tobast/m1-internship-report
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Presentation details, make fit in 14 pages

Browse source
This commit is contained in:
Théophile Bastian 2017-08-28 19:37:35 +02:00
parent c7d1841af7
commit 1802f8caa8
1 changed files with 10 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

19
report/report.tex
View file

@ -782,13 +782,14 @@ whole circuit, as signature of subgroups are always computed by default at the
order $2$, unless this particular group needs a more accurate signature. order $2$, unless this particular group needs a more accurate signature.
The measures were made for 100 consecutive runs of the program (then averaged The measures were made for 100 consecutive runs of the program (then averaged
for a single run) and measured by the command \texttt{time}. for a single run) and measured by the command \texttt{time}. The computing time
necessary for different signature levels is plotted in
Figure~\ref{fig:bench_sig_level}.
\begin{center} \begin{figure}
\centering
\begin{tikzpicture} \begin{tikzpicture}
\begin{axis}[ \begin{axis}[
title={Signature time of the processor for different levels
of signature},
xlabel={Level of signature}, xlabel={Level of signature},
ylabel={Time (ms)}, ylabel={Time (ms)},
xmin=0, xmax=16, xmin=0, xmax=16,
@ -821,7 +822,9 @@ for a single run) and measured by the command \texttt{time}.
\legend{-O3} \legend{-O3}
\end{axis} \end{axis}
\end{tikzpicture} \end{tikzpicture}
\end{center} \caption{Signature time of the processor for different levels of
signature}\label{fig:bench_sig_level}
\end{figure}
The computation time is more or less linear in in the level of signature The computation time is more or less linear in in the level of signature
required, which is coherent with the implementation. In practice, only small required, which is coherent with the implementation. In practice, only small
@ -867,8 +870,6 @@ overlapping), it takes \textbf{113\,ms}.
There were a few observed cases where the algorithm tends to be slower on There were a few observed cases where the algorithm tends to be slower on
certain configurations, and a few other such cases that could be fixed. certain configurations, and a few other such cases that could be fixed.
\todo{More corner cases}
\paragraph{I/O pins.} In Section~\ref{sec:signatures}, we introduce a term \paragraph{I/O pins.} In Section~\ref{sec:signatures}, we introduce a term
named \emph{IO adjacency} in the signatures of order higher than $0$. This is named \emph{IO adjacency} in the signatures of order higher than $0$. This is
because some sub-circuits can be told apart from their signatures only through because some sub-circuits can be told apart from their signatures only through
@ -878,7 +879,7 @@ the one in Figure~\ref{fig:io_adj_term}.
\begin{figure} \begin{figure}
\centering \centering
\includegraphics[width=0.8\textwidth]{img/io_adj_term.png} \includegraphics[width=0.6\textwidth]{img/io_adj_term.png}
\caption{A case where the I/O adjacency term is necessary}\label{fig:io_adj_term} \caption{A case where the I/O adjacency term is necessary}\label{fig:io_adj_term}
\end{figure} \end{figure}
@ -915,7 +916,7 @@ For instance, in Figure~\ref{fig:split_tree}, the orange borders are the
boundaries of what can be taken into account for the signatures of order $1$ of boundaries of what can be taken into account for the signatures of order $1$ of
the gates marked with a red dot. Thus, those signatures are exactly the same. the gates marked with a red dot. Thus, those signatures are exactly the same.
\begin{figure} \begin{figure}[hb!]
\centering \centering
\includegraphics[width=0.8\textwidth]{img/tree_local.png} \includegraphics[width=0.8\textwidth]{img/tree_local.png}
\caption{Case of a split (or merge) tree}\label{fig:split_tree} \caption{Case of a split (or merge) tree}\label{fig:split_tree}