Systems for performing instructions to quickly convert and use tiles as 1D vectors
Disclosed embodiments relate to systems for performing instructions to quickly convert and use matrices (tiles) as one-dimensional vectors. In one example, a processor includes fetch circuitry to fetch an instruction having fields to specify an opcode, locations of a two-dimensional (2D) matrix and a one-dimensional (1D) vector, and a group of elements comprising one of a row, part of a row, multiple rows, a column, part of a column, multiple columns, and a rectangular sub-tile of the specified 2D matrix, and wherein the opcode is to indicate a move of the specified group between the 2D matrix and the 1D vector, decode circuitry to decode the fetched instruction; and execution circuitry, responsive to the decoded instruction, when the opcode specifies a move from 1D, to move contents of the specified 1D vector to the specified group of elements.
1. A processor comprising:
an instruction decoder to decode an instruction, the instruction to specify:
a two-dimensional (2D) tile storage of the processor,
either of multiple rows or multiple columns of the 2D tile storage,
multiple vector registers of the processor, and
a size of elements of the multiple rows or multiple columns of the 2D tile storage as any one of 8-bits, 16-bits, 32-bits, and 64-bits; and
execution circuitry, coupled with the instruction decoder, to perform operations corresponding to the instruction, the operations to include storing elements from each row of the multiple rows or each column of the multiple columns of the 2D tile storage to a corresponding one of the multiple vector registers as a corresponding one-dimensional (1D) vector.
2. The processor of claim 1 , wherein a dimension of the 2D tile storage depends on the size of elements of the multiple rows or multiple columns of the 2D tile storage.
3. The processor of claim 1 , wherein the 2D tile storage has a same number of columns as a number of elements of each of the multiple vector registers.
4. The processor of claim 1 , further comprising a storage to configure a number of columns of the 2D tile storage.
5. The processor of claim 1 , wherein the instruction is to specify the one of the multiple rows or multiple columns of the 2D tile storage based on an immediate of the instruction.
6. The processor of claim 1 , wherein the instruction is to specify the one of the multiple rows or multiple columns of the 2D tile storage based on a value in a register specified by the instruction.
7. The processor of claim 1 , wherein the processor has a plurality of 2D tile storage, and wherein the instruction is to specify one of the plurality of 2D tile storage as the 2D tile storage.
8. The processor of claim 1 , wherein the 2D tile storage comprises a plurality of vector registers.
9. A method comprising:
decoding an instruction, the instruction to specify:
a two-dimensional (2D) tile storage,
either of multiple rows or multiple columns of the 2D tile storage,
multiple vector registers, and
a size of elements of the multiple rows or multiple columns of the 2D tile storage as any one of 8-bits, 16-bits, 32-bits, and 64-bits; and
executing the instruction, the executing to include performing operations corresponding to the instruction, the operations to include storing elements from each row of the multiple rows or each column of the multiple columns of the 2D tile storage to a corresponding one of the multiple vector registers as a corresponding one-dimensional (1D) vector.
10. The method of claim 9 , wherein a dimension of the 2D tile storage depends on the size of elements of the multiple rows or multiple columns of the 2D tile storage.
11. The method of claim 9 , wherein the 2D tile storage has a same number of columns as a number of elements of each of the multiple vector registers.
12. The method of claim 9 , wherein the instruction is to specify the one of the multiple rows or multiple columns of the 2D tile storage based on an immediate of the instruction.
13. The method of claim 9 , wherein the instruction is to specify the one of the multiple rows or multiple columns of the 2D tile storage based on a value in a register specified by the instruction.
14. The method of claim 9 , wherein the 2D tile storage comprises a plurality of vector registers.
15. A non-transitory machine-readable medium storing code which when executed by a processor causes the processor to perform a method, the method comprising:
decoding a first instruction, the first instruction to specify:
a two-dimensional (2D) tile storage of the processor,
either of multiple rows or multiple columns of the 2D tile storage,
multiple vector registers of the processor, and
a size of elements of the multiple rows or multiple columns of the 2D tile storage as any one of 8-bits, 16-bits, 32-bits, and 64-bits; and
executing the first instruction, the executing to include performing operations corresponding to the first instruction, the operations to include storing elements from each row of the multiple rows or each column of the multiple columns of the 2D tile storage to a corresponding one of the multiple vector registers as a corresponding one-dimensional (1D) vector.
16. The non-transitory machine-readable medium of claim 15 , wherein a dimension of the 2D tile storage depends on the size of elements of the multiple rows or multiple columns of the 2D tile storage.
17. The non-transitory machine-readable medium of claim 15 , wherein the 2D tile storage has a same number of columns as a number of elements of each of the multiple vector registers.
18. The non-transitory machine-readable medium of claim 15 , wherein the first instruction is to specify the one of the multiple rows or multiple columns of the 2D tile storage based on an immediate of the first instruction.
19. The non-transitory machine-readable medium of claim 15 , wherein the first instruction is to specify the one of the multiple rows or multiple columns of the 2D tile storage based on a value in a register specified by the first instruction.
20. The non-transitory machine-readable medium of claim 15 , wherein the 2D tile storage comprises a plurality of vector registers.