Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6

	- information on the 3Com EtherLink Plus (3c505) driver.

6pack.txt

	- info on the 6pack protocol, an alternative to KISS for AX.25

Configurable

	- info on some of the configurable network parameters

DLINK.txt

	- info on the D-Link DE-600/DE-620 parallel port pocket adapters

PLIP.txt

	- info on the driver for Baycom style amateur radio modems

bridge.txt

	- where to get user space programs for ethernet bridging with Linux.

comx.txt

	- info on drivers for COMX line of synchronous serial adapters.

cops.txt

	- info on the COPS LocalTalk Linux driver

cs89x0.txt

	- the Apple or Farallon LocalTalk PC card driver

multicast.txt

	- Behaviour of cards under Multicast

ncsa-telnet

	- notes on how NCSA telnet (DOS) breaks with MTU discovery enabled.

netdevices.txt

	- info on network device driver functions exported to the kernel.

olympic.txt

	- IBM PCI Pit/Pit-Phy/Olympic Token Ring driver info.

policy-routing.txt

	- IP policy-based routing

pt.txt

	- the Gracilis Packetwin AX.25 device driver

ray_cs.txt

	- Raylink Wireless LAN card driver info.

routing.txt

	- the new routing mechanism

shaper.txt

	- info on the module that can shape/limit transmitted traffic.

sk98lin.txt

There are a few network parameters that can be tuned to better match

the kernel to your system hardware and intended usage. The defaults

are usually a good choice for 99% of the people 99% of the time, but

you should be aware they do exist and can be changed.

The current list of parameters can be found in the files:

	linux/net/TUNABLE

	Documentation/networking/ip-sysctl.txt

Some of these are accessible via the sysctl interface, and many more are

scheduled to be added in this way. For example, some parameters related 

to Address Resolution Protocol (ARP) are very easily viewed and altered.

	# cat /proc/sys/net/ipv4/arp_timeout

	6000

	# echo 7000 > /proc/sys/net/ipv4/arp_timeout

	# cat /proc/sys/net/ipv4/arp_timeout

	7000

Others are already accessible via the related user space programs.

For example, MAX_WINDOW has a default of 32 k which is a good choice for

modern hardware, but if you have a slow (8 bit) Ethernet card and/or a slow

machine, then this will be far too big for the card to keep up with fast 

machines transmitting on the same net, resulting in overruns and receive errors.

A value of about 4 k would be more appropriate, which can be set via:

	# route add -net 192.168.3.0 window 4096

The remainder of these can only be presently changed by altering a #define

in the related header file. This means an edit and recompile cycle.

						Paul Gortmaker 06/96

		COMX drivers for the 2.2 kernel

Originally written by: Tivadar Szemethy, <tiv@itc.hu>

Currently maintained by: Gergely Madarasz <gorgo@itc.hu>

Last change: 21/06/1999.

INTRODUCTION

This document describes the software drivers and their use for the 

COMX line of synchronous serial adapters for Linux version 2.2.0 and

above.

The cards are produced and sold by ITC-Pro Ltd. Budapest, Hungary

For further info contact <info@itc.hu> 

or http://www.itc.hu (mostly in Hungarian).

The firmware files and software are available from ftp://ftp.itc.hu

Currently, the drivers support the following cards and protocols:

COMX (2x64 kbps intelligent board)

CMX (1x256 + 1x128 kbps intelligent board)

HiCOMX (2x2Mbps intelligent board)

LoCOMX (1x512 kbps passive board)

MixCOM (1x512 or 2x512kbps passive board with a hardware watchdog an

	optional BRI interface and optional flashROM (1-32M))

SliceCOM	(1x2Mbps channelized E1 board)

PciCOM	(X21)

At the moment of writing this document, the (Cisco)-HDLC, LAPB, SyncPPP and

Frame Relay (DTE, rfc1294 IP encapsulation with partially implemented Q933a 

LMI) protocols are available as link-level protocol. 

X.25 support is being worked on.

USAGE

Load the comx.o module and the hardware-specific and protocol-specific 

modules you'll need into the running kernel using the insmod utility.

This creates the /proc/comx directory.

See the example scripts in the 'etc' directory.

/proc INTERFACE INTRO

The COMX driver set has a new type of user interface based on the /proc 

filesystem which eliminates the need for external user-land software doing 

IOCTL calls. 

Each network interface or device (i.e. those ones you configure with 'ifconfig'

and 'route' etc.) has a corresponding directory under /proc/comx. You can

dynamically create a new interface by saying 'mkdir /proc/comx/comx0' (or you

can name it whatever you want up to 8 characters long, comx[n] is just a 

convention).

Generally the files contained in these directories are text files, which can

be viewed by 'cat filename' and you can write a string to such a file by

saying 'echo _string_ >filename'. This is very similar to the sysctl interface.

Don't use a text editor to edit these files, always use 'echo' (or 'cat'

where appropriate).

When you've created the comx[n] directory, two files are created automagically

in it: 'boardtype' and 'protocol'. You have to fill in these files correctly

for your board and protocol you intend to use (see the board and protocol 

descriptions in this file below or the example scripts in the 'etc' directory).

After filling in these files, other files will appear in the directory for 

setting the various hardware- and protocol-related informations (for example

irq and io addresses, keepalive values etc.) These files are set to default 

values upon creation, so you don't necessarily have to change all of them.

When you're ready with filling in the files in the comx[n] directory, you can

configure the corresponding network interface with the standard network 

configuration utilities. If you're unable to bring the interfaces up, look up

the various kernel log files on your system, and consult the messages for

a probable reason.

EXAMPLE

To create the interface 'comx0' which is the first channel of a COMX card:

insmod comx 

# insmod comx-hw-comx ; insmod comx-proto-ppp  (these are usually

autoloaded if you use the kernel module loader)

mkdir /proc/comx/comx0

echo comx >/proc/comx/comx0/boardtype

echo 0x360 >/proc/comx/comx0/io		<- jumper-selectable I/O port 

echo 0x0a >/proc/comx/comx0/irq		<- jumper-selectable IRQ line

echo 0xd000 >/proc/comx/comx0/memaddr	<- software-configurable memory

					   address. COMX uses 64 KB, and this

					   can be: 0xa000, 0xb000, 0xc000, 

					   0xd000, 0xe000. Avoid conflicts

					   with other hardware.

cat </etc/siol1.rom >/proc/comx/comx0/firmware <- the firmware for the card

echo HDLC >/proc/comx/comx0/protocol	<- the data-link protocol

echo 10 >/proc/comx/comx0/keepalive	<- the keepalive for the protocol

ifconfig comx0 1.2.3.4 pointopoint 5.6.7.8 netmask 255.255.255.255 <-

					   finally configure it with ifconfig

Check its status:

cat /proc/comx/comx0/status

If you want to use the second channel of this board:

mkdir /proc/comx/comx1

echo comx >/proc/comx/comx1/boardtype

echo 0x360 >/proc/comx/comx1/io	

echo 10 >/proc/comx/comx1/irq		

echo 0xd000 >/proc/comx/comx1/memaddr

echo 1 >/proc/comx/comx1/channel	<- channels are numbered 

					   as 0 (default) and 1

Now, check if the driver recognized that you're going to use the other

channel of the same adapter:

cat /proc/comx/comx0/twin

comx1

cat /proc/comx/comx1/twin

comx0

You don't have to load the firmware twice, if you use both channels of

an adapter, just write it into the channel 0's /proc firmware file.

Default values: io 0x360 for COMX, 0x320 (HICOMX), irq 10, memaddr 0xd0000

THE LOCOMX HARDWARE DRIVER

The LoCOMX driver doesn't require firmware, and it doesn't use memory either,

but it uses DMA channels 1 and 3. You can set the clock rate (if enabled by

jumpers on the board) by writing the kbps value into the file named 'clock'.

Set it to 'external' (it is the default) if you have external clock source.

(Note: currently the LoCOMX driver does not support the internal clock)

THE COMX, CMX AND HICOMX DRIVERS

On the HICOMX, COMX and CMX, you have to load the firmware (it is different for

the three cards!). All these adapters can share the same memory

address (we usually use 0xd0000). On the CMX you can set the internal

clock rate (if enabled by jumpers on the small adapter boards) by writing

the kbps value into the 'clock' file. You have to do this before initializing

the card. If you use both HICOMX and CMX/COMX cards, initialize the HICOMX

first. The I/O address of the HICOMX board is not configurable by any

method available to the user: it is hardwired to 0x320, and if you have to 

change it, consult ITC-Pro Ltd.

THE MIXCOM DRIVER

The MixCOM board doesn't require firmware, the driver communicates with

it through I/O ports. You can have three of these cards in one machine.

THE SLICECOM DRIVER

The SliceCOM board doesn't require firmware. You can have 4 of these cards

in one machine. The driver doesn't (yet) support shared interrupts, so

you will need a separate IRQ line for every board.

Read Documentation/networking/slicecom.txt for help on configuring

this adapter.

THE HDLC/PPP LINE PROTOCOL DRIVER

The HDLC/SyncPPP line protocol driver uses the kernel's built-in syncppp

driver (syncppp.o). You don't have to manually select syncppp.o when building

the kernel, the dependencies compile it in automatically.

EXAMPLE

(setting up hw parameters, see above)

# using HDLC:

echo hdlc >/proc/comx/comx0/protocol

echo 10 >/proc/comx/comx0/keepalive	<- not necessary, 10 is the default

ifconfig comx0 1.2.3.4 pointopoint 5.6.7.8 netmask 255.255.255.255

(setting up hw parameters, see above)

# using PPP:

echo ppp >/proc/comx/comx0/protocol

ifconfig comx0 up

ifconfig comx0 1.2.3.4 pointopoint 5.6.7.8 netmask 255.255.255.255

THE LAPB LINE PROTOCOL DRIVER

For this, you'll need to configure LAPB support (See 'LAPB Data Link Driver' in

'Network options' section) into your kernel (thanks to Jonathan Naylor for his 

excellent implementation). 

comx-proto-lapb.o provides the following files in the appropriate directory

(the default values in parens): t1 (5), t2 (1), n2 (20), mode (DTE, STD) and

window (7). Agree with the administrator of your peer router on these

settings (most people use defaults, but you have to know if you are DTE or

DCE).

EXAMPLE

(setting up hw parameters, see above)

echo lapb >/proc/comx/comx0/protocol

echo dce >/proc/comx/comx0/mode		<- DCE interface in this example

ifconfig comx0 1.2.3.4 pointopoint 5.6.7.8 netmask 255.255.255.255

THE FRAME RELAY PROTOCOL DRIVER

You DON'T need any other frame relay related modules from the kernel to use

COMX-Frame Relay. This protocol is a bit more complicated than the others, 

because it allows to use 'subinterfaces' or DLCIs within one physical device.

First you have to create the 'master' device (the actual physical interface)

as you would do for other protocols. Specify 'frad' as protocol type.

Now you can bring this interface up by saying 'ifconfig comx0 up' (or whatever

you've named the interface). Do not assign any IP address to this interface

and do not set any routes through it.

Then, set up your DLCIs the following way: create a comx interface for each

DLCI you intend to use (with mkdir), and write 'dlci' to the 'boardtype' file, 

and 'ietf-ip' to the 'protocol' file. Currently, the only supported 

encapsulation type is this (also called as RFC1294/1490 IP encapsulation).

Write the DLCI number to the 'dlci' file, and write the name of the physical

COMX device to the file called 'master'. 

Now you can assign an IP address to this interface and set routes using it.

See the example file for further info and example config script.

Notes: this driver implements a DTE interface with partially implemented

Q933a LMI.

You can find an extensively commented example in the 'etc' directory.

FURTHER /proc FILES

boardtype:

Type of the hardware. Valid values are:

 'comx', 'hicomx', 'locomx', 'cmx', 'slicecom'.

protocol:

Data-link protocol on this channel. Can be: HDLC, LAPB, PPP, FRAD

status:

You can read the channel's actual status from the 'status' file, for example

'cat /proc/comx/comx3/status'.

lineup_delay:

Interpreted in seconds (default is 1). Used to avoid line jitter: the system

will consider the line status 'UP' only if it is up for at least this number

of seconds.

debug: 

You can set various debug options through this file. Valid options are:

'comx_events', 'comx_tx', 'comx_rx', 'hw_events', 'hw_tx', 'hw_rx'.

You can enable a debug options by writing its name prepended by a '+' into

the debug file, for example 'echo +comx_rx >comx0/debug'. 

Disabling an option happens similarly, use the '-' prefix 

(e.g. 'echo -hw_rx >debug').

Debug results can be read from the debug file, for example: 

tail -f /proc/comx/comx2/debug

NCSA telnet doesn't work with path MTU discovery enabled. This is due to a

bug in NCSA that also stops it working with other modern networking code

such as Solaris.

The following information is courtesy of 

Marek <marekm@i17linuxb.ists.pwr.wroc.pl>

There is a fixed version somewhere on ftp.upe.ac.za (sorry, I don't

remember the exact pathname, and this site is very slow from here).

It may or may not be faster for you to get it from

ftp://ftp.ists.pwr.wroc.pl/pub/msdos/telnet/ncsa_upe/tel23074.zip

(source is in v230704s.zip).  I have tested it with 1.3.79 (with

path mtu discovery enabled - ncsa 2.3.08 didn't work) and it seems

to work.  I don't know if anyone is working on this code - this

version is over a year old.  Too bad - it's faster and often more

stable than these windoze telnets, and runs on almost anything...

This is the README for the Gracilis Packetwin device driver, version 0.5

ALPHA for Linux 1.3.43.

These files will allow you to talk to the PackeTwin (now know as PT) and

connect through it just like a pair of TNCs.  To do this you will also

require the AX.25 code in the kernel enabled.

There are four files in this archive; this readme, a patch file, a .c file

and finally a .h file.  The two program files need to be put into the

drivers/net directory in the Linux source tree, for me this is the

directory /usr/src/linux/drivers/net.  The patch file needs to be patched in

at the top of the Linux source tree (/usr/src/linux in my case).

You will most probably have to edit the pt.c file to suit your own setup,

this should just involve changing some of the defines at the top of the file. 

Please note that if you run an external modem you must specify a speed of 0.

The program is currently setup to run a 4800 baud external modem on port A

and a Kantronics DE-9600 daughter board on port B so if you have this (or

something similar) then you're right.

To compile in the driver, put the files in the correct place and patch in

the diff.  You will have to re-configure the kernel again before you

recompile it. 

The driver is not real good at the moment for finding the card.  You can

'help' it by changing the order of the potential addresses in the structure

found in the pt_init() function so the address of where the card is is put

first.

After compiling, you have to get them going, they are pretty well like any

other net device and just need ifconfig to get them going.

As an example, here is my /etc/rc.net

--------------------------

#

# Configure the PackeTwin, port A.

/sbin/ifconfig pt0a 44.136.8.87 hw ax25 vk2xlz mtu 512 

/sbin/ifconfig pt0a 44.136.8.87 broadcast 44.136.8.255 netmask 255.255.255.0

/sbin/route add -net 44.136.8.0 netmask 255.255.255.0 dev pt0a

/sbin/route add -net 44.0.0.0 netmask 255.0.0.0 gw 44.136.8.68 dev pt0a

/sbin/route add -net 138.25.16.0 netmask 255.255.240.0 dev pt0a

/sbin/route add -host 44.136.8.255 dev pt0a

#

# Configure the PackeTwin, port B.

/sbin/ifconfig pt0b 44.136.8.87 hw ax25 vk2xlz-1 mtu 512

/sbin/ifconfig pt0b 44.136.8.87 broadcast 44.255.255.255 netmask 255.0.0.0

/sbin/route add -host 44.136.8.216 dev pt0b

/sbin/route add -host 44.136.8.95  dev pt0b

/sbin/route add -host 44.255.255.255 dev pt0b

This version of the driver comes under the GNU GPL.  If you have one of my

previous (non-GPL) versions of the driver, please update to this one.

I hope that this all works well for you.  I would be pleased to hear how

many people use the driver and if it does its job.

  - Craig vk2xlz <csmall@small.dropbear.id.au>