EP0432076A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
High performance shared main storage interface
Assignee: International Business Machines Corp
•
Filing: 1990-10-31
•
Granted: —
•
Expires: 2010-10-31
Abstract
Abstract A high performance interface joins multiple processing devices configured as masters, with multiple memory cards or other devices configured as slaves. The interface includes a working data bus (70) for transmitting working information between the processors (26, 28) and memory cards. Auxiliary busses, including a command/address bus (72) for commands and address information and a communication bus (74) for status information, are connected to all of the processing devices and slave devices (38, 40, 42, 44) and operate in parallel with the working data bus. A system for distributing control of the working information bus, among all of the master devices and slave devices, includes a grant token and plural select tokens. The grant token, a line connected in common to all devices, permits a device currently controlling the interface to retain control until it completes its transmission. The select tokens, each connected to a uniquely associated slave device and to all of the master devices, consists of a command active line activated by a master device when providing a store or fetch command, a return data line activated by a selected master device to retrieve earlier requested data from a selected slave device, and a buffer full line whereby a slave device with its buffer occupied communicates this fact to all of the master devices.
Inventors
- 👤 Richard Glenn Eikill
- 👤 Sheldon Bernard Levenstein
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-30 10:15
CPC Classifications
G; G06; G06F; G06F15/00; G06F15/16; G06F15/163; G06F15/17; G06F13/00; G06F13/14; G06F13/36; G06F13/368
Claims Analysis
Total Claims: 15
Claim 1
Independent Claim
348 words
An information processing system including: a plurality of devices for transmitting and receiving bit-encoded working information, comprising a first set of devices including at least one master device, a second set of devices including at least one slave device, and an interface between the first and second sets of devices, wherein the interface includes: (a) a first data bus shared by said plurality of devices for carrying out transmissions of the working information between the first and second sets of devices, each of the devices exclusively controlling the first data bus when providing working information to the first data bus during one of said transmissions; (b) a bidirectional signal conveyance means connected to the first and second sets of devices, for transmitting a data bus hold signal from one of the devices currently in control of the first data bus, to the others of said devices, thereby to prevent another one of the devices from controlling the first data bus until the currently controlling device has completed one of said transmissions; and (c) a selection means, connected to the first and second sets of devices, for transmitting control signals, each from a selected one of the master devices to a selected one of the slave devices, thereby to transfer control of the first data bus to one of the devices after said currently controlling device completes said one transmission, said control signals including: a first control signal, operative after termination of the data bus hold signal, to allow transfer of control of the first data bus, enabling the selected master device to transmit working information to a selected slave device over the first data bus; a second control signal to enable a selected slave device to buffer working information for subsequent transmission thereof from the selected slave device to a selected master device; and a third control signal, operative after termination of the data bus hold signal to pass control of the first data bus to the selected slave device via the selected master device, for transmission of the previously buffered working information to the selected master device.
Claim 2
Dependent Claim
58 words
The information processing network of Claim 1 wherein: said selection means further includes a means for transmitting a buffer-full signal from a selected slave device to said first set of devices indicating that a buffer of the selected slave device is occupied, to prevent the selected slave device from accepting any of the first or second control signals.
Claim 3
Dependent Claim
42 words
The information processing network of Claim 1 wherein: said interface further includes a second data bus for transmissions of bit-encoded control and address information, said second bus shared by the plurality of devices and operating in parallel with the first data bus.
Claim 4
Dependent Claim
35 words
The network of Claim 3 wherein: said selection means includes a first signal transmission means for transmitting the first and second control signals, and a second signal transmission means for transmitting the third control signals.
Claim 5
Dependent Claim
126 words
The network of Claim 3 wherein: said first set includes a plurality of master devices, at least several of the master devices being processors, and the second set includes a plurality of slave devices, at least several of the slave devices being memory cards; said signal conveyance means comprises a single line grant token connected in common to said plurality of devices; and said selection means comprises a plurality of select tokens, each of the select tokens unique to one of the slave devices and connected between its associated slave device and all of the master devices; each of the select tokens includes a command active line for transmitting the first and second control signals, and a return data line for transmitting the third control signal.
Claim 6
Dependent Claim
88 words
The network of Claim 5 wherein: each of the select tokens further includes a bufferfull line for transmitting a signal from the associated slave device to each of the master devices indicating that a buffer of the associated slave device is occupied, to prevent the selected slave device from accepting the first commands or the second commands from any of the master devices, each of said select tokens consists of one of the command active lines, one of the return data lines and one of the buffer-occupied lines.
Claim 7
Dependent Claim
267 words
The network of Claim 5 wherein: the master devices, slave devices and interface operate according to a predetermined clock cycle and the first control signal is generated by a selected master device and provided to a selected slave device over an associated command active line during a designated one of said clock cycles, a store command including destination address information is generated by the selected master device and provided to the selected slave device over the second data bus during the designated cycle, and wherein working information is transmitted from the selected master device to the selected slave device over the first data bus during the next clock cycle following the designated clock cycle; said second and third control signals comprise portions of a data fetch sequence occurring in three consecutive stages including a first stage during which the second control signal is generated by a selected master device and provided to a selected slave device over an associated command active line and during which the selected master device issues a fetch command including address information to said selected slave device over the second data bus, a second stage during which the slave device, responsive to the second control signal, buffers selected working information, and a third stage during which the selected slave device, responsive to receiving the third control signal from the selected master device, transmits the buffered working information over an associated return data line to the selected master device; and said first and second data busses, during said second stage, are available to transmit information other than information involved in said data fetch sequence.
Claim 8
Dependent Claim
48 words
The data processing network of Claim 3 further including: a third data bus shared by said plurality of devices and operating in parallel with the first and second data busses, for transmission of bit-encoded status information from the second set of devices to the first set of devices.
Claim 9
Dependent Claim
53 words
The network of Claim 8 wherein: said status information corresponds to working information transmitted from a selected slave device to a selected master device when the selected slave device controls the first bus, said selected slave device also controlling the third data bus to transmit the working information and associated status information simultaneously.
Claim 10
Dependent Claim
59 words
The network of Claim 8 wherein: said status information includes an acknowledgment that command and address information was received by the selected slave from a selected master over the second data bus, said acknowledgment being sent to said selected master over the third data bus during the next clock cycle after the transfer of the command and address information.
Claim 11
Dependent Claim
31 words
The network of Claim 3 wherein: the control and address information includes a byte-aligned start address and a length of the address where data either is to be stored or fetched.
Claim 12
Dependent Claim
340 words
In a data processing system including a plurality of master devices including at least one processing device, a plurality of slave devices including at least one memory card, and an interface between the master devices and slave devices including a first data bus shared by the master devices and slave devices for transmitting bit-encoded working information between the master devices and slave devices, a process for transferring exclusive control of the first data bus among the master devices and slave devices, including the steps of: transmitting a data bus hold signal from one of the devices when currently in control of the first data bus, to the others of said devices, thereby to prevent another one of the devices from controlling the first data bus until the currently controlling device has completed a current transmission of the working information over the first data bus; transmitting control signals, each from a selected one of the master devices to a selected one of the slave devices, over a selection means connected to the master devices and slave devices, thereby to transfer control of the first data bus to one of the devices after the currently controlling device completes the current transmission, said transfers of control each including one of the alternative steps of: (a) storing data by transmitting a first control signal operative after termination of the data bus hold signal, to transfer control of the first bus to a selected master device, enabling the selected master device to transmit working information to a selected slave device over the first data bus; and (b) fetching data by transmitting a second control signal to enable a selected slave device to buffer working information for subsequent transmission thereof from the selected slave device to a selected master device, and transmitting a third control signal, operative after termination of the data bus hold signal, to transfer control of the first data bus to the selected slave device via the selected master device, for transmission of the previously buffered working information to the selected master device.
Claim 13
Dependent Claim
47 words
The process of Claim 12 wherein: said step of transferring control of the first data bus includes the step of transmitting the first control signal, or alternatively the step of transmitting the second and third control signals, while the currently controlling device controls the first data bus.
Claim 14
Dependent Claim
58 words
The process of Claim 13 including the further step of: transmitting a buffer-full signal from a selected slave device to all of the master devices, over the selection means, to indicate that a buffer of the selected slave device is occupied, thereby to prevent the selected slave device from accepting any of the first and second control signals.
Claim 15
Dependent Claim
283 words
The process of Claim 13 wherein: the selection means includes a plurality of select tokens, each of the select tokens unique to one slave device and connected between its associated slave device and all master devices, each of the select tokens including a command active line and a return data line, and wherein the storing step includes: transmitting the first control signal from a selected master device to a selected slave device over an associated command active line during a designated clock cycle, providing a store command from the selected master device to the selected slave device during the designated clock cycle and over a second data bus connected to all of the devices in parallel with the first data bus, and transmitting working information during the next clock cycling after the designated clock cycle, from the selected master to the selected slave device over the first data bus; said data fetching step includes providing a fetch command including address information from a selected master device to a selected slave device over the second data bus and simultaneously providing the second control signal from the selected master to the selected slave device over an associated command active line; buffering selected working information in the selected slave device responsive to receiving the second control signal; and responsive to receiving the third control signal from the selected master device over an associated return data line, transmitting the buffered working information over the second data bus to the selected master device from the selected slave device; the process further including the step of: transmitting bit-encoded status information from the selected slave device to the selected master device, simultaneously with said transmission of the buffered working information.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.