Motorola Corporation

National Semiconductor Corporation

Via-cyrix, Inc.

Cyrix Corporation

Antone L Fourcroy

Mark W McDermott

Christopher E White

Bradley Gene Burgess

John P Dunn

Mark E Burchfield

James C Smallwood

Xiaoli Y Mendyke

Robert W French

A system and method of stretching setup and hold times for input signals into synchronous digital circuitry.

Stretching setup and hold times in synchronous designs

A pipelined x86 processor implements a method of detecting self-modifying code in which a prefetched block of instruction bytes may contain an instruction that is modified by a store instruction preceding it in the execution pipeline. The processor includes a Prefetch unit having a multi-block prefetch buffer, a Branch unit with a branch target cache (BTC), and a Load/Store (LDST) unit having store reservation stations. Self-modifying code is detected in three ways: (a) the Prefetch unit snoops store addresses from the LDST unit which are compared with (i) an address tag for each of the prefetch blocks of instruction bytes already loaded into the prefetch buffer, and (ii) the addresses of any pending prefetch requests, (b) the LDST unit snoops prefetch addresses issued by the Prefetch unit and compares them to store addresses queued in the store reservation stations, and (c) to ensure compatibility with the 486 specification for self-modifying code (which requires that a store that modifies an instruction be followed immediately by a jump to that instruction), the LDST unit detects when a store is followed by a COF that hits in the BTC which output a target address that is the same as the preceding store address. In particular, Prefetch unit snooping and LDST unit snooping detect instances of self-modifying code conditions that do not follow the 486 specification.

Detecting self-modifying code in a pipelined processor with branch

A processing system includes circuitry and methodology for selecting clock generation modes between phase-locked loop and static delay line loop circuitries. The node may be selectable through an externally accessible pin, an internal bond wire option, a boundary test scan control point, or other programmable register or control point.

Selectable clock generation mode

An NDIRTY cache line lookahead technique is used to expedite cache flush and export operations by providing a mechanism to avoid scanning at least some cache lines that do not contain dirty data (and therefore will not have to be exported). The exemplary cache organization uses one-line lookahead where each cache line but the last has associated with it an NDIRTY bit that indicates whether the next cache line contains dirty data. For cache flush and export operations, when a cache line (N) is read to check for dirty data that must be exported, the NDIRTY bit for that cache line is also tested to determine whether the next cache line (N+1) contains dirty data--if the NDIRTY bit is clear, indicating that the next cache line is clean, then that line is skipped and the scan proceeds with the line after that (N+2). This exemplary one-line lookahead implementation is readily extendible to N-line lookahead. The cache line lookahead technique reduces the number of cache line accesses required during flush/export operations, with the attendant advantages of reduced flush/export penalty cycles and power, thereby improving overall system performance.

NDIRTY cache line lookahead

A processing system includes a clock synchromesh that receives indicia of critical activity from various functional units within the processing system and responsive to the indicia, ratchets down/up the frequency of a clock output signal to at least one of the functional units to reduce power consumption. The determination of critical activity is preferably made according to a heuristic internal to a processor under software or hardware control.

Processor having a frequency modulated core clock based on the

A pipelined x86 processor includes a prefetch unit (prefetch buffer) and a branch unit that cooperate to detect when the target of a branch (designated a short branch) is already in the prefetch buffer, thereby avoiding issuing a prefetch request to retrieve the target. The branch unit includes a branch target cache (BTC) in which each entry stores, in addition to target address information for prefetching a prefetch block of instruction bytes containing a target instruction, a prefetch block location field--when this field is valid, it provides the location of the target instruction for a short branch within a prefetch block that is already in the prefetch buffer. In response to a branch that hits in the BTC, if the associated prefetch block location field is valid, the prefetch unit is able to begin transferring instruction bytes for the target instruction without issuing a prefetch request for the prefetch block containing the target instruction. The exemplary prefetch unit uses a three-block prefetch buffer each storing a 16 byte (cache line) prefetch block--the three prefetch buffers are logically allocated for the current, next, and previous prefetch blocks, and the target of a short branch may be either forward or backward of the branch, and may reside in the same prefetch buffer as the branch (which logically will be current) or in a contiguous prefetch buffer (logically next or previous). Avoiding prefetch requests in the case of short branches reduces contention for cache access and associated bus traffic.

Detecting short branches in a prefetch buffer using target location

A digital computing system having a low power mode of operation includes a mechanism for communicating, prior to entering the low power mode, information determinative of which events shall be capable of causing the termination of the low power mode. An integrated circuit microcomputer enters a low power mode in response to executing an LPSTOP instruction. Only reset events and those interrupt events having a priority level sufficiently high to pass an interrupt mask are capable of causing the termination of the low power mode. The LPSTOP instruction causes immediate data to be loaded into a status register, resetting the interrupt mask bits. The interrupt mask is then written, by means of a special bus cycle, into an interrupt mask register in a sub-system within the microcomputer. This subsystem then shuts down the clock signals to the remainder of the microcomputer, leaving only this sub-system active. The active sub-system performs a comparison of the priority levels of received interrupt requests to the interrupt mask during the low power mode. Only if the priority level of an interrupt is sufficiently high are the clock signals resumed, thus terminating the low power mode.

Digital computing system with low power mode and special bus cycle

An input/output circuit of an integrated circuit with a programmable input sensing time. The output driver of the input/output circuit is open drain and is designed for use in a wire-OR configuration with other devices. The input/output circuit is coupled to a bonding pad and through the bonding pad to a device pin, and counts a programmable number of clock cycles between a negation of an output drive signal and when the state of the pin is sampled as an input. Since different applications use a wide range of values for external pullup resistors, the input/output circuit allows adjustment of the sample time to fit a particular application.

Input/output circuit with programmable input sensing time

A microprocessor or other integrated circuit including a clock generator circuit which is dependent on an externally-provided reference signal includes the capability of detecting the loss of this externally-provided reference signal and producing an alternate clock signal despite the loss of the reference. In a particular embodiment, the clock generator comprises a phase locked loop frequency synthesizer which normally relies on an external crystal oscillator for its reference frequency signal. The generator includes a circuit for detecting abnormalities in the crystal signal and switching the voltage controlled oscillator of the frequency synthesizer to an internally-generated reference voltage. In the particular embodiment, the clock generator is also capable of producing a reset signal in response to the loss of the reference signal.

Growing community of inventors

Austin, TX, United States of America

Antone L Fourcroy