Abstract: A multithreaded rendering software pipeline architecture utilizes a rolling texture context data structure to store multiple texture contexts that are associated with different textures that are being processed in the software pipeline. Each texture context stores state data for a particular texture, and facilitates the access to texture data by multiple, parallel stages in a software pipeline. In addition, texture contexts are capable of being “rolled”, or copied to enable different stages of a rendering pipeline that require different state data for a particular texture to separately access the texture data independently from one another, and without the necessity for stalling the pipeline to ensure synchronization of shared texture data among the stages of the pipeline.

Abstract: Methods and apparatus for reducing the impact of latency associated with decrypting encrypted data are provided. Rather than wait until an entire packet of encrypted data is validated (e.g., by checking for data transfer errors), the encrypted data may be pipelined to a decryption engine as it is received, thus allowing decryption to begin prior to validation. In some cases, the decryption engine may be notified of data transfer errors detected during the validation process, in order to prevent reporting false security violations.

Abstract: Collecting and analyzing trace data while in a software debug mode through direct interthread communication (‘DITC’) on a network on chip (‘NOC’), the NOC including integrated processor (‘IP’) blocks, routers, memory communications controllers, and network interface controllers, with each IP block adapted to a router through a memory communications controller and a network interface controller, where each memory communications controller controlling communications between an IP block and memory, and each network interface controller controlling inter-IP block communications through routers, including enabling the collection of software debug information in a selected set of IP blocks distributed through the NOC, each IP block within the selected set of IP blocks having a set of trace data; collecting software debugging information via the set of trace data; communicating the set of trace data to a destination repository; and analyzing the set of trace data at the destination repository.

Abstract: A method includes rendering an object of a three dimensional image via a pixel shader based on a render context data structure associated with the object. The method includes measuring a performance statistic associated with rendering the object. The method also includes storing the performance statistic in the render context data structure associated with the object. The performance statistic is accessible to a host interface processor to determine whether to allocate a second pixel shader to render the object in a subsequent three-dimensional image.

Abstract: A method includes assigning a priority to a ray data structure of a plurality of ray data structures based on one or more priorities. The ray data structure includes properties of a ray to be traced from an illumination source in a three-dimensional image. The method includes identifying a portion of the three-dimensional image through which the ray passes. The method also includes identifying a slave processing element associated with the portion of the three-dimensional image. The method further includes sending the ray data structure to the slave processing element.

Abstract: A circuit arrangement and method support efficient indexing into large register files by utilizing register address sequence detection, wherein register addresses to be used by an instruction are produced by concatenating a portion of the address that is contained in the instruction with another portion that is speculatively produced by sequence detection logic. The portion of the correct full address that is not contained in the instruction is stored in a software accessible special purpose register. If the end of a particular sequence of addresses is detected by the sequence detection logic, the invention speculatively assumes that the next address in the sequence will be used. Since only a portion of the full addresses are stored in the instruction, they occupy less instruction space than the full address widths. An instruction may include at least one address portion that identifies a register address.

Abstract: A method, system and computer program product are presented for providing pseudo-random input test data to a test program. A seed value is generated and stored in a seed register. Using the seed value as an input, a pseudo-random input test value is generated by a Linear Feedback Shift Register (LFSR), and stored in a GPR within a processor core. Using the pseudo-random input test value from the GPR, a test program is executed within the processor core.

Abstract: A NOC for dynamic virtual software pipelining including IP blocks, routers, memory communications controllers, and network interface controllers, each IP block adapted to a router through a memory communications controller and a network interface controller, the NOC also including: a computer software application segmented into stages, each stage comprising a flexibly configurable module of computer program instructions identified by a stage ID, each stage assigned to a thread of execution on an IP block; and each stage executing on a thread of execution on an IP block, including a first stage executing on an IP block, producing output data and sending by the first stage the produced output data to a second stage, the output data including control information for the next stage and payload data; and the second stage consuming the produced output data in dependence upon the control information.

Abstract: Graphics rendering on a network on chip (‘NOC’) including receiving, in the geometry processor, a representation of an object to be rendered; converting, by the geometry processor, the representation of the object to two dimensional primitives; sending, by the geometry processor, the primitives to the plurality of scan converters; converting, by the scan converters, the primitives to fragments, each fragment comprising one or more portions of a pixel; for each fragment: selecting, by the scan converter for the fragment in dependence upon sorting rules, a pixel processor to process the fragment; sending, by the scan converter to the pixel processor, the fragment; and processing, by the pixel processor, the fragment to produce pixels for an image.

Abstract: A breakpoint packet is dispatched to a Network On A Chip (NOC). The breakpoint packet instructs one or more specified nodes on the NOC to place the specified nodes, or a core or hardware thread within a specified node, to execute in “single step” mode, in order to enable a debugging of a work packet that is dispatched to the specific node.

Abstract: Methods and apparatus that allow recovery in the event that sequence counts used on receive and transmit sides of a communications link become out of sync are provided. In response to receiving a packet with an expected sequence count from a receiving device, a transmitting device may adjust pointers into a transmit buffer allowing the transmitting device to begin transmitting packets with the sequence count expected by the receiving device.

Abstract: Software pipelining on a network on chip (‘NOC’), the NOC including integrated processor (‘IP’) blocks, routers, memory communications controllers, and network interface controllers, each IP block adapted to a router through a memory communications controller and a network interface controller, each memory communications controller controlling communication between an IP block and memory, and each network interface controller controlling inter-IP block communications through routers.

Abstract: Embodiments of the present invention provide methods, a module, and a system for calculating a credit limit for an interface capable of receiving multiple packets simultaneously. Generally, the multiple packets are simultaneously received at an interface on the second device, each packet being one of a plurality of packet types, and a flow control credit limit to be transmitted to the first device is adjusted based on the combination of packet types of the simultaneously received packets.

Abstract: According to embodiments of the invention, separate spatial indexes may be created which correspond to dynamic objects in a three dimensional scene and static objects in the three dimensional scene. By creating separate spatial indexes for static and dynamic objects, only the dynamic spatial index may need to be rebuilt in response to movement or changes in shape of objects in the three dimensional scene. Furthermore, the static and dynamic spatial indexes may be stored in separate portions of an image processing system's memory cache. By storing the static spatial index and the dynamic spatial index in separate portions of the memory cache, the dynamic portion of the memory cache may be updated without affecting the static portion of the spatial index in the memory cache.

Abstract: A circuit arrangement and method support offset based register address indexing, wherein register addresses to be used by an instruction are calculated using offsets to the full target register address, and the offsets are contained in the instruction and occupy less instruction space than the full address widths. An instruction may include at least one offset value that identifies a register address. During decoding of the instruction, an offset and a full target address are retrieved from the instruction, and then a register address is calculated by addition of the offset to the full target address.

Abstract: By merging or adding the color contributions from objects intersected by secondary rays, the image processing system may accumulate color contributions to pixels from objects intersected by secondary rays as the further color contributions are determined. Furthermore, by associating a scaling factor of color contribution with objects and with secondary rays which intersect the objects, color contributions due to secondary ray/object intersections may be calculated at a later time than the color contribution to a pixel from original ray/object intersection. Consequently, it is not necessary for a vector throughput engine or a workload manager to wait for all secondary ray/object intersections to be determined before updating the color of a pixel.

Abstract: Buffer memories having hardware controlled buffer space regions in which the hardware controls the dimensions of the various buffer space regions to meet the demands of a particular system. The hardware monitors the usage of the buffer data regions over time and subsequently and automatically adjusts the dimensions of the buffer space regions based on the utilization of those buffer regions.

Abstract: A computer-implemented method, system and computer program product for arbitrarily aligning vector operands, which are transmitted in inter-thread communication buffer packets within a highly threaded Network On a Chip (NOC) processor, are presented. A set of multiplexers in a node in the NOC realigns and extracts data word aggregations from an incoming compressed inter-thread communication buffer packet. The extracted data word aggregations are used as operands by an execution unit within the node.

Abstract: A design structure embodied in a machine readable medium is provided. Embodiments of the design structure include a network on chip (‘NOC’), the NOC comprising: integrated processor (‘IP’) blocks, routers, memory communications controllers, and network interface controller, each IP block adapted to a router through a memory communications controller and a network interface controller, each memory communications controller controlling communication between an IP block and memory, and each network interface controller controlling inter-IP block communications through routers; the network organized into partitions, each partition including at least one IP block, each partition assigned exclusive access to a separate physical memory address space; and one or more applications executing on one or more of the partitions.

Abstract: Computer systems with direct updating of cache (e.g., primary L1 cache) memories of a processor, such as a central processing unit (CPU) or graphics processing unit (GPU). Special addresses are reserved for high speed memory. Memory access requests involving these reserved addresses are routed directly to the high speed memory. Memory access requests not involving these reserved addresses are routed to memory external to the processor.