Project

General

Profile

MMU description » History » Version 98

Alexander Kamkin, 12/04/2014 03:08 PM

1 24 Alexander Kamkin
h1. MMU Description
2 1 Taya Sergeeva
3 66 Alexander Kamkin
_~By Alexander Kamkin and Taya Sergeeva~_
4 62 Alexander Kamkin
5 63 Alexander Kamkin
{{toc}}
6
7 35 Alexander Kamkin
A _memory management unit_ (_MMU_) is known to be one of the most complex and error-prone components of a modern microprocessor. MicroTESK has a special subsystem, called _MMU subsystem_, intended for (1) specifying memory devices and (2) deriving testing knowledge from such specifications. The subsystem provides unified facilities for describing memory buffers (like _L1_ and _L2 caches_, _translation look-aside buffers_ (_TLBs_), etc.) as well as a means for connecting several buffers into a memory hierarchy.
8 1 Taya Sergeeva
9 72 Alexander Kamkin
h2. Grammar
10 66 Alexander Kamkin
11
<pre>
12
startRule 
13
    : bufferOrAddress*
14
    ;
15
16
bufferOrAddress
17
    : address
18
    | buffer
19
    ;
20
</pre>
21
22 89 Alexander Kamkin
The expression syntax is derived from nML/Sim-nML (see _MicroTESK Language Reference_).
23
24 91 Alexander Kamkin
h2. Address Description (address)
25 56 Taya Sergeeva
26 1 Taya Sergeeva
A buffer is accessed by an _address_, which is typically a _bit vector_ of a fixed length (width). Different buffers are allowed to have a common address space (e.g., L1 and L2 are usually both addressed by physical addresses). However, in general case, each buffer has its own domain.
27
28 76 Alexander Kamkin
An address space is described using a keyword @address@. The description can specify two parameters: @width@ (obligatory) and @format@ (optional).
29 1 Taya Sergeeva
30 75 Alexander Kamkin
h3. Grammar
31 69 Alexander Kamkin
32
<pre>
33
address
34 90 Alexander Kamkin
    : ''address'' ID
35
        (addressParameter)*
36 69 Alexander Kamkin
    ;
37
38
addressParameter
39
    : width
40 97 Alexander Kamkin
    | segment
41 69 Alexander Kamkin
    | format
42
    ;
43
</pre>
44
45 79 Alexander Kamkin
h3. Address Width (width)
46 1 Taya Sergeeva
47 84 Alexander Kamkin
The @width@ parameter specifies the address _width_. The parameter is obligatory; its value should be non-negative (zero-length addresses are permitted).
48 1 Taya Sergeeva
49 69 Alexander Kamkin
h4. Grammar
50
51
<pre>
52
width
53
    : ''width'' ''='' expr
54
    ;
55
</pre>
56
57 97 Alexander Kamkin
h3. Address Space Segment (segment)
58
59
The @segment@ parameter specifies the address space _segment_. Any number (&ge; 0) of segments can be specified for one address. Each segment is characterized by its _name_ and _address range_. Different segments should have different names, but address ranges are allowed to overlap, and moreover, to be the same.
60
61
h4. Grammar
62
63
<pre>
64
segment
65
    : ''segment'' segmentID ''='' ''('' expr '','' expr '')''
66
    ;
67
</pre>
68
69 68 Alexander Kamkin
h3. Address Format (format)
70 95 Alexander Kamkin
71 98 Alexander Kamkin
The @format@ parameter specifies the address _format_ (a number of named fields). Any number (&ge; 0) of formats (with different names) can be specified for one address.
72 94 Alexander Kamkin
73
A field has three attributes: a _name_, a _width_ and, optionally, an _initial value_.
74 69 Alexander Kamkin
75
h4. Grammar
76
77
<pre>
78 1 Taya Sergeeva
format
79 97 Alexander Kamkin
    : ''format'' formatID? ''='' ''(''
80 69 Alexander Kamkin
        field ('','' field)*
81
      '')''
82
    ;
83
84
field
85
    : ID '':'' expr (''='' expr)?
86
    ;
87
</pre>
88
89 72 Alexander Kamkin
h2. Examples
90 1 Taya Sergeeva
91
<pre>
92 66 Alexander Kamkin
// The singleton.
93 90 Alexander Kamkin
address Void
94 1 Taya Sergeeva
  // The address width is zero (this is admissible for single-item buffers).
95 90 Alexander Kamkin
  width = 0
96 66 Alexander Kamkin
</pre>
97 1 Taya Sergeeva
98 88 Alexander Kamkin
<pre>
99 90 Alexander Kamkin
// An unstructured 64-bit virtual addresses (VA).
100 84 Alexander Kamkin
address VA
101 90 Alexander Kamkin
  // The address width.
102 1 Taya Sergeeva
  width = 64
103
</pre>
104 66 Alexander Kamkin
105
<pre>
106 90 Alexander Kamkin
// A stuctured 40-bit physical addresses (PA).
107 1 Taya Sergeeva
address PA
108 90 Alexander Kamkin
  // The address width.
109 1 Taya Sergeeva
  width = 40
110 84 Alexander Kamkin
  // The address format: (<39..36>, TAG=<35..12>, INDEX=<11..5>, LOCAL=<4..0>).
111 97 Alexander Kamkin
  format PA_L1 = (
112 84 Alexander Kamkin
    TAG   : 24, // The tag (the <35..12> address bits).
113 1 Taya Sergeeva
    INDEX : 7,  // The set index (the <11..5> address bits).
114
    LOCAL : 5,  // The byte position (the <0..4> address bits).
115 90 Alexander Kamkin
  )
116 84 Alexander Kamkin
</pre>
117 2 Taya Sergeeva
118 91 Alexander Kamkin
h2. Buffer Description (buffer)
119 76 Alexander Kamkin
120 1 Taya Sergeeva
A buffer is described using a keyword @buffer@. The description specifies a set of parameters, including @ways@, @sets@, @format@, @index@, @match@ and @policy@. All of the parameters except @index@ (if @sets = 1@) and @policy@ are obligatory.
121 75 Alexander Kamkin
122
h3. Grammar
123 1 Taya Sergeeva
124 75 Alexander Kamkin
<pre>
125 83 Alexander Kamkin
buffer
126 90 Alexander Kamkin
    : ''buffer'' bufferTypeID ''('' addressTypeID addressArgID '')''
127
        (bufferParameter)*
128 75 Alexander Kamkin
    ;
129
130
bufferParameter
131
    : ways
132
    | sets
133
    | format
134
    | index
135
    | match
136
    | policy
137
    ;
138 1 Taya Sergeeva
</pre>
139
140 78 Alexander Kamkin
h3. Buffer Associativity (ways)
141
142 84 Alexander Kamkin
The @ways@ parameter specifies the buffer _associativity_ (the number of lines in a set). The parameter is obligatory; its value should be positive.
143 78 Alexander Kamkin
144 1 Taya Sergeeva
h4. Grammar
145 80 Alexander Kamkin
146
<pre>
147
ways
148
    : ''ways'' ''='' expr
149 1 Taya Sergeeva
    ;
150
</pre>
151
152 80 Alexander Kamkin
h3. Buffer Length (sets)
153 83 Alexander Kamkin
154 84 Alexander Kamkin
The @sets@ parameter specifies the buffer _length_ (the number of sets a cache). The parameter is obligatory; its value should be positive.
155 78 Alexander Kamkin
156 1 Taya Sergeeva
h4. Grammar
157 80 Alexander Kamkin
158
<pre>
159 1 Taya Sergeeva
sets
160 80 Alexander Kamkin
    : ''sets'' ''='' expr
161
    ;
162 1 Taya Sergeeva
</pre>
163
164
h3. Buffer Line Format (format)
165 96 Alexander Kamkin
166 98 Alexander Kamkin
The @format@ parameter specifies the buffer _line format_ (a number of named fields). Any number (&ge; 0) of formats (with different names) can be specified for one buffer.
167 83 Alexander Kamkin
168 82 Alexander Kamkin
A field has three attributes: a name, a width and, optionally, an initial value.
169
170
h4. Grammar
171 1 Taya Sergeeva
172 82 Alexander Kamkin
<pre>
173 83 Alexander Kamkin
format
174 97 Alexander Kamkin
    : ''format'' formatID? ''='' ''('' field ('','' field)* '')''
175 82 Alexander Kamkin
    ;
176 1 Taya Sergeeva
177
field
178
    : fieldID '':'' expr (''='' expr)?
179
    ;
180
</pre>
181
182
h3. Buffer Index Function (index)
183 83 Alexander Kamkin
184 84 Alexander Kamkin
The @index@ parameter specifies the _address-to-index function_, which maps an address into the set index. The function may be omitted if the number of sets is @1@.
185 1 Taya Sergeeva
186
h4. Grammar
187
188 83 Alexander Kamkin
<pre>
189 1 Taya Sergeeva
index
190 90 Alexander Kamkin
    : ''index'' ''='' expr
191 83 Alexander Kamkin
    ;
192
</pre>
193 1 Taya Sergeeva
194
h3. Buffer Match Predicate (match)
195 83 Alexander Kamkin
196 84 Alexander Kamkin
The @match@ parameter specifies the _address-line match predicate_, which checks if an address matches a line. The parameter is obligatory.
197 83 Alexander Kamkin
198
h4. Grammar
199 1 Taya Sergeeva
200 83 Alexander Kamkin
<pre>
201
index
202 90 Alexander Kamkin
    : ''match'' ''='' expr
203 1 Taya Sergeeva
    ;
204
</pre>
205 83 Alexander Kamkin
206 1 Taya Sergeeva
h3. Buffer Data Replacement Policy (policy)
207
208 57 Taya Sergeeva
The @policy@ parameters specifies the _data replacement_ (_eviction_) _policy_. The parameter is optional. The list of supported policies includes: @RANDOM@, @FIFO@, @PLRU@ and @LRU@.
209 84 Alexander Kamkin
210 1 Taya Sergeeva
h4. Grammar
211
212
<pre>
213
policy
214
    : ''policy'' ''='' policyID
215 54 Taya Sergeeva
    ;
216
</pre>
217
218 1 Taya Sergeeva
h3. Examples
219
220 84 Alexander Kamkin
<pre>
221 90 Alexander Kamkin
// A 4-way set associative cache (L1) addressed by physical addresses (PA).
222
buffer L1(PA addr)
223 84 Alexander Kamkin
  // The cache associativity.
224 90 Alexander Kamkin
  ways = 4
225 84 Alexander Kamkin
  // The number of sets.
226 90 Alexander Kamkin
  sets = 128
227 1 Taya Sergeeva
  // The line format.
228 85 Alexander Kamkin
  format = (
229
    V    : 1 = 0, // The validity flag (by default, the line is invalid).
230
    TAG  : 24,    // The tag (the <35..12> address bits).
231 84 Alexander Kamkin
    DATA : 256    // The data (4 double words).
232 90 Alexander Kamkin
  )
233
  // The address-to-index function (example: using address fields).
234
  index = addr.INDEX
235
  // The address-line predicate (example: using address bits).
236
  match = addr<35..12> == TAG
237 1 Taya Sergeeva
  // The data replacement policy (example: using predefined policy LRU - Least Recently Used).
238
  policy = LRU
239 91 Alexander Kamkin
</pre>
240
241
h2. Memory Description (memory)
242
243 92 Alexander Kamkin
A memory is described using a keyword @memory@. The description includes two obligatory parameters @read@ and @write@.
244
245 91 Alexander Kamkin
h3. Grammar
246
247
<pre>
248 1 Taya Sergeeva
memory
249 92 Alexander Kamkin
    : ''memory'' memoryTypeID ''('' addressTypeID addressArgID '')''
250 91 Alexander Kamkin
        (memoryParameter)*
251
    ;
252
253
memoryParameter
254
    : read
255
    | write
256
    ;
257 1 Taya Sergeeva
</pre>
258 91 Alexander Kamkin
259 92 Alexander Kamkin
h3. Memory Read Action (read)
260 1 Taya Sergeeva
261 92 Alexander Kamkin
The @read@ parameter specifies the _read action_, which is a sequence of statements describing how the read operation is to be performed (by means of data transfers between buffers). The parameter is obligatory.
262
263 91 Alexander Kamkin
h4. Grammar
264
265
<pre>
266
read
267
    : ''read'' ''='' ''{'' sequence ''}''
268 1 Taya Sergeeva
    ;
269
</pre>
270
271 92 Alexander Kamkin
h3. Memory Write Action (write)
272
273
The @write@ parameter specifies the _read action_, which is a sequence of statements describing how the write operation is to be performed (by means of data transfers between buffers). The parameter is obligatory.
274 91 Alexander Kamkin
275
h4. Grammar
276
277
<pre>
278
write
279
    : ''write'' ''='' ''{'' sequence ''}''
280
    ;
281
</pre>
282
283
h3. Examples
284
285
<pre>
286 93 Alexander Kamkin
// A memory unit addressed by virtual addresses (VA).
287
memory Memory(VA addr)
288
  // The read action.
289 1 Taya Sergeeva
  read = {
290 93 Alexander Kamkin
    // Some statements.
291
    ...
292 1 Taya Sergeeva
  }
293 93 Alexander Kamkin
  // The write action.
294 1 Taya Sergeeva
  write = {
295 93 Alexander Kamkin
    // Some statements.
296
    ...
297 91 Alexander Kamkin
  }
298 1 Taya Sergeeva
</pre>