G - Physics – 06 – F
Patent
G - Physics
06
F
354/242, 354/230
G06F 12/08 (2006.01) G06F 13/20 (2006.01) G06F 15/16 (2006.01)
Patent
CA 1315896
ABSTRACT OF THE DISCLOSURE A multiprocessor system includes a system of store queues and write buffers in a hierarchical first level and second level memory system including a first level store queue for storing instructions and/or data from a processor of the multiprocessor system prior to storage in a first level of cache, a second level store queue for storing the instructions and/or data from the first level store queue and a plurality of write buffers for storing the instructions and/or data from the second level store queue prior to storage in a second level of cache. The multiprocessor system includes hierarchical levels of caches, including a first level of cache associated with each processor, a single shared second level of cache shared by all the processors, and a third level of main memory connected to the shared second level cache. A first level store queue, associated with each processor, receives the data and/or instructions from its processor and stores the data and/or instructions in the first level of cache. A second level store queue, associated with each processor, receives the data and/or instructions from its first level store queue and temporarily stores the information therein. For sequential stores, the data and/or instructions are stored in corresponding second level write buffers. For non-sequential stores, the data and/or instructions bypass the corresponding second level write buffers and are stored directly in a final L2 cache write buffer. When stored in the second level write buffers, access to the shared second level cache is requested; and, when access is granted, the data and/or instructions is moved from the second level write buffers to the shared second level cache. When stored in the shared second level cache, corresponding obsolete entries in the first level of cache are invalidated before any other processor "sees" the obsolete data and the new data and/or instructions are over-written in the first level of cache.
588790
International Business Machines Corporation
Saunders Raymond H.
LandOfFree
Store queue for a tightly coupled multiple processor... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Store queue for a tightly coupled multiple processor..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Store queue for a tightly coupled multiple processor... will most certainly appreciate the feedback.
Profile ID: LFCA-PAI-O-1276258