Linux hardware Installation

1. How much hard disk space does Linux installation occupy?

It depends on what distribution Linux is to be installed, how many applications need to be loaded, and whether to install the original code.

For example, the most basic Slackware Linux requires about 10 MB of space, including 80 Mb for X Windows, if you want to install basic system tool programs, common application software, and TCP/IP network software (that is, software A, B, and N Series serial numbers), about 40-50 MB, to install all the packages, about MB.

For RedHat Linux, the minimum installation (only the most basic system) is about 60 M, and the default installation is about M. If all the installation is performed, about M is required, if it includes the original code, X-windows, some common tools, applications, and a certain amount of spool space, it will take about 450 ~ About MB.

Debian GNU/Linux installation requires 500 Mb -- 1 GB.

2. How do I install Linux hardware drivers?

Linux supports a wide variety of computer hardware, which is achieved through the software Driver. These Drivers are included in the Kernel Source code, such as learning the Driver writing method, go to the/usr/src/linux/drivers directory and find the driver source code, or refer to the Kernel Hack FAQ.

In a Linux environment, you need to "Mount" a Driver in a different way than DOS. In DOS. in SYS, the corresponding driver is attached. The Linux method is attached by Recompile Kernel. During the compilation process, to select required functions and hardware categories, all corresponding Drivers will be automatically compiled into the Kernel and become a part of the Kernel image. After Linux2.0, a Module with more powerful functions is added, allowing you to load or detach drivers at any time, making installation of Linux hardware drivers easier and safer.

3. How to install a display card?

For the i740display card, use the ax4.14254100.000.tar.gz; i740/NeoMagic display card to download an rpm package. For versions earlier than RedHat6.0 (including 6.0), because the graphics card database does not have i740, You need to download a new graphics. i386.taz, download the xf86config-glibc-1.1.0.i386.tgz XFCom-i740-glibc-1.10-1.i386.taz files, then extract them to the/tmp/usr/X11R6/bin/directory, copy to the corresponding directory of the root directory, and finally configure xf86config. For commercial i740 graphics card drivers, visit: http; // www.metrolink.com. I740 graphics card Xserver has AccelerateX 4.1, but not free, the price is very high.

For S3 V2/G2 (such as Winfast S680), use AX4.1 + 016th update. As for Riva TNT, there is XF86 Patch on Nvidia's home page, and the new XF86 3.3.3 is also supported.

For S3's Savage 3D graphics card, the Linux distribution kit can only use the X server of VGA16, which can only reach the resolution in X windows. The following method can be used to allow Linux to support the Savage 3D graphics card:

(1) download the Savage 3D graphics card X server and copy it to the temporary directory/temp.

(2) Follow SVGA to configure the XF86Config file and log on to the/usr/bin/X11R6 directory as ROOT. Enter XF86Config to start configuration and configure the mouse, keyboard, display, and display card, the provisioner asks if it is selected from its graphics card database because it only supports VGA16 for the Savage 3D card, so select NO, then select 3 use the XF86-SVGA server, establish the symbolic link and select YES.

(3) install the X server of the Savage 3D video card, enter the temporary directory/temp, and use rpm-I Xfree86-S3D-0.1-3. i386.rpm installation, in/usr/bin/X11R6, cpXF86-SVGA XSVGA, cp XF86-S3D XF86-SVGA, and finally input startx-bpp32 to start X windows with true color.

18. How do I install a video card without a driver in X windows? How can I view the display card configuration in X window?

You only need to load the driver of the main chip of the video card (for example, the main chip of the WINFAST S680 display card is S3VIRGE GX/2 ).

Use SuperProbe in the console to view the configuration of the display card.

4, how to configure the X-WINDOW PS/2 mouse? What should I pay attention to when installing RedHat Linux with the mouse? X in windows, how does the mouse not listen?

The easier way is to use XF86Setup for configuration. In addition, you can directly modify the [MOUSE] section in the XF86Config file to make sure that the MOUSE device in the file points to/dev/psaux.

Note the following before installing RedHat Linux:

First, check whether there is a/etc/sysconfig/mouse file. The content is as follows:

Mouse type = "Microsoft"

XEMU3 = yes

In addition, you must have a/etc/rc. d/init. d/gpm file. Of course, the settings must be consistent with the mouse type. Note: In popular computers, the mouse type is "PS/2 ".

The mouse does not listen to the call solution: There are several lines in the/etc/X11/XF86Config file (Redhat) or/etc/XF86Config (slackware) file:

Section "Pointer"

Protocol "MouseSystems"

Device "/dev/mouse"

Check the protocols and interfaces used by the mouse.

5. How do I set the keyboard in RedHat Linux?

If you omit the keyboard setting during installation or use another keyboard, You need:

(1) select an appropriate keyboard table from/usr/lib/kbd/keytables/. For example, it. map represents the Italian keyboard.

(2) edit/etc/sysconfig/keyboard to use: KEYTABLE = "/usr/lib/kbd/keytables/it. map"

(3) set the keyboard repetition rate and latency, and add the following content to one of the documents of/etc/profile or rc:

/Sbin/kbdrate-s-r 16-d 500 # or whatever -- or another value

(4) finally run the following commands to allow keyboard settings :~ # Chmod 666/dev/port

The keyboard will work normally at next boot.

6. How much memory is required to run Linux?

It must be at least 4 MB. If it is 4 MB, it will be more difficult to install. For character terminal mode, at least 4 MB is required; for X-windows, at least 8 MB is required. If Linux is used as the server, at least 16 MB of memory is required. 64 MB of memory is required for running Netscape and other software.

7. What should I do if the system recognizes less memory than the actual number (for example, 128 MB?

Add the mem = 128 M parameter to boot.

1) if loadlin is used, add mem = 128 m to the end of the startup line in the linux. bat file.

2) If lilo is used, add mem = 128 m to lilo. conf, run lilo, and start again.

8. How to install a sound card in RedHat Linux?

In RedHat4.2, it is still troublesome to install the sound card. in current 6.2, the kernel supports the sound card. If it is a soundblaster sound card, you only need to use the sndconfig command as the root; if it is not an sb sound card, you can use the following methods:

1) Use sndconfig to set

2) Modify/etc/conf. modules and add the following lines:

Alias sound sb

Options-k sb io = 0x220 irq = 5 dma = 1, 5

Alias midi opl3

Options-k opl3 io = 0x388

3) in win98/dos, check the current interrupt irq and I/o address settings of the sound card, and manually mount the sound card module, such as modprobe opl3 irq = 5 dma = 220 io =

4) if you still cannot work, you have to re-compile a new kernel for the sound card. in/usr/src/linux, run make menuconfig. If the original code has not been loaded, run:

Rpm-I/mnt/cdrom/Redhat/RPMS/Kernel-source *. rpm

9. Does Linux support plug-and-play sound cards?

In Linux 2.1, plug-and-play should be fully supported, and work should be done to enable plug-and-play sound cards.

1) if you have a PC with Plug and Play BIOS, you must carefully set the sound card to ensure that the I/O address, IRQ, DMA, and other settings of the core sound driver are the same as those provided by the BIOS.

2) in Linux, the plug-and-play tool used to set this type of card can be found at the Red Hat website http://www.redhat.com/or included in the Linux package.

3) if this type of card is used in Windows 98, use the Device Manager to set the card, and then use LOADLIN to start the card in Linux. Make sure that Windows 98 and Linux use the same set parameters.

4) if this type of card is used in DOS, you can use the icu public program attached to the 16PnP card of the audio card, set it in DOS, and then use LOADLIN to start Linux, make sure that the same set parameters are used for DOS and Linux.

The commercial OSS audio driver supports the audio card 16PnP and can be purchased from 4 Front Technologies.

10. How to configure the sound card

For the sound card of ISA/PnP, set isapnp first.

If you are using the SoundBlaster series and RH5.1 or later versions, you can use the sndconfig to configure the sound card. No isapnp configuration is required.

If the sound card is already working, the configuration information of the sound card is available in/etc/sndstat. If this file is not available or it is empty, the sound card is not driven (some PCI sound cards may be different in 2.2.127). Check whether files such as/dev/audio/dev/dsp/dev/mixer exist. If the Sound card type cannot be found in the Kernel configuration, you can try to configure it in the SoundBlaster compatible mode. If not, you can download OSS (Open Sound System) or OSSFree, Which is commercial software, although the trial version does not stop soundoff/soundon, It is very troublesome. 2.2.12x kernel supports some new sound cards, especially PCI sound cards.

11. After the core is compiled, reboot cannot automatically load the sound card. You can use it by typing cat/dev/sndstat at the prompt. How can this problem be solved?

You can use the following methods:

1) Modify/etc/rc. d/rc. modules

2) Search for the words "sound support"

3) Add "/sbin/modprobe sound"

4) Restart

13. Can I use a soft cat in Linux?

Because this cat does not have a DSP chip, it uses a software-driven CPU to undertake DSP work. unless the manufacturer supports it, it cannot be used in linux.

14. The PNP Modem was originally working normally in Linux, because a digital camera was connected to com1, and WIN98 PNP got a bit (the ports were automatically adjusted), and so on, it went back to Linux, what should I do?

Solution:

1) shield com2 on the motherboard

2) Run pnpdump> isapnp. conf and modify isapnp. conf as needed.

3) set pnp to directly copy isapnp. conf to/etc.

4) Try setserial/dev/cua1-port 0x2f8-irq3-UART16550A-v.

If not, Redhat 6.1 is simply reinstalled. Or try compiling the serial port into a module: Compile the serial port into the core, so that the initialization time is much earlier than the time to call isapnp, so isapnp is done in white. After being compiled into a module, isapnp is called first and then modprobe is used to solve the problem.

15. The built-in modem is set to IRQ12 and IO3e8, but in Linux, COM3 occupies IRQ4 and IO 3e8. How can I change the settings?

Disable 0x3e8 in BIOS. If the built-in modem is plug-and-play, it may be re-set to IRQ4 by the plug-and-play BIOS. IRQ12 should be set by the setup program under win/dos.

16. How do I install the built-in AT & T win Modem card?

It cannot be used because its control and DSP both use the CPU. if the manufacturer does not write the driver, no one will write the driver for it. In addition, Linux is a multi-user, multi-task operating system, and its response time is very important, the CPU should not be used as the DSP to affect the running of Linux.

17. How do I configure printers on Red Hat Linux?

The setting tool of Red Hat can be easily used to set common printers. If you do not need the Linux version of Red Hat, You can manually set it.

18. How do I install the ISA/PnP card in Linux?

This type of PnP Nic must be set with isapnp before it can be used, such as 3c509 Nic, it is best to use 3C5X9XFG from the manufacturer. the EXE program sets parameters and disables PnP, so that it can be used in Linux. For PnP sound cards, make menuconfig sometimes cannot correctly configure sound card parameters because the sound card code is relatively independent from other parts of the kernel, in the configuration process, you must call its own script to modify the code. To complete the code configuration, make menuconfig should be done again after making config and then checked again.

19. How do I install multiple NICs?

1) When two NICs are installed, append = "ether =, eth1" is added to lilo. conf"

2) if three blocks are inserted, append = "ether =, eth1 ether =, eth2"

Example:

Boot =/dev/hda

# Compact # faster, but won't work on all systems.

Delay = 50

Vga = normal # force sane state

Append = "ether =, eth1" # When the ENI irq and iobase are unknown

[Append = "ether = 9, 0xfce0, eth0 ether = 5, 0xfcc0, eth1" # When you know the NIC irq and iobase]

# Ramdisk = 0 # paranoia setting

# End LILO global section

# Linux bootable partition config begins

Image =/zImage

Root =/dev/hda1

Label = linux

Read-only # Non-UMSDOS filesystems shoshould be mounted read-only for checking

# Linux bootable partition config ends

If it has been compiled into a module, it should be in/etc/rc. d/rc. modules (in Redhat, It is/etc/conf. modules), add a similar line:/sbin/modprobe ne io = io1, io2

20. How can I transmit files in Linux without using a network card, using only two serial lines or parallel lines to connect two computers? Can I implement the "Direct cable connection" function similar to Win98?

Start pppd at both ends at the same time. You can check the configuration file carefully. If one of them wants to accept serial port logon, the original mingetty of Redhat cannot work. Instead, you must use agetty or mgetty. The two are available in the free software library, and then change the corresponding content of/etc/inittab. If you do not have the logon function, it is much simpler, but the speed is slow. Linux also supports parallel IP communication called PLIP.

21. Why cannot the system recognize the NE2000 Nic?

There are two possible causes:

(1) Kernel is not supported. The solution is to re-compile the Kernel. The specific steps are as follows:

Cd/usr/src/linux; make menuconfig select M or Y for the NIC and then:

Make mproper; make dep; make clean; make bzImage; make modules; make modules_install and then compile the new core (such as bzImage) to/boot to overwrite the old version (for security reasons, to pre-backup), and finally execute lilo, restart

(2) The NIC parameter settings are incorrect.

Use the drive disk of the NIC to find out the NIC interrupt and I/O, modify/etc/conf. modules, and add the following content:

Alias eth0 ne

Options ne io = 0x300

Or Add the following content to lilo. conf:

Append = "ether = 300 x, eth0"

22. how to configure the Accton PNP Nic?

Use the NE2000 driver to drive the Accton PNP Nic, modify/etc/conf. modules (if not, create one), and add the following content:

Alias eth0 ne

Options ne io = 0x300

Or Add the following content to lilo. conf:

Append = "ether = 300 x, eth0"