Free emulates software for linux

GXemul is a machine emulator Im developing in my spare time.
GXemul project emulates processors and surrounding hardware components, in some cases well enough to allow unmodified "guest" operating systems to run inside the emulator as if they were running on a real machine.
MIPS:
- Several machine types are emulated well enough to run guest operating systems. NetBSD, OpenBSD, Linux, and even obscure things such as Ultrix and Sprite can run, to various degrees.
ARM:
- CATS is emulated well enough to run NetBSD and OpenBSD as guest OSes.
PowerPC:
- NetBSD/prep can run (experimental).
The emulator can be used for:
- experimenting with operating systems which you have not tried yet, or operating systems for hardware which you might not have access to,
- running guest operating systems for historical purposes, e.g. Ultrix or Sprite,
developing portable software, when you want yet another target platform to try to compile your software on (you can use GXemul to run a guest OS and compile your software inside that OS),
- development of new firmware or operating system code (but then you need to be aware of some limitations),
- educational purposes...
Enhancements:
- NetBSD/pmppc can now run in the emulator (with root on NFS).
- For ARM emulation (with NetBSD/cats, OpenBSD/cats, NetBSD/evbarm, or NetBSD/netwinder as the guest OS), the emulator does not use 100% of the hosts CPU if the guest OS is idle.

Wide Area Link Emulator enables you to set and test various parameters of a WAN access link (also known as a bottleneck link) before testing on a fully-functional network.

Any ISP has a high bandwidth LAN (of the order of tens of Mbps, e.g. Ethernet) and a WAN access link with relatively low speeds (of the order of a few Kbps to a few Mbps). To be able to evaluate the ISPs performance, actual tests will have to be conducted on a live, fully functional network. This would imply enabling certain features in communication equipment which could be detrimental to the proper functioning of a commissioned network. To address such a problem of being able to characterize/quantify the network operation in terms of its performance with respect to various parameters, a test-bed emulating the ISPs connectivity to the Internet would be an ideal solution.

An important factor in setting up such a test-bed, is emulating the WAN Access Link. Using complex hardware is costly. Such a link can however be easily implemented by emulating the scenario on popular physical technology such as Ethernet. By changing the TCP/IP stack on a Personal Computer, the rate at which packets are injected into the network can be controlled and other necessary features can easily be added.

WALE was implemented in 2 phases and the first version can emulated bandwidths up to 8Mbps and uses FIFO queues with drop-tail.

WALE v2 can emulate bandwidths of up to 80Mbps(a 10-fold improvement over version 1) due to the stripped down version of Micro-second resolution timer being used in the Linux kernel. WALE also implements differentiated services are recommended by IETF. A hash-based packet classifier is used to classify packets into one of 3 queues (Expedited Forwarding, Assured Forwarding and Best Effort). A simple priority scheduling has been implemented between the queues.

Version 2 has been implemented in the Generic Device Layer of the Linux Kernel. The kernel version is 2.2.19. WALE2 kernel patch can be downloaded here. An ncurses based front end has been implemented for setting and viewing various parameters of the bottleneck link.

Jimsim Network Simulator project can emulate several routers connected via virutal networks. You can connect to the routers with your own favorite telnet program.
Connecting to the Virtual Routers
Use your favorite telnet program to connect to each of the virtual routers. For example:
For router 1 type:
C:>telnet localhost 10000
For router 2 type:
C:>telnet localhost 10001
For router 3 type:
C:>telnet localhost 10002
Network Layout
Currently the network is configured like this:
access-list [1-99] [permit/deny] [ ip wildcard-mask | host a.b.c.d | any ]
ip acccess-list [standard] [name]
ip access-group
show access-list
show ip interface
show ip interface brief
debug interface
encapsulation [hdlc,ppp] (serial interfaces only)
bandwidth
clock rate (serial interfaces)
interface loopback0
User commands:
?
enable
exit
ping ip
show cdp
show cdp neigh
show cdp neigh detail
show interface
show interface e0,eth0,etc
show ip route
show version
Enable commands:
clear counters
clear counters interface
config terminal
copy running startup
copy startup running
debug all
debug ip routing
debug ip eigrp
debug cdp
disable
exit
reload
show run
show startup
show ip eigrp neighbors
show ip eigrp topology
traceroute ip
undebug all
undebug ip routing
undebug ip eigrp
undebug cdp
Config commands:
cdp run
enable password password
end
hostname
banner motd delimiter
banner exec delimiter
banner login delimiter
ip route network netmask dest
interface e0,eth0,etc
bandwidth value
ip address addr netmask
description
cdp enable
shutdown
line con0, console0, console 0
login
login local
password passwd
router eigrp number
network a.b.c.d
service password-encryption
user username password password
Version restrictions:
- The EIGRP support is very basic. Just the network command for now. (Neighbors dont expire.) You can disable EIGRP with "no router eigrp process_id"
- The routing table isnt 100% correct. It doesnt summarize subnets. Nor does it understand multiple routes to the same destination. This will be fixed in the next version.
- Traceroute seems to count extra hops. Its counting each remote interface on a router as a hop. So, when you traceroute from router1 to router3, it counts router2 twice (once for 192.168.1.2 and 10.1.1.1)

StatsServ is a little program which emulates an ever smaller IRC server (in regard to the commands it handles) and transfers statistics from IRC into a database. You can then use the data, for example, on a Web site where you want to display the latest statistics.

Statistics consist of the uptime of each server in the network, a list of channels, including their topic (except for secret (+s) and private (+p) channels), and the number of users in the network.

You can normally get these statistics via /stats u, /list, and /lusers (if /stats u is supported on the network and enabled for normal users).

openMSX is the MSX home computer system emulator that aims for perfection.

Almost all MSX machines are emulated: MSX1, MSX2, MSX2+, Turbo-R... And also quite some extra hardware is emulated: SCC, SCC+, FM-PAC, Music Module, PSG, Moonsound, Sunrise IDE, Disk Controllers, MIDI, almost all memory mappers, etc... Actually, more than any other MSX emulator!

It also features quite some extras: rendering using openGL acceleration, full screen, built in console (based on TCL), MSX debugging facilities, scalers, using directories as disk images (still buggy though), screenshots, speed settings, frame skip, scanlines, TV-blur, afterglow (GL), and more...

For a really long description: again, please see our home page.

Do note that openMSX is alpha software, which means that some things work but not all features are implemented yet.

ImplicitThis is a syntactical sugar for OO methods.

SYNOPSIS

use ImplicitThis; ImplicitThis::imply();

sub new {
my $type = shift;
my %args = @_;
# must be blessed hash object
bless {
foo => $args{foo},
bar => $args{bar},
}, $type;
}

sub my_accessor {
# $this is read for us. $bar is aliased to $this->{bar}, similiar for $foo
$this->another_accessor($bar);
$foo++;
}

sub _another_accessor {
# this will die if called from something not derived from our package
$foo++;
}

ABSTRACT

Methods in OO Perl receive "$this" without having to read it. Instance field variables are accessable by name instead of having to dereference the hash ref. Privicy is enforced for methods starting with an understore.

This emulates other OO languages, such as Java and C++, where the compiler implicitly and invisibly passes "this" as the first argument of each method call. While Perl passes this argument invisibly, you must manually write code to read it. Java and C++ also discover, at compile time, rather a variable is an instance variable or a static variable, without you needing to distinguish them using special syntax. We remove the extra syntax, but this is learned at run time, not compile time. Unlike Alias.pm, this code is likely to have a noticable impact on performance of code that uses OO accessors heavily.

ImplicitThis::imply() places a thin wrapper is placed around methods in your
object. *this{SCALAR} is a reference to a lexical weve shifted off the argument list. Aliases are created for each key in %$this to itself value in the same way.

While this module works fine for me, your milage may very: it has not been extensively tested.

Similar to Alias.pm. However, were pure Perl, and _no_ additional syntax is introduced.

blogBuddies provides an RSS and Atom aggregator that emulates the LiveJournal Friends page.
blogBuddies gathers RSS and Atom feeds into a layout similar to the LiveJournal Friends page. It is optimized for blogs, and works with Blogger, LiveJournal, DeadJournal, GreatestJournal, Xanga, RSS, and Atom feeds.
About blogBuddies
I originally started writing blogBuddies in my spare time as a response to the lack of a LiveJournal-type friends system in Blogger. Over the course of a few months, blogBuddies grew from a paltry, one-user system to a full multi-user, themable system supporting both LiveJournal and Blogger.
Going from there, blogBuddies was expanded to offer service for GreatestJournal, DeadJournal, Xanga, and RSS/Atom feeds. In the next release (0.4), blogBuddies will support MySpace blogs. I still write blogBuddies in my spare time, and do not know what I want to do with it at this point. If anyone wishes to help, they are free to PM me.
Enhancements:
- Added RSS and Atom support (URLs)
- Code cleanup
- Changed the UserAgent so that publishers can see blogBuddies
- Added support for GreatestJournal, DeadJournal, and Xanga
- Cross-site scripting: A PATCH IS NOW AVAILALBE - patch zip contains only the changed files

Tcpreplay is a suite of BSD licensed tools written by Aaron Turner for *NIX operating systems which gives you the ability to use previously captured traffic in libpcap format to test a variety of network devices.
Tcpreplay allows you to classify traffic as client or server, rewrite Layer 2, 3 and 4 headers and finally replay the traffic back onto the network and through other devices such as switches, routers, firewalls, NIDS and IPSs.
Main features:
- tcpprep - multi-pass pcap file pre-processor which determines packets as client or server and creates cache files used by tcpreplay and tcprewrite
- tcprewrite - pcap file editor which rewrites TCP/IP and Layer 2 packet headers
- tcpreplay - replays pcap files at arbitrary speeds onto the network
- tcpbridge - bridge two network segments with the power of tcprewrite
- flowreplay - emulates a network client using a pcap file as the basis of a TCP or UDP connection (currently in alpha)
Generally speaking, most people would first run tcpprep against a pcap file to create a cache file which splits traffic between client and server if they are testing an inline device like a firewall or IPS.
Then depending on their network setup and where the pcap was captured, they would use tcprewrite to edit the packets so that the device under test will examine them properly. Finally, tcpreplay is used to replay the pcap onto the network to do the test.
Enhancements:
- Upgrade libopts tearoff to 29:0:4 so that everyone else in the world can compile tcpreplay (#189)