Abstract: A floating point unit has a control unit, a data input register and a write stage register from which an instruction is transferred from the floating point unit to a storage unit. The floating point unit typically has multiple pipeline stages for arithmetic computation, a normalization stage, and a rounding stage, each of which pipeline stages may during processing of a stream of instructions contain a separate instruction. The stages are connected in an ordered manner such that the processing of instructions occurs in the pipeline. An active instruction is a "stalled" instruction within a pipeline when forward progress is not permitted to advance to a new stage in the pipeline because data needed is not available for a prior instruction creating a data dependency.

Abstract: IEEE compliant floating point unit mechanism allows variability in the execution of floating point operations according to the IEEE 754 standard and allowing variability of the standard to co-exist in hardware or in the combination of hardware and millicode. The FPU has a detector of special conditions which dynamically detects an event that the hardware execution of an IEEE compliant Binary Floating Point instruction will require millicode emulation. The complete set of events which millicode may emulate are predetermined early in the design process of the hardware. An exception handling unit assist millicode emulation by trapping the result of an exceptional condition without invoking the trap handler. When an exceptional condition is detected during execution, the IEEE 754 standard requires two different actions under control of a mask bit. If the mask bit is on, the result is written into an FPR and the trap handler is invoked.

Abstract: There is a unique partitioning problem in determining how to execute the floating point multiply instruction defined by IEEE 754 standard for the quad word format on a S/390 processor. Several manufacturers including IBM and HP define the binary quad word format to have a 113 bit significand. IBM S/390 hexadecimal long floating point format has a 56 bit significand and most S/390 floating point units only contain a long format multiplier. Quad word format multiplication must be executed as a series of several long precision multiplications and extended precision or long precision additions. The S/390 hexadecimal quad word format is easier to implement than binary format since it has a 112 bit significand and can easily be partitioned into two 56 bit parts. But a 113 bit significand would just exceed two partitions and require a third.

Abstract: Provides a program translation and execution method which stores target routines (for execution by a target processor) corresponding to incompatible instructions, interruptions and authorizations of an incompatible program written for execution on another computer system built to a computer architecture incompatible with the architecture of the target processor's computer system. The disclosed process allows the target processor to emulate incompatible acts expected in the operation of an incompatible program when the target processor itself is incapable of performing the emulated acts. Each of the instructions, interruptions and authorizations found in the incompatible programs has one or more corresponding target routines, any of which may need to be preprocessed before it can precisely emulate the execution results required by the incompatible architecture.

Abstract: A method and system for processing instructions in a floating point unit for executing denormalized numbers in floating point pipeline via serializing uses an instruction unit and having a control unit and a pipelined data flow unit, a shifter and a rounding unit. The floating point unit has an external feedback path for providing intermediate result data from said rounding unit to an input of the pipelined data flow unit to reuse the pipeline for denormalization by passing intermediate results in the pipeline which have a denormalized condition computed after the exponent calculation of the shifting circuit directly from the rounding unit to the top of the dataflow in the pipeline via an external feedback path. The pipelined has two paths which are selected based on the presence of other instructions in the pipeline. If no other instructions are in the pipeline a first path is taken which uses the external feedback path from the rounding unit back into the top of the dataflow.

Abstract: A method and system which provides exactly rounded division and square root results for a designated rounding mode independently of a remainder, or equivalent calculation of the relationship between the remainder and zero, for predetermined combinations of the rounding mode and the guard digit of an estimate that has several more bits of precision than the exactly rounded result, and has an error tolerance magnitude less than the weight of the least significant bit of the estimate. The estimate is generated in accordance with a quadratically converging division or square root algorithm. The method and system is described in connection with IEEE 754-1985 and IBM S/390 binary floating point architectures.

Abstract: A system implementing a methodology for determining the exponent in parallel with determining the fractional shift during normalization according to partitioning the exponent into partial exponent groups according to the fractional shift data flow, determining all possible partial exponent values for each partial exponent group according to the fractional data flow, and providing the exponent by selectively combining possible partial exponents from each partial exponent group according to the fractional data flow. There is also provided a system implementing a methodology for generating the sticky bit during normalization. Sticky bit information is precalculated and multiplexed according to the fractional dataflow. In an embodiment of the invention, group sticky signals are calculated in tree form, each group sticky having a number of possible sticky bits corresponding to the shift increment amount of the multiplexing.

Abstract: A system implementing a methodology for determining the exponent in parallel with determining the fractional shift during normalization according to partitioning the exponent into partial exponent groups according to the fractional shift data flow, determining all possible partial exponent values for each partial exponent group according to the fractional data flow, and providing the exponent by selectively combining possible partial exponents from each partial exponent group according to the fractional data flow. There is also provided a system implementing a methodology for generating the sticky bit during normalization. Sticky bit information is precalculated and multiplexed according to the fractional dataflow. In an embodiment of the invention, group sticky signals are calculated in tree form, each group sticky having a number of possible sticky bits corresponding to the shift increment amount of the multiplexing.

Abstract: A system implementing a methodology for determining the exponent in parallel with determining the fractional shift during normalization according to partitioning the exponent into partial exponent groups according to the fractional shift data flow, determining all possible partial exponent values for each partial exponent group according to the fractional data flow, and providing the exponent by selectively combining possible partial exponents from each partial exponent group according to the fractional data flow. There is also provided a system implementing a methodology for generating the sticky bit during normalization. Sticky bit information is precalculated and multiplexed according to the fractional dataflow. In an embodiment of the invention, group sticky signals are calculated in tree form, each group sticky having a number of possible sticky bits corresponding to the shift increment amount of the multiplexing.

Abstract: A method and system which provides exactly rounded division and square root results for a designated rounding mode independently of a remainder, or equivalent calculation of the relationship between the remainder and zero, for predetermined combinations of the rounding mode and the least significant bit of an estimate that has one more bit of precision than the exactly rounded result, and has an error tolerance magnitude less than the weight of the least significant bit of the estimate. The estimate is generated in accordance with a quadratically converging division or square root algorithm. The method and system is described in connection with IEEE 754-1985 and IBM S/390 binary floating point architectures.

Abstract: A method and system which provides exactly rounded division and square root results for a designated rounding mode independently of a remainder, or equivalent calculation of the relationship between the remainder and zero, for predetermined combinations of the rounding mode and the least significant bit of an estimate that has one more bit of precision than the exactly rounded result, and has an error tolerance magnitude less than the weight of the least significant bit of the estimate. The estimate is generated in accordance with a quadratically converging division or square root algorithm. The method and system is described in connection with IEEE 754-1985 and IBM S/390 binary floating point architectures.

Abstract: A computer system supporting multiple floating point architectures. In an embodiment of the invention, a floating point unit (FPU) is optimized for hex format. The FPU uses a hex internal dataflow with a with an exponent and bias sufficient to support a binary floating point architecture. The FPU includes format conversion means, rounding means, sticky bit calculation means, and special number control means to execute binary floating point operations according to the IEEE 754 standard. An embodiment of the invention provides a system for executing floating point operations in either IBM S/390 hexadecimal format or IEEE 754 binary format.

Abstract: An adder which takes advantage of the early arriving bits of a time skewed operand to provide a result to an add or substract operation without additional latency. Possible partial results are calculated and then selectively combined according to the late arriving data as the late arriving data becomes available. In an embodiment of the present invention, a first operand is partitioned into groups according to the arrival time of the skewed data, and possible partial results for each group are calculated for the full range of partial inputs that affect it. In addition, the high order groups are calculated with and without a borrow (carry) which is propagated from a low order group. Once the delayed partial operands are known and the borrows (carrys) determined the partial results are gated through multiplexers according to the borrows and partial results, and thus the result is provided with a delay similar to the delay in arrival of the skewed operand.