发明名称 Vector move instruction controlled by read and write masks
摘要 A processor executes a vector move instruction to move data elements from a second vector register to a first vector register under the control of a first mask register and a second mask register. A register file within the processor includes the first vector register, the second vector register, the first mask register and the second mask register. In response to the vector move instruction, execution circuitry in the processor is to replace a given number of target data elements in the first vector register with the given number of source data elements in the second vector register. Each source data element corresponds to a mask bit in the second mask register having a second bit value, and wherein each target data element corresponds to a mask bit in the first mask register having a first bit value.
申请公布号 US9378182(B2) 申请公布日期 2016.06.28
申请号 US201213630118 申请日期 2012.09.28
申请人 Intel Corporation 发明人 Plotnikov Mikhail;Naraikin Andrey;Hughes Christopher
分类号 G06F15/00;G06F15/76;G06F15/80;G06F9/38 主分类号 G06F15/00
代理机构 Nicholson De Vos Webster & Elliot, LLP 代理人 Nicholson De Vos Webster & Elliot, LLP
主权项 1. A system comprising: memory to store an input data array including a plurality of input data elements; a register file including a first mask register, a second mask register, a first vector register and a second vector register; and execution circuitry coupled to the memory and the register file, the execution circuitry to perform a recursive computation on the first vector register for a plurality of iterations, more than one of the iterations including operations of: a vector load operation to load at least a portion of the input data elements from the input data array into the second vector register;a vector move operation to move the input data elements in the second vector register into the first vector register;a mask update operation to update the first and second mask registers to identify data elements in the first and second vector registers, respectively, that need further computation;a vector compute operation to operate over the data elements in the first vector register; anda vector store operation to store results of the vector compute operation into the memory.
地址 Santa Clara CA US