摘要:
Burst SRAMs designed for operation at a given data rate corresponding to the frequency of a first clock signal but capable of operation using a higher frequency clock signal. The burst SRAMs are preferably incorporated into the cache memory of a second level cache coupled to the processor bus in a computer system, where the computer system is preferably based on a 66-MHz P5 microprocessor. A cache controller, preferably incorporated within a memory controller, controls operation of the second level cache memory by providing the address load and address advance signals. The burst SRAMs are capable of recognizing the faster clock pulses, as well as the shorter pulses asserted on the address load and address advance signals. The address control signals are asserted and then negated during consecutive clock cycles of the faster clock signal, so that the burst SRAMs effectively operate at the same data rate corresponding to the lower frequency clock signal.
摘要:
A memory controller which makes maximum use of any processor pipelining and runs a large number of cycles concurrently. The memory controller can utilize different speed memory devices at their desired optimal speeds. The functions are performed by a plurality of simple, interdependent state machines, each responsible for one small portion of the overall operation. As each state machine reaches has completed its function, it notifies a related state machine that it can now proceed and proceeds to wait for its next start or proceed indication. The next state machine operates in a similar fashion. The state machines responsible for the earlier portions of a cycle have started their tasks on the next cycle before the state machines responsible for the later portions of the cycle have completed their tasks. The memory controller is logically organized as three main blocks, a front end block, a memory block and a host block, each being responsible for interactions with its related bus and components and interacting with the various other blocks for handshaking. The memory controller operates in system management mode to override any write protect status of memory so that the SMRAM can be located in the main memory space and be write protected during normal operations but be full usable during system management mode.
摘要:
A memory controller for a computer system provides a series of queues between the processor and a PCI bus and the memory system. Memory coherency is maintained in two different ways. Before any read operations are accepted from the PCI bus, both of the posting queues must be empty. A content addressable memory (CAM) is utilized as the PCI-to-memory queue. When the processor performs a read request, the CAM is checked to determine if one of the pending write operations in the PCI-to-memory queue is to the same address as the read operation of the processor. If so, the read operation is not executed until the PCI-to-memory queue is cleared of the write. To resolve the problem of aborting a Memory Read Multiple operation, an abort signal from the PCI bus interface is received and as soon thereafter as can be done the read ahead cycle is terminated, even though the read ahead cycle has not fully completed. The memory controller has improved prediction rules based on whether the cycle is coming from the processor or is coming from the PCI bus to allow more efficient precharging when PCI bus cycles are used. The memory controller is highly programmable for multiple speeds and types of processors and several speeds of memory devices. The memory controller includes a plurality of registers that specify number of clock periods for the particular portions of a conventional dynamic random access memory cycle which are used to control state machine operations.
摘要:
A memory controller which makes maximum use of any processor pipelining and runs a large number of cycles concurrently. The memory controller can utilize different speed memory devices at their desired optimal speeds. The functions are performed by a plurality of simple, interdependent state machines, each responsible for one small portion of the overall operation. As each state machine reaches has completed its function, it notifies a related state machine that it can now proceed and proceeds to wait for its next start or proceed indication. The next state machine operates in a similar fashion. The state machines responsible for the earlier portions of a cycle have started their tasks on the next cycle before the state machines responsible for the later portions of the cycle have completed their tasks. The memory controller is logically organized as three main blocks, a front end block, a memory block and a host block, each being responsible for interactions with its related bus and components and interacting with the various other blocks for handshaking. The memory controller operates in system management mode to override any write protect status of memory so that the SMRAM can be located in the main memory space and be write protected during normal operations but be full usable during system management mode.
摘要:
A computer system which provides for slowing the effective speed of a microprocessor. The microprocessor includes a disabling input that when deactivated disables operations of the microprocessor on the processor bus. A computer system according to the invention periodically deasserts this signal with the certain duty cycle, allowing the microprocessor to continue to perform necessary functions at an effective rate compatible with older microprocessors, but never requiring an actual clock frequency change. This periodic deassertion is performed in response to a memory refresh counter that periodically counts down to zero and is reloaded. By comparing an input/output register with the refresh counter, and by adjusting the input/output register, the deasserting signal to the processor is periodically deasserted with a selectable duty cycle.
摘要:
A system management mode address correction system for a computer provides correct address values on the address bus when the computer is in system management mode. Conventionally, bit 20 of the microprocessor's address outputs may be masked by asserting the FORCE A20 signal. The computer system also operates in a system management mode, which requires all of the address bits to be available for proper access to the system management interrupt vector. When the computer is in system management mode, the computer's microprocessor asserts a system management interrupt active (SMIACT*) signal. This signal is provided to a circuit which also receives the FORCE A20 signal. While the SMIACT signal is deactivated, the control circuit provides the true FORCE A20 signal to the computer system. When an SMI occurs, the SMIACT signal is activated and the FORCE A20 signal is disabled. As a result, the address generated by the microprocessor is asserted on the address bus.
摘要:
A memory controller which makes maximum use of any processor pipelining and runs a large number of cycles concurrently. The memory controller utilizes different speed memory devices at each memory devices optimal speed. The functions are performed by a plurality of simple, interdependent state machines, each responsible for one small portion of the overall operation. As each state machine reaches has completed its function, it notifies a related state machine that it can now proceed and proceeds to wait for its next start or proceed indication. The next state machine operates in a similar fashion. The state machines responsible for the earlier portions of a cycle have started their tasks on the next cycle before the state machines responsible for the later portions of the cycle have completed their tasks. The memory controller is logically organized as three main blocks, a front end block, a memory block and a host block, each being responsible for interactions with its related bus and components and interacting with the various other blocks for handshaking. The memory system includes a single chip which provides all of the address and control signals to a memory device so that a clock cycle can be saved because of reduced skew of the signals. The signals are provided synchronously from the chip.