摘要:
A multi-processor system including a main storage (1) for storing instructions and data, a master processor (3) for supplying to a slave processor (4) data required for the processing to be executed by the slave processor and commanding initiation of the processing, the master processor further having a function to test the operation state of the slave processor and perform processing by utilizing the result of th processing executed by the slave processor. The slave processor (4) initiates the processing under the command of the master processor and has a function to inform of the master processor of completion of the processing. The slave processor has a function to execute a pause instruction for suspending temporarily activation of processing for a succeeding instruction and setting a pause indication at an indicator (453,454,455) of the slave processor. When the pause indication is set in the slave processor, the master processor has a function to reset this indication to release the slave processor from the pause state. When the pause state indication is not set, the master processor executes a clearing instruction supplied from the main storage for suspending the function to activate the succeeding instruction. The processor is further imparted with a function for setting at the indicator an indication instruction indicating completion of execution of the succeeding instruction. The master processor is imparted with a function to reset the indication of completed execution of instruction set at the slave processor and otherwise execute an indication resetting instruction for suspending activation of a succeeding instruction.
摘要:
The vector processor according to the present invention includes a read address counter and a write address counter in an address circuit for each vector register, so that a read operation and a write operation are performed in parallel to the same vector register by these adress counters.
摘要:
@ A data processing apparatus comprises a plurality of sub-systems (30a-30d) each including at least one arithmetic unit (38a-38d), a plurality of registers (35a-35d), a first selector (33a-33d) for receiving vector data and selectively outputting the input data to the registers, and a second selector (34a-34d) for receiving the vector data from the registers and selectively outputting the input data to a plurality of output lines. The data output of the arithmetic unit in each sub-system is supplied to the first selector in the same sub-system and the first selector in other sub-system, and the arithmetic unit in each sub-system receives the output data from the second selector in the same sub-system. The data output from the second selector in at least one sub-system is supplied to a main storage unit (10), and the data output from the main storage unit is supplied to the first selector in at least one sub-system.
摘要:
A parallel processor system is disclosed which comprises a plurality of processors (1) each for executing at least one of a plurality of mutually associated programs and a transfer circuit (3) connected to the processors, for transferring the data outputted from any one of the programs during execution of one program by any one of the processors (1) to other processors (1) to which a receiving program is allotted, in response to a program identification code outputted during execution of the one program by one processor to identify the receiving program.
摘要:
Herein disclosed is a parallel computer system having a plurality of processors (2), each of which is equipped with means (5) receiving from an interconnecting network and for storing pairs of a data signal and a data identification code predetermined for the data signal and for reading a data signal belonging to one of the pairs having a data identification code designated by a data readout instruction.
摘要:
A storage control apparatus for sequentially reading out each data element of vector data stored in a main storage unit (1-1-1) and for writing the readout data elements sequentially in the corresponding register elements of the vector register (VR1) disposed in a vector processor is disclosed. In the apparatus only the data element indicated as the write data by the mask information among the readout data elements are stored from the main storage unit (1-1-1) to the pertinent register elements of the vector register (VR1) in the vector processor based on the mask information which indicates whether or not the write operation is required (for example, "1" indicates that the write operation is necessary and "0" indicates that the write operation is unnecessary). When the mask information indicates that the write operation is not required, the storage control apparatus controls operationsto prevent the memory bank of the main storage from being set to the busy state, thereby eliminating the memory bank conflict which should not take place in accordance with the intrinsic system characteristics.
摘要:
A storage control apparatus for sequentially reading out each data element of vector data stored in a main storage unit (1-1-1) and for writing the readout data elements sequentially in the corresponding register elements of the vector register (VR1) disposed in a vector processor is disclosed. In the apparatus only the data element indicated as the write data by the mask information among the readout data elements are stored from the main storage unit (1-1-1) to the pertinent register elements of the vector register (VR1) in the vector processor based on the mask information which indicates whether or not the write operation is required (for example, "1" indicates that the write operation is necessary and "0" indicates that the write operation is unnecessary). When the mask information indicates that the write operation is not required, the storage control apparatus controls operationsto prevent the memory bank of the main storage from being set to the busy state, thereby eliminating the memory bank conflict which should not take place in accordance with the intrinsic system characteristics.
摘要:
An approximate quotient-correcting circuit wherein ah approximate quotient Q H , a divisor D, and the least significant bit of the fraction part of a dividend N are read out; the approximate quotient Q H and the divisor D are multiplied; it is decided that the lower m digits of Q H x D are not all 'O' and that the m-th significant bit of Q H x D is coincident with the m-th significant bit of N; and when the result of the decision is positive, Q H - 2 -m is provided as a quotient.