Prefetching across page boundaries in hierarchically cached processors

摘要

Processors and methods for preventing lower level prefetch units from stalling at page boundaries. An upper level prefetch unit closest to the processor core issues a preemptive request for a translation of the next page in a given prefetch stream. The upper level prefetch unit sends the translation to the lower level prefetch units prior to the lower level prefetch units reaching the end of the current page for the given prefetch stream. When the lower level prefetch units reach the boundary of the current page, instead of stopping, these prefetch units can continue to prefetch by jumping to the next physical page number provided in the translation.

主权项

1. A processor comprising:
a cache hierarchy comprising a plurality of levels of caches; and a plurality of prefetch units including a first prefetch unit and a second prefetch unit; wherein the first prefetch unit of the plurality of prefetch units is configured to request an address translation for a new page corresponding to a given prefetch stream responsive to reaching a threshold location of a current page; and wherein the second prefetch unit is configured to:
maintain a prefetch table comprising a plurality of entries, wherein each entry comprises a stream ID;receive a physical page number of the new page from the first prefetch unit;store the physical page number of the new page in a corresponding entry in the prefetch table; andprefetch data corresponding to the new page based on the address translation for the new page.