Question

In SystemVerilog, $readmemb, $readmemh, $writememb and $writememh can work with multi-dimensional unpacked arrays.

The file contents are organized in row-major order, with each dimension’s entries ranging from low to high address. This is backward compatible with plain Verilog memories.

In this organization, the lowest dimension (i.e. the right-most dimension in the array declaration) varies the most rapidly. There is a hierarchical sense to the file data. The higher dimensions contain words of lowerdimension data, sorted in row-major order. Each successive lower dimension is entirely enclosed as part of higher dimension words.

As an example of file format organization, here is the layout of a file representing words for a memory declared:

reg [31:0] mem [0:2][0:4][5:8];

In the example word contents, wzyx,

• z corresponds to words of the [0:2] dimension
• y corresponds to words of the [0:4] dimension
• x corresponds to words of the [5:8] dimension

w005 w006 w007 w008
w015 w016 w017 w018
w025 w026 w027 w028
w035 w036 w037 w038
w045 w046 w047 w048
w105 w106 w107 w108
w115 w116 w117 w118
w125 w126 w127 w128
w135 w136 w137 w138
w145 w146 w147 w148
w205 w206 w207 w208
w215 w216 w217 w218
w225 w226 w227 w228
w235 w236 w237 w238
w245 w246 w247 w248

Note that the diagram would be identical if one or more of the unpacked dimension declarations were reversed, as in:

reg [31:0] mem [2:0][0:4][8:5]

Address entries in the file exclusively address the highest dimension’s words. In the above case, address entries in the file could look something as follows:

@0 w005 w006 w007 w008
   w015 w016 w017 w018
   w025 w026 w027 w028
   w035 w036 w037 w038
   w045 w046 w047 w048
@1 w105 w106 w107 w108
   w115 w116 w117 w118
   w125 w126 w127 w128
   w135 w136 w137 w138
   w145 w146 w147 w148
@2 w205 w206 w207 w208
   w215 w216 w217 w218
   w225 w226 w227 w228
   w235 w236 w237 w238
   w245 w246 w247 w248

When $readmemh or $readmemb is given a file without address entries, all data is read assuming that each dimension has complete data. i.e. each word in each dimension will be initialized with the appropriate value from the file. If the file contains incomplete data, the read operation will stop at the last initialized word, and any remaining array words or sub words will be left unchanged.

When $readmemh or $readmemb is given a file with address entries, initialization of the specified highest dimension words is done. If the file contains insufficient words to completely fill a highest dimension word, then the remaining sub words are left unchanged.

When a memory contains multiple packed dimensions, the memory words in the pattern file are composed of the sum total of all bits in the packed dimensions. The layout of packed bits in packed dimensions is defined in Section 4.3.