tobast/phd-thesis
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

Foundations: small amends

Browse source
This commit is contained in:
Théophile Bastian 2024-01-23 18:40:13 +01:00
parent 627ec583c9
commit 39d8741ae7
1 changed files with 5 additions and 4 deletions

9
manuscrit/20_foundations/20_code_analyzers.tex
View file

@ -341,14 +341,12 @@ $\cyc{\kerK} = 1.5$.
This last right-hand expression is independent of $n$, which we note $M$. This last right-hand expression is independent of $n$, which we note $M$.
Dividing by $n$, we obtain Dividing by $n$, we obtain
\[ \[
\abs{\dfrac{\ckn{n}}{n} - \cyc{\kerK}} &\leq \dfrac{M}{n} \\ \abs{\dfrac{\ckn{n}}{n} - \cyc{\kerK}} \leq \dfrac{M}{n} \\
\] \]
from which both results follow. from which both results follow.
\end{proof} \end{proof}
\todo{}
\medskip \medskip
Throughout this manuscript, we mostly use reciprocal throughput as a metric, as Throughout this manuscript, we mostly use reciprocal throughput as a metric, as
@ -362,7 +360,10 @@ stead.
\begin{definition}[Throughput of a kernel] \begin{definition}[Throughput of a kernel]
The \emph{throughput} of a kernel $\kerK$, measured in \emph{instructions The \emph{throughput} of a kernel $\kerK$, measured in \emph{instructions
per cycle}, or IPC, is defined as the number of instructions in $\kerK$, divided per cycle}, or IPC, is defined as the number of instructions in $\kerK$, divided
by the steady-state execution time of $\kerK$. by the steady-state execution time of $\kerK$:
\[
\operatorname{IPC}(\kerK) = \dfrac{\card{\kerK}}{\cyc{\kerK}}
\]
\end{definition} \end{definition}
In the literature or in analyzers' reports, the throughput of a kernel is often In the literature or in analyzers' reports, the throughput of a kernel is often