Data processing apparatuses, methods of data processing, complementary instructions and programs related to ring buffer administration are disclosed. An enqueuing operation performs an atomic compare-and-swap oper-ation to store a first processed data item indication to an enqueuing-target slot in the ring buffer contingent on an in-order marker not being present there and, when successful, determines that a ready-to-dequeue condition is true for the first processed data item indication. A dequeuing operation, when the ready-to-de-queue condition for a dequeuing-target slot is true, comprises writing a null data item to the dequeuing-target slot and, when dequeuing in-order, further comprises, dependent on whether a next contiguous slot has null content, determining a retirement condition and, when the retirement condition is true, performing a retirement process on the next contiguous slot comprising making the next con-tiguous slot available to a subsequent enqueuing operation. Further subsequent slots may also be retired.

Some embodiments of the present disclosure provide an associatively indexed circular buffer (ACB). The ACB may be viewed as a dynamically allocatable memory structure that offers in-order data access (say, first-in-first-out, or “FIFO”) or random order data access at a fixed, relatively low latency. The ACB includes a data store of non-contiguous storage. To manage the pushing of data to, and popping data from, the data store, the ACB includes a contiguous pointer generator, a content addressable memory (CAM) and a free pool.

A method can be used to process an initial set of data through a convolutional neural network that includes a convolution layer followed by a pooling layer. The initial set is stored in an initial memory along first and second orthogonal directions. The method includes performing a first filtering of the initial set of data by the convolution layer using a first sliding window along the first direction. Each slide of the first window produces a first set of data. The method also includes performing a second filtering of the first sets of data by the pooling layer using a second sliding window along the second direction.

Techniques are disclosed for power conservation. A plurality of processing elements and a plurality of instructions are configured. The plurality of processing elements is controlled by instructions contained in a plurality of circular buffers. The plurality of processing elements can comprise a dataflow processor. A first processing element, from the plurality of interconnected processing elements, is set into a sleep state by a first instruction from the plurality of instructions. The first processing element is woken from the sleep state as a result of valid data being presented to the first processing element. A subsection of the plurality of interconnected processing elements is also set into a sleep state based on the first processing element being set into a sleep state. At least one circular buffer from the plurality of circular buffers remains awake while the first processing element is in the sleep state, and the at least one circular buffer provides for data steering through a reconfigurable fabric.

A group of low-level FIFOs may be logically bound together to form a super-FIFO. The super-FIFO may treat each low-level FIFO as a storage location. The super-FIFO may enable a push to (or a pop from) every low-level FIFO, simultaneously. The super-FIFO may enable a virtual channel (VC) to use the super-FIFO, bypassing a VC FIFO for the VC, removing several cycles of latency otherwise needed for enqueuing and dequeuing messages in the VC FIFO. In addition, the super-FIFO may enable bypassing of an arbiter, further reducing latency by avoiding a penalty of the arbiter.

Techniques are described for managing virtual machines using input/output (I/O) device logging. For example, a system bus or other interface to a device may be monitored for traffic data elements. The traffic data elements may include, for example, transaction layer packets (TLPs) for communication across a PCI Express interface, or TCP/IP packets for communication over a network. These traffic data elements may be logged in an I/O device logging buffer. The I/O device logging buffer can then be used to ensure that all memory relating to a virtual machine is copied when transferring the virtual machine to another computing device. In addition, the I/O device logging buffer can be used to stop a virtual machine without waiting for the virtual machine to complete I/O processing.

An object of the present invention is to prevent occurrence of data destruction when a transfer source region and a transfer destination region of data overlap with each other and even when transfer is performed using a burst transfer function. The data read from the transfer source region is temporarily written into a ring buffer, and then the data written into the ring buffer is written into the transfer source region. In this case, reading of the data from the ring buffer is controlled, based on a magnitude relation between the number of times of wrap-arounds caused by writing of the data into the ring buffer and the number of times of wrap-arounds caused by reading of the data from the ring buffer.

Some embodiments of the present disclosure provide an associatively indexed circular buffer (ACB). The ACB may be viewed as a dynamically allocatable memory structure that offers in-order data access (say, first-in-first-out, or “FIFO”) or random order data access at a fixed, relatively low latency. The ACB includes a data store of non-contiguous storage. To manage the pushing of data to, and popping data from, the data store, the ACB includes a contiguous pointer generator, a content addressable memory (CAM) and a free pool.

Some embodiments of the present disclosure provide an associatively indexed circular buffer (ACB). The ACB may be viewed as a dynamically allocatable memory structure that offers in-order data access (say, first-in-first-out, or “FIFO”) or random order data access at a fixed, relatively low latency. The ACB includes a data store of non-contiguous storage. To manage the pushing of data to, and popping data from, the data store, the ACB includes a contiguous pointer generator, a content addressable memory (CAM) and a free pool.

The present disclosure provides a speech data processing method and apparatus, a device and a storage medium, wherein the method comprises: a client obtains speech data, uses a speech wakeup engine to perform wakeup recognition for the obtained speech data, and stores the obtained speech data in a cyclic buffer queue; the client obtains a wakeup recognition result returned by the speech wakeup engine upon completion of the wakeup recognition, and if the wakeup recognition result is acknowledged wakeup and it is determined that there occurs a case of oneshot currently, determines an interception starting position and obtains buffer data in the cyclic buffer queue from the interception staring position to an end of the queue, the buffer data at least including the content to be recognized; the client sends the buffer data to a speech recognition engine so that the speech recognition engine performs speech recognition for the content to be recognized in the buffer data. The technical solution of the present disclosure can be applied to improve accuracy of the speech recognition result and the like.