From: Magnus Holmgren <lear_at_algonet.se>
Date: Sat, 16 Jul 2005 15:08:31 +0200

Pedro Vasconcelos wrote:

> The optimizations that I made were fairly straightforward: writing short
> asm routines for 32-bit arithmetic in the hooks provided and placing
> some critical arrays in the fast IRAM.

Any idea how much the cosine table is used? It can be made to fit
(perhaps by throwing out some of the window lookup tables that aren't
used anyway), but when I tried that, I didn't notice much of a difference.

>>From discussions with Thom, I presume there should a big scope for
> optimizations in the actual algorithms used, particularly in the MDCT
> code. Unfortunately I don't know enough digital signal processing to do
> it and don't have spare time at the moment to go out and learn more :-(
>
> The other difficulty is the lack of profiling in the actual iriver
> hardware. I have done Tremor profiling on my P4 to get an idea of what
> were the critical functions, but the Coldfire is very diferent (cache,
> pipelines, etc) so it all guesswork.

Still, that should give an indication over what functions are used a
lot, to see what to focus on and what to put in IRAM... Have you asked
on the Tremor mailing list about profiling data (on ARM or ColdFire
hardware)?

Perhaps a ColdFire emulator can be used? Just being able to run
ivorbisfile_example.c might help.

   Magnus
_______________________________________________
http://cool.haxx.se/mailman/listinfo/rockbox
Received on 2005-07-16