Abstract: A scheduler and method for dynamic power reduction, e.g., in a processor core, is proposed. In conventional processor cores for example, the scheduler precharges grant lines of many instructions only to discharge a great majority of the precharged lines in one cycle. To reduce power consumption, selective precharge and/or selective evaluation are proposed. In the selective precharge, the grant lines of instructions that will evaluate to false (e.g., invalid instructions) are not precharged in a cycle. In the selective evaluation, among the precharged instructions, instructions that are not ready are not evaluated in the same cycle. In this way, power consumption is reduced by avoiding unnecessary precharge and discharge.

Abstract: A scheduler with a picker block capable of dispatching multiple instructions per cycle is disclosed. The picker block may comprise an inter-group picker and an intra-group picker. The inter-group picker may be configured to pick multiple ready groups when there are two or more ready groups among a plurality of groups of instructions, and pick a single ready group when the single ready group is the only ready group among the plurality of groups. The intra-group picker may be configured to pick one ready instruction from each of the multiple ready groups when the inter-group picker picks the multiple ready groups, and to pick multiple ready instructions from the single ready group when the inter-group picker picks the single ready group.

Abstract: Replaying speculatively dispatched load-dependent instructions in response to a cache miss for a producing load instruction in an out-of-order processor (OoP) is disclosed. To allow for a scheduler circuit to restore register dependencies in a register dependency tracking circuit for a replay operation in response to a cache miss for execution of a load instruction, the scheduler circuit includes a replay circuit. The replay circuit includes a load dependency tracking circuit. The replay circuit is configured to track dependencies of dispatched load instructions in the load dependency tracking circuit. The replay circuit uses these tracked dependencies to restore register dependencies for the dispatched load instructions in the register dependency tracking circuit in response to a replay operation. Thus, the load instruction does not have to be re-allocated to restore register dependencies in the register dependency tracking circuit used for re-dispatching load-dependent instructions.

Abstract: Efficiently generating selection masks for row selections within indexed address spaces is disclosed. In this regard, in one aspect, an indexed array circuit is provided, comprising a start indicator that indicates a start indexed array row of a row selection, and an end indicator that indicates an end indexed array row of the row selection. The indexed array circuit further comprises a plurality of indexed array rows ordered in a logical sequence, each comprising a row-level compare circuit. Each row-level compare circuit is configured to generate a selection mask indicator based on a first parallel comparison of subsets of bits of a logical address of the indexed array row with corresponding subsets of bits of the start indicator, and a second parallel comparison of subsets of bits of the logical address of the indexed array row with corresponding subsets of bits of the end indicator.

Abstract: Multiple-hot (multi-hot) bit decoding in a memory system for activating multiple memory locations in a memory for a memory access operation are disclosed. In one aspect, a multi-hot bit decoding system is provided that includes a memory access control system that includes a decoder. The decoder is configured to decode an address for a memory access operation into a single-hot bit decode word for activating a memory row at the encoded address. To automatically access another memory row(s) for a memory access operation, the memory access control system also includes a mapping circuit configured to provide an additional decode word(s) for activating another memory row(s) based on the address. The decode word and additional decode word(s) are merged to provide a multi-hot bit decode word that is asserted onto a decode wordline such that multiple memory rows are activated for a memory access operation.

Abstract: A scheduler and method for dynamic power reduction, e.g., in a processor core, is proposed. In conventional processor cores for example, the scheduler precharges grant lines of many instructions only to discharge a great majority of the precharged lines in one cycle. To reduce power consumption, selective precharge and/or selective evaluation are proposed. In the selective precharge, the grant lines of instructions that will evaluate to false (e.g., invalid instructions) are not precharged in a cycle. In the selective evaluation, among the precharged instructions, instructions that are not ready are not evaluated in the same cycle. In this way, power consumption is reduced by avoiding unnecessary precharge and discharge.

Abstract: A scheduler with a picker block capable of dispatching multiple instructions per cycle is disclosed. The picker block may comprise an inter-group picker and an intra-group picker. The inter-group picker may be configured to pick multiple ready groups when there are two or more ready groups among a plurality of groups of instructions, and pick a single ready group when the single ready group is the only ready group among the plurality of groups. The intra-group picker may be configured to pick one ready instruction from each of the multiple ready groups when the inter-group picker picks the multiple ready groups, and to pick multiple ready instructions from the single ready group when the inter-group picker picks the single ready group.

Abstract: Efficiently generating selection masks for row selections within indexed address spaces is disclosed. In this regard, in one aspect, an indexed array circuit is provided, comprising a start indicator that indicates a start indexed array row of a row selection, and an end indicator that indicates an end indexed array row of the row selection. The indexed array circuit further comprises a plurality of indexed array rows ordered in a logical sequence, each comprising a row-level compare circuit. Each row-level compare circuit is configured to generate a selection mask indicator based on a first parallel comparison of subsets of bits of a logical address of the indexed array row with corresponding subsets of bits of the start indicator, and a second parallel comparison of subsets of bits of the logical address of the indexed array row with corresponding subsets of bits of the end indicator.

Abstract: Multiple-hot (multi-hot) bit decoding in a memory system for activating multiple memory locations in a memory for a memory access operation are disclosed. In one aspect, a multi-hot bit decoding system is provided that includes a memory access control system that includes a decoder. The decoder is configured to decode an address for a memory access operation into a single-hot bit decode word for activating a memory row at the encoded address. To automatically access another memory row(s) for a memory access operation, the memory access control system also includes a mapping circuit configured to provide an additional decode word(s) for activating another memory row(s) based on the address. The decode word and additional decode word(s) are merged to provide a multi-hot bit decode word that is asserted onto a decode wordline such that multiple memory rows are activated for a memory access operation.