ARM Mali / AMD GPU - OpenCL
文章平均质量分 76
ARM Mali GPUs
AMD GPUs
Open Computing Language,OpenCL:开放计算语言
Yongqiang Cheng
既然选择了远方 便只顾风雨兼程 - 永强
展开
-
NEON Programmer’s Guide
NEON Programmer’s GuideNEON Programmer’s Guide Version: 1.0https://static.docs.arm.com/den0018/a/DEN0018A_neon_programmers_guide_en.pdfReferenceshttps://static.docs.arm.com/den0018/a/DEN0018A_neon_programmers_guide_en.pdf原创 2021-01-14 00:30:27 · 736 阅读 · 0 评论 -
Arm Neon - Resources
Arm Neon - ResourcesNeonhttps://developer.arm.com/architectures/instruction-sets/simd-isas/neonArm Neon technology is an advanced Single Instruction Multiple Data (SIMD) architecture extension for the Arm Cortex-A and Cortex-R series processors.Neon te原创 2020-11-02 23:18:10 · 603 阅读 · 0 评论 -
Introducing Neon for Armv8-A
Introducing Neon for Armv8-A1. Overview2. Before you begin3. Data processing methodologies4. Fundamentals of Armv8 Neon technology5. Check your knowledge6. Related information7. Next stepssingle page - multiple pagesReferencesIntroducing Neon fo原创 2021-01-03 23:04:04 · 413 阅读 · 3 评论 -
ARM - Advanced SIMD register - quadword (128 bits wide) and doubleword (64 bits wide)
ARM - Advanced SIMD register - quadword (128 bits wide) and doubleword (64 bits wide)1. Bytes, Halfwords, and WordsByteEight bits (8 bits).HalfwordTwo bytes (16 bits).WordFour bytes (32 bits).Quadword16 contiguous bytes (128 bits).2. Register enc原创 2021-02-22 23:46:54 · 1255 阅读 · 10 评论 -
Armv8-A and Armv8-R Architectures - half-precision (16-bit) floating-point
Armv8-A and Armv8-R Architectures - half-precision (16-bit) floating-pointFloating Pointhttps://developer.arm.com/architectures/instruction-sets/floating-point1. floating pointThe Arm architecture provides high-performance and high-efficiency hardware原创 2021-01-13 00:17:29 · 613 阅读 · 0 评论 -
AMD GPUOpen - AMD GPU 开源计划
AMD GPUOpen - AMD GPU 开源计划Referenceshttps://gpuopen.com/原创 2018-11-14 09:20:20 · 778 阅读 · 0 评论 -
OpenCL Function Qualifiers (函数限定符)
OpenCL Function Qualifiers (函数限定符)OpenCL 3.0 Reference Pages -> OpenCL Compiler -> Function Qualifiers1. Function Qualifiers (函数限定符)1.1 __kernel or kernelThe __kernel or kernel qualifier declares a function to be a kernel that can be executed by原创 2022-01-23 12:16:01 · 509 阅读 · 0 评论 -
AMD OpenCL APP SDK sample - Hello World
AMD OpenCL APP SDK sample - Hello WorldAMD OpenCL Accelerated Parallel Processing (APP) Software Development Kit (SDK)1. Overview1.1 Location$<AMDAPPSDKSamplesInstallPath>\samples\opencl\cl\1.xD:\Program Files\AMD APP SDK\3.0\samples\opencl\cl\1原创 2022-01-22 14:34:35 · 1127 阅读 · 7 评论 -
OpenCL Synchronization Functions (同步函数)
OpenCL Synchronization Functions (同步函数)OpenCL 3.0 Reference Pages -> OpenCL Compiler -> Built-in Functions -> Sync Functions对于一个内核函数,会有多个 work-groups 参与计算。每个 work-group 中会有多个 work-items 参与计算。在 OpenCL 中定义了一个相对宽松的同步机制,多个 work-groups 之间没办法同步;而在同一个原创 2022-01-22 01:02:33 · 1219 阅读 · 0 评论 -
Memory Hierarchy - 存储器层次结构
Memory Hierarchy - 存储器层次结构计算机系统将存储器分成若干层级 (memory hierarchy) ,越靠近 CPU 的存储器容量越小但访问速度越快。1. Memory hierarchy (存储器层次结构)Intel 北桥包含 2 个 channel,两组独立的线连接到各自的模块,每个 channel 包含 2 个DIMM。Shared resources in multicore processorsDRAM system organizationModu原创 2022-01-19 23:42:54 · 3542 阅读 · 0 评论 -
ARM Assembly Language: Fundamentals and Techniques - 参考代码
ARM Assembly Language: Fundamentals and Techniques - 参考代码1. ARM Assembly Language: Fundamentals and Techniques (Second Edition)https://bbooks.info/viewmore/arm-assembly-language-fundamentals-and-techniques-second-editionhttps://www.oreilly.com/library/v原创 2022-01-19 22:13:59 · 564 阅读 · 0 评论 -
AMD OpenCL University Kit - AMD OpenCL 大学课程
AMD OpenCL University Kit - AMD OpenCL 大学课程1. AMD Developer Central - University ProgramsWayback Machine (网站时光倒流机器)https://web.archive.org/web/20111219004704/http://developer.amd.com/zones/OpenCLZone/universities/Pages/default.aspxhttp://developer.amd.原创 2022-01-18 22:18:09 · 459 阅读 · 0 评论 -
AMD OpenCL Performance and Optimization for GCN Devices
AMD OpenCL Performance and Optimization for GCN Deviceshttps://rocmdocs.amd.com/en/latest/Programming_Guides/Opencl-optimization.htmlGCN 1.1http://developer.amd.com/wordpress/media/2013/07/AMD_Sea_Islands_Instruction_Set_Architecture1.pdfISA Manual for原创 2022-01-18 01:09:42 · 15872 阅读 · 0 评论 -
AMD OpenCL Programming Guide - OpenCL Architecture
AMD OpenCL Programming Guide - OpenCL Architecturehttps://rocmdocs.amd.com/en/latest/Programming_Guides/Opencl-programming-guide.html5. Memory Architecture and Access - 内存架构和访问OpenCL has four memory domains: private, local, global, and constant; the AMD原创 2022-01-17 22:42:59 · 753 阅读 · 0 评论 -
OpenCL Data Types (数据类型)
OpenCL Data Types (数据类型)1. Built-in Scalar Data Types - 内置标量数据类型The following table describes the list of built-in scalar data types.Type and Descriptionbool [1]A conditional data type which is either true or false. The value true expands to the integ原创 2022-01-16 00:55:36 · 2198 阅读 · 0 评论 -
Convolution Layer
Convolution LayerReferencesConvolution Layerhttp://caffe.berkeleyvision.org/tutorial/layers/convolution.htmlConv2Dhttps://www.paddlepaddle.org.cn/documentation/docs/zh/api/paddle/nn/Conv2D_cn.htmlConvolutional Neural Networks (CNNs / ConvNets)https:原创 2022-01-14 01:07:27 · 552 阅读 · 0 评论 -
地址空间限定符 - __global or global - __local or local - __constant or constant - __private or private
OpenCL Address Space Qualifiers (地址空间限定符) __global or global - __local or local - __constant or constant - __private or privateOpenCL 3.0 Reference Pages -> OpenCL Compiler -> Address Space QualifiersOpenCL C has a hierarchical memory architecture原创 2022-01-14 01:02:38 · 895 阅读 · 0 评论 -
AMD OpenCL Accelerated Parallel Processing (APP) - OpenCL 编程文档
AMD OpenCL Accelerated Parallel Processing (APP) - OpenCL 编程文档1. AMD Accelerated Parallel Processing OpenCL Programming GuideNovember 2013http://developer.amd.com/wordpress/media/2013/07/AMD_Accelerated_Parallel_Processing_OpenCL_Programming_Guide-re原创 2017-11-21 11:09:04 · 844 阅读 · 0 评论 -
AMD OpenCL Accelerated Parallel Processing (APP) Software Development Kit (SDK)
AMD OpenCL Accelerated Parallel Processing (APP) Software Development Kit (SDK)1. IntroductionAMD APP SDK is a software development kit by AMD for Accelerated Parallel Processing (APP). AMD APP SDK also targets Heterogeneous System Architecture (not only原创 2022-01-11 23:09:54 · 1111 阅读 · 0 评论 -
AMD GCN - Vega Instruction Set Architecture
AMD GCN - Vega Instruction Set Architecturehttps://rocmdocs.amd.com/en/latest/GCN_ISA_Manuals/testdocbook.html1. GCN ISA Manualshttps://rocmdocs.amd.com/en/latest/GCN_ISA_Manuals/GCN-ISA-Manuals.htmlGraphics Core Next,GCN:下一代图形核心Instruction Set Archit原创 2019-04-28 21:06:16 · 1495 阅读 · 0 评论 -
Introducing RDNA Architecture
Introducing RDNA ArchitectureThe RDNA architecture white paperhttps://www.amd.com/system/files/documents/rdna-whitepaper.pdfThe all new Radeon gaming architecture powering “Navi”全新 Radeon 游戏架构为 Navi 提供动力Table of ContentsIntroductionRDNA Architecture原创 2022-01-11 00:35:44 · 746 阅读 · 0 评论 -
AMD RDNA Architecture - AMD RDNA 架构
AMD RDNA Architecture - AMD RDNA 架构https://www.amd.com/en/technologies/rdnaArchitected for Gaming - 为游戏而构建The new RDNA architecture is designed for the next generation of efficient high-performance gaming. It’s the DNA that powers your games, the DNA th原创 2022-01-10 14:46:26 · 1923 阅读 · 0 评论 -
AMD ROCm Platform
AMD ROCm PlatformROCm Core Technologyhttps://github.com/RadeonOpenComputeROCm Docshttps://rocmdocs.amd.com/en/latest/ROCm Documentationhttps://github.com/RadeonOpenCompute/ROCm_DocumentationROCmhttps://github.com/RadeonOpenCompute/ROCm1. AMD ROCm™原创 2020-01-14 21:48:58 · 704 阅读 · 0 评论 -
Heterogeneous Computing with OpenCL - OpenCL 异构计算 - 书中源代码
Heterogeneous Computing with OpenCL - OpenCL 异构计算 - 书中源代码1. Heterogeneous Computehttp://www.heterogeneouscompute.org/OpenCL Programming GuideHeterogeneous Computing with OpenCL2. Figures and Codes from the first editionChapter2 - Vector addhttp://原创 2022-01-09 21:15:04 · 470 阅读 · 0 评论