nvme.h 8.1 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
/*
 * Copyright (c) 2011-2014, Intel Corporation.
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms and conditions of the GNU General Public License,
 * version 2, as published by the Free Software Foundation.
 *
 * This program is distributed in the hope it will be useful, but WITHOUT
 * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
 * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
 * more details.
 */

#ifndef _NVME_H
#define _NVME_H

#include <linux/nvme.h>
#include <linux/pci.h>
#include <linux/kref.h>
#include <linux/blk-mq.h>

C
Christoph Hellwig 已提交
22 23 24 25 26 27 28 29 30 31
enum {
	/*
	 * Driver internal status code for commands that were cancelled due
	 * to timeouts or controller shutdown.  The value is negative so
	 * that it a) doesn't overlap with the unsigned hardware error codes,
	 * and b) can easily be tested for.
	 */
	NVME_SC_CANCELLED		= -EINTR,
};

32 33 34
extern unsigned char nvme_io_timeout;
#define NVME_IO_TIMEOUT	(nvme_io_timeout * HZ)

35 36 37
extern unsigned char admin_timeout;
#define ADMIN_TIMEOUT	(admin_timeout * HZ)

38 39 40
extern unsigned char shutdown_timeout;
#define SHUTDOWN_TIMEOUT	(shutdown_timeout * HZ)

M
Matias Bjørling 已提交
41 42 43 44 45
enum {
	NVME_NS_LBA		= 0,
	NVME_NS_LIGHTNVM	= 1,
};

46
/*
47 48
 * List of workarounds for devices that required behavior not specified in
 * the standard.
49
 */
50 51 52 53 54 55
enum nvme_quirks {
	/*
	 * Prefers I/O aligned to a stripe size specified in a vendor
	 * specific Identify field.
	 */
	NVME_QUIRK_STRIPE_SIZE			= (1 << 0),
56 57 58 59 60 61

	/*
	 * The controller doesn't handle Identify value others than 0 or 1
	 * correctly.
	 */
	NVME_QUIRK_IDENTIFY_CNS			= (1 << 1),
62 63 64 65 66 67

	/*
	 * The controller deterministically returns O's on reads to discarded
	 * logical blocks.
	 */
	NVME_QUIRK_DISCARD_ZEROES		= (1 << 2),
68 69
};

70 71
struct nvme_ctrl {
	const struct nvme_ctrl_ops *ops;
72 73
	struct request_queue *admin_q;
	struct device *dev;
74
	struct kref kref;
75
	int instance;
76
	struct blk_mq_tag_set *tagset;
77
	struct list_head namespaces;
78
	struct mutex namespaces_mutex;
79
	struct device *device;	/* char device */
80
	struct list_head node;
81

82 83 84 85
	char name[12];
	char serial[20];
	char model[40];
	char firmware_rev[8];
M
Ming Lin 已提交
86
	int cntlid;
87 88 89 90

	u32 ctrl_config;

	u32 page_size;
91 92 93
	u32 max_hw_sectors;
	u32 stripe_size;
	u16 oncs;
94
	atomic_t abort_limit;
95 96
	u8 event_limit;
	u8 vwc;
97 98
	u32 vs;
	bool subsystem;
99
	unsigned long quirks;
100 101 102 103 104 105 106 107
};

/*
 * An NVM Express namespace is equivalent to a SCSI LUN
 */
struct nvme_ns {
	struct list_head list;

108
	struct nvme_ctrl *ctrl;
109 110 111 112
	struct request_queue *queue;
	struct gendisk *disk;
	struct kref kref;

113 114 115
	u8 eui[8];
	u8 uuid[16];

116 117 118 119 120
	unsigned ns_id;
	int lba_shift;
	u16 ms;
	bool ext;
	u8 pi_type;
M
Matias Bjørling 已提交
121
	int type;
122 123 124 125
	u64 mode_select_num_blocks;
	u32 mode_select_block_len;
};

126
struct nvme_ctrl_ops {
127
	struct module *module;
128
	int (*reg_read32)(struct nvme_ctrl *ctrl, u32 off, u32 *val);
129
	int (*reg_write32)(struct nvme_ctrl *ctrl, u32 off, u32 val);
130
	int (*reg_read64)(struct nvme_ctrl *ctrl, u32 off, u64 *val);
131
	bool (*io_incapable)(struct nvme_ctrl *ctrl);
132
	int (*reset_ctrl)(struct nvme_ctrl *ctrl);
133
	void (*free_ctrl)(struct nvme_ctrl *ctrl);
134 135
};

136 137 138 139 140 141 142 143 144
static inline bool nvme_ctrl_ready(struct nvme_ctrl *ctrl)
{
	u32 val = 0;

	if (ctrl->ops->reg_read32(ctrl, NVME_REG_CSTS, &val))
		return false;
	return val & NVME_CSTS_RDY;
}

145 146 147 148 149 150 151 152 153 154 155
static inline bool nvme_io_incapable(struct nvme_ctrl *ctrl)
{
	u32 val = 0;

	if (ctrl->ops->io_incapable(ctrl))
		return false;
	if (ctrl->ops->reg_read32(ctrl, NVME_REG_CSTS, &val))
		return false;
	return val & NVME_CSTS_CFS;
}

156 157 158 159 160 161 162
static inline int nvme_reset_subsystem(struct nvme_ctrl *ctrl)
{
	if (!ctrl->subsystem)
		return -ENOTTY;
	return ctrl->ops->reg_write32(ctrl, NVME_REG_NSSR, 0x4E564D65);
}

163 164 165 166 167
static inline u64 nvme_block_nr(struct nvme_ns *ns, sector_t sector)
{
	return (sector >> (ns->lba_shift - 9));
}

168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218
static inline void nvme_setup_flush(struct nvme_ns *ns,
		struct nvme_command *cmnd)
{
	memset(cmnd, 0, sizeof(*cmnd));
	cmnd->common.opcode = nvme_cmd_flush;
	cmnd->common.nsid = cpu_to_le32(ns->ns_id);
}

static inline void nvme_setup_rw(struct nvme_ns *ns, struct request *req,
		struct nvme_command *cmnd)
{
	u16 control = 0;
	u32 dsmgmt = 0;

	if (req->cmd_flags & REQ_FUA)
		control |= NVME_RW_FUA;
	if (req->cmd_flags & (REQ_FAILFAST_DEV | REQ_RAHEAD))
		control |= NVME_RW_LR;

	if (req->cmd_flags & REQ_RAHEAD)
		dsmgmt |= NVME_RW_DSM_FREQ_PREFETCH;

	memset(cmnd, 0, sizeof(*cmnd));
	cmnd->rw.opcode = (rq_data_dir(req) ? nvme_cmd_write : nvme_cmd_read);
	cmnd->rw.command_id = req->tag;
	cmnd->rw.nsid = cpu_to_le32(ns->ns_id);
	cmnd->rw.slba = cpu_to_le64(nvme_block_nr(ns, blk_rq_pos(req)));
	cmnd->rw.length = cpu_to_le16((blk_rq_bytes(req) >> ns->lba_shift) - 1);

	if (ns->ms) {
		switch (ns->pi_type) {
		case NVME_NS_DPS_PI_TYPE3:
			control |= NVME_RW_PRINFO_PRCHK_GUARD;
			break;
		case NVME_NS_DPS_PI_TYPE1:
		case NVME_NS_DPS_PI_TYPE2:
			control |= NVME_RW_PRINFO_PRCHK_GUARD |
					NVME_RW_PRINFO_PRCHK_REF;
			cmnd->rw.reftag = cpu_to_le32(
					nvme_block_nr(ns, blk_rq_pos(req)));
			break;
		}
		if (!blk_integrity_rq(req))
			control |= NVME_RW_PRINFO_PRACT;
	}

	cmnd->rw.control = cpu_to_le16(control);
	cmnd->rw.dsmgmt = cpu_to_le32(dsmgmt);
}


219 220 221 222 223 224 225 226 227 228 229 230
static inline int nvme_error_status(u16 status)
{
	switch (status & 0x7ff) {
	case NVME_SC_SUCCESS:
		return 0;
	case NVME_SC_CAP_EXCEEDED:
		return -ENOSPC;
	default:
		return -EIO;
	}
}

231 232 233 234 235 236
static inline bool nvme_req_needs_retry(struct request *req, u16 status)
{
	return !(status & NVME_SC_DNR || blk_noretry_request(req)) &&
		(jiffies - req->start_time) < req->timeout;
}

237 238 239
int nvme_disable_ctrl(struct nvme_ctrl *ctrl, u64 cap);
int nvme_enable_ctrl(struct nvme_ctrl *ctrl, u64 cap);
int nvme_shutdown_ctrl(struct nvme_ctrl *ctrl);
240 241
int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
		const struct nvme_ctrl_ops *ops, unsigned long quirks);
242
void nvme_uninit_ctrl(struct nvme_ctrl *ctrl);
243
void nvme_put_ctrl(struct nvme_ctrl *ctrl);
244
int nvme_init_identify(struct nvme_ctrl *ctrl);
245 246 247

void nvme_scan_namespaces(struct nvme_ctrl *ctrl);
void nvme_remove_namespaces(struct nvme_ctrl *ctrl);
248

249 250
void nvme_stop_queues(struct nvme_ctrl *ctrl);
void nvme_start_queues(struct nvme_ctrl *ctrl);
251

252 253
struct request *nvme_alloc_request(struct request_queue *q,
		struct nvme_command *cmd, unsigned int flags);
254
void nvme_requeue_req(struct request *req);
255 256 257
int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,
		void *buf, unsigned bufflen);
int __nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,
258 259
		struct nvme_completion *cqe, void *buffer, unsigned bufflen,
		unsigned timeout);
260 261 262
int nvme_submit_user_cmd(struct request_queue *q, struct nvme_command *cmd,
		void __user *ubuffer, unsigned bufflen, u32 *result,
		unsigned timeout);
263 264 265
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,
266
		u32 *result, unsigned timeout);
267 268
int nvme_identify_ctrl(struct nvme_ctrl *dev, struct nvme_id_ctrl **id);
int nvme_identify_ns(struct nvme_ctrl *dev, unsigned nsid,
269
		struct nvme_id_ns **id);
270 271
int nvme_get_log_page(struct nvme_ctrl *dev, struct nvme_smart_log **log);
int nvme_get_features(struct nvme_ctrl *dev, unsigned fid, unsigned nsid,
272
			dma_addr_t dma_addr, u32 *result);
273
int nvme_set_features(struct nvme_ctrl *dev, unsigned fid, unsigned dword11,
274
			dma_addr_t dma_addr, u32 *result);
C
Christoph Hellwig 已提交
275
int nvme_set_queue_count(struct nvme_ctrl *ctrl, int *count);
276 277 278 279 280 281 282

struct sg_io_hdr;

int nvme_sg_io(struct nvme_ns *ns, struct sg_io_hdr __user *u_hdr);
int nvme_sg_io32(struct nvme_ns *ns, unsigned long arg);
int nvme_sg_get_version_num(int __user *ip);

283
#ifdef CONFIG_NVM
M
Matias Bjørling 已提交
284 285 286
int nvme_nvm_ns_supported(struct nvme_ns *ns, struct nvme_id_ns *id);
int nvme_nvm_register(struct request_queue *q, char *disk_name);
void nvme_nvm_unregister(struct request_queue *q, char *disk_name);
287 288 289 290 291 292 293 294 295 296 297 298 299
#else
static inline int nvme_nvm_register(struct request_queue *q, char *disk_name)
{
	return 0;
}

static inline void nvme_nvm_unregister(struct request_queue *q, char *disk_name) {};

static inline int nvme_nvm_ns_supported(struct nvme_ns *ns, struct nvme_id_ns *id)
{
	return 0;
}
#endif /* CONFIG_NVM */
M
Matias Bjørling 已提交
300

301 302 303
int __init nvme_core_init(void);
void nvme_core_exit(void);

304
#endif /* _NVME_H */