Re: mplayer
Hi, Billy Biggs wrote:
> Maybe you'd like to educate us on how to select optimized routines at
> runtime without penalties.
You can't- You can, however, select them in such a way that it's no
problem.
> I know that in my application I have
> gratuitous use of function pointers to deal with this.
Define "gratuitous". You can dynamically load a library with the most
speedful implementation for your hardware, thus you can hide the function
pointer in the linker and replace them by an additional jump instruction.
> What do you
> recommend I do to avoid that penalty?
Optimized code typically runs for far longer than the dereferencing of a
function pointer takes, so in terms of speed there simply isn't any.
If the function-pointerized code just looks ugly, the C preprocessor is
your friend -- on the other hand you don't even need that, as foo(bar,...)
is equivalent to (*foo)(bar,...) if foo is a pointer to a function.
--
Matthias Urlichs | {M:U} IT Design @ m-u-it.de | smurf@smurf.noris.de
Disclaimer: The quote was selected randomly. Really. | http://smurf.noris.de
--
System restarting, wait...
Reply to:
- References:
- mplayer
- From: martin f krafft <madduck@debian.org>
- Re: mplayer
- From: Mike Fedyk <mfedyk@matchmail.com>
- Re: mplayer
- From: Andrew Suffield <asuffield@debian.org>
- Re: mplayer
- From: Andrew Suffield <asuffield@debian.org>
- Re: mplayer
- From: Billy Biggs <vektor@dumbterm.net>