US 5,513,331 AGrant
Method and Apparatus for Automatically Configuring System Memory Address Space of a Computer System Having a Memory Subsystem With Indeterministic Number of Memory Units of Indeterministic Sizes During System Reset
Issue Date:1996-04-30
•4 Claims
•7 Drawing Sheets
Abstract
An apparatus and method for assigning memory address information in a computer system. The present invention relates to computer systems having a plurality of ports or slots for coupling boards or other apparatus accessible to a processor of the computer system. The computer system further comprises an address decoder electrically coupled with the ports and processor. The address decoder receives, preferably at the time the computer system is first powered on, memory size identifying information from each of the memory boards coupled with the plurality of slots. The address decoder assigns system address space to each of the individual memories.
Metadata
Assignee
- Intel Corporation
Inventors
- Stephen Pawlowski
- Peter D. MacWilliams
Application Information
Application Number:US 4126076
Filing Date:1995-03-29
Priority Date:1988-12-30
Art Unit:235
Classifications
IPC:
G06F 1202
Field of Search:
395401;402;405
Patent Drawings (7 sheets)
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the field of self-identification systems for memories in computer systems and, more specifically, to methods and apparatus for identifying memory size in memory modules of a computer system.
2. Prior Art
Many known computer systems comprise a plurality of modules such as processor modules, memory modules, etc., which communicate over a bus system. Typically, the modules may be under control of a common or synchronized clock signal. Generally such a configuration provides a CPU clock signal on the bus. Individual modules generate signals on the bus synchronous to the CPU clock signal.
A deterministic relationship is assumed to exist during the design of the various modules. Modules, such as the processing unit, may then be designed to add a fixed number of wait states to their access cycles to accommodate slower modules.
Such a system design may require changes to the individual modules if the CPU clock speed is increased.
Therefore, as one objective of the present invention it is desired to develop a system architecture and bus protocol which allows CPU clock speed to be changed without affecting other modules in the system.
In computer systems, it is further necessary to provide certain configuration information about certain modules to other modules in the system. For example, the processor module may require information regarding the speed of a certain memory module, the type of memory access mode the memory module is capable of supporting (e.g. static column, fast page, burst, etc.), the page size used on dynamic random access memory (DRAM) chips, etc.
Therefore, as a second objective of the present invention it is desired to develop a computer system having a protocol for supporting a variety of module configurations; the protocol allowing for communication of such configuration information between modules in the system.
Known computer systems further typically may utilize a plurality of configuration or "dip" switches. The switches are utilized to provide information to the processor unit regarding the configuration of installed modules. For example, a particular system may be configured with a first memory board having four megabytes of memory and a second memory board having an additional eight megabytes of memory. In this configuration, configuration switches, either on the main system board (motherboard or baseboard) or on the individual add-on modules may be set to indicate four megabytes of memory are installed on the first memory board and eight megabytes of memory are installed on the second memory board.
In such systems, address decode logic may be employed on each memory board in the system. Based on the setting of the configuration switches in the examplary system described above, address decode logic may be employed such that the first board addresses memory in system memory space from memory address 0 to memory address 4M-1 and the second board addresses memory in the system memory space from memory address 4M to memory address 12M-1.
It is a third object of the present invention to develop a computer system allowing for communication of information regarding installed memory address space without the requirement of complex user settable configuration switches.
As a fourth object of the present invention it is desired to develop a computer system which does not require separate address decode logic on each memory module in the system.
These and other objects of the present invention will be described in more detail with reference to the detailed description of the present invention and the accompanying drawings.
SUMMARY OF THE INVENTION
A computer system having a self-identifying scheme for memory modules is described. The computer system comprises at least one processor and a plurality of "slots" or "ports" for accepting boards having memory accessible to the processor. Each memory board in the present invention may be one of 2.sup.n memory configurations, where "n" is a value between 0 and the number of control lines used during the initialization phase.
In one embodiment of the present invention, two control lines are used (i.e., n=2), allowing four possible memory configurations. In this embodiment, memory boards may have zero, four, eight or sixteen megabytes of memory. In a second embodiment of the present invention, four control lines are used (i.e., n=4), allowing sixteen possible memory configurations. In this embodiment, a memory board may have 0, 1, 2, 4, 6, 8, 12, 16, 24, 32, 48, 64, 80, 96, 128 or 256 megabytes of memory. Further, the "motherboard" of the computer system may have additional memory available to the processor. Memory residing on the motherboard is functionally equivilent to memory residing on a memory board and the amount of memory is identified in the same manner as for memory boards.
The computer system comprises a system bus for communicating data, address and control information between the processor and memories.
The computer system further comprises a decoder means. The decoder means accepts memory size identification information from each of the memories and assigns system memory address space to each of the memories. The decoder means of the preferred embodiment is coupled with the system bus. Preferably at system power-up, the decoder means provides a first signal indicating a request to the memories to provide memory size identification information. Responsive to the decoder means asserting the first signal, each of the memories provides memory size identification information. The memory size identification information is provided sequentially over the system bus in order by slot number. The memory size identification information identifies the size (amount of memory space) of the memory module. The decoder means then assigns system memory address space to each memory module based on the memory size identification information.
The present invention discloses two alternative embodiments for identification of memory. In a first embodiment, control lines are associated with each slot and the the motherboard at system power-up time. Signals on the lines associated with each slot are driven by the memory board in that slot to identify the slot's memory configuration. Configuration lines for each slot are driven simultaneously for all slots. Lines associated with each slot are polled in a predetermined order by the decoder means to determine the memory configuration for each slot.
In a second embodiment, during system power-up, the decoder means will drive a control signal to select a slot in a predetermined sequence. During the time the slot is selected (i.e., the control signal is driven), the decoder means will sample "n" control lines referred to as the -YCONFIG lines. These lines will be driven by the memory board in the slot (assuming one exists) to indicate the memory configuration of that slot during the time the slot is selected. After receiving memory configuration information for the first slot, the decoder means selects each of the other slots in the system. As slots are selected, the selected slot drives the -YCONFIG lines.
Once configuration information has been received by the decoder means by either of the above two methods, the decoder means will signal power-on reset circuitry that the memory identification sequence is complete. Normal system operation may then be started.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram illustrating a computer system as may be utilized by the present invention.
FIG. 2 is a timing diagram illustrating start-up memory identification signals as may be utilized by the present invention.
FIG. 3(a) is a timing diagram illustrating a arbitration protocol as may be utilized by the present invention.
FIG. 3(b) is a timing diagram illustrating a bus master exchange protocol as may be utilized by the present invention.
FIG. 3(c) is a timing diagram illustrating a preemption arbitration protocol as may be utilized by the present invention.
FIG. 4(a) is a timing diagram illustrating a deterministic read cycle as may be utilized by the present invention.
FIG. 4(b) is a timing diagram illustrating an asynchronous read cycle as may be utilized by the present invention.
FIG. 4(c) is a timing diagram illustrating a deterministic write cycle as may be utilized by the present invention.
FIG. 4(d) is a timing diagram illustrating an asynchronous write cycle as may be utilized by the present invention.
FIG. 4(e) is a timing diagram illustrating a fast paged mode operation (read or write) and static column mode write operation as may be utilized by the present invention.
FIG. 4(f) is a timing diagram illustrating a static column mode read operation as may be utilized by the present invention.
DETAILED DESCRIPTION OF THE INVENTION
A method and circuitry for identification of memory module size and assignment of system memory address space is described. In the following description, numerous specific details are set forth such as signal names, etc., in order to provide a thorough understanding Of the present invention. It will be obvious, however, to one skilled in the art that the present invention may be practiced without these specific details. In other instances, well-known circuits, structures and techniques have not been shown in detail in order not to unnecessarily obscure the present invention.
The present invention relates to a computer system architecture and bus protocol which allows for communication of certain configuration information regarding individual modules in the computer system over the system bus. The preferred embodiment of the present invention teaches providing a first plurality of configuration information on the system bus regarding memory size for each individual memory module at system power-up,time. A second plurality of configuration information is provided by the individual modules in response to any request for service by those modules.
Utilizing the circuitry and protocol of the present invention, increased CPU clock speeds may be supported without requiring changes to the design of other existing modules in the system.
Further, memory may be assigned to system memory address space without requirement for "dip" switches being set to identify memory module size. Further, address decode logic may be centrally located within the computer system and does not have to be duplicated on each memory module.
COMPUTER SYSTEM OVERVIEW
General System Description
The present invention discloses a method of "decoupling" the processor of a modern computer system from the other modules of the computer system. The processor's internal clock speed is independent of the clock speeds of other modules of the computer system.
It is known in modern computer system to allow modules, such as processors and memories, to operate asynchronously of each other. However, in such computer systems, a penalty is associated with any communication between the independent blocks. The "penalty" may be measured by the lost time or wait states imposed by synchronizing a signal incoming to a module with the clock speed of the module.
The present invention discloses operating the processor and other modules in an asynchronous mode while avoiding synchronization penalties of known computer systems.
Generally, the present invention discloses a processor (or other bus master) making a request to memory (or other resource such as I/O) over a bus. At the time the request is initiated by the processor, the bus is in a deterministic state (i.e., the processor knows the worst case time required to access a given memory). The processor waits on a response from the memory. The memory may respond to the request with information on its type, speed and mode of operation.
If the memory responds that it may be accessed in a deterministic manner, the processor waits the appropriate number of clock cycles (the number of clock cycles determined based on information returned from the module) and reads data from the bus. If the memory responds that it may not be accessed in a deterministic manner (i.e., it must be accessed asynchronously), the processor and memory communicate in an asynchronous manner with the inherent synchronization penalties.
The present invention, in its preferred embodiment, supports a 32-bit bus architecture that compliments the standard IBM compatible PC/AT interface. The bus is primarily used as a memory bus; however, it is capable of supporting multiple bus masters and I/O systems. This allows the present invention to be utilized in future embodiments having co-processors requiring high bus bandwidth.
Block Diagram of the Preferred Embodiment
A block diagram of a computer system of the presently preferred embodiment of the present invention is illustrated with reference to FIG. 1. A processor module 101 comprises a processor 102 having a cache memory 103 with its associated address tag buffer 104 and control circuitry 105. The processor 102 and address tag buffer 104 are coupled with buffer interface 106 through an address bus 107. The processor 102, along with the cache memory 103, is further coupled to the buffer interface 106 through a data bus 108.
The processor module 101 is coupled to the system bus 120 through the buffer interface 106. As stated previously, future embodiments of the present invention may comprise a plurality of coprocessors coupled with the system bus 120. The processor module 101 may access a memory, such as memory 130 over system bus 120.
As will be described in more detail with reference to FIG. 2, memory of the computer system of the present invention is configured (that is, memory modules will be assigned addresses in system memory address space) at the time the computer system is powered up. The preferred embodiment of the present invention comprises an address decode module 140 coupled with system bus 120 for configuring system memory. Each memory module in the system presents configuration information to the address decode module 140 at system power-up time.
A plurality of expansion slots, such as expansion slot 150, are coupled with the system bus 120. Additional memory or other resources may be plugged into these expansion slots and may communicate with processor module 101 over the system bus 120.
I/O modules or other resources may be coupled to the system bus 120 through interface 160 and are allowed to communicate with processor module 101 over the system bus 120.
Signal Naming Conventions
The present invention will be better understood by one of ordinary skill in the art with reference to the timing diagrams of FIGS. 2, 3(a)-(c) and 4(a)-(c). In these figures and in the accompanying text, signals will be generally referred to with signal names. It will be obvious that the particular signal names may be varied without departure from the scope and spirit of the present invention. Further, the notation used will normally refer to signal states as being active or inactive, as opposed to be high or low, one or zero, true or false, etc. Generally, in the description of the preferred embodiment, signals which are active when low are indicated with a leading minus sign ("-"). It will be obvious to one of ordinary skill that the active state of signals may be reversed from that of the present invention without departure from the inventive aspects of the present invention. As examples of this signal naming convention, see Table I, below:
In describing signals, many signals are more easily or conventionally referred to as a group. Therefore, in describing groups of signals, signals may be referred to using a decimal radix convention, such as YD(31:0) referring to 32 data lines of a bus. Within each group, the least significant bit of the group is referred to with a suffix of "0", e.g., YD0 refers to the least significant data line of the bus and YD31 refers to the most significant data line.
Further, the preferred embodiment of the present invention has a plurality of card slots for coupling cards to the system bus 120. In certain cases, signal names will have a lower case "n" following the signal name. In such cases, the "n" corresponds to one of the card slots. For example, -YREQn refers to one of a plurality of signals, each of the plurality of signals associated with a card slot. -YREQ3 refers to a signal associated with card slot number 3.
Signal Descriptions
It will be useful in understanding the present invention to generally describe certain signals utilized by the preferred embodiment.
Address Signal Group
This group of signals are address signals driven by a requestor at the beginning of an access.
YADDR or YADDR(31:2)
Refers to the 30 tri-state address lines on the address bus of the present invention. The address lines are not required to be latched by the processor and may not stay valid during the entire bus cycle.
-YBE 30!
These lines refer to each of the four bytes of a 32-bit word addressed by the YADDR(31:2) signal lines. Together, the -YBE 3:0! and YADDR(31:2) comprise the 32-bit bus addressing mechanism of the present invention.
Data Signal Group
YD(31:0) or YDATA(31:0)
Refers to the 32 tri-state data lines on the data bus of the present invention.
Request Signal Group
These signals are driven by the requestor when requesting and gaining access to the bus and by the arbiter when granting access to the bus.
-YREOn
The -YREQn signal indicates that a processor is requesting use of the system bus. For example, if a processor at slot 2 requests use of the system bus, the processor at slot 2 will activate (bring low) signal -YREQ2. The processor continues to activate signal -YREQ2 as long as the processor requires use of the system bus.
-YPRYn
The -YPRYn signal is driven by a requesting processor and is assumed to be stable whenever a processor is asserting its -YREQn signal. If the -YPRYn signal is inactive, the request is to follow a preemption exchange protocol (described in more detail with reference to FIG. 3(c) and the below description of the preemption protocol). If the -YPRYn signal is active, the request is to follow a normal exchange protocol (described in more detail with reference to FIG. 3(b) and the below description of the normal exchange protocol).
YSTAT 2:0!
These three signals define the type of transfer that is being requested by the bus agent. Transfers can be either I/O or memory operations.
-YGNTn
Responsive to a -YREQn signal being activated, a central arbiter will grant use of the system bus to the requesting processor when no requesting device with a higher priority is activating its -YREQn signal. The central arbiter activates the -YGNTn signal to signal the requesting processor that it has been granted access to the system bus. After granting access to the bus to a requester, the central arbiter may remove access by inactivating the -YGNTn signal. The requester is allowed to complete the information exchange on the current bus cycle and will then deactivate its -YREQn signal.
Response Signal Group
These signals are activated (or inactivated in the case of YRDY) by a module responding to a request from a bus agent.
YMODE 1:0!
This group of signals is used to indicate the type of access mode a particular selected memory can support.
YSPD 1:0!
This group of signals defines the access speed of a particular selected memory device.
YSIZE 1:0!
This group of signals defines the density of the particular selected memory device used for fast page mode operations.
-YCEN
When active, this line indicates that the particular memory access is cachable.
YRDY
This signal is normally high on the bus. It is pulled high during bus idle cycles. If a particular responding agent can respond in a deterministic mode, the agent will not drive the YRDY signal and it will remain high (active) on the bus. If the responding agent will respond in an asynchronous mode, then the agent will drive the YRDY signal inactive to initiate an asynchronous handshaking protocol.
Command Signal Group
These signals are used as commands to initiate and terminate accesses to the bus and to indicate a reset sequence.
-YASTB
The -YASTB signal is the bus access strobe line indicating the beginning and end of bus events.
-YCASTB
The -YCASTB signal is used by a requestor to initiate and terminate a fast page mode or static column mode memory request. The -YASTB signal is held active to keep memory active and individual cycles are initiated and terminated by the -YCASTB signal.
-RESETDRV
This is the reset signal on the system bus used to indicate to modules in expansion slots the start of a reset sequence, such as at system power-up time.
Memory Initialization Signals
-YSELn
The -YSELn line is generated by address decode logic and, when active, indicates to a selected resource which module is being accessed. Each slot has its own -YSEL signal, the particular slot being indicated by the n.
YCONFIG 3:0!
The YCONFIG 3:0! signal lines are used to define memory module size at system power-up time in one embodiment of the present invention.
YIDn 1:0!
The YIDn 1:0!
The YIDn 1:0! signals are associated with each slot of the present invention and are used to define memory module size at system power-up time in a second embodiment of the present invention.
MEMORY SIZE IDENTIFICATION (FIG. 2)
Referring now to FIG. 2, the method used by the preferred embodiment for identifying memory size is described in detail. The present invention allows for boards in the computer system which contain memory space accessible to processors of the present invention to identify the memory size which is available at the time the system is started up. It will be obvious to one of ordinary skill that memory size identification does not have to occur precisely at start-up time and that other system management functions may occur prior to memory size identification. Further, one of ordinary skill in the art can appreciate a system implementing the methods of the present which allows for reallocation of system memory after start-up of the computer system. For example, additional memory may be allocated as being accessible to the processors of the present invention after start-up.
The present invention comprises a central address decode module, shown as decode module 140 in FIG. 1. Although the preferred embodiment teaches a separate module for implementing the decode function, one of ordinary skill can appreciate other means may be utilized, such as by using a general purpose processing unit for decoding the address signals and assigning system memory address space. The address decode module latches size information received from each memory module in the system. The address decode module then assigns system address space to each memory module in the system.
System address space from memory location 0 to ( first memory module size-1!) is assigned to the first memory module. System address space from memory location ( first memory module size!) to memory location ( second memory module size-1!) is assigned to the second memory module. System address space from memory location ( first memory module size+second memory module size!) to memory location ( first memory module size+second memory module size+third memory module size-1!) is assigned to the third memory module. This pattern continues for assignment of system address space to each memory module in the system.
In the preferred embodiment, as shown by FIG. 2, a state machine in the address decode circuitry is activated by the -RESETDRV signal 201 being activated such as at the time the system is powered up or at other times during system operation when a system reset is required (*time t1). The state machine maintains the -RESETDRV signal active during the self-identification process. The state machine sets an internal address counter to zero. This counter is used for allocation of the memory address space for the various modules. The state machine then drives the -YSELn 202 signal active for slot 0 (time t2). If there is a board in slot 0, it drives configuration signals YCONFIG. 204 to indicate the amount of memory available in the slot. In the present invention the YCONFIG signals are communicated on the bus over signal lines used for other signals during non-reset period.
In one embodiment of the present invention, the -YIDn1 signal line and -YIDn2 signal line are used for providing memory size identifying information, as illustrated with reference to Table II(a), below. Note, with reference to Table II(a) and other tables in this specification, a "H" indicates a high state on the bus and a "L" indicates a low state.
In this first embodiment, all modules in all slots in the system provide memory size information on signal lines associated with the particular slot during the time the -RESETDRV signal is activated. The modules maintain these signals until the -RESETDRV is inactivated. During the time the -RESETDRV signal is activated, the decode module polls the signals associated with each slot and determines memory configuration information for that slot.
A second embodiment of the present invention, utilizes the YCONFIG signal lines during the reset period for configuration of memory. The memory size information communicated in this embodiment is understood with reference to Table II(b).
As can be seen with reference to Tables II(a) and II(b), in the case of a module not being present in a slot, or if the module has no system memory available, status information is returned indicating no memory (0 Mbytes) is available.
In the case of the second embodiment, the -YSELn signal is utilized to select a slot for providing memory configuration information. As will be seen with reference to FIG. 2, the -RESETDRV signal 201 is activated at item t1. Responsive to the -RESETDRV signal being activated, the decode module begins polling each of the slots for memory configuration information by driving the appropriate -YSELn signal 202. For example, the decode module first drives the -YSELn signal 202 to select slot 0 at time t2. The module at slot 0 then drives configuration information onto the YCONFIG lines 204 at time t3 and maintain valid information on these lines until the decode module inactivates the -YSELn signal 202 at time t4. The module at slot 0 then stops driving valid information at time t5.
If the module in slot 0 responds with a memory configuration of greater than 0 Mbytes, the decode module increments the internal counter and assigns system address space to the module as described above.
This cycle is repeated for each slot by the decode module driving the appropriate -YSELn signal 202 and the module at the corresponding slot responding by driving YCONFIG lines 204 during time t7.
The state machine disables its control of the-RESETDRV signal 202 after completion of this cycle for each memory module.
The address decode module identifies, during subsequent information exchanges between modules in the system, the appropriate memory module corresponding to a given address. This protocol allows elimination of configuration switches on the motherboard of the system for configuring memory on individual memory modules. Further, address decode logic is centrally located in address decode module 140. This eliminates the need for address decode circuitry on each of the individual memory modules.
BUS ARBITRATION CYCLE
The present invention discloses three arbitration cycles for arbitrating access to the system bus by a requestor. In general, a processor asserts its -YREQn signal and, depending on the type of arbitration desired, its -YPRYn signal. A central arbiter receives a separate -YREQn and -YPRYn signal for each potential requester in the system. The central arbiter evaluates requests and grants ownership of the bus as described in connection with FIGS. 3(a)-(c). The three arbitration cycles will be referred to as: (1) normal arbitration, (2) exchange protocol and (3) preemption protocol.
Normal Arbitration--FIG. 3(a)
Referring first to FIG. 3(a), the normal bus arbitration protocol is illustrated. In normal arbitration, a default bus owner, the motherboard processor in the preferred embodiment, has control of the bus. A coprocessor (requestor) drives its -YREQn signal 301 active and its -YPRYn signal 302 inactive at time t1. Responsive to the coprocessor driving its -YREQn signal 301 active, the central arbiter disables the default bus owner from accessing the bus by inactivating its -YGNTn signal (not shown) and then activates the requestor's -YGNTn signal 303 at time t2 (after the previous bus onwer releases its -YREQn line). The requestor senses its -YGNTn signal 303 being driven active and drives the appropriate transfer information onto the bus 304 at time t3.
The requestor continues to drive its -YREQn signal 301 active as long as it continues to transfer information onto the bus (during time t3 through t4). Unless a higher priority request is received or a refresh cycle is required, the central arbiter continues to drive active the -YGNTn signal 303. Subsequent to completing the transfer of information, the requestor inactivates its -YREQn signal (at time t5). The arbiter then inactivates the -YGNTn signal and grants access to the bus to the highest priority pending requestor.
Exchange Protocol--FIG. 3(b)
Referring to FIG. 3(b), the bus exchange protocol is illustrated. The bus exchange protocol allows exchange of control of the bus from one coprocessor to another. In the illustration of FIG. 3(b), coprocessor 1 has control of the bus at time t1 (the central arbiter is asserting the -YGNT1 signal 316 and its -YPRY signal is inactive, granting coprocessor 1 access to the bus). Coprocessor 0, which is lower priority than coprocessor 1, requests access to the bus by activating its -YREQ0 signal 310 at time t1. Coprocessor 0 will continue to assert its -YREQ0 signal 310 until it is granted control of the bus. Coprocessor 0 does not assert its -YPRY0 signal 311. Therefore, this is not a preemption request and coprocessor 1 is allowed to complete its information exchange on the bus (valid data continues to be transmitted on the bus 313 during time interval t1 through t2).
Coprocessor 1 completes its information exchange at time t2. It then inactivates the -YREQ1 signal 314 at time t3, indicating to the central arbiter it is ready to give up control of the bus. The central arbiter inactivates the -YGNT1 signal 316 at time t4 and activates the -YGNT0 signal 312 at time t5. Activating the -YGNT0 signal grants coprocessor 0 access to the bus. Coprocessor 0 begins its information exchange on the bus 313 at time t6.
Preemption Protocol--FIG. 3(c)
A processor may preempt access to the bus by another coprocessor if it has higher priority by asserting its -YPRYn signal. Priority levels in the computer system of the preferred embodiment are shown in Table III, below:
In the preferred embodiment, priority level 0 is the highest priority and priority level 9 is the lowest priority. The default bus owner is the motherboard; all other requestors have a higher priority, therefore, any pending request by any other requestor will preempt the motherboard.
The present invention discloses use of two priority levels for each coprocessor card. The first level (referred to in the preferred embodiment as the "normal level") is a lower priority level than the priority of the I/O system. This priority level is used for normal processing. The second priority level (referred to in the preferred embodiment as the "augmented level") is a higher priority level than the priority level of the I/O system. Coprocessors may elevate their bus requests above the level of the I/O system. In the preferred embodiment, a coprocessor using its augmented priority level will not interrupt an I/O cycle, however, the I/O subsystem will be prevented from preempting the coprocessor once the coprocessor has been granted access to the system bus. In the normal level, the I/O subsystem may preempt a coprocessor which has been granted access to the system bus. Augmented priority levels are useful where it is desired to allow the coprocessors to operate in an environment having an increased share of the bus bandwidth.
As can be seen with reference to Table III, in implementations in which the CPU requires uninterrupted access to the system bus, it will operate at priority level 1 (next to highest priority). In all other cases, it will operate at priority level 9 (lowest priority).
The preemption protocol of the present invention is described in more detail with reference to FIG. 3(c). Coprocessor n has control of the bus at time t1 (-YGNTn signal 322 is active) and is transferring information over the bus 323. Coprocessor m, which has a lower priority than coprocessor n when operating in normal mode, attempts to gain control of the bus; first, coprocessor m activates its -YPRYm signal 324 to indicate that it will be doing a preemption bus request at time t1. In the preferred embodiment, coprocessor m may activate either its normal or augmented -YPRYm signal. At time t2, coprocessor m activates -YREQm signal 325 to request control of the bus. Responsive to this request and the -YPRYm signal being active, the central arbiter inactivates the -YGNTn signal 322 at time t3.
Coprocessor n continues to control the bus and transfer information through the next bus cycle (until time t4) and then inactivates its -YREQn signal 321 at time t5. Responsive to coprocessor n inactivating -YREQn signal 321, the central arbiter activates signal -YGNTm 326 at time t6. Coprocessor m then controls the bus and may perform its information transfer on the bus 323 (from time t7 through t8).
Anytime after detecting the beginning of a bus event (-YASTB being driven active), coprocessor n may reactivate its -YREQn signal 321 to request control of the bus in order to complete its interrupted bus transfer. Control of the bus will be returned to coprocessor n after coprocessor m completes its information transfer, assuming there are no higher priority arbitration requests pending.
After completing its information transfer, coprocessor m inactivates the -YREQm signal 325 (at time t9). The central arbiter then inactivates the -YGNTm signal 326 at time t10 responsive to coprocessor m releases its request. Coprocessor m may inactivate its -YPRYm signal 324 after the -YGNTm signal 326 is inactivated (such as at time t11).
REQUEST/RESPONSE PROTOCOL
The present invention discloses a request/response protocol which allows memory devices to identify their access mode, (YMODE 1:0!), page size (YSIZE 1:0!), whether the memory device is cachable (-YCEN), the speed of the memory device (YSPD(1:0)) and whether the device may be accessed deterministically or asynchronously (YRDY).
Specifically, the preferred embodiment of the present invention allows four access modes: (1) standard RAS/CAS mode, (2) fast paged mode, (3) static column mode and (4) burst mode. The memory device identifies the mode it is capable of operating in by activating or inactivating the appropriate YMODE (1:0) group signals, as shown with reference to Table IV, below:
Standard RAS/CAS mode is used for I/O and other bus accesses as well as standard memory accesses. Static column mode is a mode in which the row addresses are latched by the memory device and column addresses flow through. Data changes with respect to a change in column addresses (i.e., an array may be read down its columns). Fast paged mode is similar to static column mode, however, the CAS addresses do not flow through but instead are latched on each successive access by the active going edge of the CAS. Burst mode is reserved for support of a future processor which may allow data to be "bursted" to or from the processor at a high speed. However, an understanding of the workings of any of the particular modes are not required for an understanding of the present invention.
The YSIZE 1:0! signals of the preferred embodiment is interpreted with reference to TABLE V.
The -YCEN signal is interpreted with reference to TABLE VI. A memory access is considered to be cachable when the coherency of data in memory and cache can be enforced.
The -YSPD signal is interpreted with reference to TABLE VII. The speed in the preferred embodiment is defined in terms of the RAS (row address strobe) access time.
When a processor makes a revest, it identifies the type of request using a set of request lines YSTAT 2:0!. The YSTAT 2:0! lines define the type of access currently on the bus and are coded, in the preferred embodiment, as defined by Table VIII, below:
Generally, when a coprocessor (requestor) makes a request for access to a memory device, the responding memory device will respond within a predetermined period of time to the request with information identifying its characteristics. If the memory device responds that it is able to operate in a deterministic manner, the requesting coprocessor waits a determinable period of time and then pulls data off of the bus. If the memory responds that it must operate in an asynchronous manner, the requesting processor accesses the memory device asynchronously encountering the synchronization penalties associated with the required handshaking.
FIGS. 4(a)-(d) illustrate access protocols for deterministic reads, asynchronous reads, deterministic writes and asynchronous writes, respectively.
Deterministic Read Cycle--FIG. 4(a)
Referring first to FIG. 4(a), after a processor has been granted control of the bus as described in connection with FIGS. 3(a)-(c), the processor drives an address 401 onto the bus at time t1. The processor also drives (at time t1) the appropriate request signals 402, as illustrated by Table VIII.
After setting up the ADDRESS 401 and request signals 402, the processor activates the -YASTB signal 403 at time t2. The -YASTB signal 403 indicates the beginning of a bus event.
If the YRDY signal 404 is active, a deterministic access is assumed. The YRDY signal 404 is normally in the active state and, therefore, a deterministic device does not need to drive this signal. For a deterministic access, the bus slave (memory module) asserts the appropriate response group signals. Based on the access mode and speed as defined above, the processor determines an appropriate number of wait states to delay before accessing data on the bus. In this manner (i.e., making a request and receiving a response-request/response protocol), the processor is able to customize accesses to memory based on the particular type and speed of memory being accessed. Further, this customization of memory may be accomplished without any requirement of user settable switches.
The deterministic period is calculated in the preferred embodiment based upon the access time required for the memory to respond to a given request. The deterministic period varies depending on the mode of memory operation and the speed of the memory device. Generally, the deterministic period may be calculated as follows:
Where:
T.sub.del =The amount of time from the processor reference to the -YASTB signal being active on the bus (usually with reference to a clock, e.g., -YASTB will be switched based on the processor clock; however, there is a delay from the clock to the -YASTB signal changing state. T.sub.del includes this delay);
Taccess=The period of time from -YASTB being active on the bus to valid data being on the bus (for example, in a 100 ns memory as may be utilized by the preferred embodiment, this period is 140 ns);
T.sub.data setup =The data setup period of the processor; and
T.sub.Deterministic =The deterministic memory access time for a standard RAS/CAS device.
After providing control information the YMODE, YSIZE, YCEN and YSP control lines, the data bus (YD) 405 may be driven with valid data by a bus slave. The processor waits a deterministic amount of time (time t3 through t4, determined based on the speed and type of the memory) and then may access data on the bus. Data on the bus is assumed to be valid for deterministic transfers after the deterministic period of time.
After reading the data from the bus, the processor releases the ADDRESS lines 401 and inactivates its request signals 402 at time t5. The processor then inactivates the -YASTB signal 403 at time t6. When the bus slave detects the -YASTB signal 403 going inactive, it may release the data lines (YD) 405.
This protocol is repeated for any further reads required from the memory by the processor.
Asynchronous Read Cycle--FIG. 4(b)
Referring to FIG. 4(b), the protocol for an asynchronous read is illustrated. In an asynchronous read, a processor which has been granted control of the bus drives address signals (ADDRESS) 411 onto the bus and drives its request lines 412 at time t1. The processor then activates the -YASTB signal 413 at time t2.
The selected bus slave (e.g. memory) deactivates the YRDY signal 414 at time t3 to indicate the read will be an asynchronous read. In the case of an asynchronous read, the YSIZE and YSP signals are ignored by the processor and the memory is accessed in a standard access mode, asynchronous manner.
Responsive to the -YASTB signal 413 being driven active, the selected device begins to drive data on the the data bus (YD) 415 at time t4. When valid data is available, the bus slave drives the YRDY signal 414 active (time t5). The bus master may then read data from the bus. After sensing the YRDY signal 414 being driven active, the bus master discontinues asserting its the ADDRESS signals 411 (time t6) and deactivates the -YASTB signal 413 at time t7.
The bus slave detects the -YASTB signal 413 being driven inactive and discontinues asserting data signals 415 at time t8. The YRDY signal 414 is active and will remain active throughout the bus idle state.
The bus master may request transfer of more data by repeating the above-described protocol.
Deterministic Write Cycle--FIG. 4(c)
Referring to FIG. 4(c), the deterministic write cycle of the preferred embodiment is illustrated. At time tl, the bus master which has been granted control of the bus drives an address onto the bus (ADDRESS signal 421), drives its request signal 422, and drives the data to be transferred onto the bus (YD(31:0) 425). The processor then asserts the -YASTB signal 423 at time t2.
Prior to the -YASTB signal being asserted, the address signal 421 is decoded to determine the particular memory device having the selected memory address space location. In the preferred embodiment, the selection of a particular device is dependent on system memory address space assignments based on the memory size identification information provided by each of the memory modules to the decoder module at system power-up time.
After a deterministic access time of the system bus has been met, the processor releases request signal 422 at time t3, releases address signals 421 at time t4, deactivates -YASTB signal 423 at time t5 and releases data signals 425 at time t6. As described above, the deterministic access period is based on the characteristics of the memory device; these characteristics being supplied to the processor at the time the memory device replies with its response signals. YRDY signal 424 remains active during the entire cycle. This completes the deterministic memory write cycle.
Asynchronous Write Cycle--FIG. 4(d)
Finally, referring to FIG. 4(d), an asynchronous write cycle as utilized by the present invention is illustrated. Again, the current bus master (processor granted access to the bus) drives address signals (ADDRESS 431) onto the bus, drives request signals 432 and asserts data (YD(31:0) 435) onto the bus at time t1. The -YASTB signal 433 is asserted at time t2.
Responsive to assertion of the -YASTB signal 433, the address signal is decoded and the appropriate memory selected during time t2 through t3. If the selected device cannot respond to the write request in a deterministic time (i.e., this will be an asynchronous write), the selected memory inactivates the YRDY signal 434 at time t4. The information exchange is delayed during the period of time the YRDY signal 434 is inactivated (time t4 through t5). When the selected memory device is ready to respond, the YRDY signal 434 is again activated, time t5.
Responsive to the YRDY signal 434 being activated, the bus master releases the request signal 432 at time t5, releases the address signal lines 431 at time t6 and inactivates the -YASTB signal 433 at time t7.
The selected memory detects the -YASTB signal 433 being inactivated, releases the YRDY signal 434 at time t8. However, in the preferred embodiment, YRDY signal 434 is held active by a pull-up resistor. The bus master then releases the data lines 435 at time t8.
Fast Paged Mode Operation (Read or Write) Static Column Mode Operation (Write)--FIG. 4(e)
Referring to FIG. 4(e), the protocol for fast paged mode operations, both read and write, or static column mode write operations is illustrated. During any of these operations, the processor drives the desired address on the bus, lines 443. The processor also drives the request group signals, lines 444. In the case of a write operation (either fast paged mode or static column mode), the processor also drives data lines 446.
The address decode module then generates the appropriate -YSELn signal 442 to select the memory addressed by the address on lines 443. The processor then asserts the -YASTB signal 441 to initiate the first operation. The selected memory then drives the appropriate response signals 445 indicating the type of memory mode the addressed memory is capable of operating in. The response signals 445 remain stable during the entire period the -YASTB signal 441 is asserted.
After the memory is selected and provides a response, the decode module may remove the -YSELn signal 442. The selected memory remains selected for the entire period the -YASTB signal is asserted.
As one inventive feature of the present invention, after determining, based on response signals 445, that the memory can be accessed in either a fast paged mode or static column mode, the processor activates the -YCASTB signal 448 to indicate the completion of the first operation. An event is complete either after completing placing data onto the bus for transfer to the memory on lines 446 or after completion of addressing the memory and the data being placed on the bus, lines 447.
In the case of the processor detecting the next operation is to the same page of memory, the processor deactivates the -YCASTB signal 448. The -YCASTB signal is again asserted after completion of the second operation.
At the time the processor determines it has completed all of its operations or when the processor determines the next operation will result in a page miss, the processor deasserts the -YASTB signal.
The process of asserting and deasserting the -YCASTB signal to control accesses to the same page of memory while holding the -YASTB signal active during an entire cycle of accesses a given page offers inventive advantages over the known prior art.
Static Column Mode Read Operations--FIG. 4(f)
Referring to FIG. 4(f), the process utilized by the present invention for performing static column mode read operations is described in more detail. As described above in connection with FIG. 4(e), the processor drives an address onto the bus on lines 453 along with a request type on request lines 454. The address decode logic is employed to select a particular memory. The selected memory is indicated by -YSELn line 452. The processor then asserts the -YASTB signal 451 to initiate a read operation. The selected resource responds with information on response lines 455 stating it is capable of supporting static column mode memory accesses.
The decode module may then stop asserting the -YSELn signal 452. The selected memory will remain selected as long as the -YASTB signal remains active.
The -YCASTB signal 457 is inactive at the beginning of the read cycle and remains active after each static column read request. The processor reads data on lines 456 after the first memory access cycle. The data is returned within a deterministic period of time as described above. The processor may then supply new address information on the address lines 453 and a second read cycle is initiated. Further read cycles may be initiated if required for more data on the same page. After the processor completes its data read requirements, or when a new page is to be selected, the -YASTB signal is deasserted.
In the present invention, the -YCASTB signal 457 remains inactive after a static column read and is asserted after a static column write. The -YCASTB signal is deasserted following the write cycle to indicate a new cycle has started.
The present invention utilizes two modes of operation for fast page mode operations. A first mode allows fast page mode operations to be utilized for both reads and writes. The -YASTB signal is held after an access in anticipation of another cycle to the same DRAM page as the page accessed in the last cycle. The second mode of opeation fast paged mode operations are used only for write cycles in which, at the completion of the current write cycle, another cycle is pending. In this second mode, reads are always standard RAS/CAS accesses.
Thus, a system for self-identification of memory size information and configuration of memory modules within system memory address space has been described.
Claims
We claim:
1. In a computer system comprising a processor, an address decoder, and a system bus having a plurality of connector slots, wherein the processor and the address decoder are both coupled to the bus; and one or more memory units are correspondingly connected to one or more of the connector slots, a method for automatically configuring system memory address space of the computer system during system reset, the method comprising the steps of: (a) concurrently outputting and maintaining one or more memory sizes of the one or more memory units on one or more groups of private bus signal lines of the system bus, through the connector slot(s), by the one or more memory units, while system reset is in progress, responsive to a system reset bus signal so denoting; (b) concurrently coupling the memory size(s) maintained by the memory unit(s) on the private bus signal lines to the address decoder by the bus while the memory size(s) are being output and maintained; and (c) systematically reading the private bus signal lines, one group of private bus signal line(s) at a time, incrementally configuring the system address space in accordance to the memory size(s) systematically read, and retaining the system address space configuration information by the address decoder, while system reset is in progress, responsive to the same system reset bus signals, the retained system address configuration information being used by the address decoder subsequently during operation to select the appropriate ones of the one or more memory units when the one or more memory units are address by the processor.
2. A computer system comprising: a processor; one or more memory units including one or more first circuitry correspondingly disposed on the one or more memory units for outputting memory size(s) and maintaining the output(s) while system reset is in progress, responsive to a system reset bus signal so denoting; a system bus coupled to the processor, the system bus having a plurality of connector slots against which the memory unit(s) are connected, and multiple groups of private bus signal lines correspondingly coupled to the connector signals, for concurrently transporting the memory size(s) output by the memory unit(s); as address decoder for systematically reading the private bus signal lines, one group of private bus signal line(s) at a time, incrementally configuring the system address space in accordance to the memory size(s) systematically read, and retaining the system address space configuration information by the address decoder, while system reset is in progress, responsive to the same system reset bus signal, the retained system address configuration information being used by the address decoder subsequently during operation to select the appropriate ones of the one or more memory units when the one or more memory units are address by the processor.
3. In a computer system comprising a processor, an address decoder, and a system bus having a plurality of connector slots, wherein the processor and the address decoder are both coupled to the bus, and one or more memory units are correspondingly connected to one or more of the connector slots, a method for automatically configuring system memory address space of the computer system during system reset, the method comprising the steps of: (a) outputting and maintaining one memory size of a selected one of the one or more memory units on one or more shared bus signal lines of the system bus, through one of the connector slot(s), by the selected one of the one or more memory units, while system reset is in progress, responsive to a system reset bus signal so denoting; (b) coupling the memory size maintained by the selected memory unit on the shared bus signal lines to the address decoder by the bus while the memory size is being output and maintained; and (c) systematically selecting the memory unit(s) and reading the shared bus signal lines, one memory unit at a time, incrementally configuring the system address space in accordance to the memory size(s) systematically read, and retaining the system address space configuration information by the address decoder, while system reset is in progress, responsive to the same system reset bus signal, the retained system address configuration information being used by the address decoder subsequently during operation to select the appropriate ones of the one or more memory units when the one or more memory units are address by the processor.
4. A computer system comprising: a processor; one or more memory units including one or more first circuitry correspondingly disposed on the one or more memory units for outputting memory size(s) in turn, responsive to a plurality of selection signal denoting when each memory is selected for output, and maintaining each output during each memory unit's turn, while system reset is in progress, responsive to a system reset bus signal so denoting: a system bus coupled to the processor, the system bus having a plurality of connector slots against which the memory unit(s) are connected, and one or more shared bus signal lines coupled to the connector slots, for transporting the memory size output by the selected memory unit; as address decoder for systematically selecting the one or more memory units, one memory unit at a time, reading the shared bus signal lines, incrementally configuring the system address space in accordance to the memory size(s) systematically read, and retaining the system address space configuration information by the address decoder, while system reset is in progress, responsive to the same system reset bus signal, the retained system address configuration information being used by the address decoder subsequently during operation to select the appropriate ones of the one or more memory units when the one or more memory units are address by the processor.
Patent Citations (13)
| Patent | Date | Inventor | Cited By |
|---|---|---|---|
| US3872452 | 1975-03-01 | Stoops | |
| US4001790 | 1977-01-01 | Barlow | |
| US4281392 | 1981-07-01 | Grants et al. | |
| US4744025 | 1988-05-01 | Lipcon et al. | |
| US4751671 | 1988-06-01 | Babetski et al. | |
| US4780916 | 1988-11-01 | Martin | |
| US4787060 | 1988-11-01 | Boudreau et al. | |
| EPX80626 | 1982-11-01 | ||
| EPX121381 | 1984-03-01 | ||
| EPX136178 | 1984-09-01 | ||
| EPX200198 | 1986-04-01 | ||
| EPX285986 | 1988-03-01 | ||
| GBX2215497 | 1988-09-01 |