Changeset 1381 for docs/HPCA2012/03research.tex
 Timestamp:
 Aug 25, 2011, 2:50:46 PM (8 years ago)
 File:

 1 edited
Legend:
 Unmodified
 Added
 Removed

docs/HPCA2012/03research.tex
r1377 r1381 289 289 an abstract SIMD machine based on the Inductive Doubling 290 290 Instruction Set Architecture (IDISA) \cite{CameronLin2009}. 291 The abstract machine supports all powerof2 field widths up to the full 292 SIMD register width on a target machine. 293 Let $w = 2k$ be the field width in bits. Let $f$ be a basic binary operation defined on $w$bit quantities 294 producing an $w$bit result. Let $W$ be the SIMD vector size in bits where $W = 2K$. 295 Then the C++ template notation \verb`v=simd<w>::f(a,b)` denotes the general pattern for a vertical SIMD operation yielding an output SIMD vector $v$, 296 given two input SIMD vectors $a$ and $b$. For each field $v_i$ of $v$, the value computed is $f(a_i, b_i)$. 297 For example, given 128bit SIMD vectors, \verb`simd<8>::add(a,b)` represents the simultaneous addition of sixteen 8bit fields. 298 291 299 These operations were originally developed for 128bit Altivec operations on Power PC 292 300 as well as 64bit MMX and 128bit SSE operations on Intel
Note: See TracChangeset
for help on using the changeset viewer.