Ignore:
Timestamp:
Aug 31, 2011, 3:13:36 PM (8 years ago)
Author:
ashriram
Message:

Minor bug fixes

File:
1 edited

Legend:

Unmodified
Added
Removed
  • docs/HPCA2012/03-research.tex

    r1398 r1407  
    6868point to determine other bit streams.  In particular, Parabix uses the
    6969basis bit streams to construct \emph{character-class bit streams} in
    70 which each $\tt 1$ bit indicates the presense of a significant
     70which each $\tt 1$ bit indicates the presence of a significant
    7171character (or class of characters) in the parsing process.
    7272Character-class bit streams may then be used to compute \emph{lexical
     
    138138Unlike the single-cursor approach of traditional text parsers, these allow Parabix to process multiple cursors in parallel.
    139139Error bit streams are often the byproduct or derivative of computing lexical bit streams and can be used to identify any well-formedness
    140 issues found during the parsing process. The presense of a $\tt 1$ in an error stream indicates that the lexical stream cannot be
     140issues found during the parsing process. The presence of a $\tt 1$ in an error stream indicates that the lexical stream cannot be
    141141trusted to be completely accurate and it may be necessary to perform some sequential parsing on that section to determine the cause and severity
    142142of the error. %How errors are handled depends on the logical implications of the error and go beyond the scope of this paper.
     
    347347sixteen 8-bit fields.
    348348
    349 These operations were originally developed for 128-bit Altivec operations on Power PC
    350 as well as 64-bit MMX and 128-bit SSE operations on Intel
    351 but have recently extended to support
    352 the new 256-bit AVX operations on Intel as well as the 128-bit
    353 \NEON{} operations on the ARM architecture.
    354 
     349We have ported parabix to a wide variety of processor architectures
     350demonstrating its applicability to commodity SIMD hardware. We
     351currently take advantage of the 128-bit Altivec operations on the
     352Power PC, 64-bit MMX and 128-bit SSE operations on previous generation
     353Intel platforms, the latest 256-bit AVX extensions on the Sandybridge
     354processor, and finally the 128-bit \NEON{} operations on ARM.
     355
Note: See TracChangeset for help on using the changeset viewer.