• Mid term questions are skipped
• Progress: 1,2,3,4,5,6,7,8,9,10,11,12,13 (total 13)

1. Device drivers are broadly classified into two basic categories: character devices and block devices. Please individually describe them. (10%)

Appeared in

99, 99_2, 101, 102, 104, 105_1, 105_2, 106, 108

Derivation

• Please simply describe the Device Driver concepts. (10%) 103_2
  • 驅動程式，允許High-Level的軟體與硬體溝通的程式，建立了一個硬體與硬體，或硬體與軟體溝通的介面，讓device可以交換資料。

Answer

Character devices

一個序列的資料流，以不定長度的字元來傳送資料

• Character devices are accessed as stream of data.
• e.g. Serial port

block devices

以一個block為單位，在可定址的隨機位址可讀取或寫入資料

• Read and write blocks of data from random locations on addressable medium.
• e,g, disks

2. Please refer to below example code of a Linux driver and then describe the functionalities of Section 1 and Section 2.

static int my_major = 0;
…
If (my_major) {
	// Section 1
	dev = MKDEV(my_major, my_minor);
	result = register_chrdev_region(dev, nr_devs, “mydev”);
} else {
	// Section 2
	result = alloc_chrdev_region(&dev, my_minor, nr_devs, “mydev”);
	my_major = MAJOR(dev);
}
If (result < 0){
	printk(KERN_WARNING “mydev: can’t get major %d\n”, my_major);
	return result;
}

Appeared in

103, 105_1, 106, 108

Answer

Section 1:

• pre assigned
• if my_major!=0, register device by using the assigned mymajor, myminor

Section 2:

• dynamic alloc (動態配置號碼)
• else, dynamically register device

3.What is the Proc File System? Please describe the functionalities of /proc file system. (10%)

Appeared in

101, 103_2, 105_1, 105_2, 106, 108

Answer

一個虛擬的檔案系統，實現Linux核心空間與使用者空間的通訊

• Interface that allows the kernel to communicate information about each running process on a Linux system.
• Communication between kernel space and user space

4. What is Memory Technology Devices (MTD) subsystem? (10%)

Appeared in

99, 99_2, 101, 102, 103, 103_2, 104, 105, 106, 108

Answer

Application和driver中間的虛擬層。讓Application用相同的指令對不同的flash driver做相同的事，且可以設定欲使用的driver型號。

• A type of device file to interact with flash memory.
• The MTD layer architecture enables the separation of the lowlevel device complexities from the higher-layer data organization and storage formats that use memory devices.

5. Why the engineers usually mount a NFS to develop an embedded system? Please point out the advantages of NFS for building an embedded system. (10%)

Appeared in

99_2, 101, 104, 105_1, 105_2, 106, 108

Answer

使用NFS可以把程式透過網路的方式將資料儲存在其他地方，如要使用或修改的時候一樣在透過網路的方式進行，讓儲存空間不在受限於硬體儲存空間。
可以直接在NFS上的系統安裝其他工具程式直接進行編譯，可以增加程式開發與效能的速度。

• Network File System
• NFS enables you to export a directory on an NFS server and mount that directory on a remote client machine as if it were a local file system. In other words, access file on a server machine like it is a local file system.
• developer can have access to a huge number of files, libraries, tools, and utilities during development and debugging, even if the target embedded system is resource constrained.

6. Please list the advantages of cramfs. (10%)

Appeared in

102, 104, 105_1, 106, 108

Answer

cramfs是專門針對快閃記憶體設計的唯讀壓縮的檔案系統，不是將檔案系統中的所有內容解壓到ROM中，而是系統需要訪問某個位置的資料時再將其解壓縮到ROM中，節省ROM的儲存空間，使用簡單，加載容易，速度快

• A read-only ,compressed file system. Simple and small, ideal for boot ROMS

7. What is the BusyBox in an embedded system? How to add a new command into the embedded system except the commands that BusyBox provides. (10%)

Appeared in

99, 99_2, 102,103, 103_2, 104, 105, 106, 108

Answer

把常用的指令和參數壓縮成一個檔案

• BusyBox provides compact replacements for many traditional fullblown utilities found on most desktop and embedded Linux distributions.

1.在Busybox中勾選欲增加的指令
2.將欲增加的指令Source Code編譯後放在\bin中

• add a new command
  • 1. cross compile c then copy to bin
  • 2. 修改busybox設定

8. Please describe the functionalities and their advantages of DHCP server and NFS server. (10%)

Appeared in

103

Answer

DHCP

DHCP server: 給內部網絡或網絡服務供應商自動分配IP位址、主機名、DNS伺服器、域名，方便管理。在子網移動中不需要重新設置IP，自動獲取就行。

• A DHCP server controls the IP address assignments for IP subnets for which it has been configured, and for DHCP or BOOTP clients that have been configured to participate. A DHCP server listens for requests from a DHCP client (such as your target board), and assigns addresses and other pertinent information to the client as part of the boot process.

NFS

NFS server: 讓不同的機器、不同的作業系統透過網路可以彼此分享個別的檔案。將常用的數據存放在一台服務器，可以通過網路訪問，減少本地端磁碟空間的使用。

• Network File System
• Your root file system is not size-restricted by your board's own limited resources, such as Flash memory.
• Changes made to your application files during development are immediately available to your target system.
• You can debug and boot your kernel before developing and debugging your root file
  system.

9 Fig. 1 shows a typical DHCP exchange, please explain the each step (4 steps) in Fig. 1. (10%)

Appeared in

103_2, 105_1

Answer

1. Client要求租用IP位址

• The sequence starts with the client (target) transmitting a broadcast frame attempting to discover a DHCP server. This is shown by the DHCPDISCOVER message shown.

2. DHCP提供可租用的IP位址

• The server responds (if it has been so configured and enabled) by offering an IP address for the client. This is evidenced by the DHCPOFFER message.

3. Client確認DHCP的IP租約

• The client then responds by testing this IP address locally. The testing includes sending the DHCPREQUEST packet to the DHCP server, as shown.

4. DHCP Server同意IP租約

• Finally, the server responds by acknowledging the IP address assignment to the client, thus completing the automatic target configuration.

10. What advantages of Ext3 File System compared to Ext2 File System? How to convert an Ext2 File System to the Ext3 File System? (10%)

Appeared in

103, 105_1, 106, 108

Answer

Advantage

Ext3加入了日誌檔用來記錄之前做了哪些事情，提高檔案系統的完整性，避免意外當機對檔案系統的破壞。

• Availability: By journaling, after an unexpected power failure or system crash, file system check is not necessary
• Data complete: Journaling can prevent the data damage.
• speed: ext3's journaling optimizes hard drive head motion.
• Easy to convert: Don't have to reformat to migrate from ext2 to ext3

Convert

先將要轉換的磁碟先mount上去，再輸入"tune2fs -j 磁碟名稱"，就會產生一個日誌檔

• mount ext2 file system
  mount /dev/sdb1 /mnt/flash
• create journal
  tune2fs -j /dev/sdb1

11 How to build a simple EXT2 file system and modify its content (add and delete directories, make device nodes, and so on).

Appeared in

105_1, 106, 108

Answer

在一個一般的檔案中使用Linux loopback device建立一個EXT2 file system image，建立好之後我們就能在這個檔案裡擁有一個 EXT2 file system，如此一來就能對這個一般的檔案像對block device進行操作和調整。

Create file image

• mke2fs

make directory

*　mkdir

remove directory

• rmdir or rm

make device node

• mknod

12 Please describe the functionalities of below module utilities. (10%)

​​​​(1) insmod
​​​​(2) lsmod
​​​​(3) modprobe
​​​​(4) depmod
​​​​(5) rmmod

Appeared in

106, 108

Answer

insmod

載入模組到kernel

• insert a module into a running kernel.

lsmod

列出目前已經載入的driver

• displays a formatted list of the modules that are inserted into the kernel

modprobe

Linux kernel新增或刪除模組
*和insmod不同的是此指令可以根據模組相依性載入

• discover this relationship and load the dependent modules in the proper order.

depmod

查看這支driver是否與其他driver有相依性，就是A driver要能動必須先安裝B driver，生成modules.dep檔案和對映檔案。

• When modprobe is executed, it searches for a file called modules.dep in the same location where the modules are installed

rmmod

刪除kernel中一個正在執行的模組

• It simply removes a module from a running kernel

13 Please list the advantages of JFFS2. (10%)

Appeared in

99

Answer

1. 支援 NAND flash memory 裝置。
2. 提供Hard links support，JFFS只能用soft link。
3. 提供zlib、rubin、rtime壓縮。
4. 比JFFS有更好的效能，透過演算法去收集垃圾資料將其清除。

• Support for NAND flash devices
• Hard links
• Compression
• Better performance: JFFS treated the disk as a purelycircular log. This generated a great deal of unnecessary I/O. The garbage collection algorithm in JFFS2 makes this mostly unnecessary.

14 What is the Initial RAM Disk? What is the purpose of Initial RAM Disk?

Answer

• (參考某題的答案)Initial ramdisk is a Linux kernel feature to mount an early root file system to perform certain startup-related system initialization and configuration
• The initial RAM disk is a small self-contained root file system that usually contains directives to load specific device drivers before the completion of the boot cycle.
• 中文解釋：初始RAM磁盤（Initial RAM Disk）是Linux系統中的一個臨時根文件系統，用於在系統引導過程中掛載以支持兩階段引導過程。它包含各種可執行文件和驅動程序，允許掛載真正的根文件系統，然後卸載initrd RAM磁盤並釋放其內存。在許多嵌入式Linux系統中，initrd是最終的根文件系統。初始RAM磁盤（initrd）和initramfs是Linux系統中加載臨時根文件系統的兩種方法。它們常用於在實際的根文件系統可以掛載之前進行準備工作。當使用initrd時，系統通常按以下步驟啟動：引導加載程序加載內核和初始RAM磁盤，內核將initrd轉換為“正常”的RAM磁盤，並釋放initrd使用的內存。如果根設備不是/dev/ram0，則會遵循舊的（已棄用的）change_root過程。初始RAM磁盤的目的是在系統引導過程中提供一個臨時的根文件系統，以便進行各種準備工作，例如加載驅動程序、設置設備、配置網絡等。它允許在實際的根文件系統被掛載之前執行必要的操作，從而確保系統能夠成功引導並進入正常運行狀態。一旦真正的根文件系統準備就緒，初始RAM磁盤就會被卸載，系統將轉移到真正的根文件系統上進行操作。總結起來，初始RAM磁盤是Linux系統中的一個臨時根文件系統，用於在引導過程中提供必要的準備工作，以確保系統能夠成功引導並進入正常運行狀態。

tags: Embedded system CSnote