Changes between Version 20 and Version 21 of SIMD


Ignore:
Timestamp:
Nov 11, 2011 7:44:29 PM (2 years ago)
Author:
duncan
Comment:

note FMA

Legend:

Unmodified
Added
Removed
Modified
  • SIMD

    v20 v21  
    324324 * vector constants, at least at Cmm level 
    325325 * replicating a scalar to a vector 
     326 * FMA: fused multiply add, this is supported by NEON and AVX however software fallback may not be possible with the same precision. Tricky. 
    326327 * AVX also suppports a bunch of interesting things: 
    327328   * permute, shuffle, "blend", masked moves.