当前位置:网站首页>Understanding the architecture type of mobile CPU
Understanding the architecture type of mobile CPU
2022-07-05 12:02:00 【Back end coder】
mobile phone CPU Understand the architecture type
Android The equipment CPU type , Different Android The equipment uses different CPU, Different CPU Support for different instruction sets
| CPU Architecture type | explain |
|---|---|
| armeabi/mips / mips64 | The first 5 generation 、 The first 6 Generation ARM processor , Early mobile phones used more ,NDK I used to support ARMv5 (armeabi) as well as 32 Bit and 64 position MIPS, but NDK r17 No longer support |
| armeabi-v7a | The first 7 Generation and above ARM, this ABI Applicable to 32 position ARM Of CPU processor . |
| arm64-v8a | this ABI Applicable to ARMv8-A Of CPU, The first 8 generation 、64 position ARM processor , The current mainstream version |
| x86 、x86_64 | intel CPU , Flat 、 Simulator 、64 A bit of a flat |
How to adapt in the project
It only suits armeabi-v7a, If APP Installed on mobile phones with other architectures , Such as arm64-v8a On , Can you jump ?
Can't , But the opposite will happen
because armeabi-v7a and arm64-v8a Will be downward compatible :
Fit only armeabi Of APP Can run in armeabi,x86,x86_64,armeabi-v7a,arm64-v8 Upper fitting only armeabi-v7a Can run in armeabi-v7a and arm64-v8a Fit only arm64-v8a Can run in arm64-v8a On
So how do we fit ? The following schemes are given :
Scheme 1 : Fit only armeabi
advantage : Basically fits all CPU framework ( Except for eliminated mips and mips_64)
shortcoming : Low performance , It is equivalent to the need for assistance on most mobile phones ABI Or dynamic transcoding
Option two : Fit only armeabi-v7a
Similarly, scheme I , Just sifted out some old equipment , Balance performance and compatibility
Option three : Fit only arm64-v8
advantage : Best performance
shortcoming : It can only run on arm64-v8 On , Give up some old equipment users
All three options are OK , Big factories now APP In adaptation , All three , Mostly before 2 Kind of plan . Which one to choose depends on your own considerations , Performance for compatibility arm64-v8, Trade compatibility for performance armeabi, A little balance between the two is armeabi-v7a.
For the moment , Most big factories APP Using all of these armeabi or armeabi-v7a.
边栏推荐
- Use and install RkNN toolkit Lite2 on itop-3568 development board NPU
- JS for循环 循环次数异常
- 《看完就懂系列》15个方法教你玩转字符串
- 【SingleShotMultiBoxDetector(SSD,单步多框目标检测)】
- Proof of the thinking of Hanoi Tower problem
- 投资理财适合女生吗?女生可以买哪些理财产品?
- 报错ModuleNotFoundError: No module named ‘cv2.aruco‘
- 16 channel water lamp experiment based on Proteus (assembly language)
- 全网最全的新型数据库、多维表格平台盘点 Notion、FlowUs、Airtable、SeaTable、维格表 Vika、飞书多维表格、黑帕云、织信 Informat、语雀
- Codeforces Round #804 (Div. 2)
猜你喜欢

redis主从中的Master自动选举之Sentinel哨兵机制

Wireless WiFi learning 8-channel transmitting remote control module
![[pytorch modifies the pre training model: there is little difference between the measured loading pre training model and the random initialization of the model]](/img/ad/b96e9319212cf2724e0a640109665d.png)
[pytorch modifies the pre training model: there is little difference between the measured loading pre training model and the random initialization of the model]

13.(地图数据篇)百度坐标(BD09)、国测局坐标(火星坐标,GCJ02)、和WGS84坐标系之间的转换

Matlab superpixels function (2D super pixel over segmentation of image)

Idea set the number of open file windows

Pytorch MLP

查看rancher中debug端口信息,并做IDEA Remote Jvm Debug

Intern position selection and simplified career development planning in Internet companies

Uniapp + unicloud + Unipay realize wechat applet payment function
随机推荐
Thoughts and suggestions on the construction of intelligent management and control system platform for safe production in petrochemical enterprises
12.(地图数据篇)cesium城市建筑物贴图
【云原生 | Kubernetes篇】Ingress案例实战(十三)
mmclassification 训练自定义数据
你做自动化测试为什么总是失败?
redis的持久化机制原理
Seven ways to achieve vertical centering
[untitled]
redis集群中hash tag 使用
Liunx prohibit Ping explain the different usage of traceroute
Halcon 模板匹配实战代码(一)
【PyTorch预训练模型修改、增删特定层】
Hash tag usage in redis cluster
【主流Nivida显卡深度学习/强化学习/AI算力汇总】
Pytorch weight decay and dropout
12. (map data) cesium city building map
ACID事务理论
【TFLite, ONNX, CoreML, TensorRT Export】
[deploy pytoch project through onnx using tensorrt]
Install esxi 6.0 interactively