|
About libfreevec
libfreevec is a free (LGPL) library with hand-optimized replacement routines for GLIBC, such as memcpy(), strlen(),
etc. These routines have been written specifically to take advantage of the AltiVec unit (a.k.a Velocity Engine or VMX),
and will only work on processors that include this unit. This means they will not work on older processors, such
as 603, 604, 750 (G3) or the POWER family of CPUs.
Check the Features page for more details on the exact set of functions that are
included, with comments and rough speed gain estimates, and the FAQ for additional
information about using the library itself.
News
Version 0.8 released! Download...
The author writes on the history of libfreevec:
"I have been a PowerPC user for many years, but I had never had the opportunity to work with a G4 processor until late
September 2004. At that time, I attended various AltiVec tutorials/seminars at the Euro SNDF show which were held by Dr
Sergei Larin and Chuck Corley from Freescale. I would like to thank both of them as they provided myself with the
inspiration to work on the AltiVec engine. During said seminars, I was shown how much performance is hidden in this small
and underestimated CPU. In times when everyone seems to be focused on 64-bit technology, it is no surprise that some
people dismiss the G4 processor as outdated technology. If only you knew...
Thanks to the support by Genesi and Freescale, I was given an Open Desktop Workstation for development purposes at the
show and was highly motivated to do software development on this platform. Soon after the seminars ended, I started
researching for source code on programming the AltiVec unit and its applications. To my surprise, I discovered that the G4
vector engine is basically unused by Linux itself which happens to be my platform of choice whereas it speeds up many system
components in MacOS X.
Why is this so? I have not found a definite answer to this question. One possible reason could be that AltiVec was
originally designed as a multimedia extension and for processing media data. While this simple fact cannot be denied,
it is still no convincing argument why this exciting technology has been neglected and not been used for more ordinary
tasks which might increase the performance of the whole system.
For example, did you know that you can do byte swapping with AltiVec 7 times faster than with scalar code? Or that it is
possible to sort integers and floats 4 times faster with the help of AltiVec? Were you aware that it helps to do string
searching faster? Memory hashing gets upto 7 times faster. The list could just go on and on...
With these things in mind, I was determined to take an active role in allowing Linux and other operating systems to profit
from AltiVec on a system wide scale. With the financial help of Genesi, I was able to start work on libfreevec which is
presented to you on this website. libfreevec is a library with many common glibc routines, rewritten and optimized to
use the vector engine found in the G4/G4+ PowerPC CPUs by Freescale. These processors are most commonly used in Apple Mac
computers and the Genesi Pegasos II Open Desktop Workstation (ODW). In addition to the glibc routines, it also includes
custom routines which were designed to speed up various other performance-critical tasks. For example, it includes special
AltiVec optimized versions of various string and memory functions from libstring which is part of the MySQL package.
Eventually, I hope to see the code being integrated in glibc so that the whole Linux system may benefit from AltiVec when
run on the right hardware. The current version is released under the terms of the free LGPL license and I sincerly hope that
it will enable many interested developers to give a performance boost to their applications for the PowerPC platform."
Konstantinos Margaritis
|
