Marvell Israel (misl) Ltd.

Amazon Technologies, Inc.

Mellanox Technologies, Ltd.

Other

Annapuma Labs Ltd.

Mellandx Technologies Ltd.

Gil Stoler

Tarek Rohana

Adi Habusha

Nafea Bshara

Said Bshara

Yaniv Shapira

Diego Crupnicoff

Michael Kagan

Ido Bourstein

Eitan Joshua

Shaul Chapman

Yiftach Banai

Ron Diamant

Erez Izenberg

Gil Bloch

Dror Goldenberg

Ariel Shachar

Roi Rahamim

An error-handling processing circuit and system are provided. The system can receive an error signal, such as an interrupt, and decouple (e.g., by a gate signal) a functional clock from a processing block, in some instances effectively halting the processing block's operation. This can prevent a cascade of interdependent errors, thereby avoiding producing redundant or confusing error information. The system can include the processing block, a debug clock not coupled to the processing block, and a data block (e.g., a register file) coupled to the debug clock and to an external input/output interface. The data block can be configured to continue receiving a clock signal via a multiplexer from the debug clock without disruption after the functional clock is decoupled, enabling the data block to remain operational for debugging.

Debug mechanisms for a processor circuit

An integrated circuit can include a functional unit and a local debug unit. The local debug unit can include a trace buffer, and the local debug unit is configured to track and store operation information of the functional unit in the trace buffer. The integrated circuit can also include a global debug unit coupled to the local debug unit. The integrated circuit is configured to send a debug reset command to reset the functional unit, without sending the debug reset command to the local debug unit, thereby retaining information stored in the trace buffer. The integrated circuit is also configured to send a power-up reset command to reset the local debug unit and the functional unit, thereby causing the local debug unit to clear the trace buffer.

Multiple reset types in a system

A method for writing data, the method may include: receiving or generating, by an interfacing module, a data unit coherent write request for performing a coherent write operation of a data unit to a first address; receiving, by the interfacing module and from a circuit that comprises a cache and a cache controller, a cache coherency indicator that indicates that a most updated version of the content stored at the first address is stored in the cache; and instructing, by the interfacing module, the cache controller to invalidate a cache line of the cache that stored the most updated version of the first address without sending the most updated version of the content stored at the first address from the cache to a memory module that differs from the cache if a length of the data unit equals a length of the cache line.

System and method for managing transactions

A system-on-a-chip (SoC) includes a master module and a first adapter module. The master module includes an upstream interface and a downstream interface. The upstream interface is coupled to a host unit for receiving a write burst or a read burst therefrom. The master module is configured to convert the write burst or the read burst into a series of access requests to the downstream interface. The first adapter module includes an input interface, an output interface, and an endpoint interface, and an address Base Address Register (BAR). The input interface is coupled to the downstream interface of the master module. The output interface is coupled to a second adapter module or to a termination module. The endpoint interface is coupled to a first functional unit or to a third adapter module. The first adapter module is configured to detect a respective access request corresponding to the address BAR.

Modular system on chip configuration system

Methods and apparatuses for offloading functionality in an integrated circuit are presented. Certain embodiments are described that disclose methods pertaining to implementation of a universal offload engine that can service several functional blocks, each configured to perform a different function. The offload engine can be iteratively implemented with a common interface to functional blocks. Work descriptors can be used between DMA engines and corresponding functional blocks to instruct the DMA engines how to transport data between memory locations and/or to reformat the data.

Universal offloading engine

A system includes a host debugger to carry out a debugging flow on a computing device and a debug controller to couple the host debugger to the computing device. The debug controller receives a bit stream from the host debugger, converts the incoming bit stream into a command according to a protocol, determines whether the command is a first-stage read command or a second-stage read command, and issues the first-stage read command to a data path of the computing device. If the command is a second-stage read command, the debug controller causes a reservation register of the debug controller to provide a data value or status indication to the host debugger through the interface. The reservation register contains read data returned by the first-stage read command and, in response to the second-stage read command, provides a status indication when the first-stage read command has not yet returned read data.

Debugging a computing device

A clock generator for generating a target clock with a frequency equal to the frequency of an input clock divided by a non-integer ratio is disclosed. The clock generator comprises a clock divider. The clock divider is configured to divide the input clock by a first dividing ratio during a first portion of a frame period to generate a first clock slower than the target clock, and divide the input clock by a second dividing ratio during a second portion of the frame period to generate a second clock faster than the target clock. A difference between the first dividing ratio and the second dividing ratio is 0.5 or 1. In some embodiments, the first dividing ratio and the second dividing ration are integers closest to the non-integer ratio.

Clock generation with non-integer clock dividing ratio

Embodiments of the present disclosure provide a system on a chip (SOC) comprising a processing core, and a cache including a cache instruction port, a cache data port, and a port utilization circuitry configured to selectively fetch instructions through the cache instruction port and selectively pre-fetch instructions through the cache data port. Other embodiments are also described and claimed.

Cache pre-fetch architecture and method

A system and method are described for calibrating a clock used in data transmission. In one example, dynamic phase adjustment circuitry can be used for any of a variety of different protocols to shift the clock phase with respect to a data signal. In the most typical example, the clock phase is shifted 90 degrees relative to a transmission data signal. The dynamic phase adjustment circuitry can use two cascaded programmable delay lines coupled in series. Each programmable delay line represents a half phase delay of 90 degrees. A controller can monitor an output of the programmable delay lines and incrementally add or subtract programmable delay line elements until a 180 degree phase is detected relative to a data transmission. An output clock can then be used by applying the result of the calibration delay element to the clock under discussion.

Clock phase alignment in data transmission

A clock generator for generating a clock equivalent to a target clock which is an input clock divided by a non-integer ratio is disclosed. The clock generator comprises a clock divider configured to receive the input clock and divide the input clock with a reconfigurable dividing ratio; and a control circuit controlling operations of the clock divider to divide the input clock by a first dividing ratio to generate a first number of cycles of a first clock in a frame, and divide the input clock by a second dividing ratio to generate a second number of cycles of a second clock in the frame, wherein a difference between a period of the frame and a cumulative time of the first number of cycles of the first clock and the second number of cycles of the second clock is less than a threshold value.

A command processing pipeline is coupled to a shared cache. The command processing pipeline comprises (i) a first command processing stage configured to sequentially receive and process first and second cache commands, and (ii) a second command processing stage coupled to the first command processing stage. The first and the second command processing stages are two consecutive command processing stages of the command processing pipeline. The first and second command processing stages may access different groups of cache resources, and the first and second cache commands may be processed during consecutive clock cycles of a clock signal. Processing of the second cache command may be performed independently of an outcome of processing the first cache command by the first command processing stage. A third command processing stage may write data associated with the first cache command to one of a valid memory and a data memory included in the cache.

Growing community of inventors

Nofit, Israel

Gil Stoler