, vma->vm_flags);
+
+	if (ctx->owner != current) {
+		dax_dbg("devmap called from wrong thread");
+		return -EINVAL;
+	}
+
+	if (len != DAX_MMAP_LEN) {
+		dax_dbg("len(%lu) != DAX_MMAP_LEN(%d)", len, DAX_MMAP_LEN);
+		return -EINVAL;
+	}
+
+	/* completion area is mapped read-only for user */
+	if (vma->vm_flags & VM_WRITE)
+		return -EPERM;
+	vma->vm_flags &= ~VM_MAYWRITE;
+
+	if (remap_pfn_range(vma, vma->vm_start, ctx->ca_buf_ra >> PAGE_SHIFT,
+			    len, vma->vm_page_prot))
+		return -EAGAIN;
+
+	dax_dbg("mmapped completion area at uva 0x%lx", vma->vm_start);
+	return 0;
+}
+
+/* Unlock user pages. Called during dequeue or device close */
+static void dax_unlock_pages(struct dax_ctx *ctx, int ccb_index, int nelem)
+{
+	int i, j;
+
+	for (i = ccb_index; i < ccb_index + nelem; i++) {
+		for (j = 0; j < NUM_STREAM_TYPES; j++) {
+			struct page *p = ctx->pages[i][j];
+
+			if (p) {
+				dax_dbg("freeing page %p", p);
+				unpin_user_pages_dirty_lock(&p, 1, j == OUT);
+				ctx->pages[i][j] = NULL;
+			}
+		}
+	}
+}
+
+static int dax_lock_page(void *va, struct page **p)
+{
+	int ret;
+
+	dax_dbg("uva %p", va);
+
+	ret = pin_user_pages_fast((unsigned long)va, 1, FOLL_WRITE, p);
+	if (ret == 1) {
+		dax_dbg("locked page %p, for VA %p", *p, va);
+		return 0;
+	}
+
+	dax_dbg("pin_user_pages failed, va=%p, ret=%d", va, ret);
+	return -1;
+}
+
+static int dax_lock_pages(struct dax_ctx *ctx, int idx,
+			  int nelem, u64 *err_va)
+{
+	int i;
+
+	for (i = 0; i < nelem; i++) {
+		struct dax_ccb *ccbp = &ctx->ccb_buf[i];
+
+		/*
+		 * For each address in the CCB whose type is virtual,
+		 * lock the page and change the type to virtual alternate
+		 * context. On error, return the offending address in
+		 * err_va.
+		 */
+		if (ccbp->hdr.out_addr_type == DAX_ADDR_TYPE_VA) {
+			dax_dbg("output");
+			if (dax_lock_page(ccbp->out,
+					  &ctx->pages[i + idx][OUT]) != 0) {
+				*err_va = (u64)ccbp->out;
+				goto error;
+			}
+			ccbp->hdr.out_addr_type = DAX_ADDR_TYPE_VA_ALT;
+		}
+
+		if (ccbp->hdr.pri_addr_type == DAX_ADDR_TYPE_VA) {
+			dax_dbg("input");
+			if (dax_lock_page(ccbp->pri,
+					  &ctx->pages[i + idx][PRI]) != 0) {
+				*err_va = (u64)ccbp->pri;
+				goto error;
+			}
+			ccbp->hdr.pri_addr_type = DAX_ADDR_TYPE_VA_ALT;
+		}
+
+		if (ccbp->hdr.sec_addr_type == DAX_ADDR_TYPE_VA) {
+			dax_dbg("sec input");
+			if (dax_lock_page(ccbp->sec,
+					  &ctx->pages[i + idx][SEC]) != 0) {
+				*err_va = (u64)ccbp->sec;
+				goto error;
+			}
+			ccbp->hdr.sec_addr_type = DAX_ADDR_TYPE_VA_ALT;
+		}
+
+		if (ccbp->hdr.table_addr_type == DAX_ADDR_TYPE_VA) {
+			dax_dbg("tbl");
+			if (dax_lock_page(ccbp->tbl,
+					  &ctx->pages[i + idx][TBL]) != 0) {
+				*err_va = (u64)ccbp->tbl;
+				goto error;
+			}
+			ccbp->hdr.table_addr_type = DAX_ADDR_TYPE_VA_ALT;
+		}
+
+		/* skip over 2nd 64 bytes of long CCB */
+		if (ccbp->hdr.longccb)
+			i++;
+	}
+	return DAX_SUBMIT_OK;
+
+error:
+	dax_unlock_pages(ctx, idx, nelem);
+	return DAX_SUBMIT_ERR_NOACCESS;
+}
+
+static void dax_ccb_wait(struct dax_ctx *ctx, int idx)
+{
+	int ret, nretries;
+	u16 kill_res;
+
+	dax_dbg("idx=%d", idx);
+
+	for (nretries = 0; nretries < DAX_CCB_RETRIES; nretries++) {
+		if (ctx->ca_buf[idx].status == CCA_STAT_NOT_COMPLETED)
+			udelay(DAX_CCB_USEC);
+		else
+			return;
+	}
+	dax_dbg("ctx (%p): CCB[%d] timed out, wait usec=%d, retries=%d. Killing ccb",
+		(void *)ctx, idx, DAX_CCB_USEC, DAX_CCB_RETRIES);
+
+	ret = dax_ccb_kill(ctx->ca_buf_ra + idx * sizeof(struct dax_cca),
+			   &kill_res);
+	dax_dbg("Kill CCB[%d] %s", idx, ret ? "failed" : "succeeded");
+}
+
+static int dax_close(struct inode *ino, struct file *f)
+{
+	struct dax_ctx *ctx = (struct dax_ctx *)f->private_data;
+	int i;
+
+	f->private_data = NULL;
+
+	for (i = 0; i < DAX_CA_ELEMS; i++) {
+		if (ctx->ca_buf[i].status == CCA_STAT_NOT_COMPLETED) {
+			dax_dbg("CCB[%d] not completed", i);
+			dax_ccb_wait(ctx, i);
+		}
+		dax_unlock_pages(ctx, i, 1);
+	}
+
+	kfree(ctx->ccb_buf);
+	kfree(ctx->ca_buf);
+	dax_stat_dbg("CCBs: %d good, %d bad", ctx->ccb_count, ctx->fail_count);
+	kfree(ctx);
+
+	return 0;
+}
+
+static ssize_t dax_read(struct file *f, char __user *buf,
+			size_t count, loff_t *ppos)
+{
+	struct dax_ctx *ctx = f->private_data;
+
+	if (ctx->client != current)
+		return -EUSERS;
+
+	ctx->client = NULL;
+
+	if (count != sizeof(union ccb_result))
+		return -EINVAL;
+	if (copy_to_user(buf, &ctx->result, sizeof(union ccb_result)))
+		return -EFAULT;
+	return count;
+}
+
+static ssize_t dax_write(struct file *f, const char __user *buf,
+			 size_t count, loff_t *ppos)
+{
+	struct dax_ctx *ctx = f->private_data;
+	struct dax_command hdr;
+	unsigned long ca;
+	int i, idx, ret;
+
+	if (ctx->client != NULL)
+		return -EINVAL;
+
+	if (count == 0 || count > DAX_MAX_CCBS * sizeof(struct dax_ccb))
+		return -EINVAL;
+
+	if (count % sizeof(struct dax_ccb) == 0)
+		return dax_ccb_exec(ctx, buf, count, ppos); /* CCB EXEC */
+
+	if (count != sizeof(struct dax_command))
+		return -EINVAL;
+
+	/* immediate command */
+	if (ctx->owner != current)
+		return -EUSERS;
+
+	if (copy_from_user(&hdr, buf, sizeof(hdr)))
+		return -EFAULT;
+
+	ca = ctx->ca_buf_ra + hdr.ca_offset;
+
+	switch (hdr.command) {
+	case CCB_KILL:
+		if (hdr.ca_offset >= DAX_MMAP_LEN) {
+			dax_dbg("invalid ca_offset (%d) >= ca_buflen (%d)",
+				hdr.ca_offset, DAX_MMAP_LEN);
+			return -EINVAL;
+		}
+
+		ret = dax_ccb_kill(ca, &ctx->result.kill.action);
+		if (ret != 0) {
+			dax_dbg("dax_ccb_kill failed (ret=%d)", ret);
+			return ret;
+		}
+
+		dax_info_dbg("killed (ca_offset %d)", hdr.ca_offset);
+		idx = hdr.ca_offset / sizeof(struct dax_cca);
+		ctx->ca_buf[idx].status = CCA_STAT_KILLED;
+		ctx->ca_buf[idx].err = CCA_ERR_KILLED;
+		ctx->client = current;
+		return count;
+
+	case CCB_INFO:
+		if (hdr.ca_offset >= DAX_MMAP_LEN) {
+			dax_dbg("invalid ca_offset (%d) >= ca_buflen (%d)",
+				hdr.ca_offset, DAX_MMAP_LEN);
+			return -EINVAL;
+		}
+
+		ret = dax_ccb_info(ca, &ctx->result.info);
+		if (ret != 0) {
+			dax_dbg("dax_ccb_info failed (ret=%d)", ret);
+			return ret;
+		}
+
+		dax_info_dbg("info succeeded on ca_offset %d", hdr.ca_offset);
+		ctx->client = current;
+		return count;
+
+	case CCB_DEQUEUE:
+		for (i = 0; i < DAX_CA_ELEMS; i++) {
+			if (ctx->ca_buf[i].status !=
+			    CCA_STAT_NOT_COMPLETED)
+				dax_unlock_pages(ctx, i, 1);
+		}
+		return count;
+
+	default:
+		return -EINVAL;
+	}
+}
+
+static int dax_open(struct inode *inode, struct file *f)
+{
+	struct dax_ctx *ctx = NULL;
+	int i;
+
+	ctx = kzalloc(sizeof(*ctx), GFP_KERNEL);
+	if (ctx == NULL)
+		goto done;
+
+	ctx->ccb_buf = kcalloc(DAX_MAX_CCBS, sizeof(struct dax_ccb),
+			       GFP_KERNEL);
+	if (ctx->ccb_buf == NULL)
+		goto done;
+
+	ctx->ccb_buf_ra = virt_to_phys(ctx->ccb_buf);
+	dax_dbg("ctx->ccb_buf=0x%p, ccb_buf_ra=0x%llx",
+		(void *)ctx->ccb_buf, ctx->ccb_buf_ra);
+
+	/* allocate CCB completion area buffer */
+	ctx->ca_buf = kzalloc(DAX_MMAP_LEN, GFP_KERNEL);
+	if (ctx->ca_buf == NULL)
+		goto alloc_error;
+	for (i = 0; i < DAX_CA_ELEMS; i++)
+		ctx->ca_buf[i].status = CCA_STAT_COMPLETED;
+
+	ctx->ca_buf_ra = virt_to_phys(ctx->ca_buf);
+	dax_dbg("ctx=0x%p, ctx->ca_buf=0x%p, ca_buf_ra=0x%llx",
+		(void *)ctx, (void *)ctx->ca_buf, ctx->ca_buf_ra);
+
+	ctx->owner = current;
+	f->private_data = ctx;
+	return 0;
+
+alloc_error:
+	kfree(ctx->ccb_buf);
+done:
+	kfree(ctx);
+	return -ENOMEM;
+}
+
+static char *dax_hv_errno(unsigned long hv_ret, int *ret)
+{
+	switch (hv_ret) {
+	case HV_EBADALIGN:
+		*ret = -EFAULT;
+		return "HV_EBADALIGN";
+	case HV_ENORADDR:
+		*ret = -EFAULT;
+		return "HV_ENORADDR";
+	case HV_EINVAL:
+		*ret = -EINVAL;
+		return "HV_EINVAL";
+	case HV_EWOULDBLOCK:
+		*ret = -EAGAIN;
+		return "HV_EWOULDBLOCK";
+	case HV_ENOACCESS:
+		*ret = -EPERM;
+		return "HV_ENOACCESS";
+	default:
+		break;
+	}
+
+	*ret = -EIO;
+	return "UNKNOWN";
+}
+
+static int dax_ccb_kill(u64 ca, u16 *kill_res)
+{
+	unsigned long hv_ret;
+	int count, ret = 0;
+	char *err_str;
+
+	for (count = 0; count < DAX_CCB_RETRIES; count++) {
+		dax_dbg("attempting kill on ca_ra 0x%llx", ca);
+		hv_ret = sun4v_ccb_kill(ca, kill_res);
+
+		if (hv_ret == HV_EOK) {
+			dax_info_dbg("HV_EOK (ca_ra 0x%llx): %d", ca,
+				     *kill_res);
+		} else {
+			err_str = dax_hv_errno(hv_ret, &ret);
+			dax_dbg("%s (ca_ra 0x%llx)", err_str, ca);
+		}
+
+		if (ret != -EAGAIN)
+			return ret;
+		dax_info_dbg("ccb_kill count = %d", count);
+		udelay(DAX_CCB_USEC);
+	}
+
+	return -EAGAIN;
+}
+
+static int dax_ccb_info(u64 ca, struct ccb_info_result *info)
+{
+	unsigned long hv_ret;
+	char *err_str;
+	int ret = 0;
+
+	dax_dbg("attempting info on ca_ra 0x%llx", ca);
+	hv_ret = sun4v_ccb_info(ca, info);
+
+	if (hv_ret == HV_EOK) {
+		dax_info_dbg("HV_EOK (ca_ra 0x%llx): %d", ca, info->state);
+		if (info->state == DAX_CCB_ENQUEUED) {
+			dax_info_dbg("dax_unit %d, queue_num %d, queue_pos %d",
+				     info->inst_num, info->q_num, info->q_pos);
+		}
+	} else {
+		err_str = dax_hv_errno(hv_ret, &ret);
+		dax_dbg("%s (ca_ra 0x%llx)", err_str, ca);
+	}
+
+	return ret;
+}
+
+static void dax_prt_ccbs(struct dax_ccb *ccb, int nelem)
+{
+	int i, j;
+	u64 *ccbp;
+
+	dax_dbg("ccb buffer:");
+	for (i = 0; i < nelem; i++) {
+		ccbp = (u64 *)&ccb[i];
+		dax_dbg(" %sccb[%d]", ccb[i].hdr.longccb ? "long " : "",  i);
+		for (j = 0; j < 8; j++)
+			dax_dbg("\tccb[%d].dwords[%d]=0x%llx",
+				i, j, *(ccbp + j));
+	}
+}
+
+/*
+ * Validates user CCB content.  Also sets completion address and address types
+ * for all addresses contained in CCB.
+ */
+static int dax_preprocess_usr_ccbs(struct dax_ctx *ctx, int idx, int nelem)
+{
+	int i;
+
+	/*
+	 * The user is not allowed to specify real address types in
+	 * the CCB header.  This must be enforced by the kernel before
+	 * submitting the CCBs to HV.  The only allowed values for all
+	 * address fields are VA or IMM
+	 */
+	for (i = 0; i < nelem; i++) {
+		struct dax_ccb *ccbp = &ctx->ccb_buf[i];
+		unsigned long ca_offset;
+
+		if (ccbp->hdr.ccb_version > max_ccb_version)
+			return DAX_SUBMIT_ERR_CCB_INVAL;
+
+		switch (ccbp->hdr.opcode) {
+		case DAX_OP_SYNC_NOP:
+		case DAX_OP_EXTRACT:
+		case DAX_OP_SCAN_VALUE:
+		case DAX_OP_SCAN_RANGE:
+		case DAX_OP_TRANSLATE:
+		case DAX_OP_SCAN_VALUE | DAX_OP_INVERT:
+		case DAX_OP_SCAN_RANGE | DAX_OP_INVERT:
+		case DAX_OP_TRANSLATE | DAX_OP_INVERT:
+		case DAX_OP_SELECT:
+			break;
+		default:
+			return DAX_SUBMIT_ERR_CCB_INVAL;
+		}
+
+		if (ccbp->hdr.out_addr_type != DAX_ADDR_TYPE_VA &&
+		    ccbp->hdr.out_addr_type != DAX_ADDR_TYPE_NONE) {
+			dax_dbg("invalid out_addr_type in user CCB[%d]", i);
+			return DAX_SUBMIT_ERR_CCB_INVAL;
+		}
+
+		if (ccbp->hdr.pri_addr_type != DAX_ADDR_TYPE_VA &&
+		    ccbp->hdr.pri_addr_type != DAX_ADDR_TYPE_NONE) {
+			dax_dbg("invalid pri_addr_type in user CCB[%d]", i);
+			return DAX_SUBMIT_ERR_CCB_INVAL;
+		}
+
+		if (ccbp->hdr.sec_addr_type != DAX_ADDR_TYPE_VA &&
+		    ccbp->hdr.sec_addr_type != DAX_ADDR_TYPE_NONE) {
+			dax_dbg("invalid sec_addr_type in user CCB[%d]", i);
+			return DAX_SUBMIT_ERR_CCB_INVAL;
+		}
+
+		if (ccbp->hdr.table_addr_type != DAX_ADDR_TYPE_VA &&
+		    ccbp->hdr.table_addr_type != DAX_ADDR_TYPE_NONE) {
+			dax_dbg("invalid table_addr_type in user CCB[%d]", i);
+			return DAX_SUBMIT_ERR_CCB_INVAL;
+		}
+
+		/* set completion (real) address and address type */
+		ccbp->hdr.cca_addr_type = DAX_ADDR_TYPE_RA;
+		ca_offset = (idx + i) * sizeof(struct dax_cca);
+		ccbp->ca = (void *)ctx->ca_buf_ra + ca_offset;
+		memset(&ctx->ca_buf[idx + i], 0, sizeof(struct dax_cca));
+
+		dax_dbg("ccb[%d]=%p, ca_offset=0x%lx, compl RA=0x%llx",
+			i, ccbp, ca_offset, ctx->ca_buf_ra + ca_offset);
+
+		/* skip over 2nd 64 bytes of long CCB */
+		if (ccbp->hdr.longccb)
+			i++;
+	}
+
+	return DAX_SUBMIT_OK;
+}
+
+static int dax_ccb_exec(struct dax_ctx *ctx, const char __user *buf,
+			size_t count, loff_t *ppos)
+{
+	unsigned long accepted_len, hv_rv;
+	int i, idx, nccbs, naccepted;
+
+	ctx->client = current;
+	idx = *ppos;
+	nccbs = count / sizeof(struct dax_ccb);
+
+	if (ctx->owner != current) {
+		dax_dbg("wrong thread");
+		ctx->result.exec.status = DAX_SUBMIT_ERR_THR_INIT;
+		return 0;
+	}
+	dax_dbg("args: ccb_buf_len=%ld, idx=%d", count, idx);
+
+	/* for given index and length, verify ca_buf range exists */
+	if (idx < 0 || idx > (DAX_CA_ELEMS - nccbs)) {
+		ctx->result.exec.status = DAX_SUBMIT_ERR_NO_CA_AVAIL;
+		return 0;
+	}
+
+	/*
+	 * Copy CCBs into kernel buffer to prevent modification by the
+	 * user in between validation and submission.
+	 */
+	if (copy_from_user(ctx->ccb_buf, buf, count)) {
+		dax_dbg("copyin of user CCB buffer failed");
+		ctx->result.exec.status = DAX_SUBMIT_ERR_CCB_ARR_MMU_MISS;
+		return 0;
+	}
+
+	/* check to see if ca_buf[idx] .. ca_buf[idx + nccbs] are available */
+	for (i = idx; i < idx + nccbs; i++) {
+		if (ctx->ca_buf[i].status == CCA_STAT_NOT_COMPLETED) {
+			dax_dbg("CA range not available, dequeue needed");
+			ctx->result.exec.status = DAX_SUBMIT_ERR_NO_CA_AVAIL;
+			return 0;
+		}
+	}
+	dax_unlock_pages(ctx, idx, nccbs);
+
+	ctx->result.exec.status = dax_preprocess_usr_ccbs(ctx, idx, nccbs);
+	if (ctx->result.exec.status != DAX_SUBMIT_OK)
+		return 0;
+
+	ctx->result.exec.status = dax_lock_pages(ctx, idx, nccbs,
+						 &ctx->result.exec.status_data);
+	if (ctx->result.exec.status != DAX_SUBMIT_OK)
+		return 0;
+
+	if (dax_debug & DAX_DBG_FLG_BASIC)
+		dax_prt_ccbs(ctx->ccb_buf, nccbs);
+
+	hv_rv = sun4v_ccb_submit(ctx->ccb_buf_ra, count,
+				 HV_CCB_QUERY_CMD | HV_CCB_VA_SECONDARY, 0,
+				 &accepted_len, &ctx->result.exec.status_data);
+
+	switch (hv_rv) {
+	case HV_EOK:
+		/*
+		 * Hcall succeeded with no errors but the accepted
+		 * length may be less than the requested length.  The
+		 * only way the driver can resubmit the remainder is
+		 * to wait for completion of the submitted CCBs since
+		 * there is no way to guarantee the ordering semantics
+		 * required by the client applications.  Therefore we
+		 * let the user library deal with resubmissions.
+		 */
+		ctx->result.exec.status = DAX_SUBMIT_OK;
+		break;
+	case HV_EWOULDBLOCK:
+		/*
+		 * This is a transient HV API error. The user library
+		 * can retry.
+		 */
+		dax_dbg("hcall returned HV_EWOULDBLOCK");
+		ctx->result.exec.status = DAX_SUBMIT_ERR_WOULDBLOCK;
+		break;
+	case HV_ENOMAP:
+		/*
+		 * HV was unable to translate a VA. The VA it could
+		 * not translate is returned in the status_data param.
+		 */
+		dax_dbg("hcall returned HV_ENOMAP");
+		ctx->result.exec.status = DAX_SUBMIT_ERR_NOMAP;
+		break;
+	case HV_EINVAL:
+		/*
+		 * This is the result of an invalid user CCB as HV is
+		 * validating some of the user CCB fields.  Pass this
+		 * error back to the user. There is no supporting info
+		 * to isolate the invalid field.
+		 */
+		dax_dbg("hcall returned HV_EINVAL");
+		ctx->result.exec.status = DAX_SUBMIT_ERR_CCB_INVAL;
+		break;
+	case HV_ENOACCESS:
+		/*
+		 * HV found a VA that did not have the appropriate
+		 * permissions (such as the w bit). The VA in question
+		 * is returned in status_data param.
+		 */
+		dax_dbg("hcall returned HV_ENOACCESS");
+		ctx->result.exec.status = DAX_SUBMIT_ERR_NOACCESS;
+		break;
+	case HV_EUNAVAILABLE:
+		/*
+		 * The requested CCB operation could not be performed
+		 * at this time. Return the specific unavailable code
+		 * in the status_data field.
+		 */
+		dax_dbg("hcall returned HV_EUNAVAILABLE");
+		ctx->result.exec.status = DAX_SUBMIT_ERR_UNAVAIL;
+		break;
+	default:
+		ctx->result.exec.status = DAX_SUBMIT_ERR_INTERNAL;
+		dax_dbg("unknown hcall return value (%ld)", hv_rv);
+		break;
+	}
+
+	/* unlock pages associated with the unaccepted CCBs */
+	naccepted = accepted_len / sizeof(struct dax_ccb);
+	dax_unlock_pages(ctx, idx + naccepted, nccbs - naccepted);
+
+	/* mark unaccepted CCBs as not completed */
+	for (i = idx + naccepted; i < idx + nccbs; i++)
+		ctx->ca_buf[i].status = CCA_STAT_COMPLETED;
+
+	ctx->ccb_count += naccepted;
+	ctx->fail_count += nccbs - naccepted;
+
+	dax_dbg("hcall rv=%ld, accepted_len=%ld, status_data=0x%llx, ret status=%d",
+		hv_rv, accepted_len, ctx->result.exec.status_data,
+		ctx->result.exec.status);
+
+	if (count == accepted_len)
+		ctx->client = NULL; /* no read needed to complete protocol */
+	return accepted_len;
+}