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
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
219
|
Asynchronous Transfers/Transforms API
1 INTRODUCTION
2 GENEALOGY
3 USAGE
3.1 General format of the API
3.2 Supported operations
3.3 Descriptor management
3.4 When does the operation execute?
3.5 When does the operation complete?
3.6 Constraints
3.7 Example
4 DRIVER DEVELOPER NOTES
4.1 Conformance points
4.2 "My application needs finer control of hardware channels"
5 SOURCE
---
1 INTRODUCTION
The async_tx API provides methods for describing a chain of asynchronous
bulk memory transfers/transforms with support for inter-transactional
dependencies. It is implemented as a dmaengine client that smooths over
the details of different hardware offload engine implementations. Code
that is written to the API can optimize for asynchronous operation and
the API will fit the chain of operations to the available offload
resources.
2 GENEALOGY
The API was initially designed to offload the memory copy and
xor-parity-calculations of the md-raid5 driver using the offload engines
present in the Intel(R) Xscale series of I/O processors. It also built
on the 'dmaengine' layer developed for offloading memory copies in the
network stack using Intel(R) I/OAT engines. The following design
features surfaced as a result:
1/ implicit synchronous path: users of the API do not need to know if
the platform they are running on has offload capabilities. The
operation will be offloaded when an engine is available and carried out
in software otherwise.
2/ cross channel dependency chains: the API allows a chain of dependent
operations to be submitted, like xor->copy->xor in the raid5 case. The
API automatically handles cases where the transition from one operation
to another implies a hardware channel switch.
3/ dmaengine extensions to support multiple clients and operation types
beyond 'memcpy'
3 USAGE
3.1 General format of the API:
struct dma_async_tx_descriptor *
async_<operation>(<op specific parameters>,
enum async_tx_flags flags,
struct dma_async_tx_descriptor *dependency,
dma_async_tx_callback callback_routine,
void *callback_parameter);
3.2 Supported operations:
memcpy - memory copy between a source and a destination buffer
memset - fill a destination buffer with a byte value
xor - xor a series of source buffers and write the result to a
destination buffer
xor_zero_sum - xor a series of source buffers and set a flag if the
result is zero. The implementation attempts to prevent
writes to memory
3.3 Descriptor management:
The return value is non-NULL and points to a 'descriptor' when the operation
has been queued to execute asynchronously. Descriptors are recycled
resources, under control of the offload engine driver, to be reused as
operations complete. When an application needs to submit a chain of
operations it must guarantee that the descriptor is not automatically recycled
before the dependency is submitted. This requires that all descriptors be
acknowledged by the application before the offload engine driver is allowed to
recycle (or free) the descriptor. A descriptor can be acked by one of the
following methods:
1/ setting the ASYNC_TX_ACK flag if no child operations are to be submitted
2/ setting the ASYNC_TX_DEP_ACK flag to acknowledge the parent
descriptor of a new operation.
3/ calling async_tx_ack() on the descriptor.
3.4 When does the operation execute?
Operations do not immediately issue after return from the
async_<operation> call. Offload engine drivers batch operations to
improve performance by reducing the number of mmio cycles needed to
manage the channel. Once a driver-specific threshold is met the driver
automatically issues pending operations. An application can force this
event by calling async_tx_issue_pending_all(). This operates on all
channels since the application has no knowledge of channel to operation
mapping.
3.5 When does the operation complete?
There are two methods for an application to learn about the completion
of an operation.
1/ Call dma_wait_for_async_tx(). This call causes the CPU to spin while
it polls for the completion of the operation. It handles dependency
chains and issuing pending operations.
2/ Specify a completion callback. The callback routine runs in tasklet
context if the offload engine driver supports interrupts, or it is
called in application context if the operation is carried out
synchronously in software. The callback can be set in the call to
async_<operation>, or when the application needs to submit a chain of
unknown length it can use the async_trigger_callback() routine to set a
completion interrupt/callback at the end of the chain.
3.6 Constraints:
1/ Calls to async_<operation> are not permitted in IRQ context. Other
contexts are permitted provided constraint #2 is not violated.
2/ Completion callback routines cannot submit new operations. This
results in recursion in the synchronous case and spin_locks being
acquired twice in the asynchronous case.
3.7 Example:
Perform a xor->copy->xor operation where each operation depends on the
result from the previous operation:
void complete_xor_copy_xor(void *param)
{
printk("complete\n");
}
int run_xor_copy_xor(struct page **xor_srcs,
int xor_src_cnt,
struct page *xor_dest,
size_t xor_len,
struct page *copy_src,
struct page *copy_dest,
size_t copy_len)
{
struct dma_async_tx_descriptor *tx;
tx = async_xor(xor_dest, xor_srcs, 0, xor_src_cnt, xor_len,
ASYNC_TX_XOR_DROP_DST, NULL, NULL, NULL);
tx = async_memcpy(copy_dest, copy_src, 0, 0, copy_len,
ASYNC_TX_DEP_ACK, tx, NULL, NULL);
tx = async_xor(xor_dest, xor_srcs, 0, xor_src_cnt, xor_len,
ASYNC_TX_XOR_DROP_DST | ASYNC_TX_DEP_ACK | ASYNC_TX_ACK,
tx, complete_xor_copy_xor, NULL);
async_tx_issue_pending_all();
}
See include/linux/async_tx.h for more information on the flags. See the
ops_run_* and ops_complete_* routines in drivers/md/raid5.c for more
implementation examples.
4 DRIVER DEVELOPMENT NOTES
4.1 Conformance points:
There are a few conformance points required in dmaengine drivers to
accommodate assumptions made by applications using the async_tx API:
1/ Completion callbacks are expected to happen in tasklet context
2/ dma_async_tx_descriptor fields are never manipulated in IRQ context
3/ Use async_tx_run_dependencies() in the descriptor clean up path to
handle submission of dependent operations
4.2 "My application needs finer control of hardware channels"
This requirement seems to arise from cases where a DMA engine driver is
trying to support device-to-memory DMA. The dmaengine and async_tx
implementations were designed for offloading memory-to-memory
operations; however, there are some capabilities of the dmaengine layer
that can be used for platform-specific channel management.
Platform-specific constraints can be handled by registering the
application as a 'dma_client' and implementing a 'dma_event_callback' to
apply a filter to the available channels in the system. Before showing
how to implement a custom dma_event callback some background of
dmaengine's client support is required.
The following routines in dmaengine support multiple clients requesting
use of a channel:
- dma_async_client_register(struct dma_client *client)
- dma_async_client_chan_request(struct dma_client *client)
dma_async_client_register takes a pointer to an initialized dma_client
structure. It expects that the 'event_callback' and 'cap_mask' fields
are already initialized.
dma_async_client_chan_request triggers dmaengine to notify the client of
all channels that satisfy the capability mask. It is up to the client's
event_callback routine to track how many channels the client needs and
how many it is currently using. The dma_event_callback routine returns a
dma_state_client code to let dmaengine know the status of the
allocation.
Below is the example of how to extend this functionality for
platform-specific filtering of the available channels beyond the
standard capability mask:
static enum dma_state_client
my_dma_client_callback(struct dma_client *client,
struct dma_chan *chan, enum dma_state state)
{
struct dma_device *dma_dev;
struct my_platform_specific_dma *plat_dma_dev;
dma_dev = chan->device;
plat_dma_dev = container_of(dma_dev,
struct my_platform_specific_dma,
dma_dev);
if (!plat_dma_dev->platform_specific_capability)
return DMA_DUP;
. . .
}
5 SOURCE
include/linux/dmaengine.h: core header file for DMA drivers and clients
drivers/dma/dmaengine.c: offload engine channel management routines
drivers/dma/: location for offload engine drivers
include/linux/async_tx.h: core header file for the async_tx api
crypto/async_tx/async_tx.c: async_tx interface to dmaengine and common code
crypto/async_tx/async_memcpy.c: copy offload
crypto/async_tx/async_memset.c: memory fill offload
crypto/async_tx/async_xor.c: xor and xor zero sum offload
|
