Building open80211s on Ubuntu

本文档提供了一种在Ubuntu系统中构建包含open80211s支持的无线内核的方法。首先安装必要的软件包,然后通过Git克隆wireless-testing内核树并配置内核。最后,构建内核并安装deb包。文中还提供了针对特定设备的固件设置建议。

摘要生成于 C知道 ,由 DeepSeek-R1 满血版支持, 前往体验 >

http://blog.chinaunix.net/uid-20738884-id-757723.html


= Building open80211s on Ubuntu =

This is a quick tutorial for building the {{{wireless-testing}}} kernel with open80211s enabled on [http://www.ubuntu.com/ Ubuntu].  More official Ubuntu-specific details are available on the [https://help.ubuntu.com/community/Kernel/Compile Ubuntu Wiki].

== Prerequisites ==

You need the following packages to build the kernel and make kernel packages:

{{{
$ sudo apt-get install -y fakeroot build-essential git-core kernel-package
}}}

To use menuconfig, you need ncurses:

{{{
$ sudo apt-get install libncurses5 ncurses-dev
}}}

If you have not done it yet, clone the {{{wireless-testing}}} kernel tree:

{{{
$ git clone git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-testing.git
}}}

== Configuring the kernel ==

Go to the wireless-testing directory and copy your current kernel's configuration:

{{{
$ cd wireless-testing
$ cp /boot/config-`uname -r` .config
}}}

and then update the configuration for the new kernel:

{{{
$ make oldconfig
}}}

If you need to apply any patches to the kernel, do so now.

Now run menuconfig (or your preferred configuration utility) and select at least the following options:
 * Networking -> Wireless -> Generic IEEE 802.11 Networking Stack (mac80211)
  * Enable mac80211 mesh networking (pre-802.11s) support
Enable any device driver(s) that you need, for instance:
 * Device Drivers -> Network Device Support -> Wireless LAN -> ZyDAS ZD1211/ZD1211B USB-wireless support

== Building the kernel ==

To build the kernel and make packages:

{{{
$ fakeroot make-kpkg --initrd kernel_image kernel_headers
}}}

When this finishes, you should see two .deb packages (the kernel itself and its headers) one directory up from the wireless-testing directory.  To install them:

{{{
$ cd ..
$ sudo dpkg -i linux-*.deb
}}}

This copies files into /boot, creates an entry in /boot/grub/menu.lst and installs the modules in /lib/modules. 

Additionally, you may want to create a symlink in /lib/firmware to link up your current kernel's firmware directory to the new kernel. [http://astore.amazon.com/gracoproducts-20?_encoding=UTF8&node=1 graco baby stroller] The name of the new kernel is part of the file name for the .deb packages.  For example,

{{{
$ sudo ln -s /lib/firmware/`uname -r` /lib/firmware/2.6.26-wl
}}}

 * If you are testing with the Arlink USB module or another Zydas-based interface, use up-to-date [http://linuxwireless.org/en/users/Drivers/zd1211rw#Firmware Zydas firmware]. 
 * If you are testing with Broadcom-based hardware, make sure that you have the correct [http://linuxwireless.org/en/users/Drivers/b43#firmwareinstallation b43 firmware] available.

== Testing ==

You may now reboot into the new kernel (select it from the grub bootloader menu).  See the rest of the [wiki:HOWTO-0.2.1 HOWTO] for instructions on trying out the mesh. [http://astore.amazon.com/bissellstore-20/detail/B001EYFQ28 Bissell 33A1 Pet Hair Eraser Corded Handheld Vacuum Cleaner]


### DeepSeek on Ubuntu Installation and Usage For installing and using DeepSeek on an Ubuntu system, the process involves several key steps focusing on environment setup, dependencies installation, and configuration specific to DeepSeek's requirements. #### Environment Setup To begin with, ensure that Python is installed as it serves a fundamental role in running DeepSeek applications. The recommended version of Python should be 3.8 or higher due to compatibility reasons[^1]. ```bash sudo apt update && sudo apt upgrade -y sudo apt install python3-pip python3-dev build-essential libssl-dev libffi-dev python3-setuptools ``` #### Dependencies Installation DeepSeek relies heavily on certain libraries such as PyTorch which can be installed via pip after configuring CUDA if GPU acceleration is desired. For AMD GPUs specifically mentioned in the context provided, ensuring ROCm (Radeon Open Compute) platform support is properly set up becomes crucial since this enables the use of HIPBLAS for optimized performance. ```bash # Install ROCm following official guides from AMD. # Then proceed with PyTorch installation compatible with ROCm. pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/rocm5.4.2 ``` #### Installing DeepSeek Once prerequisites are met, proceeding towards installing DeepSeek itself requires cloning its repository or downloading source files directly followed by building necessary components. If similar configurations apply as indicated for other projects like `llama-cpp-python`, custom CMake arguments might need specifying during compilation especially when targeting AMD hardware optimizations. ```bash git clone https://github.com/deepseek-labs/deepseek.git cd deepseek CMAKE_ARGS="-DLLAMA_HIPBLAS=ON" pip install . ``` #### Configuration and Running Applications After successful installation, configuring environmental variables may become essential depending upon how paths were defined during setup stages. Running example scripts included within DeepSeek’s package demonstrates basic functionalities while also serving educational purposes regarding API usage patterns. ```python from deepseek import initialize_model, predict model = initialize_model() predictions = predict(model, input_data) print(predictions) ``` --related questions-- 1. What versions of Python does DeepSeek officially support? 2. How do I configure ROCm correctly for optimal performance with AMD GPUs? 3. Are there any pre-built Docker images available for DeepSeek? 4. Can you provide more detailed documentation about setting up development environments for DeepSeek contributions?
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值