导言
While reading various research papers, it's fascinating to observe a trend where researchers use names from popular culture to label their scientific work. This practice not only adds a cool and captivating touch to the research but also leaves a lasting impression, e.g, transformer in AI. Popular culture symbols and names serve as powerful tools when incorporated into research titles, especially when they align with the core concept.
导言
随着翻译软件逐渐结合到工作流中，再加上chatgpt等润色英文写作。我相信在未来读写不会再是难点。但是与外国人直接打交道的听说还是需要积累的。（听的话假如是zoom会有实时翻译，说反而更重要。You should always reading the blog content
内容选择
科技相关的有意思的TED视频演讲视频。
导言
听师兄说国企里相对有钱点的是邮储，电网和烟草。来研究一波
默认安装了ohmytmux时的常见命令。
默认的tmux快捷键
<prefix> + maximizes the current pane to a new window
#ifndef _LIBCPP_TYPE_VIS
# if !defined(_LIBCPP_DISABLE_VISIBILITY_ANNOTATIONS)
# define _LIBCPP_TYPE_VIS __attribute__ ((__visibility__("default")))
# else
# define _LIBCPP_TYPE_VIS
# endif
#endif
#ifndef _LIBCPP_THREAD_SAFETY_ANNOTATION
# ifdef _LIBCPP_HAS_THREAD_SAFETY_ANNOTATIONS
# define _LIBCPP_THREAD_SAFETY_ANNOTATION(x) __attribute__((x))
# else
# define _LIBCPP_THREAD_SAFETY_ANNOTATION(x)
# endif
#endif // _LIBCPP_THREAD_SAFETY_ANNOTATION
class _LIBCPP_TYPE_VIS _LIBCPP_THREAD_SAFETY_ANNOTATION(capability("mutex")) mutex
{
}
It's part of code from __mutex_base
暂无
暂无
上面回答部分来自ChatGPT-3.5，没有进行正确性的交叉校验。
无
暂无
暂无
上面回答部分来自ChatGPT-3.5，没有进行正确性的交叉校验。
无
其余cmake有的, Scons 也有。
Sconstruct python file as compile entryscons commandAddOption('--buildDir',
dest='buildDir',
type='string',
default="build/",
# default=False,
nargs=1,
action='store', # meaning save the string
# or action='store', meaning True or false
metavar='DIR',
help='Base build directory'
)
baseBuildDir = GetOption('buildDir')
variant_dirusing scons debug=1 command.
SConstruct file, define the build environment by creating an Environment object. You can specify compiler options, flags, include paths, library paths, and other build settings within this object.env = Environment(CXX='g++', CCFLAGS=['-O2', '-Wall'], CPPPATH=['include'], LIBPATH=['lib'])
libEnv = env.Clone()
Program() function.source_files = ['main.cpp', 'util.cpp', 'other.cpp']
# or select the src files
Object('hello.cpp')
program = env.Program(target='my_program', source=source_files)
In this example, main.cpp, util.cpp, and other.cpp are the source files, and my_program is the name of the target executable.
# static
Library("t", Glob("src/*.cpp"))
# dynamic
source = Glob("src/*.cpp")
SharedLibrary("t", source)
Program(["hello.cpp"], LIBS=["t"], LIBPATH=".")
execute command during compilation
this is usually to print info
allSrcs, ".git/index", or "SConstruct") change.env.Command(
target='bar.out',
source='bar.in',
action=["rm -f $TARGET", "$BAR_BUILD < $SOURCES > $TARGET"],
ENV={'PATH': '/usr/local/bin/'},
)
env.Command(
versionFile,
allSrcs + [".git/index" "SConstruct"],
'printf "#define ZSIM_BUILDDATE \\"`date "+%Y-%m-%d %T"`\\"\\n#define ZSIM_BUILDVERSION \\"`python misc/getver.py`\\"" >>' + versionFile)
TODO: multipim how to add a singel head file during compilation process.
暂无
暂无
上面回答部分来自ChatGPT-3.5，没有进行正确性的交叉校验。
https://scons.org/doc/production/PDF/scons-man.pdf
工作外的内容聚焦于几个方面：
暂无
暂无
上面回答部分来自ChatGPT-3.5，没有进行正确性的交叉校验。
无
淡而不厌，简而文，温而理，知远之近，知风之自，知微之显，可与人德矣。
华为实习部门的都是手握A会的博士大佬。本人望尘莫及，我会狠狠吸收的。
体系结构量化分析方法，重点就在于量化分析开销，比较然后进行tradeoff。当前前提是你要有基本的相关概念。
具体知识来源的优先级，或者说如何使用搜索引擎：
在理解概念，量化了具体场景的数值后，就可以开心进行tradeoff了。
注意项目的可读性和可拓展性一般与性能是不兼容的。这取决于项目的checkpoint/middleValue的保存，在性能优化时往往会消除中间变量。这样会导致代码的可读性和可拓展性下降。
困难的定义可以基于以下几个要素进行评估：
上面回答部分来自ChatGPT-3.5，没有进行正确性的交叉校验。
无