Abstract: A voltage-controlled delay element utilizes a current-starved inverter configuration with a feedback path that ensures a rapid discharge of the storage node to ground once the desired delay time has elapsed. The circuit comprises a circuit path for charging the storage node (preferably rapidly), a first pull-down path capable of discharging the storage node at a rate determined by the control voltage, a second pull-down path capable of rapidly discharging the storage node, an output inverter, and a feedback connection between the output terminal of the output inverter and the second pull-down path to rapidly discharge the storage node when the output voltage starts rising. The circuit further comprises a means for programmably adjusting the delay in response to logic signals.

Abstract: A first two-input adder computes the sum of one wider and one narrower input by combining a conventional adder for the low-order bits with an incrementer and selection logic for the high-order bits. A second three-input adder computes the sum of one wider and two narrower inputs in a similar way: the low-order bits are computed with a conventional carry save adder (CSA) followed by a carry propagate adder (CPA), while the high-order bits are computed with an incrementer and selection logic. The first and second circuits are combined to form a third arithmetic circuit that takes four input operands, the first of which is wider than the other three, and speculatively computes two results: (1) the sum of the first and second input operands; and (2) the sum of the first, third, and fourth input operands. This combined circuit contains all of the elements of the first two circuits, but shares a single incrementer. A degenerate case of the third circuit occurs when the second and third inputs are common.

Abstract: AN improved branch prediction cache (BPC) scheme that utilizes a hybrid cache structure. The BPC provides two levels of branch information caching. The fully associative first level BPC is a shallow but wide structure (36 32-byte entries), which caches full prediction information for a limited number of branch instructions. The second direct mapped level BPC is a deep but narrow structure (256 2-byte entries), which caches only partial prediction information, but does so for a much larger number of branch instructions. As each branch instruction is fetched and decoded, its address is used to perform parallel look-ups in the two branch prediction caches.

Abstract: A method and apparatus for eliminating the delay in a parallel processing pipeline. In a parallel processing pipeline system, a circuitry is provided to determine the length and align two instructions in parallel. Parallel decoding circuitry is provided for decoding and executing the two instructions. A branch prediction cache stores the target instruction and next sequential instruction, and is tagged by the address of the branch instruction, as in the prior art. In addition, however, the branch prediction cache also stores the length of the first and second instructions and the address of the second instruction. This additional data allows the target and next sequential instructions to be directly aligned and presented to the parallel decoding circuits without waiting for a calculation of their lengths and starting addresses.

Abstract: The present invention provides for the updating of both the instructions in a branch prediction cache and instructions recently provided to an instruction pipeline from the cache when an instruction being executed attempts to change such instructions ("Store-Into-Instruction-Stream"). The branch prediction cache (BPC) includes a tag identifying the address of instructions causing a branch, a record of the target address which was branched to on the last occurrence of each branch instruction, and a copy of the first several instructions beginning at this target address. A separate instruction cache is provided for normal execution of instructions, and all of the instructions written into the branch prediction cache from the system bus must also be stored in the instruction cache. The instruction cache monitors the system bus for attempts to write to the address of an instruction contained in the instruction cache.

Abstract: In a first aspect of the invention, branch prediction hardware, comprising logic and interconnect, is configurable via a control line to alter the manner in which the branch prediction is generated. The configuration can be done programmatically in software. Or, the configuration can be done by hardware in response to processor events. Such processor events include the loading of the CS register and changes in the instruction workload. In a second aspect of the invention, related to speculative execution, the directions of a plurality of branches are predicted based partly on resolved branch history information. Tentative branch history information is then stored for each of the predicted branches. When a predicted branch is resolved, the resolved branch history information is updated based on the stored tentative branch history information for the branch most recently resolved. Additionally, the predictions may be partly based on preceding unresolved branch predictions if any are outstanding.

Abstract: A pipeline control system is distributed over the functional units (15, 17, 20, 25) in a processor (10). Decoder logic (12) issues operations, each with an associated tag, to the functional units, with up to n operations allowed to be outstanding. The units execute the operations and report termination information back to the decoder logic, but do not irrevocably change the state of the machine. Based on the termination information, the decoder logic retires normally terminated operations in order. If an operation terminates abnormally, the decoder logic instructs the units to back out of those operations that include and are later than the operation that terminated abnormally.

Abstract: A first two-input adder computes the sum of one wider and one narrower input by combining a conventional adder for the low-order bits with an incrementer and selection logic for the high-order bits. A second three-input adder computes the sum of one wider and two narrower inputs in a similar way: the low-order bits are computed with a conventional carry save adder (CSA) followed by a carry propagate adder (CPA), while the high-order bits are computed with an incrementer and selection logic. The first and second circuits are combined to form a third arithmetic circuit that takes four input operands, the first of which is wider than the other three, and speculatively computes two results: (1) the sum of the first and second input operands; and (2) the sum of the first, third, and fourth input operands. This combined circuit contains all of the elements of the first two circuits, but shares a single incrementer. A degenerate case of the third circuit occurs when the second and third inputs are common.

Abstract: A dual-bus architecture that includes a high-speed system bus, called the NexBus (20), and a slower peripheral bus, called the alternate bus or AB (25). The NexBus and AB are coupled by control logic (45) which includes an arbiter (50) and an alternate bus interface (ABI) (60). The ABI is treated as a master for both the NexBus and the AB. While it would be possible to have the adapter always request the AB (which also requires the NexBus), that would slow down NexBus operations to the bandwidth of the AB. This problem is avoided by providing two request lines for each adapter, -NREQ (NexBus only) and -AREQ (both buses), having the adapter normally assert -NREQ first. However, if the addressed device is on the AB, the ABI automatically detects this fact and attempts to do a crossing transfer to the AB, even though the request was for the NexBus only.

Abstract: A three-input comparator, where one of the inputs is an implicit constant, is formed with a special carry-save adder (CSA) followed by carry propagation circuitry. The special CSA uses two different bit cells depending upon whether that bit position in the constant input is a one or a zero. The three-input comparator can be modified to be a three-input adder by using a full carry-propagate adder (CPA). By taking into account a priori restrictions on the possible input operands, these arithmetic circuits are smaller and more efficient than conventional adders and comparators, which must be designed to deal with all possible input operands.