76 results about "Interrupt latency" patented technology

In one embodiment, a processor comprises one or more registers and a control unit. The registers are configured to store interrupt state describing a virtual interrupt. The control unit is configured to initiate the virtual interrupt responsive to the interrupt state. In another embodiment, a method comprises storing an interrupt state describing a virtual interrupt in a storage area allocated to a guest. A processor initiates the virtual interrupt subsequent to initiating execution of the guest, responsive to the interrupt state. In still another embodiment, a computer accessible medium stores a plurality of instructions comprising instructions which, when executed on a processor in response to a physical interrupt: determine a guest into which a virtual interrupt corresponding to the physical interrupt is to be injected; and store an interrupt state describing the virtual interrupt in a storage area allocated to the guest.

A disk drive is disclosed having a disk, a head actuated over the disk, a buffer memory for storing control routine op codes and control routine data, and a microprocessor for receiving the control routine op codes and control routine data. Control circuitry within the disk drive services an access request generated by the microprocessor by accessing the buffer memory, and monitors at least one interrupt. If the interrupt occurs while servicing the access request, the control circuitry enables the microprocessor to execute an interrupt service routine corresponding to the interrupt. Enabling the microprocessor to execute the interrupt service routine rather than wait for the access request reduces the latency in servicing the interrupt.

A method and system for controlling interrupt frequency by estimating processor load in the peripheral adapter provides adaptive interrupt latency to improve performance in a processing system. A mathematical function of the depth of one or more queues of the adapter is compared to its historical value in order to provide an estimate of processor load. The estimated processor load is then used to set a parameter that controls the frequency of an interrupt generator, which may be controlled by setting an interrupt queue depth threshold, packet frequency threshold or interrupt hold-off time value. The mathematical function may be the ratio of the transmit queue depth to the receive queue depth and the historical value may be predetermined, user-settable, obtained during a calibration interval or obtained by taking a long-term average of the mathematical function of the queue depths.

In an embodiment, a system comprises a memory system configured to store a data structure. The data structure stores at least an interrupt request state for each destination in each of a plurality of guests executable on the system. The interrupt request state identifies which interrupts have been requested at the corresponding interrupt controller in the corresponding guest of the plurality of guests. A guest interrupt manager is coupled to receive an interrupt message targeted at a first destination in a first guest of the plurality of guests, and the guest interrupt manager is configured to update the interrupt request state in the data structure that corresponds to the first destination and the first guest.

A method and system for controlling interrupt frequency by transferring processor load information to a peripheral adapter provides adaptive interrupt latency to improve performance in a processing system. A device driver obtains current processor load information from an operating system or directly from processor usage counters. The estimated processor load is then used to set a parameter in the adapter that controls the frequency of an interrupt generator, which may be controlled by setting an interrupt queue depth threshold, packet frequency threshold or interrupt hold-off time value. The result is that the relative frequency of interrupts is managed in conformity with the current processor load, provide reduced processing latency when the system is relatively idle, which avoids loading the processor with additional interrupt processing overhead when the processor is busy.

The invention discloses an interrupt processing method of a multi-PCIE (Peripheral Component Interface Express) equipment system, relating to the field of automatic control. In the invention, an interrupt register and an interrupt state register are arranged in advance, and the value of the interrupt state register changes along with the variation of a value of the interrupt register based on a preset rule; a notification message is sent when PCIE equipment generates interrupt, and an interrupt vector number and an interrupt register address which are configurated to the PCIE equipment in advance are carried in the message; after receiving the notification message, a main control end writes the interrupt vector number into the interrupt register corresponding to a mapping address; and an interrupt processing program which is registered in advance is called and executed based on the value of the interrupt state register, and the value of the interrupt state register is revised to restore an interrupt state of the PCIE equipment corresponding to the interrupt processing program into a non-interrupt state. The invention can process interrupt simultaneously generated by a plurality of PCIE equipments, and PCIE equipment which does not support an MSI (Medium-scale Integration) interrupt mode can apply the invention, therefore, the interrupt processing efficiency of the system is improved.

The invention relates to a universal data acquisition unit and a data acquisition method thereof, belonging to the technical field of data acquisition; the universal data acquisition unit comprises amicroprocessor; a multi-path discrete signal input interface is connected with the input end of a counter by a photoelectric isolating circuit, and a multi-path continuous signal input interface is connected with the input end of an A/D converter, and a microprocessor is connected with the input ends of the A/D converter and the counter. The data acquisition method comprises that: system initialization is carried out; continuous and discrete signal interrupt vector is set; interrupt latency is carried out, if the discrete signal is interrupted, the discrete signal is acquired; if the continuous signal is interrupted, the continuous signal is acquired; if a watchdog is interrupted, a system is restarted; if hibernation is interrupted, system setting data is stored, and then a hibernation state is carried out; and then, interrupt latency awaken operation is carried out, if awakening is interrupted, the system is started, and the setting is restored before interruption; if the continuoussignal and the discrete signal are acquired, interrupt latency is continuously carried out after data is transmitted to an upper computer; if not, interrupt latency is carried out directly.

The invention relates to a data processing system which comprises a memory module and a microprocessor. The memory modules comprise at least one low-speed memory and one high-speed memory; both store an interrupt vector table individually for recording the entry instruction of interrupt service routines. The microprocessor comprises a central processing unit (CPU) and a memory controller with a re-addressing device. Once an interruption occurs, the CPU generates and sends an interrupt vector address to the memory controller. If the vector is located in the range of interrupt vector table, the re-addressing device sends an enable signal to the high-speed memory to enable the CPU to fetch the entry instruction of interrupt service routines from the high-speed memory, not from the pre-determined low-speed memory. Hence, the interrupt latency is reduced.

The present invention provides a data processing apparatus and method for saving return state. The data processing apparatus comprises a processing unit for executing data processing instructions, the processing unit having a plurality of modes of operation, with each mode of operation having a corresponding stack for storing data associated with that mode. The processing unit is responsive to a return state data processing instruction to write return state data of the processing unit from its current mode of operation to a stack corresponding to a different mode of operation to the current mode of operation. This approach significantly reduces code size and improves interrupt latency over known prior art techniques.

A technique for operating a processor includes detecting an interrupt having a first core of the processor as a destination core. The technique includes handling the interrupt by a second core of the processor in response to the first core being in a low-power state. The first core may be capable of executing a greater number of instructions-per-cycle than the second core and the second core may consume less power than the first core. The first core may be coupled to a first voltage plane and the second core may be coupled to a second voltage plane having lower power than the first voltage plane.

A data processing apparatus is described which comprises processing circuitry responsive to data processing instructions to execute integer data processing operations and floating point data processing operations, a first set of integer registers useable by the processing circuitry in executing the integer data processing operations, and a second set of floating point registers useable by the processing circuitry in executing the floating point data processing operations. The processing circuitry is responsive to an interrupt request to perform one of an integer state preservation function in which at least a subset of only the integer registers are copied to a stack memory, and a floating point state preservation function in which at least a subset of both the integer registers and the floating point registers are copied to the stack memory, the one of said integer state preservation function and the floating point state preservation function being selected by the processing circuitry in dependence on state information. In this way, it is possible to reduce the memory size requirement through reduced stack sizes, and to reduce the number of memory accesses required compared with the basic solution of always preserving floating point registers. As a result, power usage and interrupt latency can be reduced.

A portable computing device includes a modem and an application processor communicatively coupled by a data bus. The modem communicates a target data throughput in accordance with an identified data call. In response, the application processor determines whether an adjustment of an interrupt latency threshold is warranted to support the target data throughput identified by the modem. Otherwise, the application processor executes no such adjustment. In addition, the modem requests a desired performance of an application processor resource. In response, the application processor adjusts a control input of the application processor controlled resource. A change in a present data transfer session triggers the modem to communicate a revised target data throughput and/or a revised request for a desired performance of an application processor resource.

A system and method are described for warming a processor from a low power state in anticipation of a time critical interrupt. For example, one embodiment of a method comprises: detecting that a time-critical interrupt will require processor resources at some point in the future; estimating a time at which the time-critical interrupt will be triggered; scheduling a timer interrupt to fire at a specified time prior to the estimated time that the time-critical interrupt will be triggered, the timer interrupt being scheduled with sufficient time to ensure that the processor is warmed to a level at which it is capable of handling the time-critical interrupt at the time that the time-critical interrupt is triggered; and responsively triggering the timer interrupt at the specified time prior to the time critical interrupt.

Disclosed are a movement control method for a terminal in an RRC idle state, a terminal and a base station. The method is applied to a terminal side and comprises: a terminal sends a neighbor cell signal measurement result to a base station to which a camp-on cell belongs, so that the base station to which the camp-on cell belongs selects a neighbor cell as a target cell according to the neighbor cell signal measurement result, and informs the target cell of configuration of group information by means of a core network; the terminal receives a switching indication carrying a target cell identifier from the base station to which the camp-on cell belongs, and switches to the target cell according to the switching indication; and the terminal receives cluster group call signaling and group call radio resource configuration signaling that are sent by a base station to which the target cell belongs, and configures a group call service according to the cluster group call signaling and the group call radio resource configuration signaling. The technical solution of the present invention can reduce the interrupt latency of a group call service during movement of a terminal.

A computer implemented method, data processing system, and processor are provided for thermal throttle control with minimal impact to interrupt latency. A setting of an interrupt status bit is monitored. A determination is made as to whether an interrupt associated with the interrupt status bit is an unmasked interrupt in response to the interrupt status bit being set. An existing throttling mode is disabled and the interrupt handled in response to the interrupt being unmasked, where the interrupt latency of the integrated circuit is reduced.

Systems, methods, and software for reducing system management interrupt (SMI) latency while operating in system management mode. The present invention implements a technique for exiting system management mode while waiting for polled hardware events, handling any pending lower-priority interrupts and then resuming polling. The present invention does this by multi-threading SMI source handlers, using an idle thread, and using protocols for software-generated system management interrupts that insure that lower priority interrupts are serviced.

A personal computer's (PC) microprocessor is used to provide both the physical layer (PHY) and media access control (MAC) processing functions required to implement a wireless local area network (WLAN) adapter. This technique uses the polling mechanism associated with the power save (PS) functionality of WLAN protocol to relieve networking stress on the host processing system. It does this while maintaining networking integrity and packet delivery. The WLAN protocol polling mechanism is used to briefly inhibit the transfer of packets from the WLAN access point (AP) during peak periods of network traffic and/or host processor loading. Because the modulation, demodulation, and MAC functions, typically implemented in dedicated hardware on existing adapters are implemented in software running on the host PC microprocessor, other host system processes and applications can interfere with these time critical functions. Conversely, latency introduced by WLAN specific processing tasks during peak periods of network traffic may cause unacceptable delays to the other processes and applications requiring microprocessor attention. In addition to its primary stated purpose of allowing WLAN mobile stations to save power, this technique will use power save polling as a method for controlling delivery of network packets when the host is heavily loaded or when peak interrupt latencies make reliable packet delivery difficult or impossible.

The invention relates to the field of communication and in particular to an interrupt signal processing system. According to the system disclosed by the invention, an interrupt processing module function is increased on CPLD between a master control board and an interface board, and the operation of interrupt state and control register between the master control board and the interface board is realized by utilizing two serial signal lines. All interrupt states of the interface board are reported to the CPLD of the master control board in a serial bus manner, the master control board processor only needs to execute once and perform bus access when interrupt happens, and the interrupt state is acquired from the CPLD of the master control board. Therefore, instruction per cycle allowing the processor to access peripherals is greatly reduced, load of the processor is reduced, and the interrupt response time is accelerated. The interrupt needs to be transferred between the interface board and the master control board by two signal lines only, and signal pins are greatly saved.

An electronic device with a CPU configured to be switched from a low power mode into a higher power mode in response to an interrupt and an interrupt relay coupled between an interrupt generator and the CPU. A functional stage is coupled to the interrupt relay and functionally linked with the interrupt so as to be used during the second mode of the CPU. The interrupt relay relays the received interrupt to the CPU only after a time needed for the functional stage to settle. Power is saved because the CPU is not powered while waiting for the functional stage to settle.