International Business Machines Corporation

Guenter Gerwig

Frank Siegfried Lehnert

Klaus Michael Kroener

Timothy John Slegel

Christian Jacobi

Chung-Lung K Shum

Scott Barnett Swaney

Tobias Webel

Juergen Haess

Tobias Ulrich Bergmann

Michael Cremer

Eric M Schwarz

Ronald Morton Smith, Sr

Erwin F Pfeffer

Ulrich Mayer

Michael Klaus Kroener

Willm Hinrichs

Holger Wetter

Eberhard Engler

Son Dao-Trong

Juergen Haas

Rolf Mueller

Barinjato Ramanandray

Peter Probst

Embodiments relate to implementing error data collection for a processor. Aspects of the embodiments include identifying a plurality of error state devices in a processor, each of the plurality of error state devices configured to hold a state indication, and organizing the plurality of error state devices as a sequence. Aspects also include collecting a plurality of state indications by serially sampling the state indication from each of the plurality of error state devices in an order corresponding to the sequence, sequentially storing the plurality of state indications as a single linear data array, and outputting the linear data array as a data structure. The data structure can include information regarding one or more error events based on one or more errors occurring in the processor.

Error checking using serial collection of error data

A mechanism is provided for completing of set of instructions while receiving interrupts. The mechanism executes a set of instructions. Responsive to receiving an interrupt and determining that the interrupt requires processing within an implementation time frame, the mechanism delays the interrupt for a predetermined time period. Responsive to completing the set of instructions within the predetermined time period, the mechanism processes the interrupt.

Delaying interrupts for a transactional-execution facility

Generating monotonically increasing time-of-day values in a multiprocessor system is provided. Synchronization impulses are received by a processor of the multiprocessor system, and an execution of a read instruction of a time-of-day value within a processor of the processors is refused, if the execution of the read instruction of the time-of-day value is requested after a predefined time after a synchronization impulse of the synchronization impulses, and if a trigger signal, indicative of new data received by a related memory system, has been received after the predefined time, wherein the memory system is external to the processor.

Generating monotonically increasing TOD values in a multiprocessor system

An indication of time that indicates at least one of the current day and the current time is received. It is determined that a raw interval pulse transmitted by a first oscillator should be adjusted based, at least partly, on the indication of time. In response to determining that the raw interval pulse should be adjusted, a steered time interval pulse is generated based, at least partly, on the raw time interval pulse and the indication of time. The steered time interval pulse is distributed to a plurality of hardware components.

Generation and distribution of steered time interval pulse to a plurality of hardware components of the computing system

Instructions within a processor are managed by receiving, at a recovery unit of the processor, an instruction that modifies a control register residing within the recovery unit. The recovery unit receives a first set of data associated with the instruction from a general register. A second set of data associated with the instruction is retrieved from the control register by the recovery unit. The recovery unit performs at least one binary logic operation on the first set of data and the second data.

Out of order millicode control operation

An improved latch arrangement for an electronic digital system is disclosed. The latch arrangement comprises a certain number of standard latches configured as configuration-switch latches which are modified only by shift operation and/or during Error Checking and Correction (ECC) action, and a corresponding number of standard latches configured as Error Checking and Correction (ECC) latches storing Error Checking and Correction (ECC) bit data used to check latch data of said configuration-switch latches.

Latch arrangement for an electronic digital system, method, data processing program, and computer program product for implementing a latch arrangement

A new method and apparatus have been taught for storing error information used for debugging as generated by the initial and subsequent error occurrences. In this invention, a register with several bit ranges is used to store error information. The first bit-range is allocated to the initial error information. If the total number of the errors exceeds the capacity of the register, the last error is kept in a last bit-range. This way, precious initial error information (as well as the last error information) will be available for debugging.

Debugging for multiple errors in a microprocessor environment

The invention proposes a Floating Point Unit with fused multiply-add, with one addend Method for calculating a result of a division with an A-register and a B-register for two multiplicand operands and a C-register for an addend operand, wherein a divide processor using a subtractive method for calculation with a divisor register and a partial remainder register and a multiplier associated to an subtractor uses the C-register as input, wherein while loading the fraction of the dividend through the divisor register into the partial remainder register of the divide processor a calculated shifting is applied for alignment by using the multiplier associated to the subtractor.

Method for calculating a result of a division with a floating point unit with fused multiply-add

Method to convert a hexadecimal floating point number (H) into a binary floating point number by using a Floating Point Unit (FPU) with fused multiply add with an A-register a B-register for two multiplicand operands and a C-register for an addend operand, wherein a leading zero counting unit (LZC) is associated to the addend C-register, wherein the difference of the leading zero result provided by the LZC and the input exponent (E) is calculated by a control unit and determines based on the Raw-Result-Exponent a force signal (F) with special conditions like 'Exponent Overflow', 'Exponent Underflow', and &#x2018;Zero Result&#x2019;.

Method and floating point unit to convert a hexadecimal floating point number to a binary floating point number

An apparatus and method for performing floating-point operations, particularly a fused multiply add operation. The apparatus includes a arithmetic logic unit adapted to produce both a high-order part (H) and a low-order part (L) of an intermediate extended result according to H, L=A*B+C, where A, B are input operands and C an addend. Each H, L part is formatted the same as the format of the input operands, and alignment of the resulting fractions is not affected by alignment of the inputs. The apparatus includes an architecture for suppressing left-alignment of the intermediate extended result, such that input operands for a subsequent A*B+C operation remain right-aligned.

Fused multiply add split for multiple precision arithmetic

Arithmetic processing circuits in a circuit in a floating point processor having a fused multiply/ADD circuitry. In order to avoid waiting cycles in the normalizer of the floating point arithmetic, control logic calculates in an extremely early state of the overall Multiply/Add processing. Parts of the intermediate add result are significant and have to be selected in the pre-normalizer multiplexer to be fed to the normalizer by counting the leading zero bits (LAB) of the addend in a dedicated circuit right at the beginning of the pipe. LAB is added to the shift amount (SA) that is calculated to align the addend and is then compared with the width of the incrementer. If the sum of (SA+LAB) is larger than the width of the incrementer, which is a constant value, then no significant bits are in the high-part of the intermediate result, and the pre-normalizer multiplexer selects the data from a second predetermined position, otherwise from a first predetermined position.

High-sticky calculation in pipelined fused multiply/add circuitry

A method and system for performing floating point additive arithmetic operations of long operands in a narrow dataflow. The operands include first and second floating point numbers having first and second mantissas, respectively, the second operand greater than the first operand. The mantissas are both separated into a low portion and a high portion, the high portions are loaded into N-bit operand registers. The high portion of the first mantissa is aligned with respect to the high portion of the second mantissa, the high portions are then moved into 2N-bit registers. The low portion of the first mantissa is aligned in accordance with the alignment of the first mantissa high portion. The low portions of both mantissas are then concatenated into the registers, the first mantissa concatenated using a hold-function circuit. A 2N-bit-wide adder performs the additive arithmetic operation on the concatenated mantissas.

Advanced execution of extended floating-point add operations in a narrow dataflow

A process is provided for monitoring the conversion of numerical values from a first to a second format, where before and after the conversion, the modulo residue of the corresponding numerical value is calculated and compared with the corresponding residue after the conversion. In this way it is possible to effect error-free monitoring of such a conversion, especially of computer data, without great hardware expenditure.

Examination of residues of data-conversions

A correction of microprocessor chip design errors is achieved by identifying selected sets of instructions and/or selected sets of instruction sequences in an erroneous control flow of the microprocessor and/or by identifying selected sets of interface control and status signals. A match selectively initiates a corrective action by interfering with the instruction flow in the microprocessor chip or by requesting external control from an associated processor unit. Alternatively, a match is used for a programmable modification of interface control and status signals to adapt the chip to changes of its environment without redesign.

Growing community of inventors

Simmozheim, Germany

Guenter Gerwig