We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

AR# 40040

LogiCORE Linear Algebra Toolkit - Release Notes and Known Issues


This Answer Record contains the Release Notes and Known Issues list for the CORE Generator LogiCORE Linear Algebra Toolkit Core.

The following information is listed for each version of the core:
  • New Features
  • Resolved Issues
  • Known Issues


LogiCORE Linear Algebra Toolkitv1.0

Initial Release in ISE 13.1.

New Features
    • Fixed-point matrix-matrix addition and subtraction, matrix-scalar multiplication, matrix-matrix multiplication supported
    • Configurable matrix dimension supported up to 32x32
    • Real or complex data support (design is optimized for both types)
    • Two's complement fixed-point maximum of 24-bit input and 48-bit output (for matrix multiplication, maximum input width support is for 24-bit x 18-bit unit multiplication)
    • Symmetric, asymmetric, convergent, and truncation options for output rounding
    • Output saturation is supported
    • Fully AXI4 real-time streaming compliant core
    • Fully synchronous, single clock domain design with AXI4-Stream Protocol compliant synchronous active low reset
    • Clock Enable (CE) input provided, core can be paused and resumed on-the-fly by de-asserting and re-asserting CE signal
    • User selectable folding factor (resource reuse factor, 1 indicating no reuse, 2 for resources reused twice, and so on) for resource optimization
    • Data I/O Support: serial (where matrix elements are fed serially) or parallel (where matrix elements are packed and fed to the core)
    • Uses SIMD mode of DSP48E1 to implement complex addition (in matrix addition/subtraction mode) on a single slice
    • Option of DSP48E1 slice or LUT (implemented on the fabric) based implementation for matrix-matrix addition/subtraction
    • Folding factors supported for matrix size MxN (for matrix addition, subtraction and scalar multiplication) are:1, 2, 3, M, N, MN
    • Folding factors supported for matrix multiplication CMxN = AMxL x BLxN are:1, 2, 3, M, MN
    • Bit-accurate C Model available

    Resolved Issues
    • None

    Known Issues(ISE)
    • None

    Known Issues (Vivado)

    • (Xilinx Answer 53465) 2012.4 Vivado Simulator - Why does my DSP Digital Communications core fail to simulate with error Error: Failed to find design work <Core name>?
    AR# 40040
    Date Created 02/10/2011
    Last Updated 01/02/2013
    Status Active
    Type Release Notes
    • Linear Algebra Toolkit