Index: /anr/section-consortium-people.tex =================================================================== --- /anr/section-consortium-people.tex (revision 329) +++ /anr/section-consortium-people.tex (revision 330) @@ -44,6 +44,6 @@ \peopletabularentry{\lip} -responsible & Alias & Christophe & junior reasearcher & HLS & ... & Memory management, HLS \\\hline -contributor & Feautrier & Paul & emeritus professor & Compil. & ... & Process scheduling and building. +responsible & Alias & Christophe & Permanent researcher at Inria & Compilers, HPC & 15 & Memory management and FIFO construction. Irregular extensions.\\\hline +contributor & Feautrier & Paul & emeritus professor & Compilers & 13 & Process scheduling and building. Automatic Parallelization \\ \hline Index: /anr/section-ressources.tex =================================================================== --- /anr/section-ressources.tex (revision 329) +++ /anr/section-ressources.tex (revision 330) @@ -75,20 +75,18 @@ workstations is evaluated to 4\% of the total requested ANR funding. \item [Personnel costs] - The faculty members involved in the project are an emeritus - professor at ENS Lyon (Paul Feautrier) and a research associate - (CR2) at INRIA Rh\^one-Alpes (Christophe Alias). The non-permanent - personel required is a PhD student that will work on network process - generation from polyhedral loops, then on extensions to - non-polyhedral loops. We are looking for a student with both - theoretical and practical skills, that will be able to get a - sufficient understanding of the polyhedral techniques and to produce - a working implementation. - \parlf - The table below summarizes the \hommemois by - deliverables and tasks for both permanent and non-permanent - personnels. The non-permanent personnels costs represent 26\% of - the personnal costs. The requested funding for non - permanent personnels is 100\% of the total ANR requested funding. - \begin{center}\input{table_inria_compsys_full.tex}\end{center} + The faculty members involved in the project are Christophe Alias + (INRIA researcher) and Paul Feautrier (emeritus professor at + ENS-Lyon). The non-permanent personel required is a post-doc that + will work on FIFO construction, then on extensions of process + construction and memory optimization to non-polyhedral loops. We + are looking for a candidate with both theoretical and practical + skills, that will be able to get a sufficient understanding of the + polyhedral techniques to produce a working implementation. \parlf + The table below summarizes the \hommemois by deliverables and tasks + for both permanent and non-permanent personnels. The non-permanent + personnels costs represents ??\% of the personnal costs. The + requested funding for non permanent personnels is 100\% of the + total ANR requested + funding. \begin{center}\input{table_compsys_short.tex}\end{center} \item [Subcontracting] No subcontracting costs. @@ -299,2 +297,16 @@ \end{figure} \end{landscape} +% +\begin{landscape} +\begin{figure} +\begin{small} +\settowidth\desclen{XILINX RTL optimisation (5)} +\def\resstablestyletitle#1{\parbox{\desclen}{\textit{#1}}} +\hfill +\begin{minipage}[b]{.47\linewidth}\center +\input{table_compsys_full.tex}\vspace{.5ex}\\ \lip \vspace{2.5ex}\\ +\end{minipage}\hfill +\end{small} +\caption{\label{table-livrables-2} Man power in \hommemois for the deliverables (3)} +\end{figure} +\end{landscape} Index: /anr/table_compsys_full.tex =================================================================== --- /anr/table_compsys_full.tex (revision 330) +++ /anr/table_compsys_full.tex (revision 330) @@ -0,0 +1,18 @@ +\begin{tabular}{|c|l||r|r|r||r|}\hline +number & \multicolumn{1}{c||}{title} & \multicolumn{3}{c||}{years } & total \\\cline{3-5} + & & \multicolumn{1}{c|}{1} & \multicolumn{1}{c|}{2} & \multicolumn{1}{c||}{3} & \\\hline\hline +\hyperlink{D112}{D112} & \resstablestyletitle{Local project management} & 0.5 & 0.5 & 0.5 & 1.5 \\\hline + & total Task-1 & 1.5 & 0.0 & 0.0 & 1.5 \\\hline +\hline \hyperlink{D210}{D210} & \resstablestyletitle{Use model specification} & 1.0 & 0.0 & 0.0 & 1.0 \\\hline +\hyperlink{D230}{D230} & \resstablestyletitle{\xcoach format specification} & 4.0 & 2.0 & 0.0 & 6.0 \\\hline + & total Task-2 & 5.0 & 2.0 & 0.0 & 7.0 \\\hline +\hline \hyperlink{D430}{D430} & \resstablestyletitle{Process generation method} & 4.0 & 0.0 & 3.0 & 7.0 \\\hline +\hyperlink{D431}{D431} & \resstablestyletitle{Arrays and FIFO} & 1.5 & 3.0 & 1.5 & 6.0 \\\hline +\hyperlink{D432}{D432} & \resstablestyletitle{Non-polyhedral extension} & 0.0 & 7.0 & 13.0 & 20.0 \\\hline + & total Task-4 & 5.5 & 10.0 & 17.5 & 33.0 \\\hline +\hline \hyperlink{D831}{D831} & \resstablestyletitle{HAS front-end user manual} & 0.0 & 0.5 & 1.0 & 1.5 \\\hline +\hyperlink{D840}{D840} & \resstablestyletitle{Publication, communication} & 0.0 & 1.0 & 1.0 & 2.0 \\\hline + & total Task-8 & 0.0 & 1.5 & 2.0 & 3.5 \\\hline +\hline + & total & 11.5 & 14.5 & 20.0 & 46.0 \\\hline +\end{tabular} Index: /anr/table_compsys_short.tex =================================================================== --- /anr/table_compsys_short.tex (revision 330) +++ /anr/table_compsys_short.tex (revision 330) @@ -0,0 +1,10 @@ +\begin{center}\begin{small}\begin{tabular}{|c|l||r|r|r||r|}\hline + & title & \multicolumn{3}{c||}{years } & total \\\cline{3-5} + & & \multicolumn{1}{c|}{1} & \multicolumn{1}{c|}{2} & \multicolumn{1}{c||}{3} & \\\hline\hline +Task-1 & Project management & 0.5 & 0.5 & 0.5 & 1.5 \\\hline +Task-2 & Backbone infrastructure & 5.0 & 2.0 & 0.0 & 7.0 \\\hline +Task-4 & HAS front-end & 5.5 & 10.0 & 17.5 & 33.0 \\\hline +Task-8 & Dissemination & 0.0 & 1.5 & 2.0 & 3.5 \\\hline +\hline + & total & 11.5 & 14.5 & 20.0 & 46.0 \\\hline +\end{tabular}\end{small}\end{center} Index: /anr/task-frontend.tex =================================================================== --- /anr/task-frontend.tex (revision 329) +++ /anr/task-frontend.tex (revision 330) @@ -59,7 +59,6 @@ \subtask{Automatic parallelization and memory optimization} - This sub-task aims at providing a source-level optimizer in front the -HLS back-end tools. The optimization are twofold: +HLS back-end tools. The optimizations are threefold: \begin{itemize} \item Extraction of parallelism in polyhedral loops and conversion @@ -70,7 +69,7 @@ We will design these methods by using polyhedral techniques, as we did in the past for pure HPC optimizations. The program model is typically -regular programs with affine loop bound conditions and array index -functions. In a second part, we will extend the program model by using -conservative approximations. +regular programs where loop bounds, conditions and array indices are +affine functions. In a second part, we will extend the program model +by using conservative approximations. \begin{livrable} \itemV{0}{6}{d}{\Slip}{Method, Preliminary Definition} @@ -85,9 +84,9 @@ \itemV{6}{12}{d}{\Slip} {Arrays and FIFO} Description and specification of the FIFO construction method - and local memory optimization algorithms. + and the local memory optimization method. \itemV{12}{18}{d+x}{\Slip} {Arrays and FIFO} Preliminary implementation in the Bee framework. Conversion of the input and output of Bee to the \xcoach format. - At this step, only memory optimizations will be available. + At this step, only local memory optimization will be available. \itemL{18}{30}{d+x}{\Slip}{Arrays and FIFO}{10:0:9} Final assessment of the method and improved version of the specification.