Abstract:
A method and apparatus for configuring a microcontroller. An XML description of the microcontroller's hardware resources may be accessed. A user may select from available hardware resources and pre-defined user modules to select a configuration. Configuration information, which may include register bit patterns and microprocessor instructions, may be automatically generated. Additionally, application programming interface calls and structure, as well as interrupt vector tables may be automatically generated. Embodiments of the present invention provide improved ease of use and the ability to manage greater complexity in the configuration of configurable microcontrollers.
Abstract:
A device for real-time correction of set-point signals intended to receive at the input set-point signals and to deliver at its output set-point signals that are modified to compensate for defects, negative effects or the like subsequently encountered during the processing and/or the application of the set-point signals. This device (1) includes at least one circuit (1′) that is based on a microprogrammed structure and composed of several subassemblies (3, 4, 5, 6, 6′) that work with digital components essentially including a micro-sequencer (3) forming a counter, a memory (4) for storing micro-instructions, and a processing unit (5) combined with at least one working memory (6, 6′) and integrating arithmetic calculation modules (7, 7′), whereby the processing unit (5) modifies the data of set-point signals in accordance with the micro-instructions that are addressed by the micro-sequencer (3) and by taking into account the correction coefficients that are provided.
Abstract:
Embodiments of apparatuses, methods, and systems for modifying the behavior of a guest installed to run within a VM are disclosed. In one embodiment, an apparatus includes virtualization logic, first storage, second storage, decode logic, and multiplexing logic. The virtualization logic is to provide a mode in which to operate a virtual machine. The first storage is to store a first plurality of micro-instructions to control the apparatus. The second storage is to store a second plurality of micro-instructions to control the apparatus. The decode logic is to decode a macro-instruction into one of a first plurality and a second plurality of micro-instructions. The multiplexing logic is to cause the macro-instruction to be decoded into the second plurality of micro-instructions instead of the first plurality of micro-instructions only when issued from the virtual machine.
Abstract:
A microprogram loaded in a microprogram memory by a host CPU is read out by a microprogram read control circuit in a vacant period existing in each cycle of operation of a hardware unit, whereby the microprogram is transmitted to each circuit of the hardware unit through a microprogram read-only bus. Each circuit of the hardware unit comprises a decoder and operation thereof is controlled based on a microinstruction decoded by the decoder.
Abstract:
An asynchronous micro-machine/interface responsive to a central processing unit (CPU) in which the CPU and the micro-machine/interface are run on clocks which are asynchronous from one another is provided. The inventive asynchronous micro-machine/interface has data path elements for receiving an incoming instruction and for performing actions requested by the incoming instruction, as well as a means for synchronizing the incoming instruction to the clock of the micro-machine/interface and for performing actions within the data path elements prior to the execution of the incoming instruction and during transfer of control, by the micro-machine/interface, to the routine that is associated with the incoming instruction.
Abstract:
A highly parallel microprocessor using a logic gating structure and a microinstruction organization which permits direct access by each of the microprocessor components to a tri-bus system. Operation is defined by a single phase clock, during which all portions of a microinstruction are executed. The system further permits overlap operation for microprocessor instructions, thereby allowing for the fetching of a next instruction while executing a current instruction. The use of general purpose, non-dedicated registers is contemplated, thereby to avoid the need for multi-phase clocking.
Abstract:
The invention relates to a microprogrammed computer whose architecture is determined by a simple and rigid format of the controlling micro-instruction. Each micro-instruction controls a data transfer and has at least four parts each of which are within a single micro-instruction. A first part always specifies the data source and data sink between which the data transfer is to take place. In a second part, conditions are stated for transfer to, or writing into the data sink. In a third part, a specific counting register out of a number of counting registers is addressed and, furthermore, it is specified how the contents of the counting register must be modified parallel to the transfer. In a fourth part, further data is contained; the fourth part may be controlled as a data source, with data transfer from the source to an arbitrary data sink being possible. In the case of transfer to the operation register of the arithmetic unit, such data indicates the logic or arithmetic operation. In the case of transfer to, for example, a counting register, the data may be employed to derive an initial address therefrom. The operation code register may advantageously have a capacity which corresponds to the length of two storage words.
Abstract:
A conventional decoder of the type that converts an m-out-of-n representation to activation of only a single one of plural output lines is modified to have powerful self-checking capabilities. The modified decoder is adapted to be included in a control system in which the words stored in a microprogram memory are coded in an m-out-of-n format. Such a system automatically detects the occurrence of any no-output or multiple-output readout from the memory.
Abstract:
Inter-microservice communications are managed through in-memory connection routing. A sending microservice writes a message over a port associated with the connection. The message is routed directly to one or more receiving microservices associated with the connection over their ports associated with the connection. The message may be converted to a different format or multiple different formats through plugins processed when the message is received over the sending microservice's port and before the converting messages are routed over the receiving microservices' ports. The inter-microservice communications are hardware and platform independent or agnostic, such that the microservices associated with the connection can be processed on different hardware and different platforms from one another.
Abstract:
In one embodiment, an apparatus includes: an instruction fetch circuit to fetch instructions; a decode circuit coupled to the instruction fetch circuit to decode the fetched instructions into micro-operations (pops); a scheduler coupled to the decode circuit to schedule the pops for execution; and an execution circuit coupled to the scheduler, the execution circuit comprising a plurality of execution ports to execute the pops. The scheduler may be configured to: schedule at least some pops of a first type for redundant execution on symmetric execution ports of the plurality of execution ports; and schedule pops of a second type for non-redundant execution on a single execution port of the plurality of execution ports. Other embodiments are described and claimed.