mirror of
https://github.com/LCTT/TranslateProject.git
synced 2026-09-01 04:27:55 +08:00
SamMa
GitHub
200
published/20190807 Trace code in Fedora with bpftrace.md
Normal file
200
published/20190807 Trace code in Fedora with bpftrace.md
Normal file
@@ -0,0 +1,200 @@
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (YungeG)
|
||||
[#]: reviewer: (wxy)
|
||||
[#]: publisher: (wxy)
|
||||
[#]: url: (https://linux.cn/article-13632-1.html)
|
||||
[#]: subject: (Trace code in Fedora with bpftrace)
|
||||
[#]: via: (https://fedoramagazine.org/trace-code-in-fedora-with-bpftrace/)
|
||||
[#]: author: (Augusto Caringi https://fedoramagazine.org/author/acaringi/)
|
||||
|
||||
在 Fedora 中用 bpftrace 追踪代码
|
||||
======
|
||||
|
||||
|
||||
|
||||
bpftrace 是一个 [基于 eBPF 的新型追踪工具][2]，在 Fedora 28 第一次引入。Brendan Gregg、Alastair Robertson 和 Matheus Marchini 在网上的一个松散的黑客团队的帮助下开发了 bpftrace。它是一个允许你分析系统在幕后正在执行的操作的追踪工具，可以告诉你代码中正在被调用的函数、传递给函数的参数、函数的调用次数等。
|
||||
|
||||
这篇文章的内容涉及了 bpftrace 的一些基础，以及它是如何工作的，请继续阅读获取更多的信息和一些有用的实例。
|
||||
|
||||
### eBPF（<ruby>扩展的伯克利数据包过滤器<rt>extended Berkeley Packet Filter</rt></ruby>）
|
||||
|
||||
[eBPF][3] 是一个微型虚拟机，更确切的说是一个位于 Linux 内核中的虚拟 CPU。eBPF 可以在内核空间以一种安全可控的方式加载和运行小型程序，使得 eBPF 的使用更加安全，即使在生产环境系统中。eBPF 虚拟机有自己的指令集架构（[ISA][4]），类似于现代处理器架构的一个子集。通过这个 ISA，可以很容易将 eBPF 程序转化为真实硬件上的代码。内核即时将程序转化为主流处理器架构上的本地代码，从而提升性能。
|
||||
|
||||
eBPF 虚拟机允许通过编程扩展内核，目前已经有一些内核子系统使用这一新型强大的 Linux 内核功能，比如网络、安全计算、追踪等。这些子系统的主要思想是添加 eBPF 程序到特定的代码点，从而扩展原生的内核行为。
|
||||
|
||||
虽然 eBPF 机器语言功能强大，由于是一种底层语言，直接用于编写代码很费力，bpftrace 就是为了解决这个问题而生的。eBPF 提供了一种编写 eBPF 追踪脚本的高级语言，然后在 clang / LLVM 库的帮助下将这些脚本转化为 eBPF，最终添加到特定的代码点。
|
||||
|
||||
### 安装和快速入门
|
||||
|
||||
在终端 [使用][5] [sudo][5] 执行下面的命令安装 bpftrace：
|
||||
|
||||
```
|
||||
$ sudo dnf install bpftrace
|
||||
```
|
||||
|
||||
使用“hello world”进行实验：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 'BEGIN { printf("hello world\n"); }'
|
||||
```
|
||||
|
||||
注意，出于特权级的需要，你必须使用 root 运行 `bpftrace`，使用 `-e` 选项指明一个程序，构建一个所谓的“单行程序”。这个例子只会打印 “hello world”，接着等待你按下 `Ctrl+C`。
|
||||
|
||||
`BEGIN` 是一个特殊的探针名，只在执行一开始生效一次；每次探针命中时，大括号 `{}` 内的操作（这个例子中只是一个 `printf`）都会执行。
|
||||
|
||||
现在让我们转向一个更有用的例子：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_enter_execve { printf("%s called %s\n", comm, str(args->filename)); }'
|
||||
```
|
||||
|
||||
这个例子打印了父进程的名字（`comm`）和系统中正在创建的每个新进程的名称。`t:syscalls:sys_enter_execve` 是一个内核追踪点，是 `tracepoint:syscalls:sys_enter_execve` 的简写，两种形式都可以使用。下一部分会向你展示如何列出所有可用的追踪点。
|
||||
|
||||
`comm` 是一个 bpftrace 内建指令，代表进程名；`filename` 是 `t:syscalls:sys_enter_execve` 追踪点的一个字段，这些字段可以通过 `args` 内建指令访问。
|
||||
|
||||
追踪点的所有可用字段可以通过这个命令列出：
|
||||
|
||||
```
|
||||
bpftrace -lv "t:syscalls:sys_enter_execve"
|
||||
```
|
||||
|
||||
### 示例用法
|
||||
|
||||
#### 列出探针
|
||||
|
||||
`bpftrace` 的一个核心概念是<ruby>探针点<rt>probe point</rt></ruby>，即 eBPF 程序可以连接到的（内核或用户空间的）代码中的测量点，可以分成以下几大类：
|
||||
|
||||
* `kprobe`——内核函数的开始处
|
||||
* `kretprobe`——内核函数的返回处
|
||||
* `uprobe`——用户级函数的开始处
|
||||
* `uretprobe`——用户级函数的返回处
|
||||
* `tracepoint`——内核静态追踪点
|
||||
* `usdt`——用户级静态追踪点
|
||||
* `profile`——基于时间的采样
|
||||
* `interval`——基于时间的输出
|
||||
* `software`——内核软件事件
|
||||
* `hardware`——处理器级事件
|
||||
|
||||
所有可用的 `kprobe` / `kretprobe`、`tracepoints`、`software` 和 `hardware` 探针可以通过这个命令列出：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -l
|
||||
```
|
||||
|
||||
`uprobe` / `uretprobe` 和 `usdt` 是用户空间探针，专用于某个可执行文件。要使用这些探针，通过下文中的特殊语法。
|
||||
|
||||
`profile` 和 `interval` 探针以固定的时间间隔触发；固定的时间间隔不在本文的范畴内。
|
||||
|
||||
#### 统计系统调用数
|
||||
|
||||
**映射** 是保存计数、统计数据和柱状图的特殊 BPF 数据类型，你可以使用映射统计每个系统调用正在被调用的次数：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_enter_* { @[probe] = count(); }'
|
||||
```
|
||||
|
||||
一些探针类型允许使用通配符匹配多个探针，你也可以使用一个逗号隔开的列表为一个操作块指明多个连接点。上面的例子中，操作块连接到了所有名称以 `t:syscalls:sysenter_` 开头的追踪点，即所有可用的系统调用。
|
||||
|
||||
`bpftrace` 的内建函数 `count()` 统计系统调用被调用的次数；`@[]` 代表一个映射（一个关联数组）。该映射的键 `probe` 是另一个内建指令，代表完整的探针名。
|
||||
|
||||
这个例子中，相同的操作块连接到了每个系统调用，之后每次有系统调用被调用时，映射就会被更新，映射中和系统调用对应的项就会增加。程序终止时，自动打印出所有声明的映射。
|
||||
|
||||
下面的例子统计所有的系统调用，然后通过 `bpftrace` 过滤语法使用 PID 过滤出某个特定进程调用的系统调用：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_enter_* / pid == 1234 / { @[probe] = count(); }'
|
||||
```
|
||||
|
||||
#### 进程写的字节数
|
||||
|
||||
让我们使用上面的概念分析每个进程正在写的字节数：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_exit_write /args->ret > 0/ { @[comm] = sum(args->ret); }'
|
||||
```
|
||||
|
||||
`bpftrace` 连接操作块到写系统调用的返回探针（`t:syscalls:sys_exit_write`），然后使用过滤器丢掉代表错误代码的负值（`/arg->ret > 0/`）。
|
||||
|
||||
映射的键 `comm` 代表调用系统调用的进程名；内建函数 `sum()` 累计每个映射项或进程写的字节数；`args` 是一个 `bpftrace` 内建指令，用于访问追踪点的参数和返回值。如果执行成功，`write` 系统调用返回写的字节数，`arg->ret` 用于访问这个字节数。
|
||||
|
||||
#### 进程的读取大小分布（柱状图）：
|
||||
|
||||
`bpftrace` 支持创建柱状图。让我们分析一个创建进程的 `read` 大小分布的柱状图的例子：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_exit_read { @[comm] = hist(args->ret); }'
|
||||
```
|
||||
|
||||
柱状图是 BPF 映射，因此必须保存为一个映射（`@`），这个例子中映射键是 `comm`。
|
||||
|
||||
这个例子使 `bpftrace` 给每个调用 `read` 系统调用的进程生成一个柱状图。要生成一个全局柱状图，直接保存 `hist()` 函数到 `@`（不使用任何键）。
|
||||
|
||||
程序终止时，`bpftrace` 自动打印出声明的柱状图。创建柱状图的基准值是通过 _args->ret_ 获取到的读取的字节数。
|
||||
|
||||
#### 追踪用户空间程序
|
||||
|
||||
你也可以通过 `uprobes` / `uretprobes` 和 USDT（用户级静态定义的追踪）追踪用户空间程序。下一个例子使用探测用户级函数结尾处的 `uretprobe` ，获取系统中运行的每个 `bash` 发出的命令行：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 'uretprobe:/bin/bash:readline { printf("readline: \"%s\"\n", str(retval)); }'
|
||||
```
|
||||
|
||||
要列出可执行文件 `bash` 的所有可用 `uprobes` / `uretprobes`， 执行这个命令：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -l "uprobe:/bin/bash"
|
||||
```
|
||||
|
||||
`uprobe` 指向用户级函数执行的开始，`uretprobe` 指向执行的结束（返回处）；`readline()` 是 `/bin/bash` 的一个函数，返回键入的命令行；`retval` 是被探测的指令的返回值，只能在 `uretprobe` 访问。
|
||||
|
||||
使用 `uprobes` 时，你可以用 `arg0..argN` 访问参数。需要调用 `str()` 将 `char *` 指针转化成一个字符串。
|
||||
|
||||
### 自带脚本
|
||||
|
||||
`bpftrace` 软件包附带了许多有用的脚本，可以在 `/usr/share/bpftrace/tools/` 目录找到。
|
||||
|
||||
这些脚本中，你可以找到：
|
||||
|
||||
* `killsnoop.bt`——追踪 `kill()` 系统调用发出的信号
|
||||
* `tcpconnect.bt`——追踪所有的 TCP 网络连接
|
||||
* `pidpersec.bt`——统计每秒钟（通过 `fork`）创建的新进程
|
||||
* `opensnoop.bt`——追踪 `open()` 系统调用
|
||||
* `bfsstat.bt`——追踪一些 VFS 调用，按秒统计
|
||||
|
||||
你可以直接使用这些脚本，比如：
|
||||
|
||||
```
|
||||
$ sudo /usr/share/bpftrace/tools/killsnoop.bt
|
||||
```
|
||||
|
||||
你也可以在创建新的工具时参考这些脚本。
|
||||
|
||||
### 链接
|
||||
|
||||
* bpftrace 参考指南——<https://github.com/iovisor/bpftrace/blob/master/docs/reference_guide.md>
|
||||
* Linux 2018 `bpftrace`（DTrace 2.0）——<http://www.brendangregg.com/blog/2018-10-08/dtrace-for-linux-2018.html>
|
||||
* BPF：通用的内核虚拟机——<https://lwn.net/Articles/599755/>
|
||||
* Linux Extended BPF（eBPF）Tracing Tools——<http://www.brendangregg.com/ebpf.html>
|
||||
* 深入 BPF：一个阅读材料列表—— [https://qmonnet.github.io/whirl-offload/2016/09/01/dive-into-bpf][6]
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://fedoramagazine.org/trace-code-in-fedora-with-bpftrace/
|
||||
|
||||
作者：[Augusto Caringi][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[YungeG](https://github.com/YungeG)
|
||||
校对：[wxy](https://github.com/wxy)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://fedoramagazine.org/author/acaringi/
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://fedoramagazine.org/wp-content/uploads/2019/08/bpftrace-816x345.jpg
|
||||
[2]: https://github.com/iovisor/bpftrace
|
||||
[3]: https://lwn.net/Articles/740157/
|
||||
[4]: https://github.com/iovisor/bpf-docs/blob/master/eBPF.md
|
||||
[5]: https://fedoramagazine.org/howto-use-sudo/
|
||||
[6]: https://qmonnet.github.io/whirl-offload/2016/09/01/dive-into-bpf/
|
||||
[7]: https://unsplash.com/@wehavemegapixels?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText
|
||||
[8]: https://unsplash.com/search/photos/trace?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText
|
||||
@@ -1,37 +1,37 @@
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (tanloong)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
[#]: reviewer: (wxy)
|
||||
[#]: publisher: (wxy)
|
||||
[#]: url: (https://linux.cn/article-13628-1.html)
|
||||
[#]: subject: (Machine learning made easy with Python)
|
||||
[#]: via: (https://opensource.com/article/21/1/machine-learning-python)
|
||||
[#]: author: (Girish Managoli https://opensource.com/users/gammay)
|
||||
|
||||
用 Python 轻松实现机器学习
|
||||
======
|
||||
用朴素贝叶斯分类器解决现实世界里的机器学习问题
|
||||
![arrows cycle symbol for failing faster][1]
|
||||
|
||||
朴素贝叶斯是一种分类技术，它是许多分类器建模算法的基础。基于朴素贝叶斯的分类器是简单、快速和易用的机器学习技术之一，而且在现实世界的应用中很有效。
|
||||
> 用朴素贝叶斯分类器解决现实世界里的机器学习问题。
|
||||
|
||||
朴素贝叶斯是从 [贝叶斯定理][2] 发展来的。贝叶斯定理由 18 世纪的统计学家 [托马斯·贝叶斯][3] 提出，它根据与一个事件相关联的其他事件来计算该事件发生的概率。比如，[帕金森氏病][4] 患者通常嗓音会发生变化，因此嗓音变化就是与预测帕金森氏病相关联的症状。贝叶斯定理提供了计算目标事件发生概率的方法，而朴素贝叶斯是对该方法的推广和简化。
|
||||
|
||||
|
||||
<ruby>朴素贝叶斯<rt>Naïve Bayes</rt></ruby>是一种分类技术，它是许多分类器建模算法的基础。基于朴素贝叶斯的分类器是简单、快速和易用的机器学习技术之一，而且在现实世界的应用中很有效。
|
||||
|
||||
朴素贝叶斯是从 <ruby>[贝叶斯定理][2]<rt>Bayes' theorem</rt></ruby> 发展来的。贝叶斯定理由 18 世纪的统计学家 [托马斯·贝叶斯][3] 提出，它根据与一个事件相关联的其他条件来计算该事件发生的概率。比如，[帕金森氏病][4] 患者通常嗓音会发生变化，因此嗓音变化就是与预测帕金森氏病相关联的症状。贝叶斯定理提供了计算目标事件发生概率的方法，而朴素贝叶斯是对该方法的推广和简化。
|
||||
|
||||
### 解决一个现实世界里的问题
|
||||
|
||||
这篇文章展示了朴素贝叶斯分类器解决现实世界问题 (与商业级应用相反) 的能力。我会假设你对机器学习有基本的了解，所以文章里会跳过一些与机器学习预测不大相关的步骤，比如 <ruby>数据打乱<rt>(date shuffling)</rt></ruby> 和 <ruby>数据切片<rt>(data splitting)</rt></ruby>。如果你是机器学习方面的新手或者需要一个进修课程，请查看 _[An introduction to machine learning today][5]_ 和 _[Getting started with open source machine learning][6]_。
|
||||
这篇文章展示了朴素贝叶斯分类器解决现实世界问题（相对于完整的商业级应用）的能力。我会假设你对机器学习有基本的了解，所以文章里会跳过一些与机器学习预测不大相关的步骤，比如 <ruby>数据打乱<rt>date shuffling</rt></ruby> 和 <ruby>数据切片<rt>data splitting</rt></ruby>。如果你是机器学习方面的新手或者需要一个进修课程，请查看 《[An introduction to machine learning today][5]》 和 《[Getting started with open source machine learning][6]》。
|
||||
|
||||
朴素贝叶斯分类器是 <ruby>[有监督的][7]<rt>(supervised)</rt></ruby>、属于 <ruby>[生成模型][8]<rt>(generative)</rt></ruby> 的、非线性的、属于 <ruby>[参数模型][9]<rt>(parametric)</rt></ruby> 的和 <ruby>[基于概率的][10]<rt>(probabilistic)</rt></ruby>。
|
||||
朴素贝叶斯分类器是 <ruby>[有监督的][7]<rt>supervised</rt></ruby>、属于 <ruby>[生成模型][8]<rt>generative</rt></ruby> 的、非线性的、属于 <ruby>[参数模型][9]<rt>parametric</rt></ruby> 的和 <ruby>[基于概率的][10]<rt>probabilistic</rt></ruby>。
|
||||
|
||||
在这篇文章里，我会演示如何用朴素贝叶斯预测帕金森氏病。需要用到的数据集来自 [UCI Machine Learning Repository][11]。这个数据集包含许多语音信号的指标，用于计算患帕金森氏病的可能性；在这个例子里我们将使用这些指标中的前 8 个：
|
||||
在这篇文章里，我会演示如何用朴素贝叶斯预测帕金森氏病。需要用到的数据集来自 [UCI 机器学习库][11]。这个数据集包含许多语音信号的指标，用于计算患帕金森氏病的可能性；在这个例子里我们将使用这些指标中的前 8 个：
|
||||
|
||||
* **MDVP:Fo(Hz)：** 平均声带基频
|
||||
* **MDVP:Fhi(Hz)：** 最高声带基频
|
||||
* **MDVP:Flo(Hz)：** 最低声带基频
|
||||
* **MDVP:Jitter(%)**、**MDVP:Jitter(Abs)**、**MDVP:RAP**、**MDVP:PPQ** 和 **Jitter:DDP：** 5 个衡量声带基频变化的指标
|
||||
* **MDVP:Fo(Hz)**：平均声带基频
|
||||
* **MDVP:Fhi(Hz)**：最高声带基频
|
||||
* **MDVP:Flo(Hz)**：最低声带基频
|
||||
* **MDVP:Jitter(%)**、**MDVP:Jitter(Abs)**、**MDVP:RAP**、**MDVP:PPQ** 和 **Jitter:DDP**：5 个衡量声带基频变化的指标
|
||||
|
||||
|
||||
|
||||
这个例子里用到的数据集，可以在我的 [GitHub 仓库][12] 里找到。数据集已经事先做了打乱和分片。
|
||||
这个例子里用到的数据集，可以在我的 [GitHub 仓库][12] 里找到。数据集已经事先做了打乱和切片。
|
||||
|
||||
### 用 Python 实现机器学习
|
||||
|
||||
@@ -41,23 +41,18 @@
|
||||
* Pandas 1.1.1
|
||||
* scikit-learn 0.22.2.post1
|
||||
|
||||
|
||||
|
||||
Python 有多个朴素贝叶斯分类器的实现，都是开源的，包括：
|
||||
|
||||
* **NLTK Naïve Bayes:** 基于标准的朴素贝叶斯算法，用于文本分类
|
||||
* **NLTK Positive Naïve Bayes:** NLTK Naïve Bayes 的变体，用于对只标注了一部分的训练集进行二分类
|
||||
* **Scikit-learn Gaussian Naïve Bayes:** 提供了 partial fit 方法来支持数据流或很大的数据集 (它们都可能无法一次性导入内存，用 partial fit 可以动态地增加数据)
|
||||
* **Scikit-learn Multinomial Naïve Bayes:** 针对离散型特征变量作了优化
|
||||
* **Scikit-learn Bernoulli Naïve Bayes:** 用于各个特征都是二元变量的情况
|
||||
|
||||
|
||||
* **NLTK Naïve Bayes**：基于标准的朴素贝叶斯算法，用于文本分类
|
||||
* **NLTK Positive Naïve Bayes**：NLTK Naïve Bayes 的变体，用于对只标注了一部分的训练集进行二分类
|
||||
* **Scikit-learn Gaussian Naïve Bayes**：提供了部分拟合方法来支持数据流或很大的数据集（LCTT 译注：它们可能无法一次性导入内存，用部分拟合可以动态地增加数据）
|
||||
* **Scikit-learn Multinomial Naïve Bayes**：针对离散型特征、实例计数、频率等作了优化
|
||||
* **Scikit-learn Bernoulli Naïve Bayes**：用于各个特征都是二元变量/布尔特征的情况
|
||||
|
||||
在这个例子里我将使用 [sklearn Gaussian Naive Bayes][13]。
|
||||
|
||||
我的 Python 实现在 `naive_bayes_parkinsons.py` 里，如下所示：
|
||||
|
||||
|
||||
```
|
||||
import pandas as pd
|
||||
|
||||
@@ -108,7 +103,6 @@ print('Accuray score on test data:', accuracy_train)
|
||||
|
||||
运行这个 Python 脚本：
|
||||
|
||||
|
||||
```
|
||||
$ python naive_bayes_parkinsons.py
|
||||
|
||||
@@ -146,21 +140,19 @@ Accuracy score on test data: 0.6666666666666666
|
||||
|
||||
### 背后原理
|
||||
|
||||
朴素贝叶斯分类器从贝叶斯定理发展来的。贝叶斯定理用于计算条件概率，或者说贝叶斯定理用于计算当与当与一个事件相关联的其他事件发生时，该事件发生的概率。简而言之，它解决了这个问题：_如果我们已经知道事件 x 发生在事件 y 之前的概率，那么当事件 x 再次发生时，事件 y 发生的概率是多少？_ 贝叶斯定理用一个先验的预测值来逐渐逼近一个最终的 [后验概率][14]。贝叶斯定理有一个基本假设，就是所有的参数重要性相同 (即相互独立)。
|
||||
朴素贝叶斯分类器从贝叶斯定理发展来的。贝叶斯定理用于计算条件概率，或者说贝叶斯定理用于计算当与一个事件相关联的其他事件发生时，该事件发生的概率。简而言之，它解决了这个问题：_如果我们已经知道事件 x 发生在事件 y 之前的概率，那么当事件 x 再次发生时，事件 y 发生的概率是多少？_ 贝叶斯定理用一个先验的预测值来逐渐逼近一个最终的 [后验概率][14]。贝叶斯定理有一个基本假设，就是所有的参数重要性相同（LCTT 译注：即相互独立）。
|
||||
|
||||
贝叶斯计算主要包括以下步骤：
|
||||
|
||||
1. 计算总的先验概率：
|
||||
$P(患病)$ 和 $P(不患病)$
|
||||
2. 计算 8 种指标各自是某个值时的后验概率 (value1,...,value8 分别是 MDVP:Fo(Hz),...,Jitter:DDP 的取值)：
|
||||
2. 计算 8 种指标各自是某个值时的后验概率 (value1,...,value8 分别是 MDVP:Fo(Hz)，...，Jitter:DDP 的取值)：
|
||||
$P(value1,\ldots,value8\ |\ 患病)$
|
||||
$P(value1,\ldots,value8\ |\ 不患病)$
|
||||
3. 将第 1 步和第 2 步的结果相乘，最终得到患病和不患病的后验概率：
|
||||
$P(患病\ |\ value1,\ldots,value8) \propto P(患病) \times P(value1,\ldots,value8\ |\ 患病)$
|
||||
$P(不患病\ |\ value1,\ldots,value8) \propto P(不患病) \times P(value1,\ldots,value8\ |\ 不患病)$
|
||||
|
||||
|
||||
|
||||
上面第 2 步的计算非常复杂，朴素贝叶斯将它作了简化：
|
||||
|
||||
1. 计算总的先验概率：
|
||||
@@ -172,18 +164,14 @@ Accuracy score on test data: 0.6666666666666666
|
||||
$P(患病\ |\ value1,\ldots,value8) \propto P(患病) \times P(value1\ |\ 患病) \times \ldots \times P(value8\ |\ 患病)$
|
||||
$P(不患病\ |\ value1,\ldots,value8) \propto P(不患病) \times P(value1\ |\ 不患病) \times \ldots \times P(value8\ |\ 不患病)$
|
||||
|
||||
|
||||
|
||||
这只是一个很初步的解释，还有很多其他因素需要考虑，比如数据类型的差异，如何解析数据，数据可能有缺失值等。
|
||||
这只是一个很初步的解释，还有很多其他因素需要考虑，比如数据类型的差异，稀疏数据，数据可能有缺失值等。
|
||||
|
||||
### 超参数
|
||||
|
||||
朴素贝叶斯作为一个简单直接的算法，不需要超参数。然而，有的版本的朴素贝叶斯实现可能提供一些高级特性 (比如超参数)。比如，[GaussianNB][13] 就有 2 个超参数：
|
||||
|
||||
* **priors:** 先验概率，可以事先指定，这样就不必让算法从数据中计算才能得出。
|
||||
* **var_smoothing:** 考虑数据的分布情况，当数据不满足标准的高斯分布时，这个超参数会发挥作用。
|
||||
|
||||
朴素贝叶斯作为一个简单直接的算法，不需要超参数。然而，有的版本的朴素贝叶斯实现可能提供一些高级特性（比如超参数）。比如，[GaussianNB][13] 就有 2 个超参数：
|
||||
|
||||
* **priors**：先验概率，可以事先指定，这样就不必让算法从数据中计算才能得出。
|
||||
* **var_smoothing**：考虑数据的分布情况，当数据不满足标准的高斯分布时，这个超参数会发挥作用。
|
||||
|
||||
### 损失函数
|
||||
|
||||
@@ -191,10 +179,10 @@ Accuracy score on test data: 0.6666666666666666
|
||||
|
||||
### 优缺点
|
||||
|
||||
**优点:** 朴素贝叶斯是最简单最快速的算法之一。
|
||||
**优点:** 在数据量较少时，用朴素贝叶斯仍可作出可靠的预测。
|
||||
**缺点:** 朴素贝叶斯的预测只是估计值，并不准确。它胜在速度而不是准确度。
|
||||
**缺点:** 朴素贝叶斯有一个基本假设，就是所有特征相互独立，但现实情况并不总是如此。
|
||||
**优点**：朴素贝叶斯是最简单、最快速的算法之一。
|
||||
**优点**：在数据量较少时，用朴素贝叶斯仍可作出可靠的预测。
|
||||
**缺点**：朴素贝叶斯的预测只是估计值，并不准确。它胜在速度而不是准确度。
|
||||
**缺点**：朴素贝叶斯有一个基本假设，就是所有特征相互独立，但现实情况并不总是如此。
|
||||
|
||||
从本质上说，朴素贝叶斯是贝叶斯定理的推广。它是最简单最快速的机器学习算法之一，用来进行简单和快速的训练和预测。朴素贝叶斯提供了足够好、比较准确的预测。朴素贝叶斯假设预测特征之间是相互独立的。已经有许多朴素贝叶斯的开源的实现，它们的特性甚至超过了贝叶斯算法的实现。
|
||||
|
||||
@@ -205,7 +193,7 @@ via: https://opensource.com/article/21/1/machine-learning-python
|
||||
作者：[Girish Managoli][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[tanloong](https://github.com/tanloong)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
校对：[wxy](https://github.com/wxy)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
@@ -0,0 +1,232 @@
|
||||
[#]: subject: "Use XMLStarlet to parse XML in your the Linux terminal"
|
||||
[#]: via: "https://opensource.com/article/21/7/parse-xml-linux"
|
||||
[#]: author: "Seth Kenlon https://opensource.com/users/seth"
|
||||
[#]: collector: "lujun9972"
|
||||
[#]: translator: "zepoch"
|
||||
[#]: reviewer: "wxy"
|
||||
[#]: publisher: "wxy"
|
||||
[#]: url: "https://linux.cn/article-13627-1.html"
|
||||
|
||||
在命令行中使用 XMLStarlet 来解析 XML
|
||||
======
|
||||
|
||||
> 借助终端上的 XML 工具包 XMLStarlet，你就是 XML 之星。
|
||||
|
||||
|
||||
|
||||
学习解析 XML 通常被认为是一件复杂的事情，但它不一定是这样。[XML 是高度严格结构化的][2]，所以也是相对来说可预测的。也有许多其他工具可以帮助你使这项工作易于管理。
|
||||
|
||||
我最喜欢的 XML 实用程序之一是 [XMLStarlet][3]，这是一个用于终端的 XML 工具包，借助这个 XML 工具包，你可以验证、解析、编辑、格式化和转换 XML 数据。XMLStarLet 是个相对较小的命令，但浏览 XML 却充满潜力，因此本文演示了如何使用它来查询 XML 数据。
|
||||
|
||||
### 安装
|
||||
|
||||
XMLStarLet 默认安装在 CentOS、Fedora，和许多其他现代 Linux 发行版上，所以你可以打开终端，输入 `xmlstarlet` 来访问它。如果 XMLStarLet 还没有被安装，你的操作系统则会为你安装它。
|
||||
|
||||
或者，你可以用包管理器安装 `xmlstarlet`：
|
||||
|
||||
```
|
||||
$ sudo dnf install xmlstarlet
|
||||
```
|
||||
|
||||
在 macOS 上，可以使用 [MacPorts][4] 或 [Homebrew][5]。在 Windows 上，可以使用 [Chocolatey][6]。
|
||||
|
||||
如果都失败了，你可以从 [Sourceforge 上的源代码][7] 手动安装它。
|
||||
|
||||
### 用 XMLStarlet 解析 XML
|
||||
|
||||
有许多工具可以帮助解析和转换 XML 数据，包括允许你 [编写自己的解析器][8] 的软件库，和复杂的命令，如 `fop` 和 `xsltproc`。不过有时你不需要处理 XML 数据；你只需要一个方便的方法从 XML 数据中来提取、更新或验证重要数据。对于随手的 XML 交互，我使用 `xmlstarlet`，这是常见的处理 XML任务的一个典型的“瑞士军刀”式应用。通过运行 `--help` 命令，你可以看到它提供哪些选项：
|
||||
|
||||
```
|
||||
$ xmlstarlet --help
|
||||
Usage: xmlstarlet [<options>] <command> [<cmd-options>]
|
||||
where <command> is one of:
|
||||
ed (or edit) - Edit/Update XML document(s)
|
||||
sel (or select) - Select data or query XML document(s) (XPATH, etc)
|
||||
tr (or transform) - Transform XML document(s) using XSLT
|
||||
val (or validate) - Validate XML document(s) (well-formed/DTD/XSD/RelaxNG)
|
||||
fo (or format) - Format XML document(s)
|
||||
el (or elements) - Display element structure of XML document
|
||||
c14n (or canonic) - XML canonicalization
|
||||
ls (or list) - List directory as XML
|
||||
[...]
|
||||
```
|
||||
|
||||
你可以通过在这些子命令的末尾附加 `-help` 来获得进一步的帮助：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet sel --help
|
||||
-Q or --quiet - do not write anything to standard output.
|
||||
-C or --comp - display generated XSLT
|
||||
-R or --root - print root element <xsl-select>
|
||||
-T or --text - output is text (default is XML)
|
||||
-I or --indent - indent output
|
||||
[...]
|
||||
```
|
||||
|
||||
#### 用 sel 命令选择数据
|
||||
|
||||
可以使用 `xmlstarlet select`（简称 `sel`）命令查看 XML 格式的数据。下面是一个简单的 XML 文档：
|
||||
|
||||
```
|
||||
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
|
||||
<xml>
|
||||
<os>
|
||||
<linux>
|
||||
<distribution>
|
||||
<name>Fedora</name>
|
||||
<release>7</release>
|
||||
<codename>Moonshine</codename>
|
||||
<spins>
|
||||
<name>Live</name>
|
||||
<name>Fedora</name>
|
||||
<name>Everything</name>
|
||||
</spins>
|
||||
</distribution>
|
||||
|
||||
<distribution>
|
||||
<name>Fedora Core</name>
|
||||
<release>6</release>
|
||||
<codename>Zod</codename>
|
||||
<spins></spins>
|
||||
</distribution>
|
||||
</linux>
|
||||
</os>
|
||||
</xml>
|
||||
```
|
||||
|
||||
在 XML 文件中查找数据时，你的第一个任务是关注要探索的节点。如果知道节点的路径，请使用 `-value of` 选项指定完整路径。你越早浏览 [文档对象模型][9]（DOM）树，就可以看到更多信息：
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
--value-of /xml/os/linux/distribution \
|
||||
--nl myfile.xml
|
||||
Fedora
|
||||
7
|
||||
Moonshine
|
||||
|
||||
Live
|
||||
Fedora
|
||||
Everything
|
||||
|
||||
Fedora Core
|
||||
6
|
||||
Zod
|
||||
```
|
||||
|
||||
`--nl` 代表“新的一行”，它插入大量的空白，以确保在输入结果后，终端在新的一行显示。我已经删除了样本输出中的一些多余空间。
|
||||
|
||||
通过进一步深入 DOM 树来凝聚关注点：
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
--value-of /xml/os/linux/distribution/name \
|
||||
--nl myfile.xml
|
||||
Fedora
|
||||
Fedora Core
|
||||
```
|
||||
|
||||
#### 条件选择
|
||||
|
||||
用于导航和解析 XML 的最强大工具之一被称为 XPath。它规范了 XML 搜索中使用的语法，并从 XML 库调用函数。XMLStarlet 能够解析 XPath 表达式，因此可以使用 XPath 函数来有条件的进行选择。XPath 具有丰富的函数，[由 W3C 提供了详细文档][10]，但我觉得 [Mozilla 的 XPath 文档][11] 更简洁。
|
||||
|
||||
可以使用方括号作为测试函数，将元素的内容与某个值进行比较。下面是对 `<name>` 元素的值的测试，它仅返回与特定匹配相关联的版本号。
|
||||
|
||||
想象一下，示例 XML 文件包含以 1 开头的所有 Fedora 版本。要查看与旧名称 “Fedora Core” 关联的所有版本号（该项目从版本 7 开始删除了名称中的 “Core”），请执行以下操作：
|
||||
|
||||
```
|
||||
$ xmlstarlet sel --template \
|
||||
--value-of '/xml/os/linux/distribution[name = "Fedora Core"]/release' \
|
||||
--nl myfile.xml
|
||||
6
|
||||
5
|
||||
4
|
||||
3
|
||||
2
|
||||
1
|
||||
```
|
||||
|
||||
通过将路径的 `--value-of` 更改为 `/xml/os/linux/distribution[name=“Fedora Core”]/codename`，你便可以查看这些版本的所有代号。
|
||||
|
||||
### 匹配路径和获取目标值
|
||||
|
||||
将 XML 标记视为节点的一个好处是，一旦找到节点，就可以将其视为当前的数据的“目录”。它不是一个真正的目录，至少不是文件系统意义上的目录，但它是一个可以查询的数据集合。为了帮助你将目标和“里面”的数据分开，XMLStarlet 把你试图用 `--match` 选项匹配的内容和用 `--value-of` 选项匹配的数据值进行了区分。
|
||||
|
||||
假设你知道 `<spin>` 节点包含几个元素。这就是你的目标节点。一旦到了这里，就可以使用 `--value-of` 指定想要哪个元素的值。要查看所有元素，可以使用点（`.`）来代表当前位置：
|
||||
|
||||
```
|
||||
$ xmlstarlet sel --template \
|
||||
--match '/xml/os/linux/distribution/spin' \
|
||||
--value-of '.' --nl myfile.xml \
|
||||
Live
|
||||
Fedora
|
||||
Everything
|
||||
```
|
||||
|
||||
与浏览 DOM 一样，可以使用 XPath 表达式来限制返回数据的范围。在本例中，我使用 `last()` 函数来检索 `spin` 节点中的最后一个元素：
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
--match '/xml/os/linux/distribution/spin' \
|
||||
--value-of '*[last()]' --nl myfile.xml
|
||||
Everything
|
||||
```
|
||||
|
||||
在本例中，我使用 `position()` 函数选择 `spin` 节点中的特定元素：
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
--match '/xml/os/linux/distribution/spin' \
|
||||
--value-of '*[position() = 2]' --nl myfile.xml
|
||||
Fedora
|
||||
```
|
||||
|
||||
`--match` 和 `--value` 选项可以重叠，因此如何将它们一起使用取决于你自己。对于示例 XML，这两个表达式执行的是相同的操作：
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
--match '/xml/os/linux/distribution/spin' \
|
||||
--value-of '.' \
|
||||
--nl myfile.xml
|
||||
Live
|
||||
Fedora
|
||||
Everything
|
||||
|
||||
$ xmlstarlet select --template \
|
||||
--match '/xml/os/linux/distribution' \
|
||||
--value-of 'spin' \
|
||||
--nl myfile.xml
|
||||
Live
|
||||
Fedora
|
||||
Everything
|
||||
```
|
||||
|
||||
### 熟悉 XML
|
||||
|
||||
XML 有时看起来过于冗长和笨拙，但为与之交互和构建的工具却总是让我吃惊。如果你想要好好使用 XML，那么 XMLStarlet 可能是一个很好的切入点。下次要打开 XML 文件查看其结构化数据时，请尝试使用 XMLStarlet，看看是否可以查询这些数据。当你对 XML 越熟悉时，它就越能作为一种健壮灵活的数据格式而为你服务。
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/21/7/parse-xml-linux
|
||||
|
||||
作者：[Seth Kenlon][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[zepoch](https://github.com/zepoch)
|
||||
校对：[wxy](https://github.com/wxy)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/seth
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/lead-images/linux_penguin_green.png?itok=ENdVzW22 "Penguin with green background"
|
||||
[2]: https://opensource.com/article/21/6/what-xml
|
||||
[3]: https://en.wikipedia.org/wiki/XMLStarlet
|
||||
[4]: https://opensource.com/article/20/11/macports
|
||||
[5]: https://opensource.com/article/20/6/homebrew-mac
|
||||
[6]: https://opensource.com/article/20/3/chocolatey
|
||||
[7]: http://xmlstar.sourceforge.net
|
||||
[8]: https://opensource.com/article/21/6/parsing-config-files-java
|
||||
[9]: https://opensource.com/article/21/6/what-xml#dom
|
||||
[10]: https://www.w3.org/TR/1999/REC-xpath-19991116
|
||||
[11]: https://developer.mozilla.org/en-US/docs/Web/XPath/Functions
|
||||
|
||||
@@ -3,46 +3,45 @@
|
||||
[#]: author: (Sumantro Mukherjee https://opensource.com/users/sumantro)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (geekpi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
[#]: reviewer: (wxy)
|
||||
[#]: publisher: (wxy)
|
||||
[#]: url: (https://linux.cn/article-13631-1.html)
|
||||
|
||||
在虚拟环境中运行 Python 应用
|
||||
pipx：在虚拟环境中运行 Python 应用
|
||||
======
|
||||
通过使用 pipx 隔离运行 Python 应用来避免版本冲突并提高安全性。
|
||||
![Digital creative of a browser on the internet][1]
|
||||
|
||||
> 通过使用 pipx 隔离运行 Python 应用来避免版本冲突并提高安全性。
|
||||
|
||||
|
||||
|
||||
如果你使用 Python，你可能会安装很多 Python 应用。有些是你只想尝试的工具。还有一些是你每天都在使用的久经考验的应用，所以你把它们安装在你使用的每一台计算机上。这两种情况下，在虚拟环境中运行你的 Python 应用是非常有用的，这可以使它们以及它们的依赖关系相互分离，以避免版本冲突，并使它们与你系统的其它部分隔离，以提高安全性。
|
||||
|
||||
这就是 [pipx][2] 出场的地方。
|
||||
|
||||
大多数 Python 应用可以使用 [pip][3] 进行安装，它只安装 Python 包。然而，Pipx 为你的 Python 应用创建并管理一个虚拟环境，并帮助你运行它们。
|
||||
大多数 Python 应用可以使用 [pip][3] 进行安装，它只安装 Python 包。然而，`pipx` 为你的 Python 应用创建并管理一个虚拟环境，并帮助你运行它们。
|
||||
|
||||
### 安装 pipx
|
||||
|
||||
Pipx 主要是一个 RPM 包，你可以在任何 Fedora、RHEL 或 CentOS 机器上安装它：
|
||||
|
||||
`pipx` 主要是一个 RPM 包，你可以在任何 Fedora、RHEL 或 CentOS 机器上安装它：
|
||||
|
||||
```
|
||||
`$ sudo dnf install pipx`
|
||||
$ sudo dnf install pipx
|
||||
```
|
||||
|
||||
### 使用 pipx
|
||||
|
||||
我将通过 Cowsay 以及 [Concentration][4] 工具演示如何使用 pipx。
|
||||
我将通过 Cowsay 以及 [Concentration][4] 工具演示如何使用 `pipx`。
|
||||
|
||||
#### 安装软件包
|
||||
|
||||
安装完 pipx 后，你可以用以下方法安装 Python 包：
|
||||
|
||||
安装完 `pipx` 后，你可以用以下方法安装 Python 包：
|
||||
|
||||
```
|
||||
`$ pipx install <python_package>`
|
||||
$ pipx install <python_package>
|
||||
```
|
||||
|
||||
要安装 Cowsay 包：
|
||||
|
||||
|
||||
```
|
||||
$ pipx install cowsay ✔ │ 20:13:41
|
||||
installed package cowsay 4.0, Python 3.9.5
|
||||
@@ -55,28 +54,25 @@ done! ✨ 🌟 ✨
|
||||
|
||||
|
||||
```
|
||||
$ cowsay "I <3 OSDC"
|
||||
_________
|
||||
| I <3 OSDC |
|
||||
=========
|
||||
\
|
||||
\
|
||||
^__^
|
||||
(oo)\\_______
|
||||
(__)\ )\/\
|
||||
||----w |
|
||||
|| ||
|
||||
$ cowsay "I <3 OSDC"
|
||||
_________
|
||||
| I <3 OSDC |
|
||||
=========
|
||||
\
|
||||
\
|
||||
^__^
|
||||
(oo)\_______
|
||||
(__)\ )\/\
|
||||
||----w |
|
||||
|| ||
|
||||
```
|
||||
|
||||
![Cowsay][5]
|
||||
|
||||
（Sumantro Mukherjee, [CC BY-SA 4.0][6]）
|
||||
|
||||
#### 以特殊权限进行安装
|
||||
|
||||
不是所有的应用都像 Cowsay 一样简单。例如，Concentration 会与你系统中的许多其他组件交互，所以它需要特殊的权限。用以下方式安装它：
|
||||
|
||||
|
||||
```
|
||||
$ pipx install concentration ✔ │ 10s │ │ 20:26:12
|
||||
installed package concentration 1.1.5, Python 3.9.5
|
||||
@@ -85,8 +81,7 @@ $ pipx install concentration ✔ │ 10s
|
||||
done! ✨ 🌟 ✨
|
||||
```
|
||||
|
||||
Concentration 通过阻止 “distractors” 文件中列出的特定网站来帮助你集中注意力。要做到这点，它需要以 sudo 或 root 权限运行。你可以用 [OpenDoas][7] 来做到这点，这是 `doas` 命令的一个版本，可以用特定的用户权限运行任何命令。要使用 `doas` 以 sudo 权限来运行 Concentration：
|
||||
|
||||
Concentration 通过阻止 `distractors` 文件中列出的特定网站来帮助你集中注意力。要做到这点，它需要以 `sudo` 或 root 权限运行。你可以用 [OpenDoas][7] 来做到这点，这是 `doas` 命令的一个版本，可以用特定的用户权限运行任何命令。要使用 `doas` 以 `sudo` 权限来运行 Concentration：
|
||||
|
||||
```
|
||||
$ doas concentration improve ✔ │ │ 20:26:54
|
||||
@@ -98,8 +93,7 @@ Concentration is now improved :D!
|
||||
|
||||
#### 列出已安装的应用
|
||||
|
||||
`pipx list` 命令显示所有用 pipx 安装的应用和它们的可执行路径：
|
||||
|
||||
`pipx list` 命令显示所有用 `pipx` 安装的应用和它们的可执行路径：
|
||||
|
||||
```
|
||||
$ pipx list
|
||||
@@ -113,16 +107,14 @@ apps are exposed on your $PATH at /home/sumantrom/.local/bin
|
||||
|
||||
#### 卸载应用
|
||||
|
||||
当你使用完毕后，知道如何卸载它们是很重要的。Pipx 有一个非常简单的卸载命令：
|
||||
|
||||
当你使用完毕后，知道如何卸载它们是很重要的。`pipx` 有一个非常简单的卸载命令：
|
||||
|
||||
```
|
||||
`$ pipx uninstall <package name>`
|
||||
$ pipx uninstall <package name>
|
||||
```
|
||||
|
||||
或者你可以删除每个软件包：
|
||||
|
||||
|
||||
```
|
||||
$ pipx uninstall-all
|
||||
|
||||
@@ -133,9 +125,9 @@ uninstalled concentration! ✨ 🌟 ✨
|
||||
|
||||
### 尝试 pipx
|
||||
|
||||
Pipx 是一个流行的 Python 应用的包管理器。它可以访问 [PyPi][8] 上的所有东西，但它也可以从包含有效 Python 包的本地目录、Python wheel 或网络位置安装应用。
|
||||
`pipx` 是一个流行的 Python 应用的包管理器。它可以访问 [PyPi][8] 上的所有东西，但它也可以从包含有效 Python 包的本地目录、Python wheel 或网络位置安装应用。
|
||||
|
||||
如果你安装了大量的 Python 应用，可以试试 pipx。
|
||||
如果你安装了大量的 Python 应用，可以试试 `pipx`。
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
@@ -144,7 +136,7 @@ via: https://opensource.com/article/21/7/python-pipx
|
||||
作者：[Sumantro Mukherjee][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[geekpi](https://github.com/geekpi)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
校对：[wxy](https://github.com/wxy)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
@@ -3,42 +3,46 @@
|
||||
[#]: author: (Ankush Das https://itsfoss.com/author/ankush/)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (geekpi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
[#]: reviewer: (wxy)
|
||||
[#]: publisher: (wxy)
|
||||
[#]: url: (https://linux.cn/article-13629-1.html)
|
||||
|
||||
在 Linux 上运行和测试 Android 应用的顶级 Android 模拟器
|
||||
Linux 上的顶级安卓模拟器
|
||||
======
|
||||
|
||||
|
||||
|
||||
安卓系统是建立在高度定制的 Linux 内核之上的。因此，使用安卓模拟器在 Linux 上运行移动应用是有意义的。
|
||||
|
||||
虽然这不是在 Linux 机器上你可以做的新事情，但在 Windows 于 2021 年推出运行安卓应用的能力后，它是一个更需要的功能。
|
||||
虽然这不是在 Linux 机器上你可以做的新鲜事，但在 Windows 于 2021 年推出运行安卓应用的能力后，它是一个更需要的功能。
|
||||
|
||||
不仅仅限于使用应用，一些安卓模拟器也可以在开发和测试中派上用场。
|
||||
|
||||
因此，我总结了一份最好的模拟器清单，你可以用它来测试或在 Linux 上运行 Android 应用/游戏。
|
||||
因此，我总结了一份最好的模拟器清单，你可以用它来测试或在 Linux 上运行安卓应用/游戏。
|
||||
|
||||
### 1\. Anbox
|
||||
### 1、Anbox
|
||||
|
||||
Anbox 是一个相当流行的让 Linux 用户运行 Android 应用的模拟器。可能 Deepin Linux 正是利用它使得开箱即可运行 Android 应用。
|
||||
|
||||
|
||||
Anbox 是一个相当流行的模拟器，可以让 Linux 用户运行 Android 应用。可能深度 Linux 正是利用它使得开箱即可运行安卓应用。
|
||||
|
||||
它使用一个容器将安卓操作系统与主机隔离，这也让它们可以使用最新的安卓版本。
|
||||
|
||||
运行的安卓应用将不能直接访问你的硬件，这是一个很好的安全决定。
|
||||
运行的安卓应用不能直接访问你的硬件，这是一个很好的安全决定。
|
||||
|
||||
与这里的其他一些选项不同，Anbox 在技术上不需要仿真层来使安卓系统工作。换句话说，它在你的 Linux 系统上最接近于原生的 Android 体验。
|
||||
与这里的其他一些选择不同，Anbox 在技术上不需要仿真层来使安卓系统工作。换句话说，它在你的 Linux 系统上最接近于原生的安卓体验。
|
||||
|
||||
由于这个原因，它可能不是最简单的选择。你不能只使用谷歌应用商店来安装应用，你需要使用安卓调试桥（ADB）。你只需要一个应用的 APK 文件就可以安装和使用它。
|
||||
|
||||
[Anbox][1]
|
||||
- [Anbox][1]
|
||||
|
||||
### 2\. Genymotion
|
||||
### 2、Genymotion
|
||||
|
||||
![][2]
|
||||
|
||||
Genymotion 是一个为测试和开发量身定做的令人印象深刻的解决方案。
|
||||
|
||||
它不是一个免费和开源的选择。它们通过云端或独立于 Android Studio 的桌面客户端，提供虚拟的 Android 体验作为服务。
|
||||
它不是一个自由开源的选择。它们通过云端或独立于 Android Studio 的桌面客户端，提供虚拟的安卓体验作为服务。
|
||||
|
||||
你可以模拟各种硬件配置和安卓版本，创建一个虚拟设备进行测试。它还让你有能力扩大规模，并有多个安卓虚拟设备运行，进行广泛的测试。
|
||||
|
||||
@@ -46,63 +50,63 @@ Genymotion 是一个为测试和开发量身定做的令人印象深刻的解决
|
||||
|
||||
虽然它是一个主要针对专业人士的高级解决方案，但它确实支持最新的 Linux 发行版，包括 Ubuntu 20.04 LTS。
|
||||
|
||||
[Genymotion][3]
|
||||
- [Genymotion][3]
|
||||
|
||||
### 3\. Android-x86
|
||||
### 3、Android-x86
|
||||
|
||||
![][4]
|
||||
|
||||
Android x86 是一个开源项目，使得 Android 在 PC 上运行，并支持 32 位。
|
||||
Android x86 是一个开源项目，使得安卓可以在 PC 上运行，并支持 32 位。
|
||||
|
||||
你可以选择在你的 Linux 系统上使用虚拟机管理器来安装它，或者直接在你的 PC 上试用它。
|
||||
|
||||
如果你需要安装，可以查看官方的[安装说明][5]。
|
||||
如果你需要安装，可以查看官方的 [安装说明][5]。
|
||||
|
||||
与其他一些选项不同，它是一个简单的试图在 PC 上工作的模拟器，没有花哨的功能。
|
||||
与其他一些选择不同，它是一个简单的试图在 PC 上工作的模拟器，没有花哨的功能。
|
||||
|
||||
[Android x86][6]
|
||||
- [Android x86][6]
|
||||
|
||||
### 4\. Android Studio （虚拟设备）
|
||||
### 4、Android Studio （虚拟设备）
|
||||
|
||||
![][7]
|
||||
|
||||
Android Studio 是一个用于开发和测试的完整工具。幸运的是，由于对 Linux 的支持，如果你需要的话，你可以用它来模拟 Android 的体验进行实验。
|
||||
Android Studio 是一个用于开发和测试的完整工具。幸运的是，由于对 Linux 的支持，如果你需要的话，你可以用它来模拟安卓的体验进行实验。
|
||||
|
||||
你只需要创建一个安卓虚拟设备（AVD），你可以对其进行配置，然后作为模拟器进行模拟。
|
||||
|
||||
也有很大的机会找到对一些最新的智能手机、电视和智能手表的支持。
|
||||
也有很大的机会能找到对一些最新的智能手机、电视和智能手表的支持。
|
||||
|
||||
它需要一定的学习曲线才能上手，但它是免费的，而且是完全开源的。
|
||||
|
||||
[Android Studio][8]
|
||||
- [Android Studio][8]
|
||||
|
||||
### 5\. ARChon
|
||||
### 5、ARChon
|
||||
|
||||
![][9]
|
||||
|
||||
一个有趣的解决方案是一个你可以在 Linux 和任何其他平台上使用的 Android 模拟器。
|
||||
一个有趣的解决方案，这是一个你可以在 Linux 和任何其他平台上使用的安卓模拟器。
|
||||
|
||||
它有助于在 Chrome OS 上运行 Android 应用，或者在任何操作系统上使用 Chrome 浏览器。与其他一些不同的是，你可能不会得到完整的安卓体验，而只是能够运行安卓应用。
|
||||
它有助于在 Chrome OS 上运行安卓应用，或者在任何操作系统上使用 Chrome 浏览器。与其他一些不同的是，你可能不会得到完整的安卓体验，而只是能够运行安卓应用。
|
||||
|
||||
你只需解压运行时，并将其加载到 Chrome 扩展中。接下来，下载 APK 文件到上面来添加你想使用的应用。
|
||||
|
||||
[ARChon][10]
|
||||
- [ARChon][10]
|
||||
|
||||
### 6\. Bliss OS
|
||||
### 6、Bliss OS
|
||||
|
||||
![][11]
|
||||
|
||||
Bliss OS 是另一个开源项目，与 Android x86 类似，旨在使 Android 在 PC 上运行。
|
||||
Bliss OS 是另一个开源项目，与 Android x86 类似，旨在使安卓在 PC 上运行。
|
||||
|
||||
与 Android x86 不同的是，它通过支持 32 位和 64 位架构提供了更多的兼容性选择。此外，你可以根据你的处理器下载兼容文件。
|
||||
|
||||
它有积极的维护，支持市场上最新的安卓版本。
|
||||
|
||||
[Bliss OS][12]
|
||||
- [Bliss OS][12]
|
||||
|
||||
### 总结
|
||||
|
||||
虽然你会发现有几个可用于 Linux 的安卓模拟器，但它们可能无法取代全面的智能手机体验。
|
||||
虽然你会能找到几个可用于 Linux 的安卓模拟器，但它们可能无法取代全面的智能手机体验。
|
||||
|
||||
每个模拟器都有一系列的功能和特定目的。请选择你需要的那个！
|
||||
|
||||
@@ -115,7 +119,7 @@ via: https://itsfoss.com/android-emulators-linux/
|
||||
作者：[Ankush Das][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[geekpi](https://github.com/geekpi)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
校对：[wxy](https://github.com/wxy)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
@@ -1,179 +0,0 @@
|
||||
[#]: subject: (F(r)iction: Or How I Learnt to Stop Worrying and Start Loving Vim)
|
||||
[#]: via: (https://news.itsfoss.com/how-i-started-loving-vim/)
|
||||
[#]: author: (Theena https://news.itsfoss.com/author/theena/)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (piaoshi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
F(r)iction: Or How I Learnt to Stop Worrying and Start Loving Vim
|
||||
======
|
||||
|
||||
It is Dec 2009, and I am ready to quit my job.
|
||||
|
||||
I wanted to focus on writing my first book; neither my commitments at work nor the state of technology was helping.
|
||||
|
||||
Writing is hard work.
|
||||
|
||||
Few tasks in the modern world can be as singular – or as daunting – a pursuit as sitting down in front of a blank piece of paper, and asking your brain to vomit out words that communicate an idea to readers. I am not suggesting that writing can’t be collaborative of course, but merely illustrating how daunting it can be for writers to set off on a new piece by themselves. This is true for fiction
|
||||
and non-fiction writing, but since I am a novelist I’d like to focus primarily on fiction in this article.
|
||||
|
||||
![][1]
|
||||
|
||||
Remember what 2009 was like?
|
||||
|
||||
Smart phones were 3 years old – I still hadn’t gone away from feature phones. Laptops were big and bulky. Meanwhile, cloud-based web applications for productivity was in their infancy, and just not good. Technologically speaking, writers like me were using their Gmail accounts (and a very young cloud-based storage called Dropbox) as an always-available option to work on their drafts, even while away from my personal computer. While this was a nice change from what writers had to go through when working with typewriters (or god forbid, pen and paper), it wasn’t much.
|
||||
|
||||
For one thing, version control of manuscripts was a nightmare. Further, the more tools I added to my tool kit to simplify the workflow, the more I had to switch context – both from a UI and a UX sense.
|
||||
|
||||
I would start writing drafts on Windows Notepad, save it on a MS Word Document on my PC at home, email myself a copy, keep another copy on Dropbox (since Dropbox wasn’t accessible at work), work on the copy of that file at work, email it back to myself at the end of the day, download it on the home computer, saving it under a new name and the respective date so that I would recognize the changes in the file were made at work (as opposed to home)…well you get the picture. If you think this workflow involving Windows Notepad, MS Word, Gmail, and Dropbox is insane, well now you know why I quit my job.
|
||||
|
||||
More soberingly, I still know writers, damn good writers too, who use variations of the workflow that I followed in 2009.
|
||||
|
||||
Over the next three years, I worked on the manuscript, completing the first draft in 2012. During the three years much had changed with the state of technology. Smart phones were actually pretty great, and some of the complications I had in 2009 had disappeared. I could still work on the same file that I had been working from at home, on my phone (not necessarily fresh writing, but editing had become considerably easier thanks to Dropbox on the phone.) My main writing tool remained Microsoft’s Windows Notepad and Word, which is how I completed the first draft.
|
||||
|
||||
The novel [**First Utterance**][2] was released in 2016 to critical and commercial acclaim.
|
||||
|
||||
The end.
|
||||
|
||||
Or so I thought.
|
||||
|
||||
As soon as I completed the manuscript and sent it to my editor, I had begun working on the second novel. I was no longer quitting my job to work on writing, but I had taken a more pragmatic approach: I’d take two weeks off at the end of ever year so that I could go to a little cabin in the mountains to write.
|
||||
|
||||
It took me half a day to realize that the things that annoyed me about my [writing tools][3] and workflow had not disappeared, but morphed into a more complex beast. As a writer, I wasn’t being productive or as efficient as I wanted.
|
||||
|
||||
### Linux in the time of Corona
|
||||
|
||||
![][4]
|
||||
|
||||
It is 2020 and the world is on the verge of mass hysteria.
|
||||
|
||||
What had started out as an isolated novel virus in China was morphing into the first global pandemic since 1911. On March 20th, Sri Lanka followed most of the rest of the world and shutdown.
|
||||
|
||||
April in Sri Lanka is the height of the dry season. Temperatures in concrete jungles like Colombo can reach the mid 30s, with humidity in the high 90s. It can drive most people to distraction at the best of times, but stuck at home with no always-on air conditioning while a global pandemic is underway? That is a good recipe for madness.
|
||||
|
||||
My madness was Linux or, as we in the open source community call it, ‘distro-hopping’.
|
||||
|
||||
The more I played around with *nix distros, the more enamoured I came to be with the idea of control. When nothing seems to be within our control – not even the simple act of shaking hands with another person – then it is only natural we lean towards things where we feel more in control.
|
||||
|
||||
Where better to get more control in my life than with my computing? Naturally, this extended to my writing tools and workflow too.
|
||||
|
||||
### The path to Vim
|
||||
|
||||
There’s a joke about [Vim][5] that describes perfectly my first experience with it. People are obsessive about Vim because they don’t know how to close it.
|
||||
|
||||
I was attempting to edit a configuration file, and the [fresh install of Ubuntu Server][6] had only Vim pre-installed. First there was panic – so much so I restarted the machine thinking the OS wasn’t picking up my keyboard. Then when it happened again, the inevitable Google search: ‘[How do I close vim?][7]‘
|
||||
|
||||
_Oh. That’s interesting_, I thought.
|
||||
|
||||
_But why?_
|
||||
|
||||
To understand why I was even remotely interested in a text editor that was too complex to close, you have to understand how much I adore Windows Notepad.
|
||||
|
||||
As a writer, I loved writing on its no-nonsense, no buttons, white-abyss like canvas. It had no spell check. It had no formatting. But I didn’t care.
|
||||
|
||||
For the writer in me, Notepad was the best writing scratch pad ever devised. Unfortunately, it isn’t powerful – so even if I start writing my drafts in Notepad, I would move it to MS Word once I had passed a 1000 words – Notepad wasn’t built for prose, and those limitations would be glaringly obvious when I passed that word limit.
|
||||
|
||||
So the first thing I installed I moved all my computing away from Windows, was a good text editor.
|
||||
|
||||
[Kate][8] was the first replacement where I felt more comfortable than I did on Windows Notepad – it was more powerful (it had spell-checker!), and hey, I could mess around with some hobbyist-type coding in the same environment.
|
||||
|
||||
It was love.
|
||||
|
||||
But then Vim happened.
|
||||
|
||||
The more I learnt about Vim, the more I watched developers live coding on Vim, the more I found myself opening Vim for my text editing needs. I use the phrase ‘text editing’ in the traditional Unix sense: editing blocks of text in configuration files, or sometimes writing basic Bash scripts.
|
||||
|
||||
I still hadn’t used Vim remotely for my prose writing needs.
|
||||
|
||||
For that I had LibreOffice.
|
||||
|
||||
Sort of.
|
||||
|
||||
While it is an adequate [replacement for MS Office][9], I found myself underwhelmed. The UI is perhaps even more distracting than MS Word, and with each distro having different packages of LibreOffice, I found myself using a hellishly fragmented tool kit and workflow, to say nothing about how different the UI can look in various distros and desktop environments.
|
||||
|
||||
Things had become even more complicated because I had also started my Masters. In this scenario, I was taking notes down on Kate, transferring them to LibreOffice, and then saving it on to my Dropbox.
|
||||
|
||||
Context switching was staring at me in the face every day.
|
||||
|
||||
Productivity dropped as I had to open and close a number of unrelated applications. I needed one writing tool to meet all my needs – as a novelist, as a student, and as a hobbyist coder.
|
||||
|
||||
And that’s when I realized that the solution to my context switching nightmare was also staring at me in the face at the same time.
|
||||
|
||||
By this point, I had used Vim often enough – even used it with Termux on my Android phone – to be pretty comfortable with the idea of moving everything to Vim. Since it supported markdown syntax, note-taking would also become even easier.
|
||||
|
||||
This was just about two months ago.
|
||||
|
||||
How am I doing?
|
||||
|
||||
It is April 2021.
|
||||
|
||||
I started this draft on my phone, [using Vim][10] via Termux (with the aid of a Bluetooth keyboard), while in a taxi. I pushed the file to a GitHub private repo for my writing, from which I pulled the file to my PC, wrote a few more lines, before heading out again. I pulled the new version of the file from GitHub to my phone, made changes, pushed it, repeat, until I emailed the final draft to the editor.
|
||||
|
||||
The context switching is now no more.
|
||||
|
||||
The distractions that come from writing in word processors is no more.
|
||||
|
||||
Editing is infinitely easier, and faster.
|
||||
|
||||
My wrists are no longer in pain because I hid my mouse from sight.
|
||||
|
||||
It is April 2021.
|
||||
|
||||
I am a novelist.
|
||||
|
||||
And I write on Vim.
|
||||
|
||||
How? I’ll discuss the specific of this workflow in the second part of this column series on how non-tech people are using free and open source technology. Stay tuned.
|
||||
|
||||
![][11]
|
||||
|
||||
I'm not interested
|
||||
|
||||
#### _Related_
|
||||
|
||||
* [Going Against Google Analytics With Plausible's Co-Founder [Interview]][12]
|
||||
* ![][13] ![Interview with Plausible founder Marco Saric][14]
|
||||
|
||||
|
||||
* [The Progress Linux has Made in Terms of Gaming is Simply Incredible: Lutris Creator][15]
|
||||
* ![][13] ![][16]
|
||||
|
||||
|
||||
* [Multi Monitor and HiDPI Setup is Looking Better on Ubuntu 21.04 [My Experience So Far]][17]
|
||||
* ![][13] ![Multimonitor setup with Ubuntu 21.04][18]
|
||||
|
||||
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://news.itsfoss.com/how-i-started-loving-vim/
|
||||
|
||||
作者：[Theena][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://news.itsfoss.com/author/theena/
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: data:image/svg+xml;base64,PHN2ZyBoZWlnaHQ9JzM5MCcgd2lkdGg9Jzc4MCcgeG1sbnM9J2h0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnJyB2ZXJzaW9uPScxLjEnLz4=
|
||||
[2]: https://www.goodreads.com/book/show/29616237-first-utterance
|
||||
[3]: https://itsfoss.com/open-source-tools-writers/
|
||||
[4]: data:image/svg+xml;base64,PHN2ZyBoZWlnaHQ9JzM5NScgd2lkdGg9Jzc4MCcgeG1sbnM9J2h0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnJyB2ZXJzaW9uPScxLjEnLz4=
|
||||
[5]: https://www.vim.org/
|
||||
[6]: https://itsfoss.com/install-ubuntu-server-raspberry-pi/
|
||||
[7]: https://itsfoss.com/how-to-exit-vim/
|
||||
[8]: https://kate-editor.org/
|
||||
[9]: https://itsfoss.com/best-free-open-source-alternatives-microsoft-office/
|
||||
[10]: https://linuxhandbook.com/basic-vim-commands/
|
||||
[11]: data:image/svg+xml;base64,PHN2ZyBoZWlnaHQ9JzI1MCcgd2lkdGg9Jzc1MCcgeG1sbnM9J2h0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnJyB2ZXJzaW9uPScxLjEnLz4=
|
||||
[12]: https://news.itsfoss.com/marko-saric-plausible/
|
||||
[13]: data:image/svg+xml;base64,PHN2ZyBoZWlnaHQ9JzIwMCcgd2lkdGg9JzM1MCcgeG1sbnM9J2h0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnJyB2ZXJzaW9uPScxLjEnLz4=
|
||||
[14]: https://i1.wp.com/news.itsfoss.com/wp-content/uploads/2021/04/Interview-plausible.jpg?fit=1200%2C675&ssl=1&resize=350%2C200
|
||||
[15]: https://news.itsfoss.com/lutris-creator-interview/
|
||||
[16]: https://i0.wp.com/news.itsfoss.com/wp-content/uploads/2021/03/lutris-interview-ft.png?fit=1200%2C675&ssl=1&resize=350%2C200
|
||||
[17]: https://news.itsfoss.com/ubuntu-21-04-multi-monitor-support/
|
||||
[18]: https://i1.wp.com/news.itsfoss.com/wp-content/uploads/2021/03/multi-monitor-ubuntu-21-itsfoss.jpg?fit=1200%2C675&ssl=1&resize=350%2C200
|
||||
@@ -1,90 +0,0 @@
|
||||
[#]: subject: (Install Shutter in Fedora 34 and Above)
|
||||
[#]: via: (https://www.debugpoint.com/2021/07/install-shutter-fedora/)
|
||||
[#]: author: (Arindam https://www.debugpoint.com/author/admin1/)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (geekpi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
Install Shutter in Fedora 34 and Above
|
||||
======
|
||||
This quick guide explains the steps required to install shutter in
|
||||
Fedora 34 and above.
|
||||
There are many alternatives and options for screenshot tool. But in my personal opinion, there is none that come close to the flexibility of Shutter. Unfortunately over the years, Linux distributions such as Ubuntu, Fedora faced problem packaging this application in official repo for various dependency problem and specially how it is designed.
|
||||
|
||||
The primary problem is it still based on GTK2 and Perl. While most of the application moved to GTK3, it remained in GTK2. This created a dependency problem as Debian/Ubuntu, Fedora they removed the dependent GTK2 version of certain packages.
|
||||
|
||||
Installing [Shutter][1] screenshot tool in Fedora 34 and above requires an alternative approach.
|
||||
|
||||
Right now, you can only install this tool via personal package archive only. Here’s how to install it in Fedora 34 and above.
|
||||
|
||||
![Shutter in Fedora][2]
|
||||
|
||||
### Install Shutter in Fedora 34 and above
|
||||
|
||||
Open a terminal in your Fedora and enable the following [Copr repository for Shutter][3]. This package archive provides a separate build for Shutter for Fedora with all the unmet dependencies.
|
||||
|
||||
```
|
||||
sudo dnf copr enable geraldosimiao/shutter
|
||||
```
|
||||
|
||||
Once this is done, you can simply install shutter in Fedora 34 and above by dnf.
|
||||
|
||||
```
|
||||
sudo dnf install shutter
|
||||
```
|
||||
|
||||
Though the latest release is v0.97 at present. The repository sadly contains the old v0.94.x at the moment. I hope the repo owner soon include the latest version.
|
||||
|
||||
After installation, you can launch it via application menu.
|
||||
|
||||
#### Uninstall Shutter
|
||||
|
||||
If you want, you can easily remove this third-party repository via command:
|
||||
|
||||
```
|
||||
sudo dnf copr remove geraldosimiao/shutter
|
||||
```
|
||||
|
||||
And followed by below to remove Shutter completely including dependencies.
|
||||
|
||||
```
|
||||
sudo dnf autoremove shutter
|
||||
```
|
||||
|
||||
#### Install Shutter in Other Linux Distributions
|
||||
|
||||
If you want to install it in Debian, Ubuntu or related distribution – [check out this guide][4].
|
||||
|
||||
[][5]
|
||||
|
||||
SEE ALSO: How to fix disabled Edit Image Option in Shutter in Ubuntu 18.04 LTS
|
||||
|
||||
### Development of Shutter
|
||||
|
||||
Recently this project [moved to GitHub][6] for better collaboration and a GTK3 port is underway. And it is fairly active and recently had a release. We hope it gets ported to gtk3 soon and available in native repo of respective distributions.
|
||||
|
||||
Let me know using the comment box below if you face any error installing Shutter.
|
||||
|
||||
* * *
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://www.debugpoint.com/2021/07/install-shutter-fedora/
|
||||
|
||||
作者：[Arindam][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://www.debugpoint.com/author/admin1/
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://www.debugpoint.com/tag/shutter
|
||||
[2]: https://www.debugpoint.com/blog/wp-content/uploads/2021/07/Shutter-in-Fedora.jpeg
|
||||
[3]: https://copr.fedorainfracloud.org/coprs/geraldosimiao/shutter/
|
||||
[4]: https://www.debugpoint.com/2020/04/shutter-install-ubuntu/
|
||||
[5]: https://www.debugpoint.com/2018/05/shutter-fix-disable-edit/
|
||||
[6]: https://github.com/shutter-project/shutter
|
||||
@@ -2,7 +2,7 @@
|
||||
[#]: via: (https://opensource.com/article/21/7/cron-linux)
|
||||
[#]: author: (Seth Kenlon https://opensource.com/users/seth)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: translator: (perfiffer)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
@@ -2,7 +2,7 @@
|
||||
[#]: via: (https://opensource.com/article/21/7/linux-kernel-ftrace)
|
||||
[#]: author: (Gaurav Kamathe https://opensource.com/users/gkamathe)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: translator: (mengxinayan)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
@@ -390,7 +390,7 @@ via: https://opensource.com/article/21/7/linux-kernel-ftrace
|
||||
|
||||
作者：[Gaurav Kamathe][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
译者：[萌新阿岩](https://github.com/mengxinayan)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
@@ -1,110 +0,0 @@
|
||||
[#]: subject: (Check used disk space on Linux with du)
|
||||
[#]: via: (https://opensource.com/article/21/7/check-disk-space-linux-du)
|
||||
[#]: author: (Seth Kenlon https://opensource.com/users/seth)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (geekpi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
Check used disk space on Linux with du
|
||||
======
|
||||
Find out how much disk space you're using with the Linux du command.
|
||||
![Check disk usage][1]
|
||||
|
||||
No matter how much storage space you have, there's always the possibility for it to fill up. On most personal devices, drives get filled up with photos and videos and music, but on servers, it's not unusual for space to diminish due to data in user accounts and log files. Whether you're in charge of managing a multi-user system or just your own laptop, you can check in on disk usage with the `du` command.
|
||||
|
||||
By default, `du` provides the amount of disk space used in your current directory, as well as the size of each subdirectory:
|
||||
|
||||
|
||||
```
|
||||
$ du
|
||||
12 ./.backups
|
||||
60 .
|
||||
```
|
||||
|
||||
In this example, my current directory takes up all of 60 KB, 12 KB of which is occupied by the subdirectory `.backups`.
|
||||
|
||||
If you find that confusing and would prefer to see all sizes separately, you can use the `--separate-dirs` (or `-S` for short) option:
|
||||
|
||||
|
||||
```
|
||||
$ du --separate-dirs
|
||||
12 ./.backups
|
||||
48 .
|
||||
```
|
||||
|
||||
It's the same information (48 and 12 is 60) but each directory is treated independently of one another.
|
||||
|
||||
To see even more detail, use the --all (or -a for short) option, which displays each file in each directory:
|
||||
|
||||
|
||||
```
|
||||
$ du --separate-dirs --all
|
||||
4 ./example.adoc
|
||||
28 ./graphic.png
|
||||
4 ./.backups/example.adoc~
|
||||
12 ./.backups
|
||||
4 ./index.html
|
||||
4 ./index.adoc
|
||||
48 .
|
||||
```
|
||||
|
||||
### See modification time of files
|
||||
|
||||
When looking through files to find out what's taking up space, it can be useful to see when a file was last modified. Something that hasn't been touched in a year is a likely candidate for archival, especially if you're running out of space.
|
||||
|
||||
To see modification times of files with du, use the `--time` option:
|
||||
|
||||
|
||||
```
|
||||
$ du --separate-dirs --all --time
|
||||
28 2021-07-21 11:12 ./graphic.png
|
||||
4 2021-07-03 10:43 ./example.adoc
|
||||
4 2021-07-13 13:03 ./index.html
|
||||
4 2021-07-23 14:18 ./index.adoc
|
||||
48 2021-07-23 14:19 .
|
||||
```
|
||||
|
||||
### Set a threshold for file size
|
||||
|
||||
When reviewing files in the interest of disk space, you may only care about files of nontrivial size. You set a threshold for the file sizes you want to see with the `--threshold` (or `-t` for short) option. For instance, to view only sizes larger than 1 GB:
|
||||
|
||||
|
||||
```
|
||||
$ \du --separate-dirs --all --time --threshold=1G ~/Footage/
|
||||
1839008 2021-07-14 13:55 /home/tux/Footage/snowfall.mp4
|
||||
1577980 2020-04-11 13:10 /home/tux/Footage/waterfall.mp4
|
||||
8588936 2021-07-14 13:55 /home/tux/Footage/
|
||||
```
|
||||
|
||||
When file sizes get particularly large, they can be difficult to read. Make file sizes easier with the `--human-readable` (or `-h` for short) option:
|
||||
|
||||
|
||||
```
|
||||
$ \du --separate-dirs --all --time \
|
||||
\--threshold=1G --human-readable ~/Footage/
|
||||
1.8G 2021-07-14 13:55 /home/tux/Footage/snowfall.mp4
|
||||
1.6G 2020-04-11 13:10 /home/tux/Footage/waterfall.mp4
|
||||
8.5G 2021-07-14 13:55 /home/tux/Footage/
|
||||
```
|
||||
|
||||
### See available disk space
|
||||
|
||||
To just get a summary of how much disk space remains on a drive, read our article about the [df command][2].
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/21/7/check-disk-space-linux-du
|
||||
|
||||
作者：[Seth Kenlon][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/seth
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/du-splash.png?itok=nRLlI-5A (Check disk usage)
|
||||
[2]: https://opensource.com/article/21/7/use-df-check-free-disk-space-linux
|
||||
@@ -2,7 +2,7 @@
|
||||
[#]: via: (https://www.debugpoint.com/2021/07/change-lock-login-screen-background-elementary-os/)
|
||||
[#]: author: (Arindam https://www.debugpoint.com/author/admin1/)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: translator: (geekpi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
@@ -0,0 +1,289 @@
|
||||
[#]: subject: (Create your own custom Raspberry Pi image)
|
||||
[#]: via: (https://opensource.com/article/21/7/custom-raspberry-pi-image)
|
||||
[#]: author: (Stephan Avenwedde https://opensource.com/users/hansic99)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
Create your own custom Raspberry Pi image
|
||||
======
|
||||
Build a Raspberry Pi image from scratch or convert your running,
|
||||
modified Raspberry Pi OS back to an image others can use.
|
||||
![Vector, generic Raspberry Pi board][1]
|
||||
|
||||
When I recently read [Alan Formy-Duval's][2] article [_Manage your Raspberry Pi with Cockpit_][3], I thought it would be a good idea to have an image with Cockpit already preinstalled. Luckily there are at least two ways to accomplish this task:
|
||||
|
||||
* Adapt the sources of the Raspberry Pi OS image building toolchain [pi-gen][4] which enables you to build a Raspberry Pi image from scratch
|
||||
* Convert your running, modified Raspberry Pi OS back to an image others can use
|
||||
|
||||
|
||||
|
||||
This article covers both methods. I'll highlight the pros and cons of each technique.
|
||||
|
||||
### Pi-gen
|
||||
|
||||
Let's begin with [pi-gen][4]. Before we start, there are a few prerequisites you'll need to consider.
|
||||
|
||||
#### Prerequisites
|
||||
|
||||
To successfully run the build process, it is recommended to use a 32bit version of Debian Buster or Ubuntu Xenial. It may work on other systems as well but to avoid unnecessary complications, I recommend to setup a virtual machine with one of the recommended systems. If you are not familiar with virtual machines, take a look at my article [Try Linux on any operating system with VirtualBox][5]. When you have everything up and running, also install the dependencies mentioned in the [repository description][6]. Also consider that you need internet access in the virtual machine and enough free disk space. I set up my virtual machine with a 40GB hard drive which seemed to be enough.
|
||||
|
||||
In order to follow the instructions in this article, make a clone of the [pi-gen][4] repository or fork it if you want to start developing you own image.
|
||||
|
||||
#### Repository Overview
|
||||
|
||||
The overall build process is separated into stages. Each stage is represented as an ordinary folder and represents a logical intermediate with regards to a full Raspberry Pi OS image.
|
||||
|
||||
* **Stage 0**: Bootstrap—Creates a usable filesystem
|
||||
* **Stage 1**: Minimal system—Creates an absolute minimal system
|
||||
* **Stage 2**: Lite system—Corresponds to Raspberry Pi OS Lite
|
||||
* **Stage 3**: Desktop system—Installs X11, LXDE, web browsers, and so on
|
||||
* **Stage 4**: Corresponds to an ordinary Raspberry Pi OS
|
||||
* **Stage 5**: Corresponds to Raspberry Pi OS Full
|
||||
|
||||
|
||||
|
||||
The stages build upon each other: It is not possible to build a higher stage without building the lower stages. You can't leave out a stage in the middle either. For example, to build a Raspberry Pi OS Lite, you have to build stages 0, 1, and 2. To build a Raspberry Pi OS with a desktop, you have to build stages 0, 1, 2, 3, 4, and 5.
|
||||
|
||||
#### Build process
|
||||
|
||||
The build process is controlled by the `build.sh`, which can be found in the root repository. If you already know how to read and write bash scripts, it won't be a hurdle to understand the process defined there. If not, reading the `build.sh` and trying to understand what is going on is a really good practice. But even without bash scripting skills, you will be able to create your own image with Cockpit preinstalled.
|
||||
|
||||
In general, the build process consists of several nested for-loops.
|
||||
|
||||
* **stage-loop:** Loop through all stage directories in ascending order
|
||||
* Skip further processing if a file named _SKIP_ is found
|
||||
* Run the script `prerun.sh`
|
||||
* **sub-loop:** Loop through each subdirectory in ascending order and process the following files if they are present:
|
||||
* * `00-run-sh`: Arbitrary instructions to run in advance
|
||||
* `00-run-chroot.sh`: Run this script in the chroot directory of the image
|
||||
* `00-debconfs`: Variables for the` debconf-set-selection`
|
||||
* `00-packages`: A list of packages to install
|
||||
* `00-packages-nr`: Similar to the _00-packages_, except that this will cause the installation with --no-install-recommends -y parameter to _apt-get_
|
||||
* `00-patches`: A directory containing patch files to be applied, using [quilt][7]
|
||||
|
||||
* Back in the stage-loop, if a file named `EXPORT_IMAGE` is found, generate an image for this stage
|
||||
|
||||
* If a file named `SKIP_IMAGE` is found, skip creating the image
|
||||
|
||||
|
||||
|
||||
|
||||
The `build.sh `also requires a file named `config` containing some specification which is read on startup.
|
||||
|
||||
#### Hands-On
|
||||
|
||||
First, we will create a basic Raspberry Pi OS Lite image. The Raspberry Pi OS Lite image will act as a base for our custom image. Create an empty file named _config_ and add the following two lines:
|
||||
|
||||
|
||||
```
|
||||
IMG_NAME='Cockpit'
|
||||
ENABLE_SSH=1
|
||||
```
|
||||
|
||||
Create an empty file named `SKIP` in the directories `stage3`, `stage4`, and `stage5`. `Stages 4` and `5` emit an image by default, therefore add an empty file named `SKIP_IMAGE` in `stage4` and `stage5`.
|
||||
|
||||
Now open a terminal and switch to the root user by typing `su`. Navigate to the root directory of the repository and start the build script by typing `./build.sh`.
|
||||
|
||||
The build process will take some time.
|
||||
|
||||
After the build process has finished, you will find two more directories in the root of the repository: `work `and `deploy`. The `work` folder contains some intermediate output. In the `deploy` folder you should find the zipped image file, ready for deployment.
|
||||
|
||||
If the overall build process was successful, we now can modify the process so that it installs Cockpit additionally.
|
||||
|
||||
#### Extending the build process
|
||||
|
||||
The Raspberry Pi OS Lite image acts as the base for our Cockpit installation. As the Raspberry Pi OS Lite image is complete with `stage2`, we will create our own `stage3` which will handle the Cockpit installation.
|
||||
|
||||
We remove the original `stage3` completely and create a new, empty `stage3`:
|
||||
|
||||
|
||||
```
|
||||
`rm -rf stage3 && mkdir stage3`
|
||||
```
|
||||
|
||||
Inside `stage3`, we create a substage for installing cockpit:
|
||||
|
||||
|
||||
```
|
||||
`mkdir stage3/00-cockpit`
|
||||
```
|
||||
|
||||
To install cockpit on the image, we simply need to add it to the package list:
|
||||
|
||||
|
||||
```
|
||||
`echo "cockpit" >> stage3/00-cockpit/00-packages`
|
||||
```
|
||||
|
||||
We also want to configure our new `stage3` to output an image, therefore we simply add this file in the `stage3` directory:
|
||||
|
||||
|
||||
```
|
||||
`touch stage3/EXPORT_IMAGE`
|
||||
```
|
||||
|
||||
As there are already intermediate images from the previous build process, we can prevent that the stages are built again by adding `skip-files` in the related directories:
|
||||
|
||||
Skip the build process for `stage0` and `stage1`:
|
||||
|
||||
|
||||
```
|
||||
`touch stage0/SKIP && touch stage1/SKIP`
|
||||
```
|
||||
|
||||
Skip the build process for `stage2` and also skip the image creation:
|
||||
|
||||
|
||||
```
|
||||
`touch stage2/SKIP && touch stage2/SKIP_IMAGE`
|
||||
```
|
||||
|
||||
Now run the build script again:
|
||||
|
||||
|
||||
```
|
||||
`./build.sh`
|
||||
```
|
||||
|
||||
In the folder `deployment` you now should find a zipped image `<date>-Cockpit-lite.zip`, which is ready for deployment.
|
||||
|
||||
#### Troubleshooting
|
||||
|
||||
If you try to apply more complex modifications, there is a lot of trial and error involved in building your own Raspberry Pi image with pi-gen. You will certainly face that the build process will stop in between for some reason. As there is no exception handling in the build process, we do have some cleanup manually in case the process stopped.
|
||||
|
||||
It is likely that the `chroot` file system is still mounted after the process stopped. You won't be able to start a new build process without unmounting it. In the case it is still mounted, unmount it manually by typing:
|
||||
|
||||
|
||||
```
|
||||
`umount work/<Build-date-&-image-name>/tmpimage/`
|
||||
```
|
||||
|
||||
Another issue I determined was that the script stopped when the `chroot` filesystem was about to be unmounted. In the file `scripts/qcow2_handling`, you can see that directly before the attempt to unmount `sync` is called. `Sync` forces the system to flush the write buffer. Running the build system as a virtual machine, the write process was not ready when `unmount` was called so the script stopped here.
|
||||
|
||||
To solve this, I just inserted a `sleep` between `sync` and `unmount` which solved the issue:
|
||||
|
||||
![Sleep in between sync and unmount - core dump example][8]
|
||||
|
||||
(I know that 30 seconds are overkill but as the whole build process takes > 20 minutes, 30 seconds are just a drop in the ocean)
|
||||
|
||||
### Modify existing image
|
||||
|
||||
In contrast to building an image with `pi-gen`, you could also directly apply the modification on a running Raspberry Pi OS. In our scenario, simply log in and install Cockpit with the following command:
|
||||
|
||||
|
||||
```
|
||||
`sudo apt install cockpit`
|
||||
```
|
||||
|
||||
Now shut down your Raspberry Pi, take out the SD card, and connect it to your PC. Check if your system has automatically mounted the partitions on the SD card by typing `lsblk -p`:
|
||||
|
||||
![Using lsblk -p to check mounting partitions][9]
|
||||
|
||||
In the screenshot above, the SD card is the device `/dev/sdc` and the `boot`\- and `rootfs`-partitions were automatically mounted at the mentioned mount points. Before you proceed, unmount them with :
|
||||
|
||||
|
||||
```
|
||||
`umount /dev/sdc1 && umount /dev/sdc2`
|
||||
```
|
||||
|
||||
Now we copy the contents of the SD card to our file system. Make sure you have enough disk space available as the image will have the same size as the SD card. Start the copy process with the following command:
|
||||
|
||||
|
||||
```
|
||||
`dd if=/dev/sdc of=~/MyImage.img bs=32M`
|
||||
```
|
||||
|
||||
![Copying image from the SD card][10]
|
||||
|
||||
Once the copy process is finished, we can shrink the image with the [PiShrink][11]. Follow the installation instructions mentioned in the repository which are:
|
||||
|
||||
|
||||
```
|
||||
wget <https://raw.githubusercontent.com/Drewsif/PiShrink/master/pishrink.sh>
|
||||
chmod +x pishrink.sh
|
||||
sudo mv pishrink.sh /usr/local/bin
|
||||
```
|
||||
|
||||
Now invoke the script by typing:
|
||||
|
||||
|
||||
```
|
||||
`sudo pishrink.sh ~/MyImage.img`
|
||||
```
|
||||
|
||||
![Invoking the pishrink.sh script][12]
|
||||
|
||||
PiShrink reduced the image size by almost a factor of ten: From the former 30GB to 3.5GB. You can still optimize the size by zipping it before you upload or share it.
|
||||
|
||||
That's it, you are now able to share and flash this image.
|
||||
|
||||
### Flashing the image
|
||||
|
||||
If you want to flash your own custom Raspberry Pi image back to the SD card using Linux, follow the steps below.
|
||||
|
||||
Put the SD card into your PC. Your system will likely automatically mount the filesystem on the SD card if there is already a previous installation. You can check this by opening a command line and typing `lsblk -p`:
|
||||
|
||||
![Checking automatic mounting with lsblk -p][13]
|
||||
|
||||
As you can see in the screenshot above, my system automatically mounted two filesystems, `boot` and `rootfs` as this SD card already contained a Raspberry Pi OS. Before we start flashing the SD card we have to unmount the file systems first by typing:
|
||||
|
||||
|
||||
```
|
||||
`umount /dev/sdc1 && umount /dev/sdc2`
|
||||
```
|
||||
|
||||
The output of `lsblk -p` should look like this in order to proceed:
|
||||
|
||||
![Output of lsblk -p][14]
|
||||
|
||||
Now you can flash the image to the SD card: Open a command line and type:
|
||||
|
||||
|
||||
```
|
||||
`dd if=/path/to/image.img of=/dev/sdc bs=32M, conv=fsync`
|
||||
```
|
||||
|
||||
With `bs=32M`, you specify that the SD card is written in 32-megabyte blocks, `conv=fsync` forces the process to physically write each block.
|
||||
|
||||
If successful, you should see this output:
|
||||
|
||||
![Successful output example][15]
|
||||
|
||||
Done! You can now put the SD card back into the Raspberry Pi and boot it.
|
||||
|
||||
### Summary
|
||||
|
||||
Both of the techniques presented in this article have their advantages and disadvantages. Whereas using `pi-gen` to create your own custom Raspberry Pi images is more error-prone than simply modifying an existing image, it is the method of choice if you plan to set up a [CICD pipeline][16]. My personal favorite is clearly to modify an existing image as you are directly able to make sure that the changes you applied are working.
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/21/7/custom-raspberry-pi-image
|
||||
|
||||
作者：[Stephan Avenwedde][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/hansic99
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/lead-images/raspberrypi_board_vector_red.png?itok=yaqYjYqI (Vector, generic Raspberry Pi board)
|
||||
[2]: https://opensource.com/users/alanfdoss
|
||||
[3]: https://opensource.com/article/21/5/raspberry-pi-cockpit
|
||||
[4]: https://github.com/RPi-Distro/pi-gen
|
||||
[5]: https://opensource.com/article/21/6/try-linux-virtualbox
|
||||
[6]: https://github.com/RPi-Distro/pi-gen/blob/master/README.md#Dependencies
|
||||
[7]: https://man7.org/linux/man-pages/man1/quilt.1.html
|
||||
[8]: https://opensource.com/sites/default/files/uploads/1_pi_gen_sleep.png (Sleep in between sync and unmount - core dump example)
|
||||
[9]: https://opensource.com/sites/default/files/uploads/pi_gen_lsblk_mounted.png (Using lsblk -p to check mounting partitions)
|
||||
[10]: https://opensource.com/sites/default/files/uploads/rpi_image_copy.png (Copying image from the SD card)
|
||||
[11]: https://github.com/Drewsif/PiShrink
|
||||
[12]: https://opensource.com/sites/default/files/uploads/rpi_pishrink.png (Invoking the pishrink.sh script)
|
||||
[13]: https://opensource.com/sites/default/files/uploads/pi_gen_lsblk_mounted_0.png (Checking automatic mounting with lsblk -p)
|
||||
[14]: https://opensource.com/sites/default/files/uploads/pi_gen_lsblk_unmounted2.png (Output of lsblk -p)
|
||||
[15]: https://opensource.com/sites/default/files/uploads/pi_gen_flash.png (Successful output example)
|
||||
[16]: https://en.wikipedia.org/wiki/CI/CD
|
||||
@@ -0,0 +1,305 @@
|
||||
[#]: subject: (Getting started with Maxima in Fedora Linux)
|
||||
[#]: via: (https://fedoramagazine.org/getting-started-with-maxima-in-fedora-linux/)
|
||||
[#]: author: (Jagat Kafle https://fedoramagazine.org/author/jkafle/)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
Getting started with Maxima in Fedora Linux
|
||||
======
|
||||
|
||||
![][1]
|
||||
|
||||
Photo by [Roman Mager][2] on [Unsplash][3]
|
||||
|
||||
[Maxima][4] is an open source computer algebra system (CAS) with powerful symbolic, numerical, and graphical capabilities. You can perform matrix operations, differentiation, integration, solve ordinary differential equations as well as plot functions and data in two and three dimensions. As such, it is helpful for anyone interested in science and math. This article goes through installing and using Maxima in Fedora Linux.
|
||||
|
||||
### Installing Maxima
|
||||
|
||||
Maxima is a command line system. You can install Maxima from the official Fedora repository using the following command:
|
||||
|
||||
```
|
||||
sudo dnf install maxima
|
||||
```
|
||||
|
||||
You can then use Maxima from the terminal by invoking the command _maxima_.
|
||||
|
||||
![Maxima session in gnome terminal in Fedora Linux 34][5]
|
||||
|
||||
### Installing wxMaxima
|
||||
|
||||
[wxMaxima][6] is a document based interface for Maxima. To install it in Fedora Linux, use the following command:
|
||||
|
||||
```
|
||||
sudo dnf install wxmaxima
|
||||
```
|
||||
|
||||
You can launch wxMaxima either by invoking the command _wxmaxima_ in the terminal or clicking its application icon from the app grid or menu.
|
||||
|
||||
![wxMaxima session in Fedora Linux 34][7]
|
||||
|
||||
### Basic Commands
|
||||
|
||||
After calling _maxima_, you should see terminal output as in the [figure above][8].
|
||||
|
||||
The _(%i1)_ is the input label where you enter the commands. Command in Maxima is an expression that can span over many lines and is closed with a semicolon (;). The _o_ labels denote the outputs. Comments are enclosed between _/*_ and _*/_. You can use the special symbol percent _(%)_ to refer to the immediately preceding result computed by Maxima. If you don’t want to print a result, you can finish your command with _$_ instead of _;_. Here are basic arithmetic commands in Maxima:
|
||||
|
||||
```
|
||||
(%i1) (19 + 7)/(52 - 2 * 13);
|
||||
(%o1) 1
|
||||
(%i2) 127 / 5;
|
||||
127
|
||||
(%o2) ---
|
||||
5
|
||||
(%i3) float (127 / 5);
|
||||
(%o3) 25.4
|
||||
(%i4) 127.0 / 5;
|
||||
(%o4) 25.4
|
||||
(%i5) sqrt(2.0);
|
||||
(%o5) 1.414213562373095
|
||||
(%i6) sin(%pi/2);
|
||||
(%o6) 1
|
||||
(%i7) abs(-12);
|
||||
(%o7) 12
|
||||
(%i8) 2+3%i + 5 - 4%i; /*complex arithmetic*/
|
||||
(%o8) 7 - %i
|
||||
```
|
||||
|
||||
To end the Maxima session, type the command:
|
||||
|
||||
```
|
||||
quit();
|
||||
```
|
||||
|
||||
### Algebra
|
||||
|
||||
Maxima can expand and factor polynomials:
|
||||
|
||||
```
|
||||
(%i1) (x+y)^3 + (x+y)^2 + (x+y);
|
||||
3 2
|
||||
(%o1) (y + x) + (y + x) + y + x
|
||||
(%i2) expand(%);
|
||||
3 2 2 2 3 2
|
||||
(%o2) y + 3 x y + y + 3 x y + 2 x y + y + x + x + x
|
||||
(%i3) factor(%);
|
||||
2 2
|
||||
(%o3) (y + x) (y + 2 x y + y + x + x + 1)
|
||||
```
|
||||
|
||||
To substitute _y_ with _z_ and _x_ with _5,_ refer the output label above and use the following command:
|
||||
|
||||
```
|
||||
(%i4) %o3, y=z, x=5;
|
||||
2
|
||||
(%o4) (z + 5) (z + 11 z + 31)
|
||||
```
|
||||
|
||||
You can easily manipulate trigonometric identities:
|
||||
|
||||
```
|
||||
(%i1) sin(x) * cos(x+y)^2;
|
||||
2
|
||||
(%o1) sin(x) cos (y + x)
|
||||
(%i2) trigexpand(%);
|
||||
2
|
||||
(%o2) sin(x) (cos(x) cos(y) - sin(x) sin(y))
|
||||
(%i3) trigreduce(%o1);
|
||||
sin(2 y + 3 x) - sin(2 y + x) sin(x)
|
||||
(%o3) ----------------------------- + ------
|
||||
4 2
|
||||
```
|
||||
|
||||
You can also solve algebraic equations in one or more variables:
|
||||
|
||||
```
|
||||
(%i1) solve(x^2+5*x+6);
|
||||
(%o1) [x = - 3, x = - 2]
|
||||
(%i2) solve(x^3 + 1);
|
||||
sqrt(3) %i - 1 sqrt(3) %i + 1
|
||||
(%o2) [x = - --------------, x = --------------, x = - 1]
|
||||
2 2
|
||||
(%i3) eqns: [x^2 + y^2 = 9, x + y = 3];
|
||||
2 2
|
||||
(%o3) [y + x = 9, y + x = 3]
|
||||
(%i4) solve(eqns, [x,y]);
|
||||
(%o4) [[x = 3, y = 0], [x = 0, y = 3]]
|
||||
```
|
||||
|
||||
### Calculus
|
||||
|
||||
Define _f_ to be a function of _x._ You can then find the limit, derivative and integral of the function:
|
||||
|
||||
```
|
||||
(%i1) f: x^2;
|
||||
2
|
||||
(%o1) x
|
||||
(%i2) limit(f,x,0);
|
||||
(%o2) 0
|
||||
(%i3) limit(1/f,x,0);
|
||||
(%o3) inf
|
||||
(%i4) diff(f, x);
|
||||
(%o4) 2 x
|
||||
(%i5) integrate(f, x);
|
||||
3
|
||||
x
|
||||
(%o5) --
|
||||
3
|
||||
```
|
||||
|
||||
To find definite integrals, slightly modify the syntax above.
|
||||
|
||||
```
|
||||
(%i6) integrate(f, x, 1, inf);
|
||||
defint: integral is divergent.
|
||||
-- an error. To debug this try: debugmode(true);
|
||||
(%i7) integrate(1/f, x, 1, inf);
|
||||
(%o7) 1
|
||||
```
|
||||
|
||||
Maxima can perform Taylor expansion. Here’s the Taylor expansion of sin(x) up to order 5 terms.
|
||||
|
||||
```
|
||||
(%i1) taylor(sin(x), x, 0, 5);
|
||||
3 5
|
||||
x x
|
||||
(%o1)/T/ x - -- + --- + . . .
|
||||
6 120
|
||||
```
|
||||
|
||||
To represent derivatives in unevaluated form, use the following syntax.
|
||||
|
||||
```
|
||||
(%i2) 'diff(y,x);
|
||||
dy
|
||||
(%o2) --
|
||||
dx
|
||||
```
|
||||
|
||||
The ode2 function can solve first and second order ordinary differential equations (ODEs).
|
||||
|
||||
```
|
||||
(%i1) 'diff(y,x,2) + y = 0;
|
||||
2
|
||||
d y
|
||||
(%o1) --- + y = 0
|
||||
2
|
||||
dx
|
||||
(%i2) ode2(%o1,y,x);
|
||||
(%o2) y = %k1 sin(x) + %k2 cos(x)
|
||||
```
|
||||
|
||||
### Matrix Operations
|
||||
|
||||
To enter a matrix, use the entermatrix function. Here’s an example of a general 2×2 matrix.
|
||||
|
||||
```
|
||||
(%i1) A: entermatrix(2,2);
|
||||
Is the matrix 1. Diagonal 2. Symmetric 3. Antisymmetric 4. General
|
||||
Answer 1, 2, 3 or 4 :
|
||||
4;
|
||||
Row 1 Column 1:
|
||||
1;
|
||||
Row 1 Column 2:
|
||||
2;
|
||||
Row 2 Column 1:
|
||||
3;
|
||||
Row 2 Column 2:
|
||||
4;
|
||||
Matrix entered.
|
||||
[ 1 2 ]
|
||||
(%o1) [ ]
|
||||
[ 3 4 ]
|
||||
```
|
||||
|
||||
You can then find the determinant, transpose, inverse, eigenvalues and eigenvectors of the matrix.
|
||||
|
||||
```
|
||||
(%i2) determinant(A);
|
||||
(%o2) - 2
|
||||
(%i3) transpose(A);
|
||||
[ 1 3 ]
|
||||
(%o3) [ ]
|
||||
[ 2 4 ]
|
||||
(%i4) invert(A);
|
||||
[ - 2 1 ]
|
||||
[ ]
|
||||
(%o4) [ 3 1 ]
|
||||
[ - - - ]
|
||||
[ 2 2 ]
|
||||
(%i5) eigenvectors(A);
|
||||
sqrt(33) - 5 sqrt(33) + 5
|
||||
(%o5) [[[- ------------, ------------], [1, 1]],
|
||||
2 2
|
||||
sqrt(33) - 3 sqrt(33) + 3
|
||||
[[[1, - ------------]], [[1, ------------]]]]
|
||||
4 4
|
||||
```
|
||||
|
||||
In the output label _(%o5)_ the first array gives the eigenvalues, the second array gives the multiplicity of the respective eigenvalues, and the next two arrays give the corresponding eigenvectors of the matrix A.
|
||||
|
||||
### Plotting
|
||||
|
||||
Maxima can use either [Gnuplot][9], [Xmaxima][10] or [Geomview][11] as graphics program. Maxima package in Fedora Linux comes with _gnuplot_ as a dependency, so Maxima uses _gnuplot_pipes_ as the plotting format. To check the plotting format, use the following command inside Maxima.
|
||||
|
||||
```
|
||||
get_plot_option(plot_format);
|
||||
```
|
||||
|
||||
Below are some plotting examples.
|
||||
|
||||
```
|
||||
(%i1) plot2d([sin(x), cos(x)], [x, -2%pi, 2%pi]);
|
||||
```
|
||||
|
||||
![2d plot using Maxima][12]
|
||||
|
||||
```
|
||||
(%i2) plot3d(sin(sqrt(x^2+y^2)), [x, -7, 7], [y, -7, 7]);
|
||||
```
|
||||
|
||||
![3d plot using Maxima][13]
|
||||
|
||||
```
|
||||
(%i3) mandelbrot ([iterations, 30], [x, -2, 1], [y, -1.2, 1.2],
|
||||
[grid,400,400]);
|
||||
```
|
||||
|
||||
![The Mandelbrot Set][14]
|
||||
|
||||
You can read more about Maxima and its capabilities in its [official website][15] and [documentation][16].
|
||||
|
||||
Fedora Linux has plethora of tools for scientific use. You can find the widely used ones in the [Fedora Scientific Guide][17].
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://fedoramagazine.org/getting-started-with-maxima-in-fedora-linux/
|
||||
|
||||
作者：[Jagat Kafle][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://fedoramagazine.org/author/jkafle/
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://fedoramagazine.org/wp-content/uploads/2021/07/Getting-started-with-Maxima-in-Fedora-Linux-816x345.png
|
||||
[2]: https://unsplash.com/@roman_lazygeek
|
||||
[3]: https://unsplash.com/s/photos/mathematics-tasks
|
||||
[4]: https://maxima.sourceforge.io/
|
||||
[5]: https://fedoramagazine.org/wp-content/uploads/2021/07/maxima-terminal.png
|
||||
[6]: https://wxmaxima-developers.github.io/wxmaxima/index.html
|
||||
[7]: https://fedoramagazine.org/wp-content/uploads/2021/07/wxmaxima.png
|
||||
[8]: tmp.LH5pctTy1x#maxima-terminal
|
||||
[9]: http://www.gnuplot.info/
|
||||
[10]: https://maxima.sourceforge.io/docs/xmaxima/xmaxima.html
|
||||
[11]: http://www.geomview.org/
|
||||
[12]: https://fedoramagazine.org/wp-content/uploads/2021/07/2d-maxima.png
|
||||
[13]: https://fedoramagazine.org/wp-content/uploads/2021/07/3d-maxima.png
|
||||
[14]: https://fedoramagazine.org/wp-content/uploads/2021/07/mandelbrot-maxima.png
|
||||
[15]: https://maxima.sourceforge.io/index.html
|
||||
[16]: https://maxima.sourceforge.io/docs/manual/maxima_toc.html
|
||||
[17]: https://fedora-scientific.readthedocs.io/en/latest/index.html
|
||||
376
sources/tech/20210728 Kernel tracing with trace-cmd.md
Normal file
376
sources/tech/20210728 Kernel tracing with trace-cmd.md
Normal file
@@ -0,0 +1,376 @@
|
||||
[#]: subject: (Kernel tracing with trace-cmd)
|
||||
[#]: via: (https://opensource.com/article/21/7/linux-kernel-trace-cmd)
|
||||
[#]: author: (Gaurav Kamathe https://opensource.com/users/gkamathe)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
Kernel tracing with trace-cmd
|
||||
======
|
||||
trace-cmd is an easy-to-use, feature-rich utility for tracing Linux
|
||||
kernel functions.
|
||||
![Puzzle pieces coming together to form a computer screen][1]
|
||||
|
||||
In my [previous article][2], I explained how to use `ftrace` to trace kernel functions. Using `ftrace` by writing and reading from files can get tedious, so I used a wrapper around it to run commands with options to enable and disable tracing, set filters, view output, clear output, and more.
|
||||
|
||||
The [trace-cmd][3] command is a utility that helps you do just this. In this article, I use `trace-cmd` to perform the same tasks I did in my `ftrace` article. Since I refer back to that article frequently, I recommend you read it before you read this one.
|
||||
|
||||
### Install trace-cmd
|
||||
|
||||
I run the commands in this article as the root user.
|
||||
|
||||
The `ftrace` mechanism is built into the kernel, and you can verify it is enabled with:
|
||||
|
||||
|
||||
```
|
||||
# mount | grep tracefs
|
||||
none on /sys/kernel/tracing type tracefs (rw,relatime,seclabel)
|
||||
```
|
||||
|
||||
However, you need to install the `trace-cmd` utility manually.
|
||||
|
||||
|
||||
```
|
||||
`# dnf install trace-cmd -y`
|
||||
```
|
||||
|
||||
### List available tracers
|
||||
|
||||
When using `ftrace`, you must view a file's contents to see what tracers are available. But with `trace-cmd`, you can get this information with:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd list -t
|
||||
hwlat blk mmiotrace function_graph wakeup_dl wakeup_rt wakeup function nop
|
||||
```
|
||||
|
||||
### Enable the function tracer
|
||||
|
||||
In my [earlier article][2], I used two tracers, and I'll do the same here. Enable your first tracer, `function`, with:
|
||||
|
||||
|
||||
```
|
||||
$ trace-cmd start -p function
|
||||
plugin 'function'
|
||||
```
|
||||
|
||||
### View the trace output
|
||||
|
||||
Once the tracer is enabled, you can view the output by using the `show` arguments. This shows only the first 20 lines to keep the example short (see my earlier article for an explanation of the output):
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd show | head -20
|
||||
## tracer: function
|
||||
#
|
||||
# entries-in-buffer/entries-written: 410142/3380032 #P:8
|
||||
#
|
||||
# _-----=> irqs-off
|
||||
# / _----=> need-resched
|
||||
# | / _---=> hardirq/softirq
|
||||
# || / _--=> preempt-depth
|
||||
# ||| / delay
|
||||
# TASK-PID CPU# |||| TIMESTAMP FUNCTION
|
||||
# | | | |||| | |
|
||||
gdbus-2606 [004] ..s. 10520.538759: __msecs_to_jiffies <-rebalance_domains
|
||||
gdbus-2606 [004] ..s. 10520.538760: load_balance <-rebalance_domains
|
||||
gdbus-2606 [004] ..s. 10520.538761: idle_cpu <-load_balance
|
||||
gdbus-2606 [004] ..s. 10520.538762: group_balance_cpu <-load_balance
|
||||
gdbus-2606 [004] ..s. 10520.538762: find_busiest_group <-load_balance
|
||||
gdbus-2606 [004] ..s. 10520.538763: update_group_capacity <-update_sd_lb_stats.constprop.0
|
||||
gdbus-2606 [004] ..s. 10520.538763: __msecs_to_jiffies <-update_group_capacity
|
||||
gdbus-2606 [004] ..s. 10520.538765: idle_cpu <-update_sd_lb_stats.constprop.0
|
||||
gdbus-2606 [004] ..s. 10520.538766: __msecs_to_jiffies <-rebalance_domains
|
||||
```
|
||||
|
||||
### Stop tracing and clear the buffer
|
||||
|
||||
Tracing continues to run in the background, and you can keep viewing the output using `show`.
|
||||
|
||||
To stop tracing, run `trace-cmd` with the `stop` argument:
|
||||
|
||||
|
||||
```
|
||||
`# trace-cmd stop`
|
||||
```
|
||||
|
||||
To clear the buffer, run it with the `clear` argument:
|
||||
|
||||
|
||||
```
|
||||
`# trace-cmd clear`
|
||||
```
|
||||
|
||||
### Enable the function_graph tracer
|
||||
|
||||
Enable the second tracer, `function_graph`, by running:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd start -p function_graph
|
||||
plugin 'function_graph'
|
||||
```
|
||||
|
||||
Once again, view the output using the `show` argument. As expected, the output is slightly different from the first trace output. This time it includes a `function calls` chain:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd show | head -20
|
||||
## tracer: function_graph
|
||||
#
|
||||
# CPU DURATION FUNCTION CALLS
|
||||
# | | | | | | |
|
||||
4) 0.079 us | } /* rcu_all_qs */
|
||||
4) 0.327 us | } /* __cond_resched */
|
||||
4) 0.081 us | rcu_read_unlock_strict();
|
||||
4) | __cond_resched() {
|
||||
4) 0.078 us | rcu_all_qs();
|
||||
4) 0.243 us | }
|
||||
4) 0.080 us | rcu_read_unlock_strict();
|
||||
4) | __cond_resched() {
|
||||
4) 0.078 us | rcu_all_qs();
|
||||
4) 0.241 us | }
|
||||
4) 0.080 us | rcu_read_unlock_strict();
|
||||
4) | __cond_resched() {
|
||||
4) 0.079 us | rcu_all_qs();
|
||||
4) 0.235 us | }
|
||||
4) 0.095 us | rcu_read_unlock_strict();
|
||||
4) | __cond_resched() {
|
||||
```
|
||||
|
||||
Use the `stop` and `clear` commands to stop tracing and clear the buffer:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd stop
|
||||
# trace-cmd clear
|
||||
```
|
||||
|
||||
### Tweak tracing to increase depth
|
||||
|
||||
If you want to see more depth in the function calls, you can tweak the tracer:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd start -p function_graph --max-graph-depth 5
|
||||
plugin 'function_graph'
|
||||
```
|
||||
|
||||
Now when you compare this output with what you saw before, you should see more nested function calls:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd show | head -20
|
||||
## tracer: function_graph
|
||||
#
|
||||
# CPU DURATION FUNCTION CALLS
|
||||
# | | | | | | |
|
||||
6) | __fget_light() {
|
||||
6) 0.804 us | __fget_files();
|
||||
6) 2.708 us | }
|
||||
6) 3.650 us | } /* __fdget */
|
||||
6) 0.547 us | eventfd_poll();
|
||||
6) 0.535 us | fput();
|
||||
6) | __fdget() {
|
||||
6) | __fget_light() {
|
||||
6) 0.946 us | __fget_files();
|
||||
6) 1.895 us | }
|
||||
6) 2.849 us | }
|
||||
6) | sock_poll() {
|
||||
6) 0.651 us | unix_poll();
|
||||
6) 1.905 us | }
|
||||
6) 0.475 us | fput();
|
||||
6) | __fdget() {
|
||||
```
|
||||
|
||||
### Learn available functions to trace
|
||||
|
||||
If you want to trace only certain functions and ignore the rest, you need to know the exact function names. You can get them with the `list` argument followed by `-f`. This example searches for the common kernel function `kmalloc`, which is used to allocate memory in the kernel:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd list -f | grep kmalloc
|
||||
bpf_map_kmalloc_node
|
||||
mempool_kmalloc
|
||||
__traceiter_kmalloc
|
||||
__traceiter_kmalloc_node
|
||||
kmalloc_slab
|
||||
kmalloc_order
|
||||
kmalloc_order_trace
|
||||
kmalloc_large_node
|
||||
__kmalloc
|
||||
__kmalloc_track_caller
|
||||
__kmalloc_node
|
||||
__kmalloc_node_track_caller
|
||||
[...]
|
||||
```
|
||||
|
||||
Here's the total count of functions available on my test system:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd list -f | wc -l
|
||||
63165
|
||||
```
|
||||
|
||||
### Trace kernel module-related functions
|
||||
|
||||
You can also trace functions related to a specific kernel module. Imagine you want to trace `kvm` kernel module-related functions. Ensure the module is loaded:
|
||||
|
||||
|
||||
```
|
||||
# lsmod | grep kvm_intel
|
||||
kvm_intel 335872 0
|
||||
kvm 987136 1 kvm_intel
|
||||
```
|
||||
|
||||
Run `trace-cmd` again with the `list` argument, and from the output, `grep` for lines that end in `]`. This will filter out the kernel modules. Then `grep` the kernel module `kvm_intel`, and you should see all the functions related to that kernel module:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd list -f | grep ]$ | grep kvm_intel
|
||||
vmx_can_emulate_instruction [kvm_intel]
|
||||
vmx_update_emulated_instruction [kvm_intel]
|
||||
vmx_setup_uret_msr [kvm_intel]
|
||||
vmx_set_identity_map_addr [kvm_intel]
|
||||
handle_machine_check [kvm_intel]
|
||||
handle_triple_fault [kvm_intel]
|
||||
vmx_patch_hypercall [kvm_intel]
|
||||
|
||||
[...]
|
||||
|
||||
vmx_dump_dtsel [kvm_intel]
|
||||
vmx_dump_sel [kvm_intel]
|
||||
```
|
||||
|
||||
### Trace specific functions
|
||||
|
||||
Now that you know how to find functions of interest, put that knowledge to work with an example. As in the earlier article, try to trace filesystem-related functions. The filesystem I had on my test system was `ext4`.
|
||||
|
||||
This procedure is slightly different; instead of `start`, you run the command with the `record` argument followed by the "pattern" of the functions you want to trace. You also need to specify the tracer you want; in this case, that's `function_graph`. The command continues recording the trace until you stop it with **Ctrl+C**. So after a few seconds, hit **Ctrl+C** to stop tracing:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd list -f | grep ^ext4_
|
||||
|
||||
# trace-cmd record -l ext4_* -p function_graph
|
||||
plugin 'function_graph'
|
||||
Hit Ctrl^C to stop recording
|
||||
^C
|
||||
CPU0 data recorded at offset=0x856000
|
||||
8192 bytes in size
|
||||
[...]
|
||||
```
|
||||
|
||||
### View the recorded trace
|
||||
|
||||
To view the trace you recorded earlier, run the command with the `report` argument. From the output, it's clear that the filter worked, and you see only the ext4-related function trace:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd report | head -20
|
||||
[...]
|
||||
cpus=8
|
||||
trace-cmd-12697 [000] 11303.928103: funcgraph_entry: | ext4_show_options() {
|
||||
trace-cmd-12697 [000] 11303.928104: funcgraph_entry: 0.187 us | ext4_get_dummy_policy();
|
||||
trace-cmd-12697 [000] 11303.928105: funcgraph_exit: 1.583 us | }
|
||||
trace-cmd-12697 [000] 11303.928122: funcgraph_entry: | ext4_create() {
|
||||
trace-cmd-12697 [000] 11303.928122: funcgraph_entry: | ext4_alloc_inode() {
|
||||
trace-cmd-12697 [000] 11303.928123: funcgraph_entry: 0.101 us | ext4_es_init_tree();
|
||||
trace-cmd-12697 [000] 11303.928123: funcgraph_entry: 0.083 us | ext4_init_pending_tree();
|
||||
trace-cmd-12697 [000] 11303.928123: funcgraph_entry: 0.141 us | ext4_fc_init_inode();
|
||||
trace-cmd-12697 [000] 11303.928123: funcgraph_exit: 0.931 us | }
|
||||
trace-cmd-12697 [000] 11303.928124: funcgraph_entry: 0.081 us | ext4_get_dummy_policy();
|
||||
trace-cmd-12697 [000] 11303.928124: funcgraph_entry: 0.133 us | ext4_get_group_desc();
|
||||
trace-cmd-12697 [000] 11303.928124: funcgraph_entry: 0.115 us | ext4_free_inodes_count();
|
||||
trace-cmd-12697 [000] 11303.928124: funcgraph_entry: 0.114 us | ext4_get_group_desc();
|
||||
```
|
||||
|
||||
### Trace a specific PID
|
||||
|
||||
Say you want to trace functions related to a specific persistent identifier (PID). Open another terminal and note the PID of the running shell:
|
||||
|
||||
|
||||
```
|
||||
# echo $$
|
||||
10885
|
||||
```
|
||||
|
||||
Run the `record` command again and pass the PID using the `-P` option. This time, let the terminal run (i.e., do not press **Ctrl+C** yet):
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd record -P 10885 -p function_graph
|
||||
plugin 'function_graph'
|
||||
Hit Ctrl^C to stop recording
|
||||
```
|
||||
|
||||
### Run some activity on the shell
|
||||
|
||||
Move back to the other terminal where you had a shell running with a specific PID and run any command, e.g., `ls` to list files:
|
||||
|
||||
|
||||
```
|
||||
# ls
|
||||
Temp-9b61f280-fdc1-4512-9211-5c60f764d702
|
||||
tracker-extract-3-files.1000
|
||||
v8-compile-cache-1000
|
||||
[...]
|
||||
```
|
||||
|
||||
Move back to the terminal where you enabled tracing and hit **Ctrl+C** to stop tracing:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd record -P 10885 -p function_graph
|
||||
plugin 'function_graph'
|
||||
Hit Ctrl^C to stop recording
|
||||
^C
|
||||
CPU1 data recorded at offset=0x856000
|
||||
618496 bytes in size
|
||||
[...]
|
||||
```
|
||||
|
||||
In the trace's output, you can see the PID and the Bash shell on the left and the function calls related to it on the right. This can be pretty handy to narrow down your tracing:
|
||||
|
||||
|
||||
```
|
||||
# trace-cmd report | head -20
|
||||
|
||||
cpus=8
|
||||
<idle>-0 [001] 11555.380581: funcgraph_entry: | switch_mm_irqs_off() {
|
||||
<idle>-0 [001] 11555.380583: funcgraph_entry: 1.703 us | load_new_mm_cr3();
|
||||
<idle>-0 [001] 11555.380586: funcgraph_entry: 0.493 us | switch_ldt();
|
||||
<idle>-0 [001] 11555.380587: funcgraph_exit: 7.235 us | }
|
||||
bash-10885 [001] 11555.380589: funcgraph_entry: 1.046 us | finish_task_switch.isra.0();
|
||||
bash-10885 [001] 11555.380591: funcgraph_entry: | __fdget() {
|
||||
bash-10885 [001] 11555.380592: funcgraph_entry: 2.036 us | __fget_light();
|
||||
bash-10885 [001] 11555.380594: funcgraph_exit: 3.256 us | }
|
||||
bash-10885 [001] 11555.380595: funcgraph_entry: | tty_poll() {
|
||||
bash-10885 [001] 11555.380597: funcgraph_entry: | tty_ldisc_ref_wait() {
|
||||
bash-10885 [001] 11555.380598: funcgraph_entry: | ldsem_down_read() {
|
||||
bash-10885 [001] 11555.380598: funcgraph_entry: | __cond_resched() {
|
||||
```
|
||||
|
||||
### Give it a try
|
||||
|
||||
These brief examples show how using `trace-cmd` instead of the underlying `ftrace` mechanism is both easy to use and rich in features, including many I didn't cover here. To learn more and get better at it, consult its man page and try out its other useful commands.
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/21/7/linux-kernel-trace-cmd
|
||||
|
||||
作者：[Gaurav Kamathe][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/gkamathe
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/lead-images/puzzle_computer_solve_fix_tool.png?itok=U0pH1uwj (Puzzle pieces coming together to form a computer screen)
|
||||
[2]: https://opensource.com/article/21/7/analyze-linux-kernel-ftrace
|
||||
[3]: https://lwn.net/Articles/410200/
|
||||
@@ -0,0 +1,114 @@
|
||||
[#]: subject: (5 reasons you should run your apps on WildFly)
|
||||
[#]: via: (https://opensource.com/article/21/7/run-apps-wildfly)
|
||||
[#]: author: (Ranabir Chakraborty https://opensource.com/users/ranabir-chakraborty)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
5 reasons you should run your apps on WildFly
|
||||
======
|
||||
WildFly is a popular choice for users and developers worldwide who
|
||||
develop enterprise-capable applications.
|
||||
![Person drinking a hot drink at the computer][1]
|
||||
|
||||
WildFly, formerly known as JBoss Application Server, is an open source Java EE application server. Its primary goal is to provide a set of vital tools for enterprise Java applications.
|
||||
|
||||
According to the Jakarta EE 2020/2021 [survey][2], WildFly is head and shoulders above in the recent application servers and in the rating categories. Here are some of the reasons why:
|
||||
|
||||
### 1. Save time with faster development
|
||||
|
||||
WildFly supports the newest standards for REST-based data access, including JAX-RS 2 and JSON-P, and because it's building on Jakarta EE, which provides rich enterprise capabilities with ease of use of frameworks that eliminate boilerplate and reduce technical burden.
|
||||
|
||||
The quick boot feature of WildFly, integrated with the easy-to-use Arquillian framework, allows for test-driven development using the actual environment in which your code runs. This test code is separate and deployed alongside the application, where it has full access to server resources.
|
||||
|
||||
### 2\. Powerful but simple to use
|
||||
|
||||
WildFly configuration setup is centralized, simple, and user-focused.
|
||||
The configuration file—organized by subsystems—is easy to understand and has no internal server wiring that will be exposed. All management capabilities appear in a unified manner across all forms of access. These include a command-line interface, a web-based administration console, a native Java API, an HTTP/JSON-based REST API, and a JMX gateway. These options allow for custom automation using the tools and languages that best suit your needs.
|
||||
|
||||
### 3. Modular and lightweight
|
||||
|
||||
WildFly does classloading right. And it does it smoothly. It uses JBoss Modules to provide true application isolation while hiding server implementation classes from the application and only connects to JARs that your application needs. Appearance rules have sensible defaults but are usually customized. The dependency resolution algorithm means that classloading performance isn't affected by the number of versions of libraries you've got installed.
|
||||
|
||||
In WildFly base, they've developed runtime services to attenuate heap allocation using standard cached indexed metadata over duplicate full parses, which reduces heap and object churn. One hundred percent of the administration console is stateless and purely client-driven. It starts immediately and requires zero memory on the server. This integrated configuration enables WildFly to run with stock JVM settings—even on small devices while leaving more headroom for application data and supporting high-level scalability.
|
||||
|
||||
### 4\. Save resources with efficient management
|
||||
|
||||
WildFly takes a more aggressive approach to memory management and relies on pluggable subsystems, installed or removed as required. Subsystems use smart and intelligent defaults but can still be customized to best suit your needs. When working with domain mode, all participating servers' configuration is laid out in a well-organized, consistent manner within the same file.
|
||||
|
||||
### 5. Leverage open source
|
||||
|
||||
WildFly is an open source community project and is out there to be used and distributed using the LGPL v2.1 license, which means it's available for you to download and use for whatever you need. This allows organizations to develop unique new technologies and federates the world of technology to help successful startups to spring up anywhere.
|
||||
|
||||
## 8 ways to contribute to WildFly
|
||||
|
||||
Now that you know a bit about WildFly, Let’s try to understand the ways you can get involved with WildFly.
|
||||
|
||||
WildFly relies on contributions from people like you. I’ve joined Red Hat and contributed to WildFly for a year now, and it’s fun to work with great minds around you, and you’ll get to learn a lot. Here are some ways by which you can be a part of and assist the community.
|
||||
|
||||
### 1\. Check out the repository.
|
||||
|
||||
Here are the [WildFly][3] and [WildFly Core][4] (WildFly Core provides the core runtime used by the Wildfly application server). If you want to get more details, you can check out this [document][5].
|
||||
|
||||
### 2\. Raise a ticket or work on existing issues.
|
||||
|
||||
After checking out the WildFly repositories, if you feel some enhancements or fixes are needed, you can create issues for [WildFly][6] and [WildFly Core][7], or work on pre-existing issues.
|
||||
|
||||
### 3\. Edit the website
|
||||
|
||||
Like the WildFly project, the website is open source too. You can check out the [repository][8] and contribute here too, with some new and attractive modifications.
|
||||
|
||||
### 4\. Blog with us
|
||||
|
||||
You have a [blog][9], and all entries are maintained in a Git repository. If you have new ideas, you can share your experience and ideas in the form of an editorial. We use [markdown][10] and [AsciiDoc][11] so that you can submit your blog post as a pull request.
|
||||
|
||||
### 5\. Edit the Documentation
|
||||
|
||||
You can also help us to [make better documentation][12]. Let us know if you find a typo or an error, and feel free to send a pull request. Your input is valuable to us, and always welcome.
|
||||
|
||||
### 6\. Help somebody out
|
||||
|
||||
You can check out our [forums][13] and if you run into an issue, post your question and check the previous issues if you see some similarities. You can also share your knowledge and answer some of the queries because your knowledge can help others.
|
||||
|
||||
### 7\. Join our chatroom and follow the latest news
|
||||
|
||||
Our Project team has an open (and open source) and active [chatroom][14] where you can ask your questions and check out the [latest news][15] section to find what are the new things we are working on. Stop by, say hello, interact with the team members, but keep in mind that basic rules of civility apply.
|
||||
|
||||
### 8\. Spread the word
|
||||
|
||||
The simplest and easiest way to help the WildFly community is to act as a project ambassador by spreading the news, educating others about the usage of WildFly, and showing up to the community events in your area.
|
||||
|
||||
## **Final thoughts**
|
||||
|
||||
WildFly is a popular choice for users and developers worldwide who develop enterprise-capable applications. WildFly is an active project, so there are always new features in the works, and we're all delighted to be a part of it.
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/21/7/run-apps-wildfly
|
||||
|
||||
作者：[Ranabir Chakraborty][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/ranabir-chakraborty
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/lead-images/coffee_tea_laptop_computer_work_desk.png?itok=D5yMx_Dr (Person drinking a hot drink at the computer)
|
||||
[2]: https://arjan-tijms.omnifaces.org/2021/02/jakarta-ee-survey-20202021-results.html
|
||||
[3]: https://github.com/wildfly/wildfly
|
||||
[4]: https://github.com/wildfly/wildfly-core
|
||||
[5]: https://developer.jboss.org/docs/DOC-48381
|
||||
[6]: https://issues.redhat.com/browse/WFLY-14541?jql=project%20%3D%20WFLY%20AND%20resolution%20%3D%20Unresolved%20ORDER%20BY%20priority%20DESC%2C%20updated%20DESC
|
||||
[7]: https://issues.redhat.com/projects/WFCORE/issues/WFCORE-4827?filter=allopenissues
|
||||
[8]: https://github.com/wildfly/wildfly.org
|
||||
[9]: https://github.com/wildfly/wildfly.org/tree/master/_posts
|
||||
[10]: https://opensource.com/article/19/9/introduction-markdown
|
||||
[11]: https://asciidoc.org/
|
||||
[12]: https://github.com/wildfly/wildfly/tree/master/docs
|
||||
[13]: https://groups.google.com/g/wildfly
|
||||
[14]: https://wildfly.zulipchat.com/#recent_topics
|
||||
[15]: https://www.wildfly.org/news/
|
||||
@@ -0,0 +1,85 @@
|
||||
[#]: subject: (Use df to check free disk space on Linux)
|
||||
[#]: via: (https://opensource.com/article/21/7/check-disk-space-linux-df)
|
||||
[#]: author: (Seth Kenlon https://opensource.com/users/seth)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
Use df to check free disk space on Linux
|
||||
======
|
||||
Find out how much Linux disk space you have left with the df command.
|
||||
![Free disk space][1]
|
||||
|
||||
Drive space isn't quite as precious as it was in the early days of computing, but no matter how much space you have, there's always the potential to run out. Computers need a little space just to operate, so it's important to check occasionally to ensure you haven't inadvertently used up literally _all_ the space available on your drive. In the Linux terminal, you can do that with the `df` command.
|
||||
|
||||
The df command displays the amount of disk space available on the file system.
|
||||
|
||||
To make the output easy for you to read, you can use the `--human-readable` (or `-h` for short) option:
|
||||
|
||||
|
||||
```
|
||||
$ df --human-readable
|
||||
Filesystem Size Used Avail Use% Mounted on
|
||||
/dev/sda1 1.0T 525G 500G 52% /
|
||||
```
|
||||
|
||||
In this example, the computer's drive is 52% full, with 500 GB free for use.
|
||||
|
||||
Because Linux views its file system holistically across all mounted devices, the df command provides you details for every storage device attached to your computer. If you have lots of drives, then the output reflects that:
|
||||
|
||||
|
||||
```
|
||||
$ df --human-readable
|
||||
Filesystem Size Used Avail Use% Mounted on
|
||||
/dev/root 110G 45G 61G 43% /
|
||||
devtmpfs 12G 0 12G 0% /dev
|
||||
tmpfs 12G 848K 12G 1% /run
|
||||
/dev/sda1 1.6T 1.3T 191G 87% /home
|
||||
/dev/sdb1 917G 184G 687G 22% /penguin
|
||||
/dev/sdc1 57G 50G 4.5G 92% /sneaker
|
||||
/dev/sdd1 3.7T 2.4T 1.3T 65% /tux
|
||||
```
|
||||
|
||||
In this example, the `/home` directory of the computer is 87% full, with 191 GB free.
|
||||
|
||||
### See total disk space available
|
||||
|
||||
If you do have a complex file system and would like to see the total space across all drives, use the `--total` option:
|
||||
|
||||
|
||||
```
|
||||
$ df --human-readable --total
|
||||
Filesystem Size Used Avail Use% Mounted on
|
||||
/dev/root 110G 45G 61G 43% /
|
||||
devtmpfs 12G 0 12G 0% /dev
|
||||
tmpfs 12G 848K 12G 1% /run
|
||||
/dev/sda1 1.6T 1.3T 191G 87% /home
|
||||
/dev/sdb1 917G 184G 687G 22% /penguin
|
||||
/dev/sdc1 57G 50G 4.5G 92% /sneaker
|
||||
/dev/sdd1 3.7T 2.4T 1.3T 65% /tux
|
||||
total 6.6T 4.0T 2.5T 62% -
|
||||
```
|
||||
|
||||
The final line of output provides the total space of the filesystem, total space used, total space available.
|
||||
|
||||
### See disk space usage
|
||||
|
||||
To just get a summary of what's occupying the space on your drives, read our article about the [du command][2].
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/21/7/check-disk-space-linux-df
|
||||
|
||||
作者：[Seth Kenlon][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/seth
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/df-splash.png?itok=AGXQm737 (Free disk space)
|
||||
[2]: https://opensource.com/article/21/7/check-used-disk-space-linux-du
|
||||
@@ -0,0 +1,247 @@
|
||||
[#]: subject: (Brave vs. Firefox: Your Ultimate Browser Choice for Private Web Experience)
|
||||
[#]: via: (https://itsfoss.com/brave-vs-firefox/)
|
||||
[#]: author: (Ankush Das https://itsfoss.com/author/ankush/)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: ( )
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
Brave vs. Firefox: Your Ultimate Browser Choice for Private Web Experience
|
||||
======
|
||||
|
||||
Web browsers have evolved over the years. From downloading files to accessing a full-fledged web application, we have come a long way.
|
||||
|
||||
For a lot of users, the web browser is the only thing they need to get their work done these days.
|
||||
|
||||
Hence, choosing the right browser becomes an important task that could help improve your workflow over the years.
|
||||
|
||||
### Brave vs. Firefox Browser
|
||||
|
||||
Brave and Mozilla’s Firefox are two of the most popular web browsers for privacy-conscious users and open-source enthusiasts.
|
||||
|
||||
Considering that both focus heavily on privacy and security, let us look at what exactly they have to offer, to help you decide what you should go with.
|
||||
|
||||
Here are the comparison pointers that I’ve used, you can directly navigate to any of them:
|
||||
|
||||
* [User Interface][1]
|
||||
* [Performance][2]
|
||||
* [Browser Engine][3]
|
||||
* [Ad & Tracking Blocking Capabilities][4]
|
||||
* [Containers][5]
|
||||
* [Rewards][6]
|
||||
* [Cross-Platform Availability][7]
|
||||
* [Synchronization][8]
|
||||
* [Service Integrations][9]
|
||||
* [Customizability][10]
|
||||
* [Extension Support][11]
|
||||
|
||||
|
||||
|
||||
### User Interface
|
||||
|
||||
The user interface is what makes the biggest difference with the workflow and experience when using the browser.
|
||||
|
||||
Of course, you can have your personal preferences, but the easier, snappier, and cleaner it looks, the better it is.
|
||||
|
||||
![Brave browser][12]
|
||||
|
||||
To start with, Brave shares a similar look and feel to Chrome and Microsoft Edge. It offers a clean experience with minimal UI elements and all the essential options accessible through the browser menu.
|
||||
|
||||
It offers a black theme as well. The subtle animations make the interaction a pleasant experience.
|
||||
|
||||
To customize it, you can choose to use themes available from the chrome web store.
|
||||
|
||||
When it comes to Mozilla Firefox, it has had a couple of major redesigns over the years, and the latest user interface tries to offer a closer experience to Chrome.
|
||||
|
||||
![Firefox browser][13]
|
||||
|
||||
The Firefox design looks impressive and provides a clean user experience. It also lets you opt for a dark theme if needed and there are several theme options to download/apply as well.
|
||||
|
||||
Both web browsers offer a good user experience.
|
||||
|
||||
If you want a familiar experience, but with a pinch of uniqueness, Mozilla’s Firefox can be a good pick.
|
||||
|
||||
But, if you want a snappier experience with a better feel for the animations, Brave gets the edge.
|
||||
|
||||
### Performance
|
||||
|
||||
Practically, I find Brave loading web pages faster. Also, the overall user experience feels snappy.
|
||||
|
||||
Firefox is not terribly slow, but it definitely felt slower than Brave.
|
||||
|
||||
To give you some perspective, I also utilized [Basemark][14] to run a benchmark to see if that is true on paper.
|
||||
|
||||
You can check with other browser benchmark tools available, but Basemark performs a variety of tests, so we’ll go with that for this article.
|
||||
|
||||
![Firefox benchmark score][15]
|
||||
|
||||
![Brave benchmark score][16]
|
||||
|
||||
Firefox managed to score **630** and Brave pulled it off better with ~**792**.
|
||||
|
||||
Do note that these benchmarks were run with default browser settings without any browser extensions installed.
|
||||
|
||||
Of course, synthetic scores may vary depending on what you have going on in the background and the hardware configuration of your system.
|
||||
|
||||
This is what I got with **i5-7400, 16 GB RAM, and GTX 1050ti GPU** on my desktop.
|
||||
|
||||
In general, Brave browser is a fast browser compared to most of the popular options available.
|
||||
|
||||
Both utilize a decent chunk of system resources and that varies to a degree with the number of tabs, types of webpages accessed, and the kind of blocking extension used.
|
||||
|
||||
For instance, Brave blocks aggressively by default but Firefox does not block display advertisements by default. And, this affects the system resource usage.
|
||||
|
||||
### Browser Engine
|
||||
|
||||
Firefox utilizes its own Gecko engine as the foundation and is using components on top of that from [servo research project][17] to improve.
|
||||
|
||||
Currently, it is essentially an improved Gecko engine dubbed by a project name “Quantum” which was introduced with the release of Firefox Quantum.
|
||||
|
||||
On the other hand, Brave uses Chromium’s engine.
|
||||
|
||||
While both are capable enough to handle modern web experiences, Chromium-based engine is just more popular and web developers often tailor their sites for the best experience on Chrome-based browsers
|
||||
|
||||
Also, some services happen to exclusively support Chrome-based browsers.
|
||||
|
||||
### Ad & Tracker Blocking Capabilities
|
||||
|
||||
![][18]
|
||||
|
||||
As I have mentioned before, Brave is aggressive in blocking trackers and advertisements. By default, it comes with the blocking feature enabled.
|
||||
|
||||
Firefox also enables the enhanced privacy protection by default but does not block display advertisements.
|
||||
|
||||
You will have to opt for the “**Strict**” privacy protection mode with Firefox if you want to get rid of display advertisements.
|
||||
|
||||
With that being said, Firefox enforces some unique tracking protection technology that includes Total Cookie Protection which isolates cookies for each site and prevents cross-site cookie tracking.
|
||||
|
||||
![][19]
|
||||
|
||||
This was introduced with [Firefox 86][20] and to use it, you need to enable a strict privacy protection mode.
|
||||
|
||||
Overall, Brave might look like a better option out of the box, and Mozilla Firefox offers better privacy protection features.
|
||||
|
||||
### Containers
|
||||
|
||||
Firefox also offers a way to isolate site activity when you use Facebook with help of a container. In other words, it prevents Facebook from tracking your offsite activity.
|
||||
|
||||
You can also use containers to organize your tabs and separate sessions when needed.
|
||||
|
||||
Brave does not offer anything similar but it does block cross-site trackers and cookies out-of-the-box.
|
||||
|
||||
### Rewards
|
||||
|
||||
![][21]
|
||||
|
||||
Unlike Firefox, Brave offers its own advertising network by blocking other advertisements on the web.
|
||||
|
||||
When you opt in to display privacy-friendly ads by Brave, you get rewarded with tokens to a crypto wallet. And you can use these tokens to give back to your favorite websites.
|
||||
|
||||
While this is a good business strategy to get away from mainstream advertising, for users who do not want any kind of advertisements, it may not be useful.
|
||||
|
||||
So, Brave offers an alternative in the form of rewards to help websites even if you block advertisements. If it is something you appreciate, Brave will be a good pick for you.
|
||||
|
||||
### Cross-Platform Availability
|
||||
|
||||
You will find both Brave and Firefox available for Linux, Windows, and macOS. Mobile apps are also available for iOS and Android.
|
||||
|
||||
For Linux users, Firefox comes baked in with most of the Linux distributions. And, you can also find it available in the software center. In addition to that, there is also a [Flatpak][22] package available.
|
||||
|
||||
Brave is not available through default repositories and the software center. Hence, you need to follow the official instructions to add the private repository and then [get Brave installed in your Linux distro][23].
|
||||
|
||||
### Synchronization
|
||||
|
||||
With Mozilla Firefox, you get to create a Firefox account to sync all your data cross-platform.
|
||||
|
||||
![][24]
|
||||
|
||||
Brave also lets you sync cross-platform but you need access to one of the devices in order to successfully do it.
|
||||
|
||||
![][25]
|
||||
|
||||
Hence, Firefox sync is more convenient.
|
||||
|
||||
Also, you get access to Firefox’s VPN, data breach monitor, email relay, and password manager with the Firefox account.
|
||||
|
||||
### Service Integrations
|
||||
|
||||
Right off the bat, Firefox offers more service integrations that include Pocket, VPN, password manager, and also some of its new offerings like Firefox relay.
|
||||
|
||||
If you want access to these services through your browser, Firefox will be the convenient option for you.
|
||||
|
||||
While Brave does offer crypto wallets, it is not for everyone.
|
||||
|
||||
![][26]
|
||||
|
||||
Similarly, if you like using [Brave Search][27], you may have a seamless experience when using it with Brave browser because of the user experience.
|
||||
|
||||
### Customizability & Security
|
||||
|
||||
Firefox shines when it comes to customizability. You get more options to tweak the experience and also take control of the privacy/security of your browser.
|
||||
|
||||
The ability to customize lets you make Firefox more secure than the Brave browser.
|
||||
|
||||
While hardening Firefox is a separate topic which we’ll talk about. To give you an example, [Tor Browser][28] is just a customized Firefox browser.
|
||||
|
||||
However, that does not make Brave less secure. It is a secure browser overall but you do get more options with Firefox.
|
||||
|
||||
### Extension Support
|
||||
|
||||
There’s no doubt that the Chrome web store offers way more extensions.
|
||||
|
||||
So, Brave gets a clear edge over Firefox if you are someone who utilizes a lot of extensions (or constantly try new ones).
|
||||
|
||||
Firefox may not have the biggest catalog of extensions, it does support most of the extensions. For common use-cases, you will rarely find an extension that is not available as an addon for Firefox.
|
||||
|
||||
### What Should You Choose?
|
||||
|
||||
If you want the best compatibility with the modern web experience and want access to more extensions, Brave browser seems to make more sense.
|
||||
|
||||
On the other hand, Firefox is an excellent choice for everyday browsing with industry-first privacy features, and a convenient sync option for non-tech savvy users.
|
||||
|
||||
You will have a few trade-offs when selecting either of them. So, your will have to prioritize what you want the most.
|
||||
|
||||
Let me know about your final choice for your use case in the comments down below!
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://itsfoss.com/brave-vs-firefox/
|
||||
|
||||
作者：[Ankush Das][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[译者ID](https://github.com/译者ID)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://itsfoss.com/author/ankush/
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: tmp.5yJseRG2rb#ui
|
||||
[2]: tmp.5yJseRG2rb#perf
|
||||
[3]: tmp.5yJseRG2rb#engine
|
||||
[4]: tmp.5yJseRG2rb#ad
|
||||
[5]: tmp.5yJseRG2rb#container
|
||||
[6]: tmp.5yJseRG2rb#reward
|
||||
[7]: tmp.5yJseRG2rb#cp
|
||||
[8]: tmp.5yJseRG2rb#sync
|
||||
[9]: tmp.5yJseRG2rb#service
|
||||
[10]: tmp.5yJseRG2rb#customise
|
||||
[11]: tmp.5yJseRG2rb#extensions
|
||||
[12]: https://i2.wp.com/itsfoss.com/wp-content/uploads/2021/07/brave-ui-new.jpg?resize=800%2C450&ssl=1
|
||||
[13]: https://i0.wp.com/itsfoss.com/wp-content/uploads/2021/07/firefox-ui.jpg?resize=800%2C450&ssl=1
|
||||
[14]: https://web.basemark.com
|
||||
[15]: https://i0.wp.com/itsfoss.com/wp-content/uploads/2021/07/firefox-basemark.png?resize=800%2C598&ssl=1
|
||||
[16]: https://i1.wp.com/itsfoss.com/wp-content/uploads/2021/07/basemark-brave.png?resize=800%2C560&ssl=1
|
||||
[17]: https://servo.org
|
||||
[18]: https://i2.wp.com/itsfoss.com/wp-content/uploads/2021/07/brave-blocker.png?resize=800%2C556&ssl=1
|
||||
[19]: https://i2.wp.com/itsfoss.com/wp-content/uploads/2021/07/firefox-blocker.png?resize=800%2C564&ssl=1
|
||||
[20]: https://news.itsfoss.com/firefox-86-release/
|
||||
[21]: https://i1.wp.com/itsfoss.com/wp-content/uploads/2021/07/brave-rewards.png?resize=800%2C560&ssl=1
|
||||
[22]: https://itsfoss.com/what-is-flatpak/
|
||||
[23]: https://itsfoss.com/brave-web-browser/
|
||||
[24]: https://i0.wp.com/itsfoss.com/wp-content/uploads/2021/07/firefox-sync.png?resize=800%2C651&ssl=1
|
||||
[25]: https://i2.wp.com/itsfoss.com/wp-content/uploads/2021/07/brave-sync.png?resize=800%2C383&ssl=1
|
||||
[26]: https://i0.wp.com/itsfoss.com/wp-content/uploads/2021/07/brave-crypto-wallet.png?resize=800%2C531&ssl=1
|
||||
[27]: https://itsfoss.com/brave-search-features/
|
||||
[28]: https://itsfoss.com/install-tar-browser-linux/
|
||||
@@ -0,0 +1,151 @@
|
||||
[#]: subject: (F(r)iction: Or How I Learnt to Stop Worrying and Start Loving Vim)
|
||||
[#]: via: (https://news.itsfoss.com/how-i-started-loving-vim/)
|
||||
[#]: author: (Theena https://news.itsfoss.com/author/theena/)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (piaoshi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
虚构还是摩擦: 我如何学会克服焦虑并开始爱上 Vim
|
||||
======
|
||||
|
||||
时间：2009 年 12 月。我准备辞去工作。
|
||||
|
||||
我希望专心写我的第一本书；我在工作中的承诺和当是技术状况都阻止不了我。
|
||||
|
||||
写作是件苦差事。
|
||||
|
||||
在现代世界中，很少有工作像写作这样奇特或者说艰巨的追求 -- 面对一张白纸，坐下来，迫使你的大脑吐出文字，向读者传达一个想法。当然，我并不是说写作不能与他人合作完成，而只是想说明，对于作家来说，自己着手写一部新作品是多么令人生畏。虚构和非虚构写作都是如此。但由于我是一名小说家，我在这篇文章中主要想关注是虚构作品的写作。
|
||||
|
||||
![][1]
|
||||
|
||||
还记得2009年是什么样子吗？
|
||||
|
||||
智能手机已经诞生 3 年了 -- 我还没有摆脱功能机。笔记本电脑又大又笨重。同时，用于生产的基于云的网络应用还处于起步阶段，并不那么好用。从技术上讲，像我这样的作家们正在以 Gmail 账户（和一个非常年轻的基于云的存储服务 Dropbox）作为一个始终可用的选项来处理自己的草稿，即使我的个人电脑不在身边。虽然这与作家们必须要使用打字机（或，上帝保佑，笔和纸）工作时相比已经是一个很好的变化了，但并没有好多少。
|
||||
|
||||
首先，对手稿的版本控制是一场噩梦。此外，我为简化工作流程而在工具包中添加的工具越多，我转换写作环境（无论是用户界面还是用户体验）的次数就越多。
|
||||
|
||||
我可能是在 Windows 记事本上开始写草稿，然后把它保存在家里电脑上的 MS Word 文档中，用电子邮件发给自己一份副本，同时在 Dropbox 上保留另一份副本（因为在上班时无法访问 Dropbox），在公司时对该文件的副本进行处理，在一天结束时用电子邮件发给自己，在家里的电脑上下载它，用一个新的名字和相应的日期保存它，这样我就能认出该文件是在工司（而不是家里）进行修改的……好吧，你知道这是怎样一个画面。如果你能感受到这种涉及 Windows 记事本、MS Word、Gmail 和 Dropbox 的工作流程有多么疯狂，那么现在你就知道我为什么辞职了。
|
||||
|
||||
让我更清醒的是，我仍然知道一些作家，其中竟然有些还是好作家，依然在使用我 2009 年遵循的工作流程的各种变体。
|
||||
|
||||
在接下来的三年里，我致力于我的手稿，在 2012 年完成了初稿。在这三年里，技术状况发生了很大变化。智能手机确实相当给力，我在 2009 年遇到的一些复杂情况已经消失了。我仍然可以用手机处理我在家里外理的文件（不一定是新鲜的写作，但由于手机上的 Dropbox，编辑变得相当容易）。我的主要写作工具仍然是微软的 Windows 记事本和 Word，我就是这样完成初稿的。
|
||||
|
||||
小说 [**《第一声》**][2] 于 2016 年出版，获得了评论界和商业界的好评。
|
||||
|
||||
结束了。
|
||||
|
||||
或许我是这么想的。
|
||||
|
||||
刚完成手稿并发给我的编辑，我就已经开始着手第二部小说的写作。我不再离职从事写作，而是采取了一种更务实的方法：我会在每年年底请两个星期的假，这样我就可以到山上的一个小木屋里去写作。
|
||||
|
||||
花了半天时间我才意识到，那些让我讨厌的 [写作工具][3] 和工作流程并没有消失，而是演变成了一个更复杂的野兽。作为一个作家，我并不像我想像的那样高产或高效。
|
||||
|
||||
### 新冠期间的 Linux
|
||||
|
||||
![][4]
|
||||
|
||||
时间：2020 年。世界正处于集体疯狂的边缘。
|
||||
|
||||
起初在中国分离出的一种新型病毒正在演变成 1911 年以来的第一次全球大流行疾病。3 月 20 日，斯里兰卡，跟随世界上大多数国家的脚步，关闭了。
|
||||
|
||||
四月是斯里兰卡旱季的高峰。在像科伦坡这样的混凝土丛林中，温度可以达到三十四五度，湿度高达九十八九度。在最好的情况下，它也可以使大多数人精神错乱，更别说在全球大流行病正在进行的时候，被困在没有一直开着空调的家里？真是一个让人疯狂的好温床。
|
||||
|
||||
让我的疯狂是 Linux，或者说是"发行版跳跃"，像我们在开源社区中所说的。
|
||||
|
||||
我在各种 *nix 发行版间跳的次数越多，我就对控制的想法越迷恋。当任何事情似乎都不在我们的控制之中时 -- 即使是与另一个人握手这样的简单行为 -- 我们自然会倾向于做那些我们感觉更有控制力的事。
|
||||
|
||||
在我的生活中，还有什么比计算机更容易被控制的呢？自然，这也延伸到我的写作工具和工作流程。
|
||||
|
||||
### 通往 Vim 之路
|
||||
|
||||
有一个关于 [Vim][5] 的笑话完美地描述了我对它的第一次体验：人们对 Vim 难以割舍是因为他们不知道怎么关掉它。
|
||||
|
||||
我试图编辑一个配置文件，而 [新安装的 Ubuntu 服务器][6] 只预装了 Vim 文本编辑器。第一次是恐慌 -- 以至于我重新启动了机器，以为操作系统没有识别出我的键盘。然而当它再次发生时，不可避免地，我谷歌搜索：“[我该如何关闭 Vim？][7]”
|
||||
|
||||
_哦。这真有趣_，我想。
|
||||
|
||||
_但为什么呢？_
|
||||
|
||||
要理解我为什么会对一个复杂到无法关闭的文本编辑器有点兴趣，你必须了解我是多么崇拜 Windows 记事本。
|
||||
|
||||
作为一个作家，我喜欢在它的没有废话、没有按钮、白纸一样的画布上写作。它没有拼写检查。它没有格式。但这些我并不关心。
|
||||
|
||||
对于我这个作家来说，记事本是有史以来最好的草稿写作板。不幸的是，它并不强大 -- 所以即使我会先用记事本写草稿，一旦超过 1000 字，我就会把它移到 MS Word 上 -- 记事本不是为散文而生的，当超过这个字数限制时，这些局限就会凸显出来。
|
||||
|
||||
因此，当我把我所有的计算从 Windows 上迁移走时，我第一个要安装的就是一个好的文本编辑器。
|
||||
|
||||
[Kate][8] 是第一个让我感到比用 Windows 记事本更舒服的替代品 -- 它更强大（它有拼写检查功能！），而且，我可以在同一个环境中搞一些业余爱好式的编程。
|
||||
|
||||
当时它是我的爱。
|
||||
|
||||
但后来 Vim 出现了。
|
||||
|
||||
我对 Vim 了解得越多，看开发者在 Vim 上现场进行编码的次数越多，我就越发现自己在编辑文本时更想打开 Vim。我在传统的 Unix 意义上使用“文本编辑”这一短语：编辑配置文件中的文本块，或者有时编写基本的 Bash 脚本。
|
||||
|
||||
我仍然没有用 Vim 来满足我的散文写作需求。
|
||||
|
||||
在这方面我有 Libre Office。
|
||||
|
||||
算是吧。
|
||||
|
||||
虽然它是一个适当的 [MS Office 替代品][9]，但我发现自己没有被它打动。它的用户界面可能比 MS Word 更让人分心，而且每个发行版都有不同的 Libre Office 软件包，我发现自己使用的是一个非常零散的工具包和工作流程，更不用说用户界面在不同的发行版和桌面环境中差异是多么大。
|
||||
|
||||
当我开始读我的硕士学位时，事情变得更加复杂了。这是，我要在 Kate 上做笔记，把它们转移到 Libre Office 上，然后保存到我的 Dropbox 上。
|
||||
|
||||
情境转换每天都在盯着我看。
|
||||
|
||||
生产力下降，因为我不得不打开和关闭一些不相关的应用程序。我需要一个写作工具来满足我所有的需求 -- 作为一个小说家，作为一个学生，作为一个业余的程序员。
|
||||
|
||||
这时我意识到，解决我场景切换噩梦的方法也同时在盯着我的脸。
|
||||
|
||||
这时，我已经经常使用 Vim -- 甚至在我的安卓手机上利用 Termux 使用它。这使我对要把所有东西都搬到 Vim 上的想法感到相当舒服。由于它支持 Markdown 语法，记笔记也会变得更加容易。
|
||||
|
||||
这仅仅是大约两个月前的事。
|
||||
|
||||
现在怎么样了？
|
||||
|
||||
时间：2021 年 4 月。
|
||||
|
||||
坐在出租车上，我通过 Termux（借助蓝牙键盘）[用 Vim][10] 在手机上开始写这个草稿。我把文件推送到 GitHub 上我的用于写作使用的私人仓库，我从那里把文件拉到我的电脑上，又写了几行，然后再次出门。我把新版本的文件从 GitHub 拉到我的手机上，修改，推送，如此往复，直到我把最后的草稿用电子邮件发给编辑。
|
||||
|
||||
现在，场景切换的情况已经不复存在。
|
||||
|
||||
在文字处理器中写作所带来的分心问题也没有了。
|
||||
|
||||
编辑工作变得无比简单，而且更快了。
|
||||
|
||||
我的手腕不再疼痛，因为我不再需要鼠标了。
|
||||
|
||||
现在是 2021 年 4 月。
|
||||
|
||||
我是一名小说家。
|
||||
|
||||
而我在 Vim 上写作。
|
||||
|
||||
怎么做的？我将在本专栏系列的第二部分讨论这个工作流程的具体内容，即非技术人员如何使用免费和开源技术。敬请关注。
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://news.itsfoss.com/how-i-started-loving-vim/
|
||||
|
||||
作者：[Theena][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[piaoshi](https://github.com/piaoshi)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://news.itsfoss.com/author/theena/
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: data:image/svg+xml;base64,PHN2ZyBoZWlnaHQ9JzM5MCcgd2lkdGg9Jzc4MCcgeG1sbnM9J2h0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnJyB2ZXJzaW9uPScxLjEnLz4=
|
||||
[2]: https://www.goodreads.com/book/show/29616237-first-utterance
|
||||
[3]: https://itsfoss.com/open-source-tools-writers/
|
||||
[4]: data:image/svg+xml;base64,PHN2ZyBoZWlnaHQ9JzM5NScgd2lkdGg9Jzc4MCcgeG1sbnM9J2h0dHA6Ly93d3cudzMub3JnLzIwMDAvc3ZnJyB2ZXJzaW9uPScxLjEnLz4=
|
||||
[5]: https://www.vim.org/
|
||||
[6]: https://itsfoss.com/install-ubuntu-server-raspberry-pi/
|
||||
[7]: https://itsfoss.com/how-to-exit-vim/
|
||||
[8]: https://kate-editor.org/
|
||||
[9]: https://itsfoss.com/best-free-open-source-alternatives-microsoft-office/
|
||||
[10]: https://linuxhandbook.com/basic-vim-commands/
|
||||
@@ -1,210 +0,0 @@
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (YungeG)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
[#]: subject: (Trace code in Fedora with bpftrace)
|
||||
[#]: via: (https://fedoramagazine.org/trace-code-in-fedora-with-bpftrace/)
|
||||
[#]: author: (Augusto Caringi https://fedoramagazine.org/author/acaringi/)
|
||||
|
||||
在 Fedora 中用 bpftrace 追踪代码
|
||||
======
|
||||
|
||||
![][1]
|
||||
|
||||
bpftrace 是一个[基于 eBPF 的新型追踪工具][2]，在 Fedora 28 第一次引入。Brendan Gregg，Alastair Robertson 和 Matheus Marchini 在分散于全网络的黑客团队的帮助下开发了 bpftrace，一个允许你分析系统在幕后正在执行的操作的追踪工具，告诉你代码中正在被调用的函数、传递给函数的参数、函数的调用次数等。
|
||||
|
||||
这篇文章的内容涉及了 bpftrace 的一些基础，以及它是如何工作的，请继续阅读获取更多的信息和一些有用的实例。
|
||||
|
||||
### eBPF (extended Berkeley Packet Filter)
|
||||
|
||||
[eBPF][3] 是一个微型虚拟机，更确切的说是一个虚拟 CPU，位于 Linux 内核中。eBPF 可以在内核空间以一种安全可控的方式加载和运行体积较小的程序，保证 eBPF 的使用更加安全，即使在生产环境系统中。eBPF 虚拟机有自己的指令集（[ISA][4]），类似于现代处理器架构的一个子集。通过这个 ISA，可以很容易将 eBPF 程序转化为真实硬件上的代码。内核即时将程序转化为主流处理器架构上的本地代码，从而提升性能。
|
||||
|
||||
eBPF 虚拟机允许通过编程扩展内核，目前已经有一些内核子系统使用这一新型强大的 Linux Kernel 功能，比如网络，安全计算、追踪等。这些子系统的主要思想是添加 eBPF 程序到特定的代码点，从而扩展原生的内核行为。
|
||||
|
||||
虽然 eBPF 机器语言功能强大，由于是一种底层语言，直接用于编写代码很费力，bpftrace 就是为了解决这个问题而生的。eBPF 提供了一种高级语言编写 eBPF 追踪脚本，然后在 clang / LLVM 库的帮助下将这些脚本转化为 eBPF，最终添加到特定的代码点。
|
||||
|
||||
## 安装和快速入门
|
||||
|
||||
在终端 [使用][5] _[sudo][5]_ 执行下面的命令安装 bpftrace：
|
||||
|
||||
```
|
||||
$ sudo dnf install bpftrace
|
||||
```
|
||||
|
||||
使用“hello world”进行实验：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 'BEGIN { printf("hello world\n"); }'
|
||||
```
|
||||
|
||||
注意，出于特权级的需要，你必须使用 _root_ 运行 _bpftrace_，使用 _-e_ 选项指明一个程序，构建一个所谓的“单行程序”。这个例子只会打印 _hello world_，接着等待你按下 **Ctrl+C**。
|
||||
|
||||
_BEGIN_ 是一个特殊的探针名，只在执行一开始生效一次；每次探针命中时，大括号 _{}_ 内的操作——这个例子中只是一个 _printf_——都会执行。
|
||||
|
||||
现在让我们转向一个更有用的例子：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_enter_execve { printf("%s called %s\n", comm, str(args->filename)); }'
|
||||
```
|
||||
|
||||
这个例子打印系统中正在创建的每个新进程的父进程名 _(comm)_。_t:syscalls:sys_enter_execve_ 是一个内核追踪点，是 _tracepoint:syscalls:sys_enter_execve_ 的简写，两种形式都可以使用。下一部分会向你展示如何列出所有可用的追踪点。
|
||||
|
||||
_comm_ 是一个 bpftrace 内建指令，代表进程名；_filename_ 是 _t:syscalls:sys_enter_execve_ 追踪点的一个域，这些域可以通过 _args_ 内建指令访问。
|
||||
|
||||
追踪点的所有可用域可以通过这个命令列出：
|
||||
|
||||
```
|
||||
bpftrace -lv "t:syscalls:sys_enter_execve"
|
||||
```
|
||||
|
||||
## 示例用法
|
||||
|
||||
### 列出探针
|
||||
|
||||
_bpftrace_ 的一个核心概念是 **探针点**，即 eBPF 程序可以连接到的（内核或用户空间）代码中的测量点，可以分成以下几大类：
|
||||
|
||||
* _kprobe_——内核函数的开始处
|
||||
* _uprobe_——内核函数的返回处
|
||||
* _uprobe_——用户级函数的开始处
|
||||
* _uretprobe_——用户级函数的返回处
|
||||
* _tracepoint_——内核静态追踪点
|
||||
* _usdt_——用户级静态追踪点
|
||||
* _profile_——基于时间的采样
|
||||
* _interval_——基于时间的输出
|
||||
* _software_——内核软件事件
|
||||
* _hardware_——用户级事件
|
||||
|
||||
所有可用的 _kprobe / kretprobe_、_tracepoints_、_software_ 和 _hardware_ 探针可以通过这个命令列出：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -l
|
||||
```
|
||||
|
||||
_uprobe / uretprobe_ 和 _usdt_ 是用户空间探针，专用于某个可执行文件。要使用这些探针，通过下文中的特殊语法。
|
||||
|
||||
_profile_ 和 _interval_ 探针以固定的时间间隔触发；固定的时间间隔不在本文的范畴内。
|
||||
|
||||
|
||||
### 统计系统调用数
|
||||
|
||||
### Counting system calls
|
||||
|
||||
**Maps** 是保存计数、统计数据和柱状图的特殊 BPF 数据类型，你可以使用映射统计每个系统调用正在被调用的次数：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_enter_* { @[probe] = count(); }'
|
||||
```
|
||||
|
||||
一些探针类型允许使用通配符匹配多个探针，你也可以使用一个逗号隔开的列表为一个操作块指明多个连接点。上面的例子中，操作块连接到了所有名称以 _t:syscalls:sysenter__ 开头的追踪点，即所有可用的系统调用。
|
||||
|
||||
bpftrace 的内建函数 _count()_ 统计系统调用被调用的次数；_@[]_ 代表一个映射（一个关联的数组）。映射的键是另一个内建指令 _probe_，代表完整的探针名。
|
||||
|
||||
这个例子中，相同的操作块连接到了每个系统调用，之后每次有系统调用被调用时，映射就会被更新，映射中和系统调用对应的项就会增加。程序终止时，自动打印出所有声明的映射。
|
||||
|
||||
下面的例子统计所有的系统调用，然后通过 bpftrace 过滤语法使用 _PID_ 过滤出某个特定进程调用的系统调用：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_enter_* / pid == 1234 / { @[probe] = count(); }'
|
||||
```
|
||||
|
||||
### 进程写的字节数
|
||||
|
||||
让我们使用上面的概念分析每个进程正在写的字节数：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_exit_write /args->ret > 0/ { @[comm] = sum(args->ret); }'
|
||||
```
|
||||
|
||||
_bpftrace_ 连接操作块到写系统调用的返回探针（_t:syscalls:sys_exit_write_），然后使用过滤器丢掉代表错误代码的负值（_/arg->ret > 0/_）。
|
||||
|
||||
映射的键 _comm_ 代表调用系统调用的进程名；内建函数 _sum()_ 累计每个映射项或进程写的字节数；_args_ 是一个 `bpftrace` 内建指令，用于访问追踪点的参数和返回值。如果执行成功，_write_ 系统调用返回写的字节数，_arg->ret_ 用于访问这个字节数。
|
||||
|
||||
### 进程的读取大小分布（柱状图）：
|
||||
|
||||
_bpftrace_ 支持创建柱状图。让我们分析一个创建进程的 _read_ 大小分布的柱状图的例子：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 't:syscalls:sys_exit_read { @[comm] = hist(args->ret); }'
|
||||
```
|
||||
|
||||
柱状图是 BPF 映射，因此必须保存为一个映射（_@_），这个例子中映射键是 _comm_。
|
||||
|
||||
这个例子使 _bpftrace_ 给每个调用 _read_ 系统调用的进程生成一个柱状图。要生成一个全局柱状图，直接保存 _hist()_ 函数到 _'@'_（不使用任何键）。
|
||||
|
||||
程序终止时，bpftrace 自动打印出声明的柱状图。创建柱状图的基准值是通过 _args->ret_ 获取到的读取的字节数。
|
||||
|
||||
### 追踪用户空间程序
|
||||
|
||||
你也可以通过 _uprobes / uretprobes_ 和 _USDT_（用户级静态定义的追踪）追踪用户空间程序。下一个例子使用探测用户级函数结尾处的 _uretprobe_ ，获取系统中运行的每个 _bash_ 发出的命令行：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -e 'uretprobe:/bin/bash:readline { printf("readline: \"%s\"\n", str(retval)); }'
|
||||
```
|
||||
|
||||
要列出可执行文件 _bash_ 的所有可用 _uprobes / uretprobes_， 执行这个命令：
|
||||
|
||||
```
|
||||
$ sudo bpftrace -l "uprobe:/bin/bash"
|
||||
```
|
||||
|
||||
_uprobe_ 指向用户级函数执行的开始，_uretprobe_ 指向执行的结束（返回处）；_readline()_ 是 _/bin/bash_ 的一个函数，返回键入的命令行；_retval_ 是被探测的指令的返回值，只能在 _uretprobe_ 访问。
|
||||
|
||||
使用 _uprobes_ 时，你可以用 _arg0..argN_ 访问参数。需要调用 _str()_ 将 _char *_ 指针转化成一个 _string_。
|
||||
|
||||
## 自带脚本
|
||||
|
||||
bpftrace 软件包附带了许多有用的脚本，可以在 _/usr/share/bpftrace/tools/_ 目录找到。
|
||||
|
||||
这些脚本中，你可以找到：
|
||||
|
||||
* _killsnoop.bt_——追踪 `kill()` 系统调用发出的信号
|
||||
* _tcpconnect.bt_——追踪所有的 TCP 网络连接
|
||||
* _pidpersec.bt_——统计每秒钟（通过 fork）创建的新进程
|
||||
* _opensnoop.bt_——追踪 `open()` 系统调用
|
||||
* _bfsstat.bt_——追踪一些 VFS 调用，按秒统计
|
||||
|
||||
你可以直接使用这些脚本，比如：
|
||||
|
||||
```
|
||||
$ sudo /usr/share/bpftrace/tools/killsnoop.bt
|
||||
```
|
||||
|
||||
你也可以在创建新的工具时参考这些脚本。
|
||||
|
||||
## 链接
|
||||
|
||||
* bpftrace 参考指南——<https://github.com/iovisor/bpftrace/blob/master/docs/reference_guide.md>
|
||||
* Linux 2018 `bpftrace`（DTrace 2.0）——<http://www.brendangregg.com/blog/2018-10-08/dtrace-for-linux-2018.html>
|
||||
* BPF：通用的内核虚拟机——<https://lwn.net/Articles/599755/>
|
||||
* Linux Extended BPF（eBPF）Tracing Tools——<http://www.brendangregg.com/ebpf.html>
|
||||
* 深入 BPF：一个阅读材料列表—— [https://qmonnet.github.io/whirl-offload/2016/09/01/dive-into-bpf][6]
|
||||
|
||||
|
||||
|
||||
|
||||
* * *
|
||||
|
||||
_Photo by _[_Roman Romashov_][7]_ on _[_Unsplash_][8]_._
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://fedoramagazine.org/trace-code-in-fedora-with-bpftrace/
|
||||
|
||||
作者：[Augusto Caringi][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[YungeG](https://github.com/YungeG)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://fedoramagazine.org/author/acaringi/
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://fedoramagazine.org/wp-content/uploads/2019/08/bpftrace-816x345.jpg
|
||||
[2]: https://github.com/iovisor/bpftrace
|
||||
[3]: https://lwn.net/Articles/740157/
|
||||
[4]: https://github.com/iovisor/bpf-docs/blob/master/eBPF.md
|
||||
[5]: https://fedoramagazine.org/howto-use-sudo/
|
||||
[6]: https://qmonnet.github.io/whirl-offload/2016/09/01/dive-into-bpf/
|
||||
[7]: https://unsplash.com/@wehavemegapixels?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText
|
||||
[8]: https://unsplash.com/search/photos/trace?utm_source=unsplash&utm_medium=referral&utm_content=creditCopyText
|
||||
@@ -1,243 +0,0 @@
|
||||
[#]: subject: "Use XMLStarlet to parse XML in your the Linux terminal"
|
||||
[#]: via: "https://opensource.com/article/21/7/parse-xml-linux"
|
||||
[#]: author: "Seth Kenlon https://opensource.com/users/seth"
|
||||
[#]: collector: "lujun9972"
|
||||
[#]: translator: "zepoch"
|
||||
[#]: reviewer: " "
|
||||
[#]: publisher: " "
|
||||
[#]: url: " "
|
||||
|
||||
在你的 Linux 终端中使用 XMLStarlet 来解析 XML
|
||||
======
|
||||
|
||||
借助 XMLStarlet，一个终端上的 XML 工具包，你就是 XML 之星。
|
||||
![Penguin with green background][1]
|
||||
|
||||
学习解析 XML 通常被认为是一件复杂的事情，但其实大可不必。[XML 是高度严格结构化的][2]，所以也是相对来说可预测的。也有许多其他工具可以帮助你管理工作。
|
||||
|
||||
我最喜欢的 XML 实用程序之一是 [XMLStarlet][3]， 用于终端的 XML 工具包，借助 XML 工具包，你可以验证、解析、编辑、格式化和转换 XML 数据。XMLStarLet 是个相对最小的命令，但指导 XML 却充满潜力，因此本文演示了如何使用它来查询 XML 数据。
|
||||
|
||||
### 安装
|
||||
|
||||
XMLStarLet 默认安装在 CentOS，Fedora，和许多其他现代 Linux 发行版上，所以你可以打开终端，输入 `xmlstarlet` 来访问它。如果 XMLStarLet 还没有被安装，你的操作系统则会为你安装它。
|
||||
|
||||
或者，你可以用包管理器安装 `xmlstarlet`：
|
||||
|
||||
|
||||
```
|
||||
`$ sudo dnf install xmlstarlet`
|
||||
```
|
||||
|
||||
在 macOS 上，使用 [MacPorts][4] 或 [Homebrew][5]。在 Windows 上，使用 [Chocolatey][6]。
|
||||
|
||||
如果都失败了，您可以从 [Sourceforge 上的源代码][7]手动安装它。
|
||||
|
||||
### 用 XMLStarlet 解析 XML
|
||||
|
||||
有许多工具可以帮助解析和转换XML数据，包括允许您[编写自己的解析器][8]和复杂命令的软件库，如 `fop` 和 `xsltproc`。尽管有时您不需要处理XML数据；您可以很方便的从 XML 数据中来提取、更新或验证重要数据。对于自发的XML交互，我使用 `xmlstarlet`，这是常见的处理 XML任务的一个典型的“瑞士军刀”式应用。通过运行 `--help` 命令，您可以看到它提供哪些选项：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet --help
|
||||
Usage: xmlstarlet [<options>] <command> [<cmd-options>]
|
||||
where <command> is one of:
|
||||
ed (or edit) - Edit/Update XML document(s)
|
||||
sel (or select) - Select data or query XML document(s) (XPATH, etc)
|
||||
tr (or transform) - Transform XML document(s) using XSLT
|
||||
val (or validate) - Validate XML document(s) (well-formed/DTD/XSD/RelaxNG)
|
||||
fo (or format) - Format XML document(s)
|
||||
el (or elements) - Display element structure of XML document
|
||||
c14n (or canonic) - XML canonicalization
|
||||
ls (or list) - List directory as XML
|
||||
[...]
|
||||
```
|
||||
|
||||
您可以通过在这些子命令的末尾附加 `-help` 来获得进一步的帮助：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet sel --help
|
||||
-Q or --quiet - do not write anything to standard output.
|
||||
-C or --comp - display generated XSLT
|
||||
-R or --root - print root element <xsl-select>
|
||||
-T or --text - output is text (default is XML)
|
||||
-I or --indent - indent output
|
||||
[...]
|
||||
```
|
||||
|
||||
#### 用 sel 命令选择数据
|
||||
|
||||
可以使用 `xmlstarlet select`（简称 `sel`）命令查看XML格式的数据。下面是一个简单的XML文档：
|
||||
|
||||
|
||||
```
|
||||
<?xml version="1.0" encoding="UTF-8" standalone="no"?>
|
||||
<xml>
|
||||
<os>
|
||||
<linux>
|
||||
<distribution>
|
||||
<name>Fedora</name>
|
||||
<release>7</release>
|
||||
<codename>Moonshine</codename>
|
||||
<spins>
|
||||
<name>Live</name>
|
||||
<name>Fedora</name>
|
||||
<name>Everything</name>
|
||||
</spins>
|
||||
</distribution>
|
||||
|
||||
<distribution>
|
||||
<name>Fedora Core</name>
|
||||
<release>6</release>
|
||||
<codename>Zod</codename>
|
||||
<spins></spins>
|
||||
</distribution>
|
||||
</linux>
|
||||
</os>
|
||||
</xml>
|
||||
```
|
||||
|
||||
在 XML 文件中查找数据时，您的第一个任务是关注要探索的节点。如果知道节点的路径，请使用 `-value of` 选项指定完整路径。您越早浏览 [Document Object Model][9]（DOM）树，就可以看到更多信息：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
\--value-of /xml/os/linux/distribution \
|
||||
\--nl myfile.xml
|
||||
Fedora
|
||||
7
|
||||
Moonshine
|
||||
|
||||
Live
|
||||
Fedora
|
||||
Everything
|
||||
|
||||
Fedora Core
|
||||
6
|
||||
Zod
|
||||
```
|
||||
|
||||
`--nl` 代表“新的一行”，它插入大量的空白，以确保在输入结果后，终端在新的一行显示。我已经删除了样本输出中的一些多余空间。
|
||||
|
||||
通过进一步深入 DOM 树来凝聚焦点：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
\--value-of /xml/os/linux/distribution/name \
|
||||
\--nl myfile.xml
|
||||
Fedora
|
||||
Fedora Core
|
||||
```
|
||||
|
||||
#### 条件选择
|
||||
|
||||
用于导航和解析 XML 的最强大工具之一被称为 XPath。它控制 XML 搜索中使用的语法，并从 XML 库调用函数。XMLStarlet 能够解析 XPath 表达式，因此可以使用 XPath 函数来有条件的进行选择。XPath 具有丰富的函数，[由 W3C 详细记录][10]，但我觉得 [Mozilla 的 XPath 文档][11]更简洁。
|
||||
|
||||
可以使用方括号作为测试函数，将元素的内容与某个值进行比较。下面是对 `<name>` 元素的值的测试，它仅返回与特定匹配相关联的版本号。
|
||||
|
||||
想象一下，示例 XML 文件包含以 1 开头的所有 Fedora 版本。要查看与旧名称 “Fedora Core” 关联的所有版本号（项目从版本 7 开始名称中的 “Core” 被删除掉了），请执行以下操作：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet sel --template \
|
||||
\--value-of '/xml/os/linux/distribution[name = "Fedora Core"]/release' \
|
||||
\--nl myfile.xml
|
||||
6
|
||||
5
|
||||
4
|
||||
3
|
||||
2
|
||||
1
|
||||
```
|
||||
|
||||
通过将 `--path 的值`更改为 `/xml/os/linux/distribution[name=“Fedora Core”]/codename`，您便可以查看这些版本的所有代号。
|
||||
|
||||
### 匹配路径和获取目标值
|
||||
|
||||
将 XML 标记视为节点的一个优点是，一旦找到节点，就可以将其视为当前的数据“目录”。它不是一个真正的目录，至少不是文件系统意义上的目录，但它是一个可以查询的数据集合。为了帮助您将目标和“其他”的数据分开，XMLStarlet 把您试图用 `--match` 选项匹配的内容和用 `--value` 选项匹配的数据值进行了区分。
|
||||
|
||||
假设你知道 `<spin>` 节点包含几个元素。这就是你的目的节点了。一旦到了这一步，就可以使用 `-value of` 指定要为哪个元素赋值。要查看所有元素，可以使用点（`.`）展示当前位置的所有元素：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet sel --template \
|
||||
\--match '/xml/os/linux/distribution/spin' \
|
||||
\--value-of '.' --nl myfile.xml \
|
||||
Live
|
||||
Fedora
|
||||
Everything
|
||||
```
|
||||
|
||||
与导航 DOM 一样，可以使用 XPath 表达式来限制返回数据的范围。在本例中，我使用 `last（）` 函数来检索 `spin` 节点中的最后一个元素：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
\--match '/xml/os/linux/distribution/spin' \
|
||||
\--value-of '*[last()]' --nl myfile.xml
|
||||
Everything
|
||||
```
|
||||
|
||||
在本例中，我使用 `position（）` 函数选择 `spin` 节点中的特定元素：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
\--match '/xml/os/linux/distribution/spin' \
|
||||
\--value-of '*[position() = 2]' --nl myfile.xml
|
||||
Fedora
|
||||
```
|
||||
|
||||
The `--match` and `--value-of` options can overlap, so it's up to you how you want to use them together. These two expressions, in the case of the sample XML, do the same thing:
|
||||
|
||||
`--match` 和 `--value` 选项可以重叠，因此如何将它们一起使用取决于您自己。对于示例 XML，这两个表达式执行的是相同的操作：
|
||||
|
||||
|
||||
```
|
||||
$ xmlstarlet select --template \
|
||||
\--match '/xml/os/linux/distribution/spin' \
|
||||
\--value-of '.' \
|
||||
\--nl myfile.xml
|
||||
Live
|
||||
Fedora
|
||||
Everything
|
||||
|
||||
$ xmlstarlet select --template \
|
||||
\--match '/xml/os/linux/distribution' \
|
||||
\--value-of 'spin' \
|
||||
\--nl myfile.xml
|
||||
Live
|
||||
Fedora
|
||||
Everything
|
||||
```
|
||||
|
||||
### 熟悉XML
|
||||
|
||||
XML 有时看起来过于冗长和笨拙，但为与之交互和构建的工具却总是让我吃惊。如果您想要好好使用 XML，那么XMLStarlet 可能是一个很好的切入点。下次要打开 XML 文件查看结构化数据时，请尝试使用 XMLStarlet，看看是否可以改为查询该数据。当 XML 越适合你时，它就越能作为一种健壮灵活的数据格式而存在。
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/21/7/parse-xml-linux
|
||||
|
||||
作者：[Seth Kenlon][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[zepoch](https://github.com/zepoch)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/seth
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/lead-images/linux_penguin_green.png?itok=ENdVzW22 "Penguin with green background"
|
||||
[2]: https://opensource.com/article/21/6/what-xml
|
||||
[3]: https://en.wikipedia.org/wiki/XMLStarlet
|
||||
[4]: https://opensource.com/article/20/11/macports
|
||||
[5]: https://opensource.com/article/20/6/homebrew-mac
|
||||
[6]: https://opensource.com/article/20/3/chocolatey
|
||||
[7]: http://xmlstar.sourceforge.net
|
||||
[8]: https://opensource.com/article/21/6/parsing-config-files-java
|
||||
[9]: https://opensource.com/article/21/6/what-xml#dom
|
||||
[10]: https://www.w3.org/TR/1999/REC-xpath-19991116
|
||||
[11]: https://developer.mozilla.org/en-US/docs/Web/XPath/Functions
|
||||
|
||||
@@ -0,0 +1,83 @@
|
||||
[#]: subject: (Install Shutter in Fedora 34 and Above)
|
||||
[#]: via: (https://www.debugpoint.com/2021/07/install-shutter-fedora/)
|
||||
[#]: author: (Arindam https://www.debugpoint.com/author/admin1/)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (geekpi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
在 Fedora 34 及以上版本中安装 Shutter
|
||||
======
|
||||
这个快速指南解释了在 Fedora 34 及以上版本中安装 shutter 所需的步骤。
|
||||
|
||||
截图工具有很多替代和选择。但在我个人看来，没有一个能接近 Shutter 的灵活性。不幸的是，多年来，Linux 发行版，如 Ubuntu、Fedora，由于各种依赖性问题，特别是它的设计方式，面临着将这个应用打包到官方仓库的问题。
|
||||
|
||||
主要问题是它仍然基于 GTK2 和 Perl。当大多数应用转移到 GTK3 时，它仍然是 GTK2。这就造成了一个依赖性问题，因为 Debian/Ubuntu、Fedora 删除了某些包的依赖的 GTK2 版本。
|
||||
|
||||
在 Fedora 34 及以上版本中安装 [Shutter][1] 截图工具需要采用另一种方法。
|
||||
|
||||
现在，你只能通过个人包存档来安装这个工具。下面是如何在 Fedora 34 及以上版本中安装它。
|
||||
|
||||
![Shutter in Fedora][2]
|
||||
|
||||
### 在 Fedora 34 及以上版本中安装 Shutter
|
||||
|
||||
在你的 Fedora 中打开一个终端，启用以下 [Shutter 的 copr 仓库][3]。这个包存档为 Fedora 的 Shutter 提供了一个单独的构建，其中包含了所有未满足的依赖项。
|
||||
|
||||
```
|
||||
sudo dnf copr enable geraldosimiao/shutter
|
||||
```
|
||||
|
||||
完成后，你就可以通过 dnf 在 Fedora 34 及以上版本中简单地安装 shutter。
|
||||
|
||||
```
|
||||
sudo dnf install shutter
|
||||
```
|
||||
|
||||
尽管目前最新的版本是 v0.97。遗憾的是，该仓库目前包含旧的 v0.94.x。我希望版本库的所有者尽快包括最新的版本。
|
||||
|
||||
安装后，你可以通过应用菜单启动它。
|
||||
|
||||
#### 卸载 Shutter
|
||||
|
||||
如果你愿意，你可以通过命令轻松地删除这个第三方仓库：
|
||||
|
||||
```
|
||||
sudo dnf copr remove geraldosimiao/shutter
|
||||
```
|
||||
|
||||
然后按照下面的方法，完全删除 Shutter，包括依赖关系。
|
||||
|
||||
```
|
||||
sudo dnf autoremove shutter
|
||||
```
|
||||
|
||||
#### 在其他 Linux 发行版中安装 Shutter
|
||||
|
||||
如果你想在 Debian、Ubuntu 或相关发行版中安装它，请[查看此指南][4]。
|
||||
|
||||
### Shutter 的开发
|
||||
|
||||
最近，这个项目[转移到了 GitHub][6]，以便更好地合作，并且正在进行 GTK3 移植。而且它相当活跃，最近还发布了一个版本。我们希望它能尽快被移植到 gtk3 上，并在各发行版的原生仓库中可用。
|
||||
|
||||
如果你在安装 Shutter 时遇到任何错误，请在评论栏告诉我。
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://www.debugpoint.com/2021/07/install-shutter-fedora/
|
||||
|
||||
作者：[Arindam][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[geekpi](https://github.com/geekpi)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://www.debugpoint.com/author/admin1/
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://www.debugpoint.com/tag/shutter
|
||||
[2]: https://www.debugpoint.com/blog/wp-content/uploads/2021/07/Shutter-in-Fedora.jpeg
|
||||
[3]: https://copr.fedorainfracloud.org/coprs/geraldosimiao/shutter/
|
||||
[4]: https://www.debugpoint.com/2020/04/shutter-install-ubuntu/
|
||||
[6]: https://github.com/shutter-project/shutter
|
||||
@@ -0,0 +1,110 @@
|
||||
[#]: subject: (Check used disk space on Linux with du)
|
||||
[#]: via: (https://opensource.com/article/21/7/check-disk-space-linux-du)
|
||||
[#]: author: (Seth Kenlon https://opensource.com/users/seth)
|
||||
[#]: collector: (lujun9972)
|
||||
[#]: translator: (geekpi)
|
||||
[#]: reviewer: ( )
|
||||
[#]: publisher: ( )
|
||||
[#]: url: ( )
|
||||
|
||||
用 du 检查 Linux 上已使用的磁盘空间
|
||||
======
|
||||
用 Linux 的 du 命令了解你正在使用多少磁盘空间。
|
||||
![Check disk usage][1]
|
||||
|
||||
无论你有多少存储空间，它总是有可能被填满。在大多数个人设备上，驱动器被照片、视频和音乐填满，但在服务器上，由于用户账户和日志文件中的数据，空间减少是很正常的。无论你是负责管理一个多用户系统，还是只负责自己的笔记本电脑，你都可以用 `du` 命令检查磁盘的使用情况。
|
||||
|
||||
默认情况下，`du` 提供了你当前目录中使用的磁盘空间，以及每个子目录的大小。
|
||||
|
||||
|
||||
```
|
||||
$ du
|
||||
12 ./.backups
|
||||
60 .
|
||||
```
|
||||
|
||||
在这个例子中，我的当前目录总共占用了 60KB，其中 12KB 被子目录 `.backups` 占用。
|
||||
|
||||
如果你觉得这很混乱，并希望分别看到所有的大小，你可以使用 `--separate-dirs`（或简写 `S`）选项：
|
||||
|
||||
|
||||
```
|
||||
$ du --separate-dirs
|
||||
12 ./.backups
|
||||
48 .
|
||||
```
|
||||
|
||||
这是相同的信息（48 加 12 是 60），但每个目录被独立处理。
|
||||
|
||||
要想看到更多的细节，可以使用 --all（简写 -a）选项，它显示每个目录中的每个文件：
|
||||
|
||||
|
||||
```
|
||||
$ du --separate-dirs --all
|
||||
4 ./example.adoc
|
||||
28 ./graphic.png
|
||||
4 ./.backups/example.adoc~
|
||||
12 ./.backups
|
||||
4 ./index.html
|
||||
4 ./index.adoc
|
||||
48 .
|
||||
```
|
||||
|
||||
### 查看文件的修改时间
|
||||
|
||||
当查看文件以找出占用空间的内容时，查看文件最后一次被修改的时间是很有用的。一年内没有使用的文件很可能是归档的候选文件，特别是当你的空间快用完时。
|
||||
|
||||
用 du 查看文件的修改时间，使用 `--time` 选项：
|
||||
|
||||
|
||||
```
|
||||
$ du --separate-dirs --all --time
|
||||
28 2021-07-21 11:12 ./graphic.png
|
||||
4 2021-07-03 10:43 ./example.adoc
|
||||
4 2021-07-13 13:03 ./index.html
|
||||
4 2021-07-23 14:18 ./index.adoc
|
||||
48 2021-07-23 14:19 .
|
||||
```
|
||||
|
||||
### 为文件大小设置一个阈值
|
||||
|
||||
当为了磁盘空间而查看文件时，你可能只关心较大的文件。你可以用 `--threshold`（简写 `-t`）选项为你想看的文件大小设置一个阈值。例如，只查看大于 1GB 的文件：
|
||||
|
||||
|
||||
```
|
||||
$ \du --separate-dirs --all --time --threshold=1G ~/Footage/
|
||||
1839008 2021-07-14 13:55 /home/tux/Footage/snowfall.mp4
|
||||
1577980 2020-04-11 13:10 /home/tux/Footage/waterfall.mp4
|
||||
8588936 2021-07-14 13:55 /home/tux/Footage/
|
||||
```
|
||||
|
||||
当文件特别大时，它们可能难以阅读。使用 `--human-readable`（简写 `-h`）选项可以使文件大小更容易阅读：
|
||||
|
||||
|
||||
```
|
||||
$ \du --separate-dirs --all --time \
|
||||
\--threshold=1G --human-readable ~/Footage/
|
||||
1.8G 2021-07-14 13:55 /home/tux/Footage/snowfall.mp4
|
||||
1.6G 2020-04-11 13:10 /home/tux/Footage/waterfall.mp4
|
||||
8.5G 2021-07-14 13:55 /home/tux/Footage/
|
||||
```
|
||||
|
||||
### 查看可用磁盘空间
|
||||
|
||||
如果只是想获得一个驱动器上还剩下多少磁盘空间的摘要，请阅读我们关于 [df 命令][2]的文章。
|
||||
|
||||
--------------------------------------------------------------------------------
|
||||
|
||||
via: https://opensource.com/article/21/7/check-disk-space-linux-du
|
||||
|
||||
作者：[Seth Kenlon][a]
|
||||
选题：[lujun9972][b]
|
||||
译者：[geekpi](https://github.com/geekpi)
|
||||
校对：[校对者ID](https://github.com/校对者ID)
|
||||
|
||||
本文由 [LCTT](https://github.com/LCTT/TranslateProject) 原创编译，[Linux中国](https://linux.cn/) 荣誉推出
|
||||
|
||||
[a]: https://opensource.com/users/seth
|
||||
[b]: https://github.com/lujun9972
|
||||
[1]: https://opensource.com/sites/default/files/styles/image-full-size/public/du-splash.png?itok=nRLlI-5A (Check disk usage)
|
||||
[2]: https://opensource.com/article/21/7/use-df-check-free-disk-space-linux
|
||||
Reference in New Issue
Block a user