1fi_rma(3) Libfabric v1.12.0rc1 fi_rma(3)
2
6 fi_rma - Remote memory access operations
7 fi_read / fi_readv / fi_readmsg
9 Initiates a read from remote memory
10 fi_write / fi_writev / fi_writemsg fi_inject_write / fi_writedata :
12 Initiate a write to remote memory
13
15 #include <rdma/fi_rma.h>
16 ssize_t fi_read(struct fid_ep *ep, void *buf, size_t len, void *desc,
18 fi_addr_t src_addr, uint64_t addr, uint64_t key, void *context);
19 ssize_t fi_readv(struct fid_ep *ep, const struct iovec *iov, void **desc,
21 size_t count, fi_addr_t src_addr, uint64_t addr, uint64_t key,
22 void *context);
23 ssize_t fi_readmsg(struct fid_ep *ep, const struct fi_msg_rma *msg,
25 uint64_t flags);
26 ssize_t fi_write(struct fid_ep *ep, const void *buf, size_t len,
28 void *desc, fi_addr_t dest_addr, uint64_t addr, uint64_t key,
29 void *context);
30 ssize_t fi_writev(struct fid_ep *ep, const struct iovec *iov, void **desc,
32 size_t count, fi_addr_t dest_addr, uint64_t addr, uint64_t key,
33 void *context);
34 ssize_t fi_writemsg(struct fid_ep *ep, const struct fi_msg_rma *msg,
36 uint64_t flags);
37 ssize_t fi_inject_write(struct fid_ep *ep, const void *buf, size_t len,
39 fi_addr_t dest_addr, uint64_t addr, uint64_t key);
40 ssize_t fi_writedata(struct fid_ep *ep, const void *buf, size_t len,
42 void *desc, uint64_t data, fi_addr_t dest_addr, uint64_t addr,
43 uint64_t key, void *context);
44 ssize_t fi_inject_writedata(struct fid_ep *ep, const void *buf, size_t len,
46 uint64_t data, fi_addr_t dest_addr, uint64_t addr, uint64_t key);
47
49 ep Fabric endpoint on which to initiate read or write operation.
50 buf Local data buffer to read into (read target) or write from
52 (write source)
53 len Length of data to read or write, specified in bytes. Valid
55 transfers are from 0 bytes up to the endpoint's max_msg_size.
56 iov Vectored data buffer.
58 count Count of vectored data entries.
60 addr Address of remote memory to access. This will be the virtual
62 address of the remote region in the case of FI_MR_BASIC, or the
63 offset from the starting address in the case of FI_MR_SCALABLE.
64 key Protection key associated with the remote memory.
66 desc Descriptor associated with the local data buffer See fi_mr(3).
68 data Remote CQ data to transfer with the operation.
70 dest_addr
72 Destination address for connectionless write transfers. Ignored
73 for connected endpoints.
74 src_addr
76 Source address to read from for connectionless transfers. Ig‐
77 nored for connected endpoints.
78 msg Message descriptor for read and write operations.
80 flags Additional flags to apply for the read or write operation.
82 context
84 User specified pointer to associate with the operation. This
85 parameter is ignored if the operation will not generate a suc‐
86 cessful completion, unless an op flag specifies the context pa‐
87 rameter be used for required input.
88
90 RMA (remote memory access) operations are used to transfer data direct‐
91 ly between a local data buffer and a remote data buffer. RMA transfers
92 occur on a byte level granularity, and no message boundaries are main‐
93 tained.
94 The write functions -- fi_write, fi_writev, fi_writemsg, fi_in‐
96 ject_write, and fi_writedata -- are used to transmit data into a remote
97 memory buffer. The main difference between write functions are the
98 number and type of parameters that they accept as input. Otherwise,
99 they perform the same general function.
100 The read functions -- fi_read, fi_readv, and fi_readmsg -- are used to
102 transfer data from a remote memory region into local data buffer(s).
103 Similar to the write operations, read operations operate asynchronous‐
104 ly. Users should not touch the posted data buffer(s) until the read
105 operation has completed.
106 Completed RMA operations are reported to the user through one or more
108 completion queues associated with the endpoint. Users provide context
109 which are associated with each operation, and is returned to the user
110 as part of the completion. See fi_cq for completion event details.
111 By default, the remote endpoint does not generate an event or notify
113 the user when a memory region has been accessed by an RMA read or write
114 operation. However, immediate data may be associated with an RMA write
115 operation. RMA writes with immediate data will generate a completion
116 entry at the remote endpoint, so that the immediate data may be deliv‐
117 ered.
118 fi_write
120 The call fi_write transfers the data contained in the user-specified
121 data buffer to a remote memory region.
122 fi_writev
124 The fi_writev call adds support for a scatter-gather list to fi_write.
125 The fi_writev transfers the set of data buffers referenced by the iov
126 parameter to the remote memory region.
127 fi_writemsg
129 The fi_writemsg call supports data transfers over both connected and
130 connectionless endpoints, with the ability to control the write opera‐
131 tion per call through the use of flags. The fi_writemsg function takes
132 a struct fi_msg_rma as input.
133 struct fi_msg_rma {
135 const struct iovec *msg_iov; /* local scatter-gather array */
136 void **desc; /* operation descriptor */
137 size_t iov_count; /* # elements in msg_iov */
138 fi_addr_t addr; /* optional endpoint address */
139 const struct fi_rma_iov *rma_iov;/* remote SGL */
140 size_t rma_iov_count;/* # elements in rma_iov */
141 void *context; /* user-defined context */
142 uint64_t data; /* optional immediate data */
143 };
144 struct fi_rma_iov {
146 uint64_t addr; /* target RMA address */
147 size_t len; /* size of target buffer */
148 uint64_t key; /* access key */
149 };
150 fi_inject_write
152 The write inject call is an optimized version of fi_write. It provides
153 similar completion semantics as fi_inject fi_msg(3).
154 fi_writedata
156 The write data call is similar to fi_write, but allows for the sending
157 of remote CQ data (see FI_REMOTE_CQ_DATA flag) as part of the transfer.
158 fi_inject_writedata
160 The inject write data call is similar to fi_inject_write, but allows
161 for the sending of remote CQ data (see FI_REMOTE_CQ_DATA flag) as part
162 of the transfer.
163 fi_read
165 The fi_read call requests that the remote endpoint transfer data from
166 the remote memory region into the local data buffer.
167 fi_readv
169 The fi_readv call adds support for a scatter-gather list to fi_read.
170 The fi_readv transfers data from the remote memory region into the set
171 of data buffers referenced by the iov parameter.
172 fi_readmsg
174 The fi_readmsg call supports data transfers over both connected and
175 connectionless endpoints, with the ability to control the read opera‐
176 tion per call through the use of flags. The fi_readmsg function takes
177 a struct fi_msg_rma as input.
178
180 The fi_readmsg and fi_writemsg calls allow the user to specify flags
181 which can change the default data transfer operation. Flags specified
182 with fi_readmsg / fi_writemsg override most flags previously configured
183 with the endpoint, except where noted (see fi_endpoint.3). The follow‐
184 ing list of flags are usable with fi_readmsg and/or fi_writemsg.
185 FI_REMOTE_CQ_DATA
187 Applies to fi_writemsg and fi_writedata. Indicates that remote
188 CQ data is available and should be sent as part of the request.
189 See fi_getinfo for additional details on FI_REMOTE_CQ_DATA.
190 FI_COMPLETION
192 Indicates that a completion entry should be generated for the
193 specified operation. The endpoint must be bound to a completion
194 queue with FI_SELECTIVE_COMPLETION that corresponds to the spec‐
195 ified operation, or this flag is ignored.
196 FI_MORE
198 Indicates that the user has additional requests that will imme‐
199 diately be posted after the current call returns. Use of this
200 flag may improve performance by enabling the provider to opti‐
201 mize its access to the fabric hardware.
202 FI_INJECT
204 Applies to fi_writemsg. Indicates that the outbound data buffer
205 should be returned to user immediately after the write call re‐
206 turns, even if the operation is handled asynchronously. This
207 may require that the underlying provider implementation copy the
208 data into a local buffer and transfer out of that buffer. This
209 flag can only be used with messages smaller than inject_size.
210 FI_INJECT_COMPLETE
212 Applies to fi_writemsg. Indicates that a completion should be
213 generated when the source buffer(s) may be reused.
214 FI_TRANSMIT_COMPLETE
216 Applies to fi_writemsg. Indicates that a completion should not
217 be generated until the operation has been successfully transmit‐
218 ted and is no longer being tracked by the provider.
219 FI_DELIVERY_COMPLETE
221 Applies to fi_writemsg. Indicates that a completion should be
222 generated when the operation has been processed by the destina‐
223 tion.
224 FI_COMMIT_COMPLETE
226 Applies to fi_writemsg when targeting persistent memory regions.
227 Indicates that a completion should be generated only after the
228 result of the operation has been made durable.
229 FI_FENCE
231 Applies to transmits. Indicates that the requested operation,
232 also known as the fenced operation, and any operation posted af‐
233 ter the fenced operation will be deferred until all previous op‐
234 erations targeting the same peer endpoint have completed. Oper‐
235 ations posted after the fencing will see and/or replace the re‐
236 sults of any operations initiated prior to the fenced operation.
237 The ordering of operations starting at the posting of the fenced opera‐
239 tion (inclusive) to the posting of a subsequent fenced operation (ex‐
240 clusive) is controlled by the endpoint's ordering semantics.
241
243 Returns 0 on success. On error, a negative value corresponding to fab‐
244 ric errno is returned. Fabric errno values are defined in rdma/fi_er‐
245 rno.h.
246
248 -FI_EAGAIN
249 See fi_msg(3) for a detailed description of handling FI_EAGAIN.
250
252 fi_getinfo(3), fi_endpoint(3), fi_domain(3), fi_cq(3)
253
255 OpenFabrics.
256Libfabric Programmer's Manual 2020-10-14 fi_rma(3)