Systems, methods, and apparatus for tile configuration
Embodiments detailed herein relate to matrix (tile) operations. For example, decode circuitry to decode an instruction having fields for an opcode and a memory address; and execution circuitry to execute the decoded instruction to set a tile configuration for the processor to utilize tiles in matrix operations based on a description retrieved from the memory address, wherein a tile a set of 2-dimensional registers are discussed.
1. A processor comprising:
decode circuitry to decode an instruction having fields for an opcode and a memory address, the opcode to indicate execution circuitry is to load a tile configuration stored at the memory address, the tile configuration information to at least specify a first and a second dimension for a two-dimensional data structure tile; and
execution circuitry to execute the decoded instruction to load the tile configuration for the processor to utilize tiles in matrix operations.
2. The processor of claim 1 , wherein the tile configuration is to define a number of bytes per tile and bytes per row of the tile.
3. The processor of claim 1 , wherein the execution circuitry to generate a fault upon a value of a number of rows for a tile being zero and a number of columns for the row being non-zero.
4. The processor of claim 1 , wherein the configuration is 64 bytes in size.
5. A method comprising:
decoding an instruction having fields for an opcode and a memory address, the opcode to indicate execution circuitry is to load a tile configuration stored at the memory address, the tile configuration information to at least specify a first and a second dimension for a two-dimensional data structure tile; and
executing the decoded instruction to load the tile configuration for the processor to utilize tiles in matrix operations.
6. The method of claim 5 , wherein the tile configuration defines a number of rows and columns per tile.
7. The method of claim 5 , further comprising:
faulting upon a value of a number of rows for a tile being zero and a number of columns for the row being non-zero.
8. The method of claim 5 , wherein the tile configuration is 64 bytes in size.
9. A non-transitory machine-readable medium storing an instruction which causes a processor to perform a method, the method comprising:
decoding an instruction having fields for an opcode and a memory address, the opcode to indicate execution circuitry is to load a tile configuration stored at the memory address, the tile configuration information to at least specify a first and a second dimension for a two-dimensional data structure tile; and
executing the decoded instruction to load the tile configuration for the processor to utilize tiles in matrix operations.
10. The non-transitory machine-readable medium of claim 9 , wherein the tile configuration details a number of rows and columns per tile.
11. The non-transitory machine-readable medium of claim 9 , further comprising:
faulting upon a value of a number of rows for a tile being zero and a number of columns for the row being non-zero.