Samsung Electronics Co., Ltd.

Sun Microsystems, Inc.

Oracle America, Inc.

Hewlett-packard Development Company, L.p.

Hewlett-packard Company

Santosh Abraham

Lawrence Spracklen

Yuan C Chou

John Wakefield Brothers

Karthik Ramani

Abhinav Golas

Woong Seo

Vinod Kumar Kathail

Christopher Thomas Cheng

Sreekumar Ramakrishnan Nair

Stevan A Vlaovic

Scott A Mahlke

Liangjun Zhang

Ki Fung Chow

Robert S Schreiber

Shail Aditya Gupta

Adam R Talcott

Jeongae Park

Bantwal R Rau

Kwontaek Kwon

Michael S Schlansker

Sanjay Patel

Joohoon Lee

Boris Beylin

Darryl J Gove

Craig Anderson

B Ramakrishna Rau

Seunghun Jin

Maxim Lukyanov

Lingjie Xu

Farnad Sajjadian

Sushant Kondguli

Seonggun Kim

Nicholas Sohre

Brian Fahs

Greg Snider

Alex Grosul

Sothea Soun

A method includes determining that a new draw call is received; comparing a state identity (ID) of a graphics state stored in the ring storage with a state ID of a graphics state associated with the new draw call; determining if the ring storage has available space to store the graphics state associated with the new draw call; storing the graphics state associated with the new draw call in the ring storage, based on determining that the ring storage has available space; determining a location of a first valid and non-default entry and a last valid and non-default entry of the graphics state associated with the new draw call stored in the ring storage; and collecting data from one or more valid entries of the graphics state associated with the new draw call stored in the ring storage to complete a task associated with the new draw call.

Variable entry transitional ring storage for efficiently accessing graphics states

Inventive aspects include a binning unit for coarse depth culling during binning of pixel geometries. The binning unit includes a rasterizer to receive primitives, and generates pixel coverage information and depth information. The binning unit includes one or more local culling stages to perform local culling within a window of primitives. The local culling unit outputs a set of surviving coverage and surviving depth information. The binning unit includes one or more global culling stages to use the set of the surviving coverage and the surviving depth information to further cull based on an entirety of previously received coverage information and depth information.

Coarse depth culling during binning

An apparatus, system and method is provided to determine a motion of pixels in local regions of a scene, classify the motion into a speed category, and make decisions on how to render blocks of pixels. In one implementation the motion in a tile is classified into at least three different speed regimes. If the pixels in a tile are in a quasi-static speed regime, a determination is made whether or not to reuse a fraction of pixels from the previous frame. If the pixels are determined to be in a high speed regime, a decision is made whether or not a sampling rate may be reduced.

Motion based adaptive rendering

A method of managing graphics data in a graphics processing device may include: receiving a first draw call having a first identifier, generating a first lookup table having the first identifier mapped in association with a first handle value by allocating the first handle value to the first identifier, generating a second lookup table having the first handle value mapped in association with a first graphics state setting value by allocating the first handle value to the first graphics state setting value, wherein the first graphics state setting value corresponds to the first identifier, and performing at least one graphics pipeline operation to process the first draw call by using the first graphics state setting value obtained from the second lookup table.

Methods and apparatuses for managing graphics data using two-stage lookup tables in cache

A graphics processing unit (GPU) configured to perform a graphics pipeline may generate operation data based on base operation data representing a base state of a plurality of components and implement the plurality of components to perform the graphics pipeline according to the generated operation data. The GPU may determine a priority of a plurality of instances of state version data based on frequencies of use associated with the plurality of instances of state version data, maintain first state version data having a determined highest priority from among the plurality of instances of state version data, and control second state version data having a determined lower priority than the first state version data to be updated based on a graphics pipeline being performed. The state version data may include code associated with performing the graphics pipeline in each of a plurality of states of the plurality of components.

Graphics processing devices and graphics processing methods

A texture cache architecture includes a first texture cache to store compressed texel data and a second texture cache to store decompressed texel data. A controller schedules accesses to access texel data from the first or second texture cache. The second texture cache permits decompressed texel data to be reused for more than one texel access request.

Efficient low-power texture cache architecture

A method for reducing execution of redundant threads in a processing environment. The method includes detecting threads that include redundant work among many different threads. Multiple threads from the detected threads are grouped into one or more thread clusters based on determining same thread computation results. Execution of all but a particular one thread in each of the one or more thread clusters is suppressed. The particular one thread in each of the one or more thread clusters is executed. Results determined from execution of the particular one thread in each of the one or more thread clusters are broadcasted to other threads in each of the one or more thread clusters.

Redundancy elimination in single instruction multiple data/thread (SIMD/T) execution processing

A texture cache architecture facilitates access of compressed texture data in non-power of two formats, such as the Adaptive Scalable Texture Compression (ASTC) codec. In one implementation, the texture cache architecture includes a controller, a first buffer, a second buffer, and a texture decompressor. A first buffer stores one or more blocks of compressed texel data fetched, in response to a first request, from a first texture cache, where the one or more blocks of compressed texel data including at least requested texel data. The second buffer stores decompressed one or more blocks of compressed texel data and provides the decompressed requested texel data as output to a second texture cache. The one or more blocks of compressed texel data stored by the first buffer includes second texel data in addition to the requested texel data.

Cache architecture for efficiently accessing texture data using buffers

In a graphics system regions of a frame are analyzed to determine local regions of the frame in which adaptive desampling may be performed. In one implementation a standard sampling scheme includes at least one sample per pixel and regions that are adaptively desampled have one sample for a block of pixels having a size of at least four pixels. A level of detail map is generating to identify regions in which desampling may be performed. The level of detail map may be based on detecting motion, detecting an edge, and detecting a content frequency.

Adaptive desampling in a graphics system with composited level of detail map

A graphics processing operation may include a set of render target operations, in which render targets are read and one or more intermediate computations are performed before generating final render target output. A method of performing graphics processing includes determining a dependency between render targets and defining a scheduling of tiles to reduce or eliminate a need to write intermediate computations to external memory. An interleaved order may be determined to maintain intermediate computations of dependent render target operations in an on-chip cache hierarchy.

Forward late predictive rendering in a graphics system

In a pipelined application having different stages of processing, such as a graphics application or an image processing application, there may be a dependence of one compute kernel upon another. Data associated with individual kernels needs to be written and read. A technique to minimize a need to read and write kernel data to external memory utilize at least one of fusing kernels, resizing workgroups, and performing interleaving of kernels.

Automated compute kernel fusion, resizing, and interleave

Intelligent prediction of critical sections is implemented using a method comprising updating a critical section estimator based on historical analysis of atomic/store instruction pairs during runtime and performing lock elision when the critical section estimator indicates that the atomic/store instruction pairs define a critical section.

Method and apparatus for critical section prediction for intelligent lock elision

In one embodiment, a processor comprises execution circuitry and a translation lookaside buffer (TLB) coupled to the execution circuitry. The execution circuitry is configured to execute a store instruction having a data operand; and the execution circuitry is configured to generate a virtual address as part of executing the store instruction. The TLB is coupled to receive the virtual address and configured to translate the virtual address to a first physical address. Additionally, the TLB is coupled to receive the data operand and to translate the data operand to a second physical address. A hardware accelerator is also contemplated in various embodiments, as is a processor coupled to the hardware accelerator, a method, and a computer readable medium storing instruction which, when executed, implement a portion of the method.

Efficient on-chip accelerator interfaces to reduce software overhead

In one embodiment, a method is contemplated. Access to a hardware accelerator is requested by a user-privileged thread. Access to the hardware accelerator is granted to the user-privileged thread by a higher-privileged thread responsive to the requesting. One or more commands are communicated to the hardware accelerator by the user-privileged thread without intervention by higher-privileged threads and responsive to the grant of access. The one or more commands cause the hardware accelerator to perform one or more tasks. Computer readable media comprises instructions which, when executed, implement portions of the method are also contemplated in various embodiments, as is a hardware accelerator and a processor coupled to the hardware accelerator.

Low overhead access to shared on-chip hardware accelerator with memory-based interfaces

Growing community of inventors

Pleasanton, CA, United States of America

Santosh Abraham