Abstract: Disclosed embodiments relate to systems and methods for performing 16-bit floating-point vector dot product instructions. In one example, a processor includes fetch circuitry to fetch an instruction having fields to specify an opcode and locations of first source, second source, and destination vectors, the opcode to indicate execution circuitry is to multiply N pairs of 16-bit floating-point formatted elements of the specified first and second sources, and accumulate the resulting products with previous contents of a corresponding single-precision element of the specified destination, decode circuitry to decode the fetched instruction, and execution circuitry to respond to the decoded instruction as specified by the opcode.

Abstract: Implementations of using tiles for caching are detailed In some implementations, an instruction execution circuitry executes one or more instructions, a register state cache coupled to the instruction execution circuitry holds thread register state in a plurality of registers, and backing storage pointer storage stores a backing storage pointer, wherein the backing storage pointer is to reference a state backing storage area in external memory to store the thread register state stored in the register state cache.

Abstract: Embodiments detailed herein relate to systems and methods to zero a tile register pair. In one example, a processor includes decode circuitry to decode a matrix pair zeroing instruction having fields for an opcode and an identifier to identify a destination matrix having a PAIR parameter equal to TRUE; and execution circuitry to execute the decoded matrix pair zeroing instruction to zero every element of a left matrix and a right matrix of the identified destination matrix.

Abstract: 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.

Abstract: Disclosed embodiments relate to systems and methods for performing instructions specifying ternary tile operations. In one example, a processor includes fetch and decode circuitry to fetch and decode an instruction specifying a ternary tile operation, and locations of destination and first, second, and third source matrices, each of the matrices having M rows by N columns; and execution circuitry to respond to the decoded instruction by, for each equal-sized group of K elements of the specified first, second, and third source matrices, generate K results by performing the ternary tile operation in parallel on K corresponding elements of the specified first, second, and third source matrices, and store each of the K results to a corresponding element of the specified destination matrix, wherein corresponding elements of the specified source and destination matrices occupy a same relative position within their associated matrix.

Abstract: Disclosed embodiments relate to computing dot products of nibbles in tile operands. In one example, a processor includes decode circuitry to decode a tile dot product instruction having fields for an opcode, a destination identifier to identify a M by N destination matrix, a first source identifier to identify a M by K first source matrix, and a second source identifier to identify a K by N second source matrix, each of the matrices containing doubleword elements, and execution circuitry to execute the decoded instruction to perform a flow K times for each element (m, n) of the specified destination matrix to generate eight products by multiplying each nibble of a doubleword element (M,K) of the specified first source matrix by a corresponding nibble of a doubleword element (K,N) of the specified second source matrix, and to accumulate and saturate the eight products with previous contents of the doubleword element.

Abstract: Disclosed embodiments relate to systems and methods for performing instructions to transform matrices into a row-interleaved format. In one example, a processor includes fetch and decode circuitry to fetch and decode an instruction having fields to specify an opcode and locations of source and destination matrices, wherein the opcode indicates that the processor is to transform the specified source matrix into the specified destination matrix having the row-interleaved format; and execution circuitry to respond to the decoded instruction by transforming the specified source matrix into the specified RowInt-formatted destination matrix by interleaving J elements of each J-element sub-column of the specified source matrix in either row-major or column-major order into a K-wide submatrix of the specified destination matrix, the K-wide submatrix having K columns and enough rows to hold the J elements.

Abstract: An apparatus is described that includes an execution unit to execute a first instruction and a second instruction. The execution unit includes input register space to store a first data structure to be replicated when executing the first instruction and to store a second data structure to be replicated when executing the second instruction. The first and second data structures are both packed data structures. Data values of the first packed data structure are twice as large as data values of the second packed data structure. The execution unit also includes replication logic circuitry to replicate the first data structure when executing the first instruction to create a first replication data structure, and, to replicate the second data structure when executing the second data instruction to create a second replication data structure.

Abstract: Disclosed embodiments relate to instructions for fast element unpacking. In one example, a processor includes fetch circuitry to fetch an instruction whose format includes fields to specify an opcode and locations of an Array-of-Structures (AOS) source matrix and one or more Structure of Arrays (SOA) destination matrices, wherein: the specified opcode calls for unpacking elements of the specified AOS source matrix into the specified Structure of Arrays (SOA) destination matrices, the AOS source matrix is to contain N structures each containing K elements of different types, with same-typed elements in consecutive structures separated by a stride, the SOA destination matrices together contain K segregated groups, each containing N same-typed elements, decode circuitry to decode the fetched instruction, and execution circuitry, responsive to the decoded instruction, to unpack each element of the specified AOS matrix into one of the K element types of the one or more SOA matrices.

Abstract: Embodiments detailed herein relate to matrix operations. In particular, tile diagonal support is described. For example, a processor is detailed having decode circuitry to decode an instruction having fields for an opcode, a source operand identifier, and a destination matrix operand identifier; and execution circuitry to execute the decoded instruction to write the identified source operand to each element along a main diagonal of the identified destination matrix operand.

Abstract: Disclosed embodiments relate to systems and methods for performing instructions to transpose rectangular tiles. In one example, a processor includes fetch circuitry to fetch an instruction having fields to specify an opcode and locations of first destination, second destination, first source, and second source matrices, the specified opcode to cause the processor to process each of the specified source and destination matrices as a rectangular matrix, decode circuitry to decode the fetched rectangular matrix transpose instruction, and execution circuitry to respond to the decoded rectangular matrix transpose instruction by transposing each row of elements of the specified first source matrix into a corresponding column of the specified first destination matrix and transposing each row of elements of the specified second source matrix into a corresponding column of the specified second destination matrix.

Abstract: Embodiments of systems, apparatuses, and methods for chained fused multiply add. In some embodiments, an apparatus includes a decoder to decode a single instruction having an opcode, a destination field representing a destination operand, a first source field representing a plurality of packed data source operands of a first type that have packed data elements of a first size, a second source field representing a plurality of packed data source operands that have packed data elements of a second size, and a field for a memory location that stores a scalar value. A register file having a plurality of packed data registers includes registers for the plurality of packed data source operands that have packed data elements of a first size, the source operands that have packed data elements of a second size, and the destination operand.

Abstract: An apparatus is described having instruction execution logic circuitry. The instruction execution logic circuitry has input vector element routing circuitry to perform the following for each of three different instructions: for each of a plurality of output vector element locations, route into an output vector element location an input vector element from one of a plurality of input vector element locations that are available to source the output vector element. The output vector element and each of the input vector element locations are one of three available bit widths for the three different instructions. The apparatus further includes masking layer circuitry coupled to the input vector element routing circuitry to mask a data structure created by the input vector routing element circuitry. The masking layer circuitry is designed to mask at three different levels of granularity that correspond to the three available bit widths.

Abstract: Disclosed embodiments relate to matrix compress/decompress instructions. In one example, a processor includes fetch circuitry to fetch a compress instruction having a format with fields to specify an opcode and locations of decompressed source and compressed destination matrices, decode circuitry to decode the fetched compress instructions, and execution circuitry, responsive to the decoded compress instruction, to: generate a compressed result according to a compress algorithm by compressing the specified decompressed source matrix by either packing non-zero-valued elements together and storing the matrix position of each non-zero-valued element in a header, or using fewer bits to represent one or more elements and using the header to identify matrix elements being represented by fewer bits; and store the compressed result to the specified compressed destination matrix.

Abstract: An apparatus and method for performing a vector bit reversal and crossing. For example, one embodiment of a processor comprises: a first source vector register to store a first plurality of source bit groups, wherein a size for the bit groups is to be specified in an immediate of an instruction; a second source vector to store a second plurality of source bit groups; vector bit reversal and crossing logic to determine a bit group size from the immediate and to responsively reverse positions of contiguous bit groups within the first source vector register to generate a set of reversed bit groups, wherein the vector bit reversal and crossing logic is to additionally interleave the set of reversed bit groups with the second plurality of bit groups; and a destination vector register to store the reversed bit groups interleaved with the first plurality of bit groups.

Abstract: A method of an aspect includes receiving a packed data operation mask shift instruction. The packed data operation mask shift instruction indicates a source having a packed data operation mask, indicates a shift count number of bits, and indicates a destination. The method further includes storing a result in the destination in response to the packed data operation mask shift instruction. The result includes a sequence of bits of the packed data operation mask that have been shifted by the shift count number of bits. Other methods, apparatus, systems, and instructions are disclosed.

Abstract: A method of an aspect includes receiving a packed data operation mask shift instruction. The packed data operation mask shift instruction indicates a source having a packed data operation mask, indicates a shift count number of bits, and indicates a destination. The method further includes storing a result in the destination in response to the packed data operation mask shift instruction. The result includes a sequence of bits of the packed data operation mask that have been shifted by the shift count number of bits. Other methods, apparatus, systems, and instructions are disclosed.

Abstract: Disclosed embodiments relate to efficient complex vector multiplication. In one example, an apparatus includes execution circuitry, responsive to an instruction having fields to specify multiplier, multiplicand, and summand complex vectors, to perform two operations: first, to generate a double-even multiplicand by duplicating even elements of the specified multiplicand, and to generate a temporary vector using a fused multiply-add (FMA) circuit having A, B, and C inputs set to the specified multiplier, the double-even multiplicand, and the specified summand, respectively, and second, to generate a double-odd multiplicand by duplicating odd elements of the specified multiplicand, to generate a swapped multiplier by swapping even and odd elements of the specified multiplier, and to generate a result using a second FMA circuit having its even product negated, and having A, B, and C inputs set to the swapped multiplier, the double-odd multiplicand, and the temporary vector, respectively.

Abstract: An apparatus is described having instruction execution logic circuitry. The instruction execution logic circuitry has input vector element routing circuitry to perform the following for each of three different instructions: for each of a plurality of output vector element locations, route into an output vector element location an input vector element from one of a plurality of input vector element locations that are available to source the output vector element. The output vector element and each of the input vector element locations are one of three available bit widths for the three different instructions. The apparatus further includes masking layer circuitry coupled to the input vector element routing circuitry to mask a data structure created by the input vector routing element circuitry. The masking layer circuitry is designed to mask at three different levels of granularity that correspond to the three available bit widths.

Abstract: An apparatus is described having instruction execution logic circuitry to execute first, second, third and fourth instruction. Both the first instruction and the second instruction insert a first group of input vector elements to one of multiple first non overlapping sections of respective first and second resultant vectors. The first group has a first bit width. Each of the multiple first non overlapping sections have a same bit width as the first group. Both the third instruction and the fourth instruction insert a second group of input vector elements to one of multiple second non overlapping sections of respective third and fourth resultant vectors. The second group has a second bit width that is larger than said first bit width. Each of the multiple second non overlapping sections have a same bit width as the second group.