Abstract: A computer system includes a main memory that is able to make use of DRAM memory devices having a relatively high level of bad cells (hard faults). An EDC circuit is provided which uses combinatorial logic to perform a BCH code type of error detection and correction. A primary feature is the recognition that due to use of high density integrated circuits - gate arrays - it is no longer necessary to use sequential logic to decode the multiple-bit error correcting codes. An EDC with 128-bits of data and a check bit field 41-bits wide, using a BCH code, constructed in ASIC sea-of-gates technology using about 87,000 logic gates, can correct 5-bits in error and can detect 6-bits in error. By using multiple-bit EDC in the controller for main memory, it is no longer necessary that all DRAM devices be ostensibly "perfect." A certain density of non-functional memory cells can be tolerated, yet the memory system will still return perfect data.

Abstract: A read buffering system and method employs a bank of FIFOs to hold sequential read data for a number of data streams being fetched by a computer. The FIFOs are located in the memory controller, so the system bus is not used in the memory accesses needed to fill the stream buffer. The buffer system stores addresses used for read requests made by a CPU, and if a next sequential address is then detected in a subsequent read request, this is designated to be a stream (i.e., sequential reads). When a stream is thus detected, data is fetched from DRAM memory for addresses following the sequential address, and this prefetched data is stored in one of the FIFOs. A FIFO is selected using a least-recently-used algorithm. When the CPU subsequently makes a read request for data in a FIFO, this data can be returned without making a memory access, and so the access time seen by the CPU is shorter.

Abstract: A read buffering system employs a bank of FIFOs to hold sequential read data for a number of data streams being fetched by a computer. The FIFOs are located in the memory controller, so the system bus is not used in the memory accesses needed to fill the stream buffer. The buffer system stores addresses used for read requests made by a CPU, and if a next sequential address is then detected in a subsequent read request, this is designated to be a stream (i.e., sequential reads). When a stream is thus detected, data is fetched from DRAM memory for addresses following the sequential address, and this prefetched data is stored in one of the FIFOs. A FIFO is selected using a least-recently-used algorithm. When the CPU subsequently makes a read request for data in a FIFO, this data can be returned without making a memory access, and so the access time seen by the CPU is shorter.

Abstract: A computer system includes a main memory that is able to make use of DRAM memory devices having a relatively high level of bad cells (hard faults). An EDC circuit is provided which uses combinatorial logic to perform a BCH code type of error detection and correction. A primary feature is the recognition that due to use of high density integrated circuits--gate arrays--it is no longer necessary to use sequential logic to decode the multiple-bit error correcting codes. An EDC with 128-bits of data and a check bit field 41-bits wide, using a BCH code, constructed in ASIC sea-of-gates technology using about 87,000 logic gates, can correct 5-bits in error and can detect 6-bits in error. By using multiple-bit EDC in the controller for main memory, it is no longer necessary that all DRAM devices be ostensibly "perfect." A certain density of non-functional memory cells can be tolerated, yet the memory system will still return perfect data.

Abstract: A read buffering system employs a bank of FIFOs to hold sequential read data for a number of data streams being fetched by a computer. The FIFOs are located in the memory controller, so the system bus is not used in the memory accesses needed to fill the stream buffer. The buffer system stores addresses used for read requests made by a CPU, and if a next sequential address is then detected in a subsequent read request, this is designated to be a stream (i.e., sequential reads). When a stream is thus detected, data is fetched from DRAM memory for addresses following the sequential address, and this prefetched data is stored in one of the FIFOs. A FIFO is selected using a least-recently-used algorithm. When the CPU subsequently makes a read request for data in a FIFO, this data can be returned without making a memory access, and so the access time seen by the CPU is shorter.

Abstract: A cache memory exchange protocol that utilizes an exchange memory access request is provided A cache controller issues the exchange request over a system bus to a main memory in response to a cache miss if the cache location is dirty. Upon receiving the request, the main memory decodes the request and fetches a block of information desired by the transfers a dirty cache block of information over the system bus and stores the dirty block in a temporary register located in the main memory. The block of information desired by the CPU is thereafter transferred by the main memory to the CPU via the system bus. The dirty cache block residing in the temporary register is then stored in the main memory storage elements.