Inventors:
Omid AZIZI - Redwood City CA, US
Guy BOUDOUKH - Ramat Hasharon, IL
Tony WERNER - Los Altos CA, US
Andrew YANG - Cupertino CA, US
Michael ROTZIN - Santa Clara CA, US
Chen KOREN - Hadera, IL
Eriko NURVITADHI - Hillsboro OR, US
International Classification:
G06F 9/30
G06F 17/16
G06F 9/38
Abstract:
Disclosed embodiments relate to sparse matrix multiplication (SMM) acceleration using column folding and squeezing. In one example, a processor, in response to a SMM instruction having fields to specify locations of first, second, and output matrices, the second matrix being a sparse matrix, uses execution circuitry to pack the second matrix by replacing one or more zero-valued elements with non-zero elements yet to be processed, each of the replaced elements further including a field to identify its logical position within the second matrix, and, the execution circuitry further to, for each non-zero element at row M and column K of the specified first matrix, generate a product of the element and each corresponding non-zero element at row K, column N of the packed second matrix, and accumulate each generated product with a previous value of a corresponding element at row M and column N of the specified output matrix.