ECF：e.g. hardware timer, packet arrived, request data from disk and sleep until the data are ready, 子进程terminate
-
nonlocal jumps (provided in C via the
setjumpandlongjump) -
概念event：change in state
-
handler的返回方式：1.返回到$I_{curr}$ 2.返回到$I_{next}$ 3.abort
- interrupt - 2.
- trap - 2.
- fault - 1. or 3.
- abort - 3.
-
思考：返回到$I_{next}$的情形值得注意，例如一个空的handler可以中断sleep。是否可能会丢失一些需要curr指令处理的状态，然后出现不可预知的问题？
-
exception number, exception table base register
exception和procedure call的区别
- The processor also pushes some additional processor state onto the stack that will be necessary to restart the interrupted program when the handler returns. For example, an x86-64 system pushes the EFLAGS register containing the current condition codes, among other things, onto the stack.
- When control is being transferred from a user program to the kernel, all of these items are pushed onto the kernel’s stack rather than onto the user’s stack. e.g. timer_interrupt存user register到kernel stack
- Exception handlers run in kernel mode (Section 8.2.4), which means they have complete access to all system resources.
exceptions的类别：interrupts, traps, faults, aborts，分为同步和异步，只有interrupts是异步的
- trap是syscall
- faults例如page fault handler
汇编：All arguments to Linux system calls are passed through general-purpose registers rather than the stack. By convention, register %rax contains the syscall number, with up to six arguments in %rdi, %rsi, %rdx, %r10, %r8, and %r9. The first argument is in %rdi, the second in %rsi
特点：
- An independent logical control flow that provides the illusion that our pro- gram has exclusive use of the processor.
- A private address space that provides the illusion that our program has exclu- sive use of the memory system.
概念：
- parallel flow: 运行在不同的cores或computers上
Linux x86-64进程地址空间
mode bit (in control register) is set => 内核态
Linux能获取内核信息的文件系统
- /proc /proc/cpuinfo /proc/process-id/maps
- /sys: additional low-level information about system buses and devices
context switches
进程上下文：
- kernel's data structure: page/process/file table
- values of objects: the general-purpose registers, the floating-point registers, the program counter, user’s stack, status registers, kernel’s stack
处理error的，<errno.h>
- error-reporting function
- error-handling wrappers
设置exit status的方式：1）输入exit参数 2）main function的return value
fork的其中一个特性是share files，比如共享stdout file，print到同一个地方
画process graph辅助理解nested fork的行为
8.4.3: reap child processes, p779, waitpid的详细介绍
如果parent不reap，由kernel创建的pid为1的init进程reap zombie children
execve(p786)
unlike fork, which is called once but returns twice, execve is called once and never returns.
load file-name之后，call the start-up code
p788: user-stack结构图 %rdi %rsi %rdx: argc argv envp
env相关的函数
char *getenv(const char *name);int setenv(const char *name, const char *newvalue, int overwrite);void unsetenv(const char *name);
p793: Linux Signals man 7 signal
signal体现了kernel exception handlers和user space的交互
pending signal，后续的相同signal会被discard，process可以block signal
- linux中，
pending和blockvector
process groups: getpgrp(), setgpid(pid,pgid)
kill和alarm函数 p798
install the handler -> catch the signal -> handle the signal
e.g. 一个空return handler可以让sleep进程return
实现handler的关键细节
p801: block机制，async-signal-safe functions，特征：reentrant，或者不会被signal handler interrupt
Declare global variables with volatile: 这样不会被编译器优化，确保handler更新的global值能被main知道
p811: Signal wrapper
p817: 用sigsuspend处理spin loop问题
user-level层面的ECF，setjmp类比catch，sigsetjmp类比throw
#include <setjmp.h>
int setjmp(jmp_buf env);
int sigsetjmp(sigjmp_buf env, int savesigs);
//Returns: 0 from setjmp, nonzero from longjmps