[tremor] Tremor on strongarm CPU

[dilb]

On Monday 9 September 2002 18:25, Nicolas Pitre wrote:
> Depends how many bits you're using in the final output samples.  This is in
> fact a clever way to eliminate a second shift instruction when scaling back
> the multiplication result.  You're losing one (or two) bit of accuracy each
> time a particular value is multiplied, but if your final samples are only
> 16 bits wide you then have some room out of the 16 remaining bits for
> accuracy loss.

If these bits are not useful, I will update my assembly, to save some more 
cycles ; at the beginning, I thought this trick was only used in arm_asm.h, 
but I've seen the same in misc.h, so I suppose these bits are not too 
important :).

<p>> If applying the same set of compiler flags to Tremor that I used in my own
> version, I also get that 5% or more improvement as well.  Would you mind
> posting your assembly version (or making it available somewhere if too big)
> and I could compare the performance gain as well?
I did play with compiler flags too, and so far, my best combination is:
CFLAGS = -O2 -D_ARM_ASSEM_ -fsigned-char  -D_REENTRANT -ffast-math 
-mcpu=strongarm1100 -fomit-frame-pointer -fthread-jumps  -fregmove 
-DUSE_MEMORY_H
when using gcc 2.95.4
I will generate an OVD_0.6 version of the libraries soon (we use it in one of 
the familiar projects), including my assembly in Tremor.

> So far, inspecting the generated assembly from a recent gcc (currently
> testing with gcc 3.2), the compiler is increasingly doing a pretty good job
> out of plain C code.  It just looks like writing assembly code is less
> worth it than it was before.
did you compile as a cross-compiler ? If so, is the pakage available 
somewhere ?
Do you know if profiling is working on this version ?

<p>> Oh parfait, pas de problème alors.
he he :)
>
> Nicolas
Dilb
--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to 'tremor-request at xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is needed.
Unsubscribe messages sent to the list will be ignored/filtered.