[macstl-dev] macstl 0.3.0.32

Glen Low glen.low at pixelglow.com
Thu Aug 11 23:14:23 WST 2005

  • Previous message: [macstl-dev] Re: -faltivec without -maltivec copying; gcc's altivec attributes
  • Next message: [macstl-dev] macstl 0.3.0.32
  • Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]


Dear All

Just committed the new version of macstl to the Subversion  
repository. This has extensive but largely transparent changes to  
support better CSE and inlining, especially when compiling -faltivec  
without -maltivec on Apple gcc 4.0. I have optimized literal terms in  
valarray expressions as well as element access to valarray.

I'm aiming to get complete Linux PPC and x86 support before  
finalizing the public release of 0.3.1, but some of the other things  
on the to-do list:

1.    refarrays (allowing arbitrary pieces of memory to support  
valarray operations) for Paul and Maxim. (I refactored the array_term  
class in 0.3.0.32 to support this.)
2.    optimizing v >> k expressions for SSE for Maxim.
3.    dropping the scalar tail code for sequential expressions --  
this should boost relative speed for small valarrays.
4.    valarrays of sized booleans for Ilya.

If I can squeeze it in, then

5.    either the static interleave or deinterleave functions for Andrew.

Cheers, Glen Low


---
pixelglow software | simply brilliant stuff
www.pixelglow.com
aim: pixglen

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://www.pixelglow.com/lists/archive/macstl-dev/attachments/20050811/805a4776/attachment.html

  • Previous message: [macstl-dev] Re: -faltivec without -maltivec copying; gcc's altivec attributes
  • Next message: [macstl-dev] macstl 0.3.0.32
  • Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the macstl-dev mailing list