summaryrefslogtreecommitdiffstats
path: root/vlib/vlib/buffer_node.h
blob: 5985ccf500dba3e3259cc5a779688fb2b137ae83 (plain)
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
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
/*
 * Copyright (c) 2015 Cisco and/or its affiliates.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at:
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */
/*
 * buffer_node.h: VLIB buffer handling node helper macros/inlines
 *
 * Copyright (c) 2008 Eliot Dresselhaus
 *
 * Permission is hereby granted, free of charge, to any person obtaining
 * a copy of this software and associated documentation files (the
 * "Software"), to deal in the Software without restriction, including
 * without limitation the rights to use, copy, modify, merge, publish,
 * distribute, sublicense, and/or sell copies of the Software, and to
 * permit persons to whom the Software is furnished to do so, subject to
 * the following conditions:
 *
 * The above copyright notice and this permission notice shall be
 * included in all copies or substantial portions of the Software.
 *
 *  THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
 *  EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
 *  MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
 *  NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
 *  LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
 *  OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
 *  WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 */

#ifndef included_vlib_buffer_node_h
#define included_vlib_buffer_node_h

/** \file 
    vlib buffer/node functions
*/

/** \brief Finish enqueueing two buffers forward in the graph.
 Standard dual loop boilerplate element. This is a MACRO,
 with MULTIPLE SIDE EFFECTS. In the ideal case, 
 <code>next_index == next0 == next1</code>,
 which means that the speculative enqueue at the top of the dual loop
 has correctly dealt with both packets. In that case, the macro does
 nothing at all.

 @param vm vlib_main_t pointer, varies by thread
 @param node current node vlib_node_runtime_t pointer
 @param next_index speculated next index used for both packets
 @param to_next speculated vector pointer used for both packets
 @param n_left_to_next number of slots left in speculated vector
 @param bi0 first buffer index
 @param bi1 second buffer index
 @param next0 actual next index to be used for the first packet
 @param next1 actual next index to be used for the second packet

 @return @c next_index -- speculative next index to be used for future packets
 @return @c to_next -- speculative frame to be used for future packets
 @return @c n_left_to_next -- number of slots left in speculative frame
*/

#define vlib_validate_buffer_enqueue_x2(vm,node,next_index,to_next,n_left_to_next,bi0,bi1,next0,next1) \
do {									\
  int enqueue_code = (next0 != next_index) + 2*(next1 != next_index);	\
									\
  if (PREDICT_FALSE (enqueue_code != 0))				\
    {									\
      switch (enqueue_code)						\
	{								\
	case 1:								\
	  /* A B A */							\
	  to_next[-2] = bi1;						\
	  to_next -= 1;							\
	  n_left_to_next += 1;						\
	  vlib_set_next_frame_buffer (vm, node, next0, bi0);		\
	  break;							\
									\
	case 2:								\
	  /* A A B */							\
	  to_next -= 1;							\
	  n_left_to_next += 1;						\
	  vlib_set_next_frame_buffer (vm, node, next1, bi1);		\
	  break;							\
									\
	case 3:								\
	  /* A B B or A B C */						\
	  to_next -= 2;							\
	  n_left_to_next += 2;						\
	  vlib_set_next_frame_buffer (vm, node, next0, bi0);		\
	  vlib_set_next_frame_buffer (vm, node, next1, bi1);		\
	  if (next0 == next1)						\
	    {								\
	      vlib_put_next_frame (vm, node, next_index,		\
				   n_left_to_next);			\
	      next_index = next1;					\
	      vlib_get_next_frame (vm, node, next_index, to_next, n_left_to_next); \
	    }								\
	}								\
    }									\
} while (0)

/** \brief Finish enqueueing one buffer forward in the graph.
 Standard single loop boilerplate element. This is a MACRO,
 with MULTIPLE SIDE EFFECTS. In the ideal case,
 <code>next_index == next0</code>,
 which means that the speculative enqueue at the top of the single loop
 has correctly dealt with the packet in hand. In that case, the macro does
 nothing at all.

 @param vm vlib_main_t pointer, varies by thread
 @param node current node vlib_node_runtime_t pointer
 @param next_index speculated next index used for both packets
 @param to_next speculated vector pointer used for both packets
 @param n_left_to_next number of slots left in speculated vector
 @param bi0 first buffer index
 @param next0 actual next index to be used for the first packet

 @return @c next_index -- speculative next index to be used for future packets
 @return @c to_next -- speculative frame to be used for future packets
 @return @c n_left_to_next -- number of slots left in speculative frame
*/
#define vlib_validate_buffer_enqueue_x1(vm,node,next_index,to_next,n_left_to_next,bi0,next0) \
do {									\
  if (PREDICT_FALSE (next0 != next_index))				\
    {									\
      vlib_put_next_frame (vm, node, next_index, n_left_to_next + 1);	\
      next_index = next0;						\
      vlib_get_next_frame (vm, node, next_index, to_next, n_left_to_next); \
									\
      to_next[0] = bi0;							\
      to_next += 1;							\
      n_left_to_next -= 1;						\
    }									\
} while (0)

always_inline uword
generic_buffer_node_inline (vlib_main_t * vm,
			    vlib_node_runtime_t * node,
			    vlib_frame_t * frame,
			    uword sizeof_trace,
			    void * opaque1,
			    uword opaque2,
			    void (* two_buffers) (vlib_main_t * vm,
						  void * opaque1,
						  uword opaque2,
						  vlib_buffer_t * b0, vlib_buffer_t * b1,
						  u32 * next0, u32 * next1),
			    void (* one_buffer) (vlib_main_t * vm,
						 void * opaque1,
						 uword opaque2,
						 vlib_buffer_t * b0,
						 u32 * next0))
{
  u32 n_left_from, * from, * to_next;
  u32 next_index;

  from = vlib_frame_vector_args (frame);
  n_left_from = frame->n_vectors;
  next_index = node->cached_next_index;

  if (node->flags & VLIB_NODE_FLAG_TRACE)
    vlib_trace_frame_buffers_only (vm, node, from, frame->n_vectors,
				   /* stride */ 1, sizeof_trace);

  while (n_left_from > 0)
    {
      u32 n_left_to_next;

      vlib_get_next_frame (vm, node, next_index,
			   to_next, n_left_to_next);

      while (n_left_from >= 4 && n_left_to_next >= 2)
	{
	  vlib_buffer_t * p0, * p1;
	  u32 pi0, next0;
	  u32 pi1, next1;

	  /* Prefetch next iteration. */
	  {
	    vlib_buffer_t * p2, * p3;

	    p2 = vlib_get_buffer (vm, from[2]);
	    p3 = vlib_get_buffer (vm, from[3]);

	    vlib_prefetch_buffer_header (p2, LOAD);
	    vlib_prefetch_buffer_header (p3, LOAD);

	    CLIB_PREFETCH (p2->data, 64, LOAD);
	    CLIB_PREFETCH (p3->data, 64, LOAD);
	  }

	  pi0 = to_next[0] = from[0];
	  pi1 = to_next[1] = from[1];
	  from += 2;
	  to_next += 2;
	  n_left_from -= 2;
	  n_left_to_next -= 2;

	  p0 = vlib_get_buffer (vm, pi0);
	  p1 = vlib_get_buffer (vm, pi1);

	  two_buffers (vm, opaque1, opaque2, p0, p1, &next0, &next1);

	  vlib_validate_buffer_enqueue_x2 (vm, node, next_index,
					   to_next, n_left_to_next,
					   pi0, pi1, next0, next1);
	}
    
      while (n_left_from > 0 && n_left_to_next > 0)
	{
	  vlib_buffer_t * p0;
	  u32 pi0, next0;

	  pi0 = from[0];
	  to_next[0] = pi0;
	  from += 1;
	  to_next += 1;
	  n_left_from -= 1;
	  n_left_to_next -= 1;

	  p0 = vlib_get_buffer (vm, pi0);

	  one_buffer (vm, opaque1, opaque2, p0, &next0);

	  vlib_validate_buffer_enqueue_x1 (vm, node, next_index,
					   to_next, n_left_to_next,
					   pi0, next0);
	}

      vlib_put_next_frame (vm, node, next_index, n_left_to_next);
    }

  return frame->n_vectors;
}

#endif /* included_vlib_buffer_node_h */