Changeset 126 for anr/section-4.4.tex


Ignore:
Timestamp:
Feb 10, 2010, 5:59:35 PM (14 years ago)
Author:
coach
Message:

OM modification pendant la réunion du 10 février

File:
1 edited

Legend:

Unmodified
Added
Removed

  • anr/section-4.4.tex

    r113 r126  
    4747    supported.
    4848    The main restriction are:
    49     1) The HAS tools have not been yet enhanced,
     49    1) The backend HAS tools have not been yet enhanced,
    5050    2) dynamic partial reconfiguration is not supported,
    51     3) \mustbecompleted{FIXME:ALL .....}
     51    3)
     52        4)\mustbecompleted{FIXME:ALL .....}
    5253\item[Final Release ($T0+36$)]
    5354\end{description}
    … …  
    5758at the integration phase is significantly reduced.
    5859\par
    59 The project has several critical issues:
     60The risks that have been identified at the beginning of the project are the following:
    6061\begin{description}
    61 \item[\xcoachplus format (\novers{\specXcoachDoc}, \novers{\specXcoachToCA})]
    62     Because all the HAS tools rely on it, it is a
    63     crucial task. There are no work-arround but as mentionned in
    64     section~\ref{xcoach-problem} (page~\pageref{xcoach-problem}) we have worked on it
    65         for a year and are confident.
    66 \item[\xcoachplus format (\novers{\specXcoachDoc},
    67       \novers{\specXcoachToSystemC}, \novers{\specXcoachToVhdl})]
    68     Its aim is the generation of the coprocessors (hardware \& prototyping model).
    69     By centralizing the coprocessor generation, it guarantees their functioning
    70     independently of the used HAS tools.
    71         Our experience with UGH and GAUT give us confidence in the succes of this
    72         task.
    73 \item[prototyping of ALTERA \& XILINX architectural templates ({\csgAlteraSystemC},
    74      {\csgXilinxSystemC}]
    75      The SocLib component library contains most of the SystemC models used for the
    76      prototyping description of the ALTERA and XILINX architectural templates.
    77      Nevertheless, at this time we do not know how many are missing and if the existing
    78      are really useables.
    79      If the work of theses tasks is too important, they will be abandoned.
    80      In this case the work-arround to prototype the XILINX and ALTERA architectural
    81      templates is to use the COACH one. These architectures being very similar, the
    82      simulation results must be proportional. Theses tasks will be changed by measuring
    83      the deviance.
     62\item[\xcoach format (\novers{\specXcoachDoc}, \novers{\specXcoachToCA})]
     63        Partners have to agree on a convenient exchange format for all tools involved.
     64        Because all the HAS tools rely on it, the \xcoach format specification is a
     65    crucial step. There are no work-arround but as mentionned in
     66    section~\ref{xcoach-problem} (page~\pageref{xcoach-problem}) the five academic partners have worked on it
     67        for a full year and a first document already exists.
     68%\item[\xcoachplus format (\novers{\specXcoachDoc},
     69%      \novers{\specXcoachToSystemC}, \novers{\specXcoachToVhdl})]
     70%    Its aim is the generation of the coprocessors (hardware \& prototyping model).
     71%    By centralizing the coprocessor generation, it guarantees their functioning
     72%    independently of the used HAS tools.
     73%       Our experience with UGH and GAUT give us confidence in the succes of this
     74%       task.
     75\item[Virtual prototyping of ALTERA \& XILINX architectural templates ({\csgAlteraSystemC},
     76     {\csgXilinxSystemC})]
     77     The SocLib component library contains several SystemC models used for the virtual
     78     prototyping of the ALTERA and XILINX architectural templates (NIOS and Microblaze processor cores).
     79     Nevertheless, at this time we do not know how many IP cores SystemC simulation models have to be developped.
     80     If the workload of this simulation model development is too important, virtual prototyping
     81         of those architectural templates will not be directly supported.
     82         The three architectural templates being quite similar, the virtual
     83         prototyping will use the neutral architectural template.
    8484\item[VCI/AVALON \& VCI/PLB bridges (\novers{\hpcAvalonBridge}, \novers{\hpcPlbBridge})]
    8585     If one of these tasks is impossible or too important or leads to inefficiency,
    … …  
    8787     In this case, the COACH architectural template will not be available for HPC and
    8888     a SystemC VCI model corresponding to the PCI/X IP will be developped to allow
    89      prototyping.
     89     virtual prototyping.
    9090\end{description}
    9191
Note: See TracChangeset for help on using the changeset viewer.