powerpc
Sponsored Links
Sponsored Links
Secleted [ 0 ] software to compare
Results 1 - 15 of about 79
Opera 9.64 / 10.00 Beta Build 4453
Surf the Internet in a safer, faster, and easier way with Opera browser. more>>
Opera 9.64 / 10.00 Beta Build 4453 offers you a powerful and very convenient tool which is an easy way to help you surf the Internet in a safer, faster, and easier way with Opera browser. The most full-featured Internet power tool on the market, Opera includes pop-up blocking, tabbed browsing, integrated searches, E-mail, RSS Newsfeeds and IRC chat. Opera supports all major Linux distributions. RedHat, SuSE, Mandrake, Debian, etc. Intel, PowerPC and Sparc versions.
Major Features:
- Pop-up Blocking:
- No more annoying pop-up advertisements. Opera lets you control whether Web sites can use pop-ups. Select to block them all, or let the browser open only pop-ups that you have requested.
- Integrated Search:
- Search your favorite sites, for example Google, eBay, or Amazon without having to go to their Web pages. Use the integrated search window or shortcuts (e.g. "g" for Google) in the address field.
- Skins:
- Give your browser the look you want with Operas skins. Make the browser your own by giving it the colors, icons, and buttons of your choice.
- E-mail with RSS Newsfeed:
- Operas built-in POP/IMAP E-mail client is a combined e-mail program, news reader, mailing list organizer and RSS newsfeed reader. A safer, faster, and more intelligent way of handling your e-mails, Operas e-mail client is database driven, enabling you to organize and find your e-mails in a matter of seconds through easy searches, labeling, and filtering rather than traditional folder storage.
- Tabbed Browsing:
- Surf the Web easier and faster by opening multiple Web pages within the same application window. Save a collection of pages as a session and open them all with one click every time.
- Advanced Security:
- Protect yourself from ad-ware, spy-ware, viruses, and other third-party malicious software applications that silently attack your computer while you are surfing the Web.
- IRC Chat:
- Communicate with people all over the world using Operas IRC chat client. Use Opera to connect to IRC servers to chat privately or in rooms, or share files and photos with your friends and family.
- Opera is from now 100% FREE and with NO ADS !
Enhancements:
- Unix support:
- On the Linux/Unix front, we are discontinuing support for builds made with gcc 2.95. Also, Sparc Solaris builds are now made on Solaris 10 (with gcc 3) they should run on Solaris 9 as well. If these changes affect you, wed love to hear your feedback.
- Skinning improvements:
- [DSK-253666] "new tab" button is too tall in multi-line setup
- Partial [DSK-253753] Closed tabs icon on the right: When set to "text only", the text is nearly unreadable (partially black on black), and touches the right part of the screen)
- [DSK-253759] Cancel button in mail panel way to unobtrusive
- [DSK-253818] Locked tabs change size when being hovered
- Fix for highlight of head and tail bar images
- Fixes for the head and tail gaps in Windows Native skin
- Fixes for pagebar head and tail icons in Windows Native skin
- Opacity changes to Speed Dial on standard skin
- New skin sections: "Pagebar Thumbnail Head Skin", "Pagebar Thumbnail Tail Skin", "Pagebar Thumbnail Floating Skin", "Pagebar Thumbnail Head Button Skin", "Pagebar Thumbnail Tail Button Skin", "Pagebar Thumbnail Floating Button Skin"
- Fix for attention states on tabs
- Updated smilies. They are now more smiley.
- Tweaks to borders on addressbar and splitter to be consistent color
- Tweaks to dialogs
- Revised panel icons and buttons
- Revised icons for mail toolbar
- User Interface:
- [DSK-194184] Cant find show Speed Dial at startup option)
- [DSK-253774] Crash in action handling
- [DSK-226623] Ctrl+Ins doesnt copy outside of edit fields
- [DSK-253386] Crash when clicking help in startup dialog
- [DSK-253202] Replace Language with Dictionary in dictionary download wizard
- [DSK-253550] Turbo notification is cropped if the turbo icon is placed on the right of the window
- Removing obsolete ini-entry "view hotlist"
- Mail:
- [DSK-253731] Opens broken Compose Message view when creating new account
- Core:
- [CORE-21309] Full language names in spell check context menu
- [CORE-21052] XSLT: Using keys in match expressions causes crash
- [DSK-253743] Operator cache deleted when Opera starts
- [DSK-251805] Hang when doubleclicking text while flash is running
- [DSK-253773] Plug-in crash
- Unix:
- [DSK-253567] Crashes on Solaris
- [DSK-253761] Crash in Plug-in manager
Added: 2009-06-27 License: Freeware Price: FREE
26747 downloads
Other version of Opera
Intel, PowerPC and Sparc versions. Main features: Pop-up Blocking: No more annoying pop-up advertisements. Opera lets you control whether Web sites can use pop-ups. Select to block them allLicense:Freeware
BottomFeeder 4.4
BottomFeeder is a viewer for RSS feeds. more>>
BottomFeeder is a news aggregator client for RSS and Atom feeds, written in VisualWorks Smalltalk. BottomFeeder runs on x86 Linux (also FreeBSD), PowerPC linux, Sparc Linux, Windows (98/ME/NT/2000/XP), Mac OS8/9, Mac OS X, AIX, SGI Irix, Compaq UNIX, HP-UX, and Solaris.
Main features:
- Full support for CSS, including user defined CSS
- View news in 3 pane or 2 pane modes
- Subscribe to any RSS or Atom format in use
- View items in a summary Newspaper View
- Synchronize 2 or more BottomFeeders via HTTP or file import
- Subscribe to feeds or feedlists
- Supports HTTPS, HTTP Authentication, and HTTP Digest Authentication
- Plugins for blogging, IRC, and MSN Messenger contacts
- Easy to update or upgrade from within BottomFeeder
- Save as many or as few feed items for as long as you want
- Import or Export in common OPML format
- Binary compatible on every platform. No need to recompile
Enhancements:
- The Feedlists folder has been eliminated.
- Importing a feedlist now adds feeds directly to your subscription list.
- The separate "Searches" folder has been eliminated.
- Search feeds are now part of your subscription list, and are marked with a new icon.
- Tabs now open empty and remember their previous state.
- There are many other changes and enhancements.
<<lessMain features:
- Full support for CSS, including user defined CSS
- View news in 3 pane or 2 pane modes
- Subscribe to any RSS or Atom format in use
- View items in a summary Newspaper View
- Synchronize 2 or more BottomFeeders via HTTP or file import
- Subscribe to feeds or feedlists
- Supports HTTPS, HTTP Authentication, and HTTP Digest Authentication
- Plugins for blogging, IRC, and MSN Messenger contacts
- Easy to update or upgrade from within BottomFeeder
- Save as many or as few feed items for as long as you want
- Import or Export in common OPML format
- Binary compatible on every platform. No need to recompile
Enhancements:
- The Feedlists folder has been eliminated.
- Importing a feedlist now adds feeds directly to your subscription list.
- The separate "Searches" folder has been eliminated.
- Search feeds are now part of your subscription list, and are marked with a new icon.
- Tabs now open empty and remember their previous state.
- There are many other changes and enhancements.
Download (2.3MB)
Added: 2007-07-10 License: Artistic License Price:
2046 downloads
Chicken 1.92
Chicken is a Scheme to C compiler. more>>
CHICKEN is a compiler for the Scheme programming language. CHICKEN produces portable, efficient C, supports almost all of the current Scheme language standard, the Revised5 Report on the Algorithmic Language Scheme (R5RS ), and includes many enhancements and extensions. CHICKEN runs on MacOS X, Windows, and many Unix flavours.
Main features:
- Compiles Scheme to efficient and portable C
- Includes a full-featured interactive interpreter as well as an optimising batch compiler
- Full support for tail recursion, first-class continuations, high-level macros and dynamic-wind
- Highly portable and known to run on many platforms, including IA-32 (80x86), IA-64, PowerPC, SPARC and UltraSPARC, Alpha, MIPS, PA-RISC, ARM, AMD64 and S/390
- Distributed free for use and modification under the terms of the BSD License
- Transparent support for shared objects and linkage to C
- A sophisticated but easy-to-use Foreign Function Interface for linking C and C++ libraries and classes to Scheme
- Lightweight threads based on first-class continuations
- An object system with multiple dispatch, multiple inheritance and a meta-object protocol
- Common Lisp-style format
- Provides the syntax-case enhanced high-level macro system, R5RS syntax-rules , and define-macro
- Support for syntactic pattern-matching via Andrew Wrights match package
- Execution profiling, debugging and backtrace support
- A clean POSIX interface that covers environment and filesystem access, pipes, processes, signals, locks, sockets, and low-level and memory-mapped I/O
- Support for interpreted or compiled shell scripts under Unix and Windows
- Support for a large number of Scheme Requests For Implementation (SRFIs), some available separately as extensions:
SRFI 0 (cond-expand )
SRFI 1 (List library)
SRFI 2 (and-let* )
SRFI 4 (Homogeneous numeric vector datatypes)
SRFI 6 (Basic string ports)
SRFI 8 (receive )
SRFI 9 (Record types)
SRFI 10 (#, external form)
SRFI 13 (String library)
SRFI 14 (Character set library)
SRFI 16 (case-lambda )
SRFI 17 (Generalized set! )
SRFI 18 (Multithreading support)
SRFI 23 (error )
SRFI 25 (Multidimensional array primitives)
SRFI 26 (cut , cute )
SRFI 28 (format )
SRFI 30 (Block comments)
SRFI 37 (Program argument processor)
SRFI 38 (External representation for data with shared structure)
SRFI 39 (Parameters)
SRFI 40 (Stream library)
SRFI 42 (Eager comprehensions)
SRFI 43 (Vector library)
SRFI 55 (require-extension )
Version restrictions:
- No unlimited-precision integers (bignums), rationals or complex numbers
- Procedures are limited to 126 arguments
- No support for unicode
- Compilation of large files can be slow
Enhancements:
- chicken-setup: better handling when no remote repository file exists
<<lessMain features:
- Compiles Scheme to efficient and portable C
- Includes a full-featured interactive interpreter as well as an optimising batch compiler
- Full support for tail recursion, first-class continuations, high-level macros and dynamic-wind
- Highly portable and known to run on many platforms, including IA-32 (80x86), IA-64, PowerPC, SPARC and UltraSPARC, Alpha, MIPS, PA-RISC, ARM, AMD64 and S/390
- Distributed free for use and modification under the terms of the BSD License
- Transparent support for shared objects and linkage to C
- A sophisticated but easy-to-use Foreign Function Interface for linking C and C++ libraries and classes to Scheme
- Lightweight threads based on first-class continuations
- An object system with multiple dispatch, multiple inheritance and a meta-object protocol
- Common Lisp-style format
- Provides the syntax-case enhanced high-level macro system, R5RS syntax-rules , and define-macro
- Support for syntactic pattern-matching via Andrew Wrights match package
- Execution profiling, debugging and backtrace support
- A clean POSIX interface that covers environment and filesystem access, pipes, processes, signals, locks, sockets, and low-level and memory-mapped I/O
- Support for interpreted or compiled shell scripts under Unix and Windows
- Support for a large number of Scheme Requests For Implementation (SRFIs), some available separately as extensions:
SRFI 0 (cond-expand )
SRFI 1 (List library)
SRFI 2 (and-let* )
SRFI 4 (Homogeneous numeric vector datatypes)
SRFI 6 (Basic string ports)
SRFI 8 (receive )
SRFI 9 (Record types)
SRFI 10 (#, external form)
SRFI 13 (String library)
SRFI 14 (Character set library)
SRFI 16 (case-lambda )
SRFI 17 (Generalized set! )
SRFI 18 (Multithreading support)
SRFI 23 (error )
SRFI 25 (Multidimensional array primitives)
SRFI 26 (cut , cute )
SRFI 28 (format )
SRFI 30 (Block comments)
SRFI 37 (Program argument processor)
SRFI 38 (External representation for data with shared structure)
SRFI 39 (Parameters)
SRFI 40 (Stream library)
SRFI 42 (Eager comprehensions)
SRFI 43 (Vector library)
SRFI 55 (require-extension )
Version restrictions:
- No unlimited-precision integers (bignums), rationals or complex numbers
- Procedures are limited to 126 arguments
- No support for unicode
- Compilation of large files can be slow
Enhancements:
- chicken-setup: better handling when no remote repository file exists
Download (3.6MB)
Added: 2005-04-18 License: BSD License Price:
1649 downloads
GNU Lightning 1.2
GNU Lightning is a library that generates assembly code at run time. more>>
GNU lightning is a library that generates assembly language code at run-time; it is very fast, making it ideal for Just-In-Time compilers, and it abstracts over the target CPU, as it exposes to the clients a standardized RISC instruction set inspired by the MIPS and SPARC chips.
GNU lightning 1.0 has been released and is usable in complex code generation tasks. The available backends cover the x86, SPARC and PowerPC architectures; the floating point interface is still experimental though, and developed for the x86 only.
There is also interested in adding more high-level features to lightning, such as a BURG-style tool, a tool to simplify the porting and make it less tedious (for example interfacing with Red Hats CGEN tool) or a standardized library to synthesize instruction sequences such as multiplications as shifts and adds, or divisions as multiplication by the inverse.
<<lessGNU lightning 1.0 has been released and is usable in complex code generation tasks. The available backends cover the x86, SPARC and PowerPC architectures; the floating point interface is still experimental though, and developed for the x86 only.
There is also interested in adding more high-level features to lightning, such as a BURG-style tool, a tool to simplify the porting and make it less tedious (for example interfacing with Red Hats CGEN tool) or a standardized library to synthesize instruction sequences such as multiplications as shifts and adds, or divisions as multiplication by the inverse.
Download (0.39MB)
Added: 2005-04-22 License: (FDL) GNU Free Documentation License Price:
1646 downloads
Peachtree Linux Atlanta
Peachtree Linux is a powerful, robust, and scalable distribution of the open source Linux operating system. more>>
Peachtree Linux is a new Linux distribution being developed by several students/former students at the Georgia Institute of Technology.
Since its inception in the fall of 2002, Peachtree Linux has aimed to be a small system for the seasoned Linux user. You wont find GNOME or KDE among Peachtree Linuxs packages, so it might not be the system for you. There are several simple rules we try to follow:
1. No GNOME or KDE.
2. All X11 programs have their binaries, libraries, and headers installed to the /usr/X11R6 tree.
3. All configuration files are placed in /etc.
4. /bin and /lib do not exist, they are soft links to /usr/bin and /usr/lib, respectively.
5. /sbin contains statically compiled programs that can bring a system up from a cold boot, nothing more.
6. We have one program per task. We do not like distributions which offer dozens of MP3 players, or mail clients, or window managers. We do have exceptions to this rule, but we generally stick to it.
Thats more or less the guiding ideas behind the distribution. It is developed in parallel on both PowerPC and i386. Work is being done to port the distribution to Alpha, PA-RISC, and MIPS platforms.
Were not out to change the world with this distribution. We just want to make a system that we like to use. If you like the sound of what were trying to put together, then Peachtree Linux may be for you. If the ideas really dont get you excited, you may want to find another distribution.
Enhancements:
- os/branches/RELENG_1/src/pkgs/httpd/files/httpd.htdigest.patch,
- os/branches/RELENG_1/src/pkgs/httpd/patchlist: (* Security Fix *)
- Fixed potential buffer overflow in htdigest (CAN-2005-1344).
<<lessSince its inception in the fall of 2002, Peachtree Linux has aimed to be a small system for the seasoned Linux user. You wont find GNOME or KDE among Peachtree Linuxs packages, so it might not be the system for you. There are several simple rules we try to follow:
1. No GNOME or KDE.
2. All X11 programs have their binaries, libraries, and headers installed to the /usr/X11R6 tree.
3. All configuration files are placed in /etc.
4. /bin and /lib do not exist, they are soft links to /usr/bin and /usr/lib, respectively.
5. /sbin contains statically compiled programs that can bring a system up from a cold boot, nothing more.
6. We have one program per task. We do not like distributions which offer dozens of MP3 players, or mail clients, or window managers. We do have exceptions to this rule, but we generally stick to it.
Thats more or less the guiding ideas behind the distribution. It is developed in parallel on both PowerPC and i386. Work is being done to port the distribution to Alpha, PA-RISC, and MIPS platforms.
Were not out to change the world with this distribution. We just want to make a system that we like to use. If you like the sound of what were trying to put together, then Peachtree Linux may be for you. If the ideas really dont get you excited, you may want to find another distribution.
Enhancements:
- os/branches/RELENG_1/src/pkgs/httpd/files/httpd.htdigest.patch,
- os/branches/RELENG_1/src/pkgs/httpd/patchlist: (* Security Fix *)
- Fixed potential buffer overflow in htdigest (CAN-2005-1344).
Download (700MB)
Added: 2005-05-17 License: GPL (GNU General Public License) Price:
1634 downloads
JamVM 1.3.1
JamVM is a compact Java Virtual Machine. more>>
JamVM is a new Java Virtual Machine which conforms to the JVM specification version 2 (blue book). In comparison to most other VMs (free and commercial) it is extremely small, with a stripped executable on PowerPC of only ~130K, and Intel 100K.
However, unlike other small VMs (e.g. KVM) it is designed to support the full specification, and includes support for object finalisation, the Java Native Interface (JNI) and the Reflection API.
Main features:
- Uses native threading (posix threads). Full thread implementation including Thread.interrupt()
- Object references are direct pointers (i.e. no handles)
- Supports class loaders
- Efficient thin locks for fast locking in uncontended cases (the majority of locking) without using spin-locking
- Two word object header to minimise heap overhead (lock word and class pointer)
- Execution engine supports basic switched interpreter and threaded interpreter, to minimise dispatch overhead (requires gcc value labels)
- Stop-the-world mark and sweep garbage collector
- Thread suspension uses signals to reduce suspend latency and improve performance (no suspension checks during normal execution)
- Full object finalisation support within the garbage collector (with finaliser thread)
- Garbage collector can run synchronously or asynchronously within its own thread
- String constants within class files are stored in hash table to minimise class data overhead (string constants shared between all classes)
- Supports JNI and dynamic loading for use with standard libraries
- Uses its own lightweight native interface for internal native methods without overhead of JNI
- JamVM is written in C, with a small amount of platform dependent assembler, and is easily portable to other architectures.
Enhancements:
- Re-worked thread suspension enable/disable code
- implemented a fast enable/disable suspend mechanism for critical code sections, which is now used in hash table access and heap allocation :
- much faster object allocation
- much faster primitive array allocation
- faster string interning
- removed some minor potential (as yet unseen) race conditions
- Integrated all outstanding patches
- generics branch patch, implementing VMClass.isEnum(), isAnnotation() and isSynthetic()
- Boot library path patch (support for system property gnu.classpath.boot.library.path)
- Enabled different min/max heap values for each architecture
- ARM values are back to the defaults for JamVM 1.2.5. Default max heap of 128Mb is too large for typical embedded ARM systems.
- Fixed the new compiler warnings from gcc-4.0 (under Mac OS X) related to char/signed char usage.
- Fixed compiler error on gcc-4.0 under Mac OS X 10.4 (duplicate definitions of wait).
- Fixed rare, potential dead-lock in direct.c when preparing methods
- Fixed a bug in VMClassLoader.defineClass(), where offset or length is wrong.
- Fixed bug in DCONST_1 on mixed-endian ARM architectures (this bug was fixed in JamVM 1.2.2 but it crept back in, due to the new interpreter in JamVM 1.2.5).
- Improved thread dump (produced via SIGQUIT, or ctrl-).
- Several other minor bug-fixes, and code tidy-ups.
<<lessHowever, unlike other small VMs (e.g. KVM) it is designed to support the full specification, and includes support for object finalisation, the Java Native Interface (JNI) and the Reflection API.
Main features:
- Uses native threading (posix threads). Full thread implementation including Thread.interrupt()
- Object references are direct pointers (i.e. no handles)
- Supports class loaders
- Efficient thin locks for fast locking in uncontended cases (the majority of locking) without using spin-locking
- Two word object header to minimise heap overhead (lock word and class pointer)
- Execution engine supports basic switched interpreter and threaded interpreter, to minimise dispatch overhead (requires gcc value labels)
- Stop-the-world mark and sweep garbage collector
- Thread suspension uses signals to reduce suspend latency and improve performance (no suspension checks during normal execution)
- Full object finalisation support within the garbage collector (with finaliser thread)
- Garbage collector can run synchronously or asynchronously within its own thread
- String constants within class files are stored in hash table to minimise class data overhead (string constants shared between all classes)
- Supports JNI and dynamic loading for use with standard libraries
- Uses its own lightweight native interface for internal native methods without overhead of JNI
- JamVM is written in C, with a small amount of platform dependent assembler, and is easily portable to other architectures.
Enhancements:
- Re-worked thread suspension enable/disable code
- implemented a fast enable/disable suspend mechanism for critical code sections, which is now used in hash table access and heap allocation :
- much faster object allocation
- much faster primitive array allocation
- faster string interning
- removed some minor potential (as yet unseen) race conditions
- Integrated all outstanding patches
- generics branch patch, implementing VMClass.isEnum(), isAnnotation() and isSynthetic()
- Boot library path patch (support for system property gnu.classpath.boot.library.path)
- Enabled different min/max heap values for each architecture
- ARM values are back to the defaults for JamVM 1.2.5. Default max heap of 128Mb is too large for typical embedded ARM systems.
- Fixed the new compiler warnings from gcc-4.0 (under Mac OS X) related to char/signed char usage.
- Fixed compiler error on gcc-4.0 under Mac OS X 10.4 (duplicate definitions of wait).
- Fixed rare, potential dead-lock in direct.c when preparing methods
- Fixed a bug in VMClassLoader.defineClass(), where offset or length is wrong.
- Fixed bug in DCONST_1 on mixed-endian ARM architectures (this bug was fixed in JamVM 1.2.2 but it crept back in, due to the new interpreter in JamVM 1.2.5).
- Improved thread dump (produced via SIGQUIT, or ctrl-).
- Several other minor bug-fixes, and code tidy-ups.
Download (0.23MB)
Added: 2005-06-17 License: GPL (GNU General Public License) Price:
1593 downloads
Oroborus 2.0.15
Oroborus is a small and simple but configurable and themeable window manager. more>>
Oroborus basically is a really minimalistic window manager, which is about 75kb small on a PowerPC machine and about 63kb on an x86 machine. "Minimalistic" means that the Oroborus window manager doesnt provide some things other window managers provide.
So you will need some external tools also provided by this project.
The main improvement with this is that you gain much flexibility.
Main features:
- good default key bindings
- windows are moveable by keyboard
- highly configurable
- xpm-themable
- GNOME compatible
- NET_WM compatible
- sort of KDE compatible
- and many, many more...
Enhancements:
- New Release.
- Add some Themes (defold, bluecurve, mkultra, Elberg_Red, Elberg_Green, Elberg_Blue)
- Fix "oroborus -V".
<<lessSo you will need some external tools also provided by this project.
The main improvement with this is that you gain much flexibility.
Main features:
- good default key bindings
- windows are moveable by keyboard
- highly configurable
- xpm-themable
- GNOME compatible
- NET_WM compatible
- sort of KDE compatible
- and many, many more...
Enhancements:
- New Release.
- Add some Themes (defold, bluecurve, mkultra, Elberg_Red, Elberg_Green, Elberg_Blue)
- Fix "oroborus -V".
Download (0.33MB)
Added: 2005-07-04 License: GPL (GNU General Public License) Price:
1573 downloads
pyglplot 0.1.0
pyglplot is an interactive 3D plotting and visualization tool written as a Python module that uses OpenGL and the GIMP Toolkit. more>>
pyglplot is a program to 3D-plot mathematical functions and datafiles. It is written as a module for python and as such features the full power and flexibility of the python programming language.
Main features:
Data Sources
- Plot python functions
- Plot python dictionaries containing the coordinates of the points and their position in the grid
- Plot plain datafiles as understood for example by gnuplot
Graphics
- Can make 3D plots of surfaces in normal or parametric mode
- Support coordinate transforms, i.e. can make e.g. polar plots
- Support for pm3d-style coloring of the surface
- Support for wireframe plotting atop of the surface
- Support for transparency: translucent plots
- Support for different types of fog and arbitrary background colors
- Support for dynamic light and specular reflexes etc.
- Use either perspective projection or parallel projection
- Save the rendered plots as portable network graphics (.png)
- Adjust all graphics parameters via dialog windows
Platforms
- This program is developed and tested on Max OS X 10.4 on powerpc with fink and Debian GNU/Linux on i386. It should compile and run everywhere where its prerequisites and requisites thereof are available. These are gtkglext-1.0 and gthread-2.0. Some minimum versions of the libriaries must be met, unfortunately, the gtk+-2.2 packages provided by fink for Mac OS X 10.3 are not sufficient -- perhaps I will port it there.
<<lessMain features:
Data Sources
- Plot python functions
- Plot python dictionaries containing the coordinates of the points and their position in the grid
- Plot plain datafiles as understood for example by gnuplot
Graphics
- Can make 3D plots of surfaces in normal or parametric mode
- Support coordinate transforms, i.e. can make e.g. polar plots
- Support for pm3d-style coloring of the surface
- Support for wireframe plotting atop of the surface
- Support for transparency: translucent plots
- Support for different types of fog and arbitrary background colors
- Support for dynamic light and specular reflexes etc.
- Use either perspective projection or parallel projection
- Save the rendered plots as portable network graphics (.png)
- Adjust all graphics parameters via dialog windows
Platforms
- This program is developed and tested on Max OS X 10.4 on powerpc with fink and Debian GNU/Linux on i386. It should compile and run everywhere where its prerequisites and requisites thereof are available. These are gtkglext-1.0 and gthread-2.0. Some minimum versions of the libriaries must be met, unfortunately, the gtk+-2.2 packages provided by fink for Mac OS X 10.3 are not sufficient -- perhaps I will port it there.
Download (0.23MB)
Added: 2005-08-05 License: GPL (GNU General Public License) Price:
1542 downloads
DotGNU Portable.NET 0.7.2
DotGNU Portable.NET is a tools kit for .NET applications on non-Microsoft platforms. more>>
The goal of this project is to build a suite of free software tools to build and execute .NET applications, including a C# compiler, assembler, disassembler, and runtime engine.
While the initial target platform was GNU/Linux, it is also known to run under Windows, Solaris, NetBSD, FreeBSD, and MacOS X. The runtime engine has been tested on the x86, PowerPC, ARM, Sparc, PARISC, s390, Alpha, and IA-64 processors.
DotGNU Portable.NET is part of the DotGNU project, built in accordance with the requirements of the GNU Project.
DotGNU Portable.NET is focused on compatibility with the ECMA specifications for CLI. There are other projects under the DotGNU meta-project to build other necessary pieces of infrastructure, and to explore non-CLI approaches to virtual machine implementation.
Building
The three primary components of the system must be built in the following order: treecc, pnet, and then pnetlib.
You will need flex and bison to build DotGNU Portable.NET. We use flex 2.5.4 and bison 1.28. Some support is provided for the BSD version of yacc, but bison is recommended. Other versions of lex or yacc may not work.
Each source package can be unpacked, built, and installed in the usual manner:
$ zcat<<less
While the initial target platform was GNU/Linux, it is also known to run under Windows, Solaris, NetBSD, FreeBSD, and MacOS X. The runtime engine has been tested on the x86, PowerPC, ARM, Sparc, PARISC, s390, Alpha, and IA-64 processors.
DotGNU Portable.NET is part of the DotGNU project, built in accordance with the requirements of the GNU Project.
DotGNU Portable.NET is focused on compatibility with the ECMA specifications for CLI. There are other projects under the DotGNU meta-project to build other necessary pieces of infrastructure, and to explore non-CLI approaches to virtual machine implementation.
Building
The three primary components of the system must be built in the following order: treecc, pnet, and then pnetlib.
You will need flex and bison to build DotGNU Portable.NET. We use flex 2.5.4 and bison 1.28. Some support is provided for the BSD version of yacc, but bison is recommended. Other versions of lex or yacc may not work.
Each source package can be unpacked, built, and installed in the usual manner:
$ zcat<<less
Download (3.4MB)
Added: 2005-08-18 License: GPL (GNU General Public License) Price:
1538 downloads
SDL_sound 1.0.1
SDL_sound is a library that handles the decoding of several popular sound file formats, such as .WAV and .MP3. more>>
SDL_sound is a library that handles the decoding of several popular sound file formats, such as .WAV and .MP3. It is meant to make the programmers sound playback tasks simpler. The programmer gives SDL_sound a filename, or feeds it data directly from one of many sources, and then reads the decoded waveform data back at her leisure.
If resource constraints are a concern, SDL_sound can process sound data in programmer-specified blocks. Alternately, SDL_sound can decode a whole sound file and hand back a single pointer to the whole waveform. SDL_sound can also handle sample rate, audio format, and channel conversion on-the-fly and behind-the-scenes, if the programmer desires.
As the name implies, SDL_sound is an add-on to Simple Directmedia Layer, and as such, youll need it to build and use SDL_sound. SDL gives us lots of convenience for porting and implementing some elements of the library, not to mention that it is a powerful, cross platform answer to DirectX. You should definitely look into it, whether you use SDL_sound or not.
Places SDL_sound is known to be used:
- The Ambient Music System uses SDL_sound in its player module.
- The popular game glTron uses SDL_sound for sound playback.
- Dominik Haumann wrote in to mention C++ wrappers for SDL_sound, which can be found here.
- MusicBox, a GNUstep music manager, uses SDL_sound to decode various audio formats.
- The SuSE Linux distribution has been know to package SDL_sound (at least version 0.1.5 has been spotted in the wild).
- Gentoo Linux distributes SDL_sound via their "portage" system. Just "emerge sdl-sound".
- FreeBSD has SDL_sound in their Ports tree: herere the details.
- NetBSD Appears to have packaged SDL_sound, too: details.
- Debian has packaged SDL_sound, too: details.
- ROCK Linux has an SDL_sound package: details.
- Probably other places. Email me if you want to be listed here.
What works:
- Support for "decoding" raw audio samples.
- Support for decoding Microsoft WAVE audio files (uncompressed and MS-ADPCM encoded waveforms are supported, currently) (.WAV).
- Support for decoding various MPEG audio files through SMPEG (.MP3, .MPG, .MPEG).
- Support for decoding MPEG-1 Layer 3 audio files internally (makes use of LGPLd code from mpg123.) (.MP3).
- Support for decoding MIDI music files (.MIDI, .MID).
- Support for decoding Ogg Vorbis audio files through libvorbis (.OGG).
- Support for decoding Speex voice files through libspeex (.SPX).
- Support for decoding FLAC audio files through libFLAC (.FLAC, .FLA).
- Support for decoding Creative Labs Voice files (.VOC).
- Support for decoding Audio Interchange File Format files (.AIFF).
- Support for decoding Shorten-compressed audio data (.SHN).
- Support for Suns audio format (.AU).
- Support for decoding through MikMod. (.MOD, .IT, .XM, .S3M, .MTM, .669, .STM, .ULT, .FAR, .MED, .AMF, .DSM, .IMF, .GDM, .STX, .OKT).
- Support for decoding through ModPlug. (.669, .AMF, .AMS, .DBM, .DMF, .DSM, .FAR, .IT, .MDL, .MED, .MOD, .MT2, .MTM, .OKT, .PTM, .PSM, .S3M, .STM, .ULT, .UMX, .XM)
- Experimental support for decoding through Apple Quicktime (MacOS only).
- Compiles/runs on Linux (x86 tested).
- Compiles/runs on Linux (x86-64 tested).
- Compiles/runs on Linux (PowerPC tested).
- Compiles/runs on FreeBSD (x86 tested).
- Compiles/runs on Win32 (x86 tested).
- Compiles/runs on MacOS Classic (PowerPC tested).
- Compiles/runs on MacOS X (PowerPC tested).
- Compiles/runs on BeOS (x86 tested).
- Compiles/runs on WinCE/PocketPC (StrongARM tested).
- May compile and run elsewhere with little to no modification. Success stories and patches are welcome.
What doesnt work:
- Support for more forms of compressed .WAV files is planned, but not yet implemented.
- Support for various forms of compressed .AIFF files is planned, but not yet implemented.
- Contributions of other ports are, of course, welcome.
<<lessIf resource constraints are a concern, SDL_sound can process sound data in programmer-specified blocks. Alternately, SDL_sound can decode a whole sound file and hand back a single pointer to the whole waveform. SDL_sound can also handle sample rate, audio format, and channel conversion on-the-fly and behind-the-scenes, if the programmer desires.
As the name implies, SDL_sound is an add-on to Simple Directmedia Layer, and as such, youll need it to build and use SDL_sound. SDL gives us lots of convenience for porting and implementing some elements of the library, not to mention that it is a powerful, cross platform answer to DirectX. You should definitely look into it, whether you use SDL_sound or not.
Places SDL_sound is known to be used:
- The Ambient Music System uses SDL_sound in its player module.
- The popular game glTron uses SDL_sound for sound playback.
- Dominik Haumann wrote in to mention C++ wrappers for SDL_sound, which can be found here.
- MusicBox, a GNUstep music manager, uses SDL_sound to decode various audio formats.
- The SuSE Linux distribution has been know to package SDL_sound (at least version 0.1.5 has been spotted in the wild).
- Gentoo Linux distributes SDL_sound via their "portage" system. Just "emerge sdl-sound".
- FreeBSD has SDL_sound in their Ports tree: herere the details.
- NetBSD Appears to have packaged SDL_sound, too: details.
- Debian has packaged SDL_sound, too: details.
- ROCK Linux has an SDL_sound package: details.
- Probably other places. Email me if you want to be listed here.
What works:
- Support for "decoding" raw audio samples.
- Support for decoding Microsoft WAVE audio files (uncompressed and MS-ADPCM encoded waveforms are supported, currently) (.WAV).
- Support for decoding various MPEG audio files through SMPEG (.MP3, .MPG, .MPEG).
- Support for decoding MPEG-1 Layer 3 audio files internally (makes use of LGPLd code from mpg123.) (.MP3).
- Support for decoding MIDI music files (.MIDI, .MID).
- Support for decoding Ogg Vorbis audio files through libvorbis (.OGG).
- Support for decoding Speex voice files through libspeex (.SPX).
- Support for decoding FLAC audio files through libFLAC (.FLAC, .FLA).
- Support for decoding Creative Labs Voice files (.VOC).
- Support for decoding Audio Interchange File Format files (.AIFF).
- Support for decoding Shorten-compressed audio data (.SHN).
- Support for Suns audio format (.AU).
- Support for decoding through MikMod. (.MOD, .IT, .XM, .S3M, .MTM, .669, .STM, .ULT, .FAR, .MED, .AMF, .DSM, .IMF, .GDM, .STX, .OKT).
- Support for decoding through ModPlug. (.669, .AMF, .AMS, .DBM, .DMF, .DSM, .FAR, .IT, .MDL, .MED, .MOD, .MT2, .MTM, .OKT, .PTM, .PSM, .S3M, .STM, .ULT, .UMX, .XM)
- Experimental support for decoding through Apple Quicktime (MacOS only).
- Compiles/runs on Linux (x86 tested).
- Compiles/runs on Linux (x86-64 tested).
- Compiles/runs on Linux (PowerPC tested).
- Compiles/runs on FreeBSD (x86 tested).
- Compiles/runs on Win32 (x86 tested).
- Compiles/runs on MacOS Classic (PowerPC tested).
- Compiles/runs on MacOS X (PowerPC tested).
- Compiles/runs on BeOS (x86 tested).
- Compiles/runs on WinCE/PocketPC (StrongARM tested).
- May compile and run elsewhere with little to no modification. Success stories and patches are welcome.
What doesnt work:
- Support for more forms of compressed .WAV files is planned, but not yet implemented.
- Support for various forms of compressed .AIFF files is planned, but not yet implemented.
- Contributions of other ports are, of course, welcome.
Download (0.097MB)
Added: 2005-09-16 License: GPL (GNU General Public License) Price:
1504 downloads
Ubuntu Linux 5.10
Ubuntu is a complete Linux-based operating system, freely available with both community and professional support. more>>
Ubuntu is a complete Linux-based operating system, freely available with both community and professional support.
The Ubuntu community is built on the ideas enshrined in the Ubuntu Manifesto: that software should be available free of charge, that software tools should be usable by people in their local language and despite any disabilities, and that people should have the freedom to customise and alter their software in whatever way they see fit.
These freedoms make Ubuntu fundamentally different from traditional proprietary software: not only are the tools you need available free of charge, you have the right to modify your software until it works the way you want it to.
The team behind Ubuntu makes the following public commitment to its users:
* Ubuntu will always be free of charge, and there is no extra fee for the "enterprise edition", we make our very best work available to everyone on the same Free terms.
* Ubuntu includes the very best in translations and accessibility infrastructure that the Free Software community has to offer, to make Ubuntu usable by as many people as possible.
* Ubuntu is released regularly and predictably; a new release is made every six months. You can use the current stable release or the current development release. Each release is supported for at least 18 months.
* Ubuntu is entirely committed to the principles of free and open source software development; we encourage people to use free and open source software, improve it and pass it on.
Ubuntu is suitable for both desktop and server use. The current Ubuntu release supports Intel x86 (IBM-compatible PC), AMD64 (Hammer) and PowerPC (Apple iBook and Powerbook, G4 and G5) architectures.
Ubuntu includes more than 1,000 pieces of software, starting with the Linux kernel version 2.6 and Gnome 2.8, and covering every standard desktop application from word processing and spreadsheet applications to internet access applications, web server software, email software, programming languages and tools and of course several games.
<<lessThe Ubuntu community is built on the ideas enshrined in the Ubuntu Manifesto: that software should be available free of charge, that software tools should be usable by people in their local language and despite any disabilities, and that people should have the freedom to customise and alter their software in whatever way they see fit.
These freedoms make Ubuntu fundamentally different from traditional proprietary software: not only are the tools you need available free of charge, you have the right to modify your software until it works the way you want it to.
The team behind Ubuntu makes the following public commitment to its users:
* Ubuntu will always be free of charge, and there is no extra fee for the "enterprise edition", we make our very best work available to everyone on the same Free terms.
* Ubuntu includes the very best in translations and accessibility infrastructure that the Free Software community has to offer, to make Ubuntu usable by as many people as possible.
* Ubuntu is released regularly and predictably; a new release is made every six months. You can use the current stable release or the current development release. Each release is supported for at least 18 months.
* Ubuntu is entirely committed to the principles of free and open source software development; we encourage people to use free and open source software, improve it and pass it on.
Ubuntu is suitable for both desktop and server use. The current Ubuntu release supports Intel x86 (IBM-compatible PC), AMD64 (Hammer) and PowerPC (Apple iBook and Powerbook, G4 and G5) architectures.
Ubuntu includes more than 1,000 pieces of software, starting with the Linux kernel version 2.6 and Gnome 2.8, and covering every standard desktop application from word processing and spreadsheet applications to internet access applications, web server software, email software, programming languages and tools and of course several games.
Download (641MB)
Added: 2005-10-13 License: GPL (GNU General Public License) Price:
1500 downloads
AROS-Max 0.4.8
AROS-Max is a AROS-based live-CD. more>>
AROS-Max is a AROS-based live-CD.
AROS Max is a pre-configured live bootable CD image, made to show off the best that AROS has to offer. It requires an AROS capable PC, your mileage may vary.
If you require help getting AROS-Max to run please ask for help on the Max area at the AROS-Exec messageboard, please do not email us directly with support problems!
AROS is a portable and free desktop operating system aiming at being compatible with AmigaOS 3.1, while improving on it in many areas. The source code is available under an open source license, which allows anyone to freely improve upon it.
The goals of the AROS project is it to create an OS which:
1. Is as compatible as possible with AmigaOS 3.1.
2. Can be ported to different kinds of hardware architectures and processors, such as x86, PowerPC, Alpha, Sparc, HPPA and other.
3. Should be binary compatible on Amiga and source compatible on any other hardware.
4. Can run as a standalone version which boots directly from hard disk and as an emulation which opens a window on an existing OS to develop software and run Amiga and native applications at the same time.
5. Improves upon the functionality of AmigaOS.
To reach this goal, we use a number of techniques. First of all, we make heavy use of the Internet. You can participate in our project even if you can write only one single OS function. The most current version of the source is accessible 24 hours per day and patches can be merged into it at any time. A small database with open tasks makes sure work is not duplicated.
Some time back in the year 1993, the situation for the Amiga looked somewhat worse than usual and some Amiga fans got together and discussed what should be done to increase the acceptance of our beloved machine. Immediately the main reason for the missing success of the Amiga became clear: it was propagation, or rather the lack thereof. The Amiga should get a more widespread basis to make it more attractive for everyone to use and to develop for. So plans were made to reach this goal. One of the plans was to fix the bugs of the AmigaOS, another was to make it an modern operating system. The AOS project was born.
But exactly what was a bug? And how should the bugs be fixed? What are the features a so-called modern OS must have? And how should they be implemented into the AmigaOS?
Two years later, people were still arguing about this and not even one line of code had been written (or at least no one had ever seen that code). Discussions were still of the pattern where someone stated that "we must have ..." and someone answered "read the old mails" or "this is impossible to do, because ..." which was shortly followed by "youre wrong because ..." and so on.
In the winter of 1995, Aaron Digulla got fed up with this situation and posted an RFC (request for comments) to the AOS mailing list in which I asked what the minimal common ground might be. Several options were given and the conclusion was that almost everyone would like to see an open OS which is compatible to AmigaOS 3.1 (kickstart 40.68) on which further discussions could be based upon to see what is possible and what is not.
So the work began and AROS was born.
<<lessAROS Max is a pre-configured live bootable CD image, made to show off the best that AROS has to offer. It requires an AROS capable PC, your mileage may vary.
If you require help getting AROS-Max to run please ask for help on the Max area at the AROS-Exec messageboard, please do not email us directly with support problems!
AROS is a portable and free desktop operating system aiming at being compatible with AmigaOS 3.1, while improving on it in many areas. The source code is available under an open source license, which allows anyone to freely improve upon it.
The goals of the AROS project is it to create an OS which:
1. Is as compatible as possible with AmigaOS 3.1.
2. Can be ported to different kinds of hardware architectures and processors, such as x86, PowerPC, Alpha, Sparc, HPPA and other.
3. Should be binary compatible on Amiga and source compatible on any other hardware.
4. Can run as a standalone version which boots directly from hard disk and as an emulation which opens a window on an existing OS to develop software and run Amiga and native applications at the same time.
5. Improves upon the functionality of AmigaOS.
To reach this goal, we use a number of techniques. First of all, we make heavy use of the Internet. You can participate in our project even if you can write only one single OS function. The most current version of the source is accessible 24 hours per day and patches can be merged into it at any time. A small database with open tasks makes sure work is not duplicated.
Some time back in the year 1993, the situation for the Amiga looked somewhat worse than usual and some Amiga fans got together and discussed what should be done to increase the acceptance of our beloved machine. Immediately the main reason for the missing success of the Amiga became clear: it was propagation, or rather the lack thereof. The Amiga should get a more widespread basis to make it more attractive for everyone to use and to develop for. So plans were made to reach this goal. One of the plans was to fix the bugs of the AmigaOS, another was to make it an modern operating system. The AOS project was born.
But exactly what was a bug? And how should the bugs be fixed? What are the features a so-called modern OS must have? And how should they be implemented into the AmigaOS?
Two years later, people were still arguing about this and not even one line of code had been written (or at least no one had ever seen that code). Discussions were still of the pattern where someone stated that "we must have ..." and someone answered "read the old mails" or "this is impossible to do, because ..." which was shortly followed by "youre wrong because ..." and so on.
In the winter of 1995, Aaron Digulla got fed up with this situation and posted an RFC (request for comments) to the AOS mailing list in which I asked what the minimal common ground might be. Several options were given and the conclusion was that almost everyone would like to see an open OS which is compatible to AmigaOS 3.1 (kickstart 40.68) on which further discussions could be based upon to see what is possible and what is not.
So the work began and AROS was born.
Download (MB)
Added: 2005-12-17 License: Freeware Price:
1422 downloads
LFMat 0.1.1
LFMats goal is to furnish convenient matrix tools for the finite element methods. more>>
LFMats goal is to furnish convenient matrix tools for the finite element methods. Actually, theres a lot of libraries for linear algebra on the net, but it seems that its still hard to find flexible and high performance free software for the required procedures (genericity, speed, adaptated storage, ...).
LFMat is a generic purpose, fully templated open source C++ matrix library. Particular attention has been furnished to get convenient storage for SIMD instructions like 3Dnow! and SSE2 on x86 processors and Altivec on PowerPC ones. It means that theres specializations for severals important types like float or double in order to get the deserving performances.
Furthermore, important routines make careful use of cache, leading -- as example -- to solvers up to 8 times faster than standard lapack ones in the same situation (see benchmarks).
Matrices can contain any kind of data (double, float, symbolic expressions, ...) and user can choose orientation, storage style and structure (see tutorial). Furthermore, matrices can be of fixed size (known at compilation time), allowing compilers to make additional optimizations.
Main features:
For now, storage styles can be:
- dense (n*m elements for a rectangular matrix, n*(n+1)/2 for a square symmetric matrix),
- dense uncompressed (n*n for a symmetric matrix),
- sky line (user gives the beginning and/or the end of each lines),
- sparse, row or column compressed,
- band.
Structures can be:
- generic (no particular properties),
- diagonal.
- symmetric,
- antisymmetric,
- hermitian,
- triangular, upper or lower,
- The number of reserved elements depends on both storage and structure.
Furthermore, matrices can be:
- row oriented,
- column oriented,
- diagonal oriented (still in progress).
- Some useful procedures have been coded for different kind of matrices:
- solvers (cholesky, ... see Table 1.2 ),
- operators (*, ... see Table 1.3),
- eigen values finders.
- converter between different kind of matrices
All these procedures have been designed to be fast, using cache and SIMD instruction where possible.
<<lessLFMat is a generic purpose, fully templated open source C++ matrix library. Particular attention has been furnished to get convenient storage for SIMD instructions like 3Dnow! and SSE2 on x86 processors and Altivec on PowerPC ones. It means that theres specializations for severals important types like float or double in order to get the deserving performances.
Furthermore, important routines make careful use of cache, leading -- as example -- to solvers up to 8 times faster than standard lapack ones in the same situation (see benchmarks).
Matrices can contain any kind of data (double, float, symbolic expressions, ...) and user can choose orientation, storage style and structure (see tutorial). Furthermore, matrices can be of fixed size (known at compilation time), allowing compilers to make additional optimizations.
Main features:
For now, storage styles can be:
- dense (n*m elements for a rectangular matrix, n*(n+1)/2 for a square symmetric matrix),
- dense uncompressed (n*n for a symmetric matrix),
- sky line (user gives the beginning and/or the end of each lines),
- sparse, row or column compressed,
- band.
Structures can be:
- generic (no particular properties),
- diagonal.
- symmetric,
- antisymmetric,
- hermitian,
- triangular, upper or lower,
- The number of reserved elements depends on both storage and structure.
Furthermore, matrices can be:
- row oriented,
- column oriented,
- diagonal oriented (still in progress).
- Some useful procedures have been coded for different kind of matrices:
- solvers (cholesky, ... see Table 1.2 ),
- operators (*, ... see Table 1.3),
- eigen values finders.
- converter between different kind of matrices
All these procedures have been designed to be fast, using cache and SIMD instruction where possible.
Download (0.40MB)
Added: 2006-03-02 License: GPL (GNU General Public License) Price:
1333 downloads
SIMD Cross-platform headers 2004.10.26
SIMD Cross-platform headers is a cross- platform, cross-compiler, cross CPU C/C++ header collection. more>>
SIMD Cross-platform headers is a cross- platform, cross-compiler, cross CPU C/C++ header collection that aids the creation portable vectorized (SIMD) C/C++ code.
SIMD Cross-platform headerst supports (or partially supports) x86 (MMX/SSE/SSE2) GCC and MSVC, PPC Altivec GCC and CodeWarrior, ARM GCC, and software-emulated SIMD.
NOTE: Code must be 16-byte aligned. Align to 16 when allocating memory.
X86/XSCALE (Intel) vs. PowerPC/MIPS
While the PowerPC and MIPS SIMD instructions take 2 source vectors and a destination vector, the Intel platforms only take a source and destination. Example:
PPC/MIPS can do:
C = A + B
X86 can only do:
A = A + B (or A+=B)
Code written either way will work on the X86, and still be faster than 387 math, but preserving the registers takes significant overhead (Disassemble the test program for an example. The prints preserve, the disassembly test does not.) For the fastest code between systems, write your SIMD math as the X86 expects, manually preserving SIMD variables.
At least GCC for PPC doesnt seem to have any issues figuring out how to deal with a source and destination memory address being the same.
Enhancements:
- Created file with some i386, GCC dialect
<<lessSIMD Cross-platform headerst supports (or partially supports) x86 (MMX/SSE/SSE2) GCC and MSVC, PPC Altivec GCC and CodeWarrior, ARM GCC, and software-emulated SIMD.
NOTE: Code must be 16-byte aligned. Align to 16 when allocating memory.
X86/XSCALE (Intel) vs. PowerPC/MIPS
While the PowerPC and MIPS SIMD instructions take 2 source vectors and a destination vector, the Intel platforms only take a source and destination. Example:
PPC/MIPS can do:
C = A + B
X86 can only do:
A = A + B (or A+=B)
Code written either way will work on the X86, and still be faster than 387 math, but preserving the registers takes significant overhead (Disassemble the test program for an example. The prints preserve, the disassembly test does not.) For the fastest code between systems, write your SIMD math as the X86 expects, manually preserving SIMD variables.
At least GCC for PPC doesnt seem to have any issues figuring out how to deal with a source and destination memory address being the same.
Enhancements:
- Created file with some i386, GCC dialect
Download (0.008MB)
Added: 2006-03-17 License: zlib/libpng License Price:
1319 downloads
glibc-audit 2.4-4
glibc-audit is a modified version of glibc for application developers who check their code with an automatic memory access check more>>
glibc-audit is a modified version of glibc for application developers who check their code with an automatic memory access checker such as valgrind, Insure++, or Purify.
glibc-audit has been audited and cleaned up so that reports from the developers use of a memory access checker are more likely to be interesting to the developer, with less "noise" from the C library itself. Typically, glibc-audit initializes all of its local variables and structs before use. Ordinary glibc uses uninitialized dummy variables that are "dont-care" to its logic but reported by the memory access checker.
Also, the r_debug.r_brk protocol has been enhanced to co-operate with a co-resident auditor. If the auditor sets .r_brk, then the runtime loader will call the auditor directly whenever a shared library event occurs.
This is much more convenient than using breakpoints. By default the old breakpoint protocol works just like before. The new protocol is binary compatible with the old on machines where a pointer to a function is the same size as an ordinary pointer. Platforms where a pointer to a function is larger (such as HP-PA RISC, Alpha processor, or PowerPC) are not binary backward compatible, and will have to increment r_debug.r_version. Exising clients (such as gdb) also will see an ignorable type mismatch error when they are built. But for now, it is worth more not to antagonize gdb at runtime on x86.
The patch modifies 91 files. Compared to glibc-2.3.2-27.9, the additional code occupies 18 more bytes of .text, and 24 fewer bytes in the .so. On a nano-scopic scale, the typical execution cost is 0 to 3 CPU cycles per affected routine; the estimated median total impact is less than 1 second per machine per day. In the case of *printf(), glibc-audit is faster than glibc because the cleaned-up source helps gcc-3.2 avoid generating atrocious code when initializing printf_spec.info for parse_one_spec() in stdio-common/printf-parse.h.
Glibc-audit was constructed by running a memory access checker on the internal testcases of glibc, then analyzing the reported errors and modifying the source. The process revealed 10 memory access bugs in glibc-2.3.2-11.9. Seven were fixed in glibc-2.3.2-27.9, two more have been fixed in CVS, and one is a design flaw that probably will not be fixed.
Predecessor patches to glibc-audit-1 were submitted to the glibc project, but those patches were ignored [user "guest", password "guest"], declined, or rejected. There is enough improvement in usability and reliability to publish glibc-audit-1 separately.
The unmodified glibc-2.3.2-27.9.src.rpm is available from RedHat mirrors. rpmbuild -ba --target i686 took about 4 hours and 2.5GB of disk space on a machine with 1.1GHz CPU, 384MB RAM, UDMA100 disk.
Enhancements:
- The patches were updated to glibc-2.4-4.
- A glibc bug that interfered with gdb stop-on-solib-events was fixed.
- On x86, x86-64, and PowerPC, the __NR_open system call was improved to avoid leaking information from the user to the kernel.
<<lessglibc-audit has been audited and cleaned up so that reports from the developers use of a memory access checker are more likely to be interesting to the developer, with less "noise" from the C library itself. Typically, glibc-audit initializes all of its local variables and structs before use. Ordinary glibc uses uninitialized dummy variables that are "dont-care" to its logic but reported by the memory access checker.
Also, the r_debug.r_brk protocol has been enhanced to co-operate with a co-resident auditor. If the auditor sets .r_brk, then the runtime loader will call the auditor directly whenever a shared library event occurs.
This is much more convenient than using breakpoints. By default the old breakpoint protocol works just like before. The new protocol is binary compatible with the old on machines where a pointer to a function is the same size as an ordinary pointer. Platforms where a pointer to a function is larger (such as HP-PA RISC, Alpha processor, or PowerPC) are not binary backward compatible, and will have to increment r_debug.r_version. Exising clients (such as gdb) also will see an ignorable type mismatch error when they are built. But for now, it is worth more not to antagonize gdb at runtime on x86.
The patch modifies 91 files. Compared to glibc-2.3.2-27.9, the additional code occupies 18 more bytes of .text, and 24 fewer bytes in the .so. On a nano-scopic scale, the typical execution cost is 0 to 3 CPU cycles per affected routine; the estimated median total impact is less than 1 second per machine per day. In the case of *printf(), glibc-audit is faster than glibc because the cleaned-up source helps gcc-3.2 avoid generating atrocious code when initializing printf_spec.info for parse_one_spec() in stdio-common/printf-parse.h.
Glibc-audit was constructed by running a memory access checker on the internal testcases of glibc, then analyzing the reported errors and modifying the source. The process revealed 10 memory access bugs in glibc-2.3.2-11.9. Seven were fixed in glibc-2.3.2-27.9, two more have been fixed in CVS, and one is a design flaw that probably will not be fixed.
Predecessor patches to glibc-audit-1 were submitted to the glibc project, but those patches were ignored [user "guest", password "guest"], declined, or rejected. There is enough improvement in usability and reliability to publish glibc-audit-1 separately.
The unmodified glibc-2.3.2-27.9.src.rpm is available from RedHat mirrors. rpmbuild -ba --target i686 took about 4 hours and 2.5GB of disk space on a machine with 1.1GHz CPU, 384MB RAM, UDMA100 disk.
Enhancements:
- The patches were updated to glibc-2.4-4.
- A glibc bug that interfered with gdb stop-on-solib-events was fixed.
- On x86, x86-64, and PowerPC, the __NR_open system call was improved to avoid leaking information from the user to the kernel.
Download (4.2MB)
Added: 2006-03-29 License: LGPL (GNU Lesser General Public License) Price:
1315 downloads
Secleted [ 0 ] software to compare
Copyright Notice:
Software piracy is theft, Using crack, password, serial numbers, registration codes, key generators is illegal and prevent future software development. The above powerpc search only lists software in full, demo and trial versions for free download. Download links are directly from our mirror sites or publisher sites, torrent files or links from rapidshare.com, yousendit.com or megaupload.com are not allowed
