当前位置:网站首页>Vector types and variables built in CUDA
Vector types and variables built in CUDA
2022-07-02 06:27:00 【Little Heshang sweeping the floor】
CUDA The built-in Vector Types and variables
Built-in Vector Types
char, short, int, long, longlong, float, double
These are vector types derived from basic integer and floating-point types . They are structures , first 、 the second 、 The third and fourth components can be passed through fields x、y、z and w visit . They all carry make_<type name> Constructor of form ; for example ,
int2 make_int2(int x, int y);
It creates a with value(x, y) Of int2 Vector of type .
The alignment requirements of vector types are detailed in the following table .
| Type | Alignment |
|---|---|
| char1, uchar1 | 1 |
| char2, uchar2 | 2 |
| char3, uchar3 | 1 |
| char4, uchar4 | 4 |
| short1, ushort1 | 2 |
| short2, ushort2 | 4 |
| short3, ushort3 | 2 |
| short4, ushort4 | 8 |
| int1, uint1 | 4 |
| int2, uint2 | 8 |
| int3, uint3 | 4 |
| int4, uint4 | 16 |
| long1, ulong1 | 4 if sizeof(long) is equal to sizeof(int) 8, otherwise |
| long2, ulong2 | 8 if sizeof(long) is equal to sizeof(int), 16, otherwise |
| long3, ulong3 | 4 if sizeof(long) is equal to sizeof(int), 8, otherwise |
| long4, ulong4 | 16 |
| longlong1, ulonglong1 | 8 |
| longlong2, ulonglong2 | 16 |
| longlong3, ulonglong3 | 8 |
| longlong4, ulonglong4 | 16 |
| float1 | 4 |
| float2 | 8 |
| float3 | 4 |
| float4 | 16 |
| double1 | 8 |
| double2 | 16 |
| double3 | 8 |
| double4 | 16 |
dim3
This type is based on uint3 Integer vector type of , Used to specify the dimension . Definition dim3 When a variable of type , Any unspecified components will be initialized to 1.
Built-in Variables
1. gridDim
The type of the variable is dim3( see also dim3) And include the size of the grid .
2. blockIdx
The variable is uint3 type ( Please see the char、short、int、long、longlong、float、double) And include the block index in the grid .
3. blockDim
The type of the variable is dim3( see also dim3) And include the size of the block .
4. threadIdx
This variable is uint3 type ( Please see the char、short、int、long、longlong、float、double ) And include the thread index in the block .
5. warpSize
The variable is int type , Contains warp size ( of warp The definition of , Please see the SIMT Architecture).
边栏推荐
- Name six schemes to realize delayed messages at one go
- virtualenv和pipenv安装
- CUDA user object
- Distributed transactions: the final consistency scheme of reliable messages
- AtCoder Beginner Contest 253 F - Operations on a Matrix // 树状数组
- CUDA中的Warp matrix functions
- CUDA中的函数执行空间说明符
- 20201025 Visual Studio2019 QT5.14 信号和槽功能的使用
- 记录一次RDS故障排除--RDS容量徒增
- Sudo right raising
猜你喜欢
随机推荐
Redis---1.数据结构特点与操作
Pbootcms collection and warehousing tutorial quick collection release
实现strStr() II
Hydration failed because the initial UI does not match what was rendered on the server.问题原因之一
FE - Weex 使用简单封装数据加载插件为全局加载方法
压力测试修改解决方案
PgSQL学习笔记
Redis——大Key问题
深入学习JVM底层(五):类加载机制
BGP routing optimization rules and notification principles
MySql索引
eslint配置代码自动格式化
Is there a really free applet?
Golang--map扩容机制(含源码)
Decryption skills of encrypted compressed files
The Chinese word segmentation task is realized by using traditional methods (n-gram, HMM, etc.), neural network methods (CNN, LSTM, etc.) and pre training methods (Bert, etc.)
In depth understanding of JUC concurrency (I) what is JUC
重载全局和成员new/delete
CUDA与Direct3D 一致性
Invalid operation: Load into table ‘sources_ orderdata‘ failed. Check ‘stl_ load_ errors‘ system table









