Abstract: An apparatus and method for supporting deprecated instructions.

Abstract: An apparatus and method for a more efficient system management mode. For example, one embodiment of a processor comprises: a plurality of cores, at least a first core of the plurality of cores to perform operations to cause the plurality of cores to enter into a system management mode (SMM), the operations comprising: allocating a memory region for a system management RAM (SMRAM); writing an SMRAM state save location to a first register; and generating a page table in the SMRAM, including mapping a virtual address space a physical address space.

Abstract: An apparatus and method for booting a processor directly into a paged 64-bit execution environment. For example, one embodiment of an a processor comprises: a register to store a first value and a second value related to a secure boot process; a plurality of cores, at least one of which performs operations comprising: receiving a first initialization message, the core to clear a plurality of registers responsively; receiving a second initialization message and reading the first and second values responsively, the first value indicating whether a first initialization mode is supported, and the second value comprising an address pointer identifying a data structure comprising a plurality of state values; and initializing a paged 64-bit execution environment using the state values from the data structure responsive to the first value indicating the first initialization mode is supported and the data structure indicating enabling the paged 64-bit execution environment.

Abstract: Techniques for accessing 32 general purpose registers, suppressing flags, and/or using a new data destination for an instance of a single instruction are described. An example of a single instruction to at least include a prefix and an opcode to indicate execution circuitry is to do perform a particular operation, wherein the prefix comprises at least two bytes and a second of the two bytes of the prefix is to provide most significant bits for at least register identifier.

Abstract: Techniques for conditional test or comparison using a single instruction are described. An example instruction includes one or more fields to identify a first source operand location, one or more fields to identify a first source operand location, and an opcode to indicate execution circuitry is to conditionally perform a comparison of data from the identified first source operand to the identified second source operand based at least in part on an evaluation of a source condition code and update a flags register, wherein a payload of the prefix is to provide most significant bits to identify at least one of the first and second source operand locations.

Abstract: Techniques for push or pop operations using a single instruction are described. An example instruction at least include a prefix, one or more fields to identify a first source operand location, one or more fields to identify a second source operand location, and an opcode to indicate execution circuitry is to do push data from the identified first source operand and the identified second source operand onto a stack, wherein a payload of the prefix to provide most significant bits to identify at least one of the first and second source operand locations.

Abstract: Techniques for conditional move operations using a single instruction are described. An example instruction at least includes a prefix, one or more fields to identify a first source operand location, one or more fields to identify a destination operand location, and an opcode to indicate execution circuitry is to conditionally move data from the identified first source operand to the identified destination operand based at least in part on evaluation of a condition code, wherein a payload of the prefix is to provide most significant bits to identify at least one of the first and second source operand locations.

Abstract: An apparatus and method for switching between different types of paging using separate control registers and without disabling paging. For example, one embodiment of a processor comprises: a first control register to store a first base address of a first paging structure associated with a first type of paging having a first number of paging structure levels; a second control register to store a second base address of a second paging structure associated with a first type of paging having a second number of paging structure levels greater than the first number of paging structure levels; page walk circuitry to select either the first base address from the first control register or the second base address from the second control register responsive to a first address translation request, the selection based on a characteristic of program code initiating the address translation request.

Abstract: An apparatus and method for implementing a new virtualized execution environment while supporting instructions and operations of a legacy virtualized execution environment.

Abstract: Techniques for CPUID are described. In some examples, a CPUID instruction is to include at least one field for an opcode, the opcode to indicate execution circuitry is to return processor identification and feature information determined by input into a first register and a second register, wherein the processor identification and feature information is to include an indication of an availability of a second execution mode that at least deprecates features of a first execution.

Abstract: Techniques for using CPUID for showing features that are deprecated are described. In some examples, CPUID is to include at least one field for an opcode, one or more fields to identify a source operand which is to store a LSL selector value, and one or more fields to identify a destination register operand, wherein the opcode is to indicate that execution circuitry is to, when the single instruction has been enabled by a setting of a bit in a control register, write a LSL value stored in the control register to the destination operand when the LSL selector value of the first source register operand matches a LSL selector value stored in the control register, and set a flag in a flags register.

Abstract: Techniques for supporting a far jump and IRET are described. An example far jump instruction support includes support for a single instruction to include at least one field for an opcode and one or more fields for an operand, wherein the opcode is to indicate execution circuitry is to perform a far jump and the operand is to specify an address to be jumped to, wherein an operand size attribute of the instance of the instruction is 32-bit or greater and the instruction has been enabled by a setting of a bit in a compatibility control register.

Abstract: An apparatus and method for supporting deprecated instructions.

Abstract: An apparatus and method for performing efficient, adaptable tensor operations. For example, one embodiment of a processor comprises: front end circuitry to schedule matrix operations responsive to a matrix multiplication instruction; a plurality of lanes to perform parallel execution of the matrix operations, wherein a lane comprises an arithmetic logic unit to multiply a block of a first matrix with a block of a second matrix to generate a product and to accumulate the product with a block of a third matrix, and wherein the matrix blocks are to be stored in registers within the lane; and broadcast circuitry to broadcast one or more invariant matrix blocks to at least one of different registers within the lane and different registers across different lanes.

Abstract: An apparatus and method for efficiently processing invariant operations on a parallel execution engine.

Abstract: An apparatus and method for offloading iterative, parallel work to a data parallel cluster. For example, one embodiment of a processor comprises: a host processor to execute a primary thread; a data parallel cluster coupled to the host processor over a high speed interconnect, the data parallel cluster comprising a plurality of execution lanes to perform parallel execution of one or more secondary threads related to the primary thread; and a data parallel cluster controller integral to the host processor to offload processing of the one or more secondary threads to the data parallel cluster in response to one of the cores executing a parallel processing call instruction from the primary thread.

Abstract: An apparatus and method for performing efficient, adaptable tensor operations. For example, one embodiment of a processor comprises: front end circuitry to schedule matrix operations responsive to a matrix multiplication instruction; a plurality of lanes to perform parallel execution of the matrix operations, wherein a lane comprises an arithmetic logic unit to multiply a block of a first matrix with a block of a second matrix to generate a product and to accumulate the product with a block of a third matrix, and wherein the matrix blocks are to be stored in registers within the lane; and broadcast circuitry to broadcast one or more invariant matrix blocks to at least one of different registers within the lane and different registers across different lanes.

Abstract: An apparatus and method for data parallel single program multiple data (SPMD) execution. For example, one embodiment of a processor comprises: instruction fetch circuitry to fetch instructions of one or more primary threads; a decoder to decode the instructions to generate uops; a data parallel cluster (DPC) to execute microthreads comprising a subset of the uops, the DPC further comprising: a plurality of execution lanes to perform parallel execution of the microthreads; an instruction decode queue (IDQ) to store the uops prior to execution; and a scheduler to evaluate the microthreads based on associated variables including instruction pointer (IP) values, the scheduler to gang microthreads into fragments for parallel execution on the execution lanes based on the evaluation.

Abstract: An apparatus and method for performing efficient, adaptable tensor operations.

Abstract: An apparatus and method for offloading iterative, parallel work to a data parallel cluster. For example, one embodiment of a processor comprises: a host processor to execute a primary thread; a data parallel cluster coupled to the host processor over a high speed interconnect, the data parallel cluster comprising a plurality of execution lanes to perform parallel execution of one or more secondary threads related to the primary thread; and a data parallel cluster controller integral to the host processor to offload processing of the one or more secondary threads to the data parallel cluster in response to one of the cores executing a parallel processing call instruction from the primary thread.