参考： https://www.cnblogs.com/charlesblc/p/6263665.htmlmmap使用细节1、使用mmap需要注意的一个关键点是，mmap映射区域大小必须是物理页大小(page_size)的整倍数（32位系统中通常是4k字节）。原因是，内存的最小粒度是页，而进程虚拟地址空间和内存的映射也是以页为单位。为了匹配内存的操作，mmap从磁盘到虚拟地址空间的映射也必须是页。2、内核可以跟踪被内存映射的底层对象（文件）的大小，进程可以合法的访问在当前文件大小...
COMPUTER OR GANIZATION AND ARCHITECTURE DESIGNING FOR PERFORMANCE NINTH EDITION 原文：http://www.cnblogs.com/yuanjiangw/p/6253854.html
问题描述：
自动关联后回放报错：Action.c(16): Error: C interpreter run time error: Action.c (16): Error -- memory violation : Exception ACCESS_VIOLATION received.
问题定位：
经过定位发现生成的关联函数为web_reg_save_param_attrib()，而非期望的web_reg_save_param_regexp()，
原因分析：
关联设置时定义的关联类型错误所致，此处类型为Attribute Based导致的。解决办法：
取消关联，设置关联类型为“Boundary Base...
Coalesced Memory access 连续当要获取的Memory首地址是cache line的倍数时，就是Aligned Memory Access，如果是非对齐的，就会导致浪费带宽。至于Coalesced Memory Access则是warp的32个thread请求的是连续的内存块。
下图就是很好的符合了连续和对齐原则，只有128-byte Memory传输的消耗：
下图则没有遵守连续和对齐原则，有三次传输消耗发生，一次是从偏移地址0开始，一次是从偏移地址256开始，还有一次是从偏移128开始，而这...
dma_pool *dma_pool_create(const char *name, struct device *dev, size_t size, size_t align, size_t allocation);
void dma_pool_destroy(struct dma_pool *pool);name是DMA池的名字，dev是device结构，size是从该池中分配的缓冲区的大小，align是该池分配操作所必须遵守的硬件对齐原则(用字节表示)，如果allocation不为零，表示内存边界不能超越allocation。比如说传入的allocation是4K，表示从该池分配的缓冲区不能跨越4KB的...
Computer Systems A Programmer‘s Perspective Second Edition
As we will discuss, the extension of IA32 to 64 bits, termed x86-64, was origi-
nally developed by Advanced Micro Devices (AMD), Intel’s biggest competitor.
Whereas a 32-bit machine can only make use of around 4 gigabytes (2
32 bytes) of random-access memory, current 64-bit machines can use up to 256 terabytes (2
48 bytes). The computer ...
memory model:
programming model:
CUDA[3] Samples for accessing shared/global memory标签：amp 技术分享 global model com access .com mode glob 本文系统来源：http://www.cnblogs.com/pdev/p/6344722.html
什么是RDMA? 简单来说，RDMA就是指不通过操作系统(OS)内核以及TCP/IP协议栈在网络上传输数据，因此延迟(latency)非常低，CPU消耗非常少。 下面给出一篇简单介绍RDMA的文章之中英文对照翻译。
Introduction to Remote Direct Memory Access (RDMA) | RDMA概述
1. What is RDMA? | 什么是RDMA
Direct memory access (DMA) is an ability of a device to access host memory directly, without the intervention of the CPU(s).
RDMA ...
This blog post is motivated by a conversation with Frits Hoogland on his great blog post The curious case of the missing semctl call about how he managed to find a useful memory address (suspecting a fixed SGA variable) used by a process in his investigation.So here i will show how we can easily generates a trace of all/range of memory addresses referenced by a program with an acceptable overhead...
