摘要:
Read buffers (RB0-RB3) are capable of holding data read out from a plurality of memory blocks (BNK0-BNK7) that are capable of parallel operation in response to a state in which the read data cannot be externally outputted from an external interface means; and, selection means (40, 41, 42) are provided for selecting data read out from one of the memory blocks, or data read out from one of the read buffers, and for feeding it to the external interface means, while the external-output-incapable state is not present. In this way, when there is a possibility that an output of read data will cause a resource competition, this read data is stored in a read buffer, and when there is no such possibility, then the read data can be externally outputted directly, thereby improving the throughput of read data output operations.
摘要:
Read buffers (RB0-RB3) are capable of holding data read out from a plurality of memory blocks (BNK0-BNK7) that are capable of parallel operation in response to a state in which the read data cannot be externally outputted from an external interface means; and, selection means (40, 41, 42) are provided for selecting data read out from one of the memory blocks, or data read out from one of the read buffers, and for feeding it to the external interface means, while the external-output-incapable state is not present. In this way, when there is a possibility that an output of read data will cause a resource competition, this read data is stored in a read buffer, and when there is no such possibility, then the read data can be externally outputted directly, thereby improving the throughput of read data output operations.
摘要:
A semiconductor integrated circuit device such as a memory device with logic function comprises a plurality of RAM macrocells and gate arrays. The RAM macrocells are constituted by bipolar CMOS RAMs having a total memory capacity of at least 100 kilobits, and the gate arrays contain at least 4000 gates. The logic circuits in the memory device with logic function or the like are constructed by selectively combining CMOS, bipolar CMOS or ECL gate circuits depending on the output load capacity, transmission characteristic requirement, power dissipation and required layout area. The level of signals at various circuits is set to the ECL level or MOS level depending on the local circuit configuration and other factors. The memory device further incorporates sequence control circuits required to be installed downstream of buffer storages of computers.
摘要:
A semiconductor integrated circuit device such as a memory device with logic function comprises a plurality of RAM macrocells and gate arrays. The RAM macrocells are constituted by bipolar CMOS RAMs having a total memory capacity of at least 100 kilobits, and the gate arrays contain at least 4000 gates. The logic circuits in the memory device with logic function or the like are constructed by selectively combining CMOS, bipolar CMOS or ECL gate circuits depending on the output load capacity, transmission characteristic requirement, power dissipation and required layout area. The level of signals at various circuits is set to the ECL level or MOS level depending on the local circuit configuration and other factors. The memory device further incorporates sequence control circuits required to be installed downstream of buffer storages of computers.
摘要:
A method and system for increasing speeds of transferring data in a data transfer system which includes a data source and data sink. Both the data source and data sink include clocks which are synchronized to a common clock frequency. A buffer is provided at the data sink and this buffer is utilized to received data from the data source. A control circuit is provided at the data sink and this control circuit receives a bus clock signal from the data source. An N segment dynamic shift register is provided within the data sink which includes at least two segments. A selectable shift control is provided for passing the data through an M segment subset of the N segment shift register, where M is less than N. Additionally, the length of the M segment subset is determined by the phase of a clock within the data sink at the time which the bus clock signal from the data source is received at the data sink. By selectively passing the data through an M segment subset of the N segment shift register, the data is accessible at the data sink at a controllable predetermined time.
摘要:
A data processor has a first cache memory with a large capacity and one port and a second cache memory with a small capacity and two ports disposed between a main memory and an instruction processing section. Data which is frequently used is stored in the first cache memory and data which is less frequently used is stored in the second cache memory under control of a controller responsive to prefetch instructions. One of the cache memories may be a set associative cache memory composed of a plurality of memory chips each having at least two memory banks and an output part to gain access to data sets consecutively and one at a time within the memory banks. On the basis of an address sent from the instruction processing section, a memory bank is selected, and a data set from the selected memory bank is supplied to the processing section.
摘要:
An information processor is capable of eliminating register conflict in short and long latency processes and for attaining high-speed pipeline processing through efficient use of registers. The scale of the necessary hardware is reduced by the processor using a register conflict detector and a scoreboard. The register conflict detector detects register conflict over a period of short latency processing, and the scoreboard checks for register conflict beyond the short latency process period and into a period of long latency processing. The processor controls the issue of instructions based on the detected register conflict status.
摘要:
To offer a floating-point addition/subtraction processing apparatus and a method thereof, capable of shortening the computation time, the floating-point calculation processing apparatus includes an approximate shift mount predicting unit for predicting a shift amount for normalization by using the input floating-point data to be addition/subtraction processed within an error of 1 bit, a shift error detecting unit for detecting a difference between the predicted shift amount and a correct shift amount, and an bit shifter for correcting a result, obtained by normalization using the predicted shift amount, by the detected difference of the two shift amounts, wherein a round-off determination and a shift amount calculation are processed in parallel before a normalization shift processing is executed.
摘要:
In a multiple virtual storage system and more particularly in an address control apparatus, there are provided two kinds of holding devices a designation holding device for holding a segment table designations in association with access registers and a translation buffer for holding translation pairs of the access register and segment table designation. With this arrangement, the segment table designation designating the virtual address space possessing an operand of an instruction can be supplied quickly and efficiently.
摘要:
A multiple virtual space control in a multiple virtual storage system having an address translation table used to translate a logical address to a real address, a control register for holding a start address of the address translation table or a space identifier (hereinafter represented by address translation table start address) and an address translation buffer containing a pair of logical address and real address and an address translation table start address for translating a logical address to a real address, in order to update the content of the control register to switch the virtual space. A group identifier comprising a plurality of bits for identifying an area common to a group of virtual spaces is added to an entry of the address translation table, an entry of the address translation buffer and the control register. When a logical address is to be translated to a real address, if there is an entry having a logical address and an address translation table start address equal to the memory request logical address and the address translation table start address of the control register, or an entry having a logical address and a group identifier equal to the memory request logical address and the group identifier of the control register, in the address translation buffer, the real address of the entry is rendered valid and used for memory access.