1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
|
/*
* NVMe over Fabrics common host code.
* Copyright (c) 2015-2016 HGST, a Western Digital Company.
*
* 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_FABRICS_H
#define _NVME_FABRICS_H 1
#include <linux/in.h>
#include <linux/inet.h>
#define NVMF_MIN_QUEUE_SIZE 16
#define NVMF_MAX_QUEUE_SIZE 1024
#define NVMF_DEF_QUEUE_SIZE 128
#define NVMF_DEF_RECONNECT_DELAY 10
/* default to 600 seconds of reconnect attempts before giving up */
#define NVMF_DEF_CTRL_LOSS_TMO 600
/*
* Define a host as seen by the target. We allocate one at boot, but also
* allow the override it when creating controllers. This is both to provide
* persistence of the Host NQN over multiple boots, and to allow using
* multiple ones, for example in a container scenario. Because we must not
* use different Host NQNs with the same Host ID we generate a Host ID and
* use this structure to keep track of the relation between the two.
*/
struct nvmf_host {
struct kref ref;
struct list_head list;
char nqn[NVMF_NQN_SIZE];
uuid_t id;
};
/**
* enum nvmf_parsing_opts - used to define the sysfs parsing options used.
*/
enum {
NVMF_OPT_ERR = 0,
NVMF_OPT_TRANSPORT = 1 << 0,
NVMF_OPT_NQN = 1 << 1,
NVMF_OPT_TRADDR = 1 << 2,
NVMF_OPT_TRSVCID = 1 << 3,
NVMF_OPT_QUEUE_SIZE = 1 << 4,
NVMF_OPT_NR_IO_QUEUES = 1 << 5,
NVMF_OPT_TL_RETRY_COUNT = 1 << 6,
NVMF_OPT_KATO = 1 << 7,
NVMF_OPT_HOSTNQN = 1 << 8,
NVMF_OPT_RECONNECT_DELAY = 1 << 9,
NVMF_OPT_HOST_TRADDR = 1 << 10,
NVMF_OPT_CTRL_LOSS_TMO = 1 << 11,
NVMF_OPT_HOST_ID = 1 << 12,
NVMF_OPT_DUP_CONNECT = 1 << 13,
};
/**
* struct nvmf_ctrl_options - Used to hold the options specified
* with the parsing opts enum.
* @mask: Used by the fabrics library to parse through sysfs options
* on adding a NVMe controller.
* @transport: Holds the fabric transport "technology name" (for a lack of
* better description) that will be used by an NVMe controller
* being added.
* @subsysnqn: Hold the fully qualified NQN subystem name (format defined
* in the NVMe specification, "NVMe Qualified Names").
* @traddr: The transport-specific TRADDR field for a port on the
* subsystem which is adding a controller.
* @trsvcid: The transport-specific TRSVCID field for a port on the
* subsystem which is adding a controller.
* @host_traddr: A transport-specific field identifying the NVME host port
* to use for the connection to the controller.
* @queue_size: Number of IO queue elements.
* @nr_io_queues: Number of controller IO queues that will be established.
* @reconnect_delay: Time between two consecutive reconnect attempts.
* @discovery_nqn: indicates if the subsysnqn is the well-known discovery NQN.
* @kato: Keep-alive timeout.
* @host: Virtual NVMe host, contains the NQN and Host ID.
* @max_reconnects: maximum number of allowed reconnect attempts before removing
* the controller, (-1) means reconnect forever, zero means remove
* immediately;
*/
struct nvmf_ctrl_options {
unsigned mask;
char *transport;
char *subsysnqn;
char *traddr;
char *trsvcid;
char *host_traddr;
size_t queue_size;
unsigned int nr_io_queues;
unsigned int reconnect_delay;
bool discovery_nqn;
bool duplicate_connect;
unsigned int kato;
struct nvmf_host *host;
int max_reconnects;
};
/*
* struct nvmf_transport_ops - used to register a specific
* fabric implementation of NVMe fabrics.
* @entry: Used by the fabrics library to add the new
* registration entry to its linked-list internal tree.
* @module: Transport module reference
* @name: Name of the NVMe fabric driver implementation.
* @required_opts: sysfs command-line options that must be specified
* when adding a new NVMe controller.
* @allowed_opts: sysfs command-line options that can be specified
* when adding a new NVMe controller.
* @create_ctrl(): function pointer that points to a non-NVMe
* implementation-specific fabric technology
* that would go into starting up that fabric
* for the purpose of conneciton to an NVMe controller
* using that fabric technology.
*
* Notes:
* 1. At minimum, 'required_opts' and 'allowed_opts' should
* be set to the same enum parsing options defined earlier.
* 2. create_ctrl() must be defined (even if it does nothing)
*/
struct nvmf_transport_ops {
struct list_head entry;
struct module *module;
const char *name;
int required_opts;
int allowed_opts;
struct nvme_ctrl *(*create_ctrl)(struct device *dev,
struct nvmf_ctrl_options *opts);
};
static inline bool
nvmf_ctlr_matches_baseopts(struct nvme_ctrl *ctrl,
struct nvmf_ctrl_options *opts)
{
if (strcmp(opts->subsysnqn, ctrl->opts->subsysnqn) ||
strcmp(opts->host->nqn, ctrl->opts->host->nqn) ||
memcmp(&opts->host->id, &ctrl->opts->host->id, sizeof(uuid_t)))
return false;
return true;
}
int nvmf_reg_read32(struct nvme_ctrl *ctrl, u32 off, u32 *val);
int nvmf_reg_read64(struct nvme_ctrl *ctrl, u32 off, u64 *val);
int nvmf_reg_write32(struct nvme_ctrl *ctrl, u32 off, u32 val);
int nvmf_connect_admin_queue(struct nvme_ctrl *ctrl);
int nvmf_connect_io_queue(struct nvme_ctrl *ctrl, u16 qid);
int nvmf_register_transport(struct nvmf_transport_ops *ops);
void nvmf_unregister_transport(struct nvmf_transport_ops *ops);
void nvmf_free_options(struct nvmf_ctrl_options *opts);
int nvmf_get_address(struct nvme_ctrl *ctrl, char *buf, int size);
bool nvmf_should_reconnect(struct nvme_ctrl *ctrl);
static inline blk_status_t nvmf_check_init_req(struct nvme_ctrl *ctrl,
struct request *rq)
{
struct nvme_command *cmd = nvme_req(rq)->cmd;
/*
* We cannot accept any other command until the connect command has
* completed, so only allow connect to pass.
*/
if (!blk_rq_is_passthrough(rq) ||
cmd->common.opcode != nvme_fabrics_command ||
cmd->fabrics.fctype != nvme_fabrics_type_connect) {
/*
* Reconnecting state means transport disruption, which can take
* a long time and even might fail permanently, fail fast to
* give upper layers a chance to failover.
* Deleting state means that the ctrl will never accept commands
* again, fail it permanently.
*/
if (ctrl->state == NVME_CTRL_RECONNECTING ||
ctrl->state == NVME_CTRL_DELETING) {
nvme_req(rq)->status = NVME_SC_ABORT_REQ;
return BLK_STS_IOERR;
}
return BLK_STS_RESOURCE; /* try again later */
}
return BLK_STS_OK;
}
#endif /* _NVME_FABRICS_H */
|
