On Fri, Jun 10, 2016 at 07:16:12AM +0900, Akira Yokosawa wrote:
> On 2016/06/10 1:30, Paul E. McKenney wrote:
> > On Fri, Jun 10, 2016 at 12:19:33AM +0900, Akira Yokosawa wrote:
> >> >From 5e22d5cb28e5db204c95386ea4427bacc985a7c9 Mon Sep 17 00:00:00 2001
> >> From: Akira Yokosawa <akiyks@xxxxxxxxx>
> >> Date: Fri, 10 Jun 2016 00:02:59 +0900
> >> Subject: [PATCH] cpu: Align numbers at decimal points in tables
> >>
> >> Signed-off-by: Akira Yokosawa <akiyks@xxxxxxxxx>
> >> ---
> >> cpu/overheads.tex | 56 +++++++++++++++++++++++++++++--------------------------
> >> 1 file changed, 30 insertions(+), 26 deletions(-)
> >>
> >> diff --git a/cpu/overheads.tex b/cpu/overheads.tex
> >> index 82853a8..2c89830 100644
> >> --- a/cpu/overheads.tex
> >> +++ b/cpu/overheads.tex
> >> @@ -111,24 +111,26 @@ This simplified sequence is just the beginning of a discipline called
> >>
> >> \begin{table}
> >> \centering
> >> -\begin{tabular}{l||r|r}
> >> - & & Ratio \\
> >> - Operation & Cost (ns) & (cost/clock) \\
> >> +\begin{tabular}{l||r@{.}l|r@{.}l}
> >> + & \multicolumn{2}{r}{} & \multicolumn{2}{r}{Ratio} \\
> >> + Operation & \multicolumn{2}{r}{Cost (ns)} & \multicolumn{2}{r}{(cost/clock)} \\
> >> \hline
> >> \hline
> >> - Clock period & 0.6 & 1.0 \\
> >> + Clock period & 0 & 6 & 1 & 0 \\
> >> \hline
> >> - Best-case CAS & 37.9 & 63.2 \\
> >> + Best-case CAS & 37 & 9 & 63 & 2 \\
> >> \hline
> >> - Best-case lock & 65.6 & 109.3 \\
> >> + Best-case lock & 65 & 6 & 109 & 3 \\
> >> \hline
> >> - Single cache miss & 139.5 & 232.5 \\
> >> + Single cache miss & 139 & 5 & 232 & 5 \\
> >> \hline
> >> - CAS cache miss & 306.0 & 510.0 \\
> >> + CAS cache miss & 306 & 0 & 510 & 0 \\
> >> \hline
> >> - Comms Fabric & 3,000 & 5,000 \\
> >> + Comms Fabric & \multicolumn{1}{r@{ }}{3,000} & &
> >> + \multicolumn{1}{r@{ }}{5,000} & \\
> >> \hline
> >> - Global Comms & 130,000,000 & 216,000,000 \\
> >> + Global Comms & \multicolumn{1}{r@{ }}{130,000,000} & &
> >> + \multicolumn{1}{r@{ }}{216,000,000} & \\
> >> \end{tabular}
> >> \caption{Performance of Synchronization Mechanisms on 4-CPU 1.8GHz AMD Opteron 844 System}
> >> \label{tab:cpu:Performance of Synchronization Mechanisms on 4-CPU 1.8GHz AMD Opteron 844 System}
> >> @@ -193,36 +195,38 @@ global agreement.
> >>
> >> \begin{table}
> >> \centering
> >> -\begin{tabular}{l||r|r}
> >> - & & Ratio \\
> >> - Operation & Cost (ns) & (cost/clock) \\
> >> +\begin{tabular}{l||r@{.}l|r@{.}l}
> >> + & \multicolumn{2}{r}{} & \multicolumn{2}{r}{Ratio} \\
> >> + Operation & \multicolumn{2}{r}{Cost (ns)} & \multicolumn{2}{r}{(cost/clock)} \\
> >> \hline
> >> \hline
> >> - Clock period & 0.4 & 1.0 \\
> >> + Clock period & 0 & 4 & 1 & 0 \\
> >> \hline
> >> - ``Best-case'' CAS & 12.2 & 33.8 \\
> >> + ``Best-case'' CAS & 12 & 2 & 33 & 8 \\
> >> \hline
> >> - Best-case lock & 25.6 & 71.2 \\
> >> + Best-case lock & 25 & 6 & 71 & 2 \\
> >> \hline
> >> - Single cache miss & 12.9 & 35.8 \\
> >> + Single cache miss & 12 & 9 & 35 & 8 \\
> >> \hline
> >> - CAS cache miss & 7.0 & 19.4 \\
> >> + CAS cache miss & 7 & 0 & 19 & 4 \\
> >> \hline
> >> - Off-Core & & \\
> >> + Off-Core & \multicolumn{1}{r@{ }}{} & & \multicolumn{1}{r@{ }}{} & \\
> >> \hline
> >> - Single cache miss & 31.2 & 86.6 \\
> >> + Single cache miss & 31 & 2 & 86 & 6 \\
> >> \hline
> >> - CAS cache miss & 31.2 & 86.5 \\
> >> + CAS cache miss & 31 & 2 & 86 & 5 \\
> >> \hline
> >> - Off-Socket & & \\
> >> + Off-Socket & \multicolumn{1}{r@{ }}{} & & \multicolumn{1}{r@{ }}{} & \\
> >> \hline
> >> - Single cache miss & 92.4 & 256.7 \\
> >> + Single cache miss & 92 & 4 & 256 & 7 \\
> >> \hline
> >> - CAS cache miss & 95.9 & 266.4 \\
> >> + CAS cache miss & 95 & 9 & 266 & 4 \\
> >> \hline
> >> - Comms Fabric & 4,500 & 7,500 \\
> >> + Comms Fabric & \multicolumn{1}{r@{ }}{4,500} & &
> >> + \multicolumn{1}{r@{ }}{7,500} & \\
> >> \hline
> >> - Global Comms & 195,000,000 & 324,000,000 \\
> >> + Global Comms & \multicolumn{1}{r@{ }}{195,000,000} & &
> >> + \multicolumn{1}{r@{ }}{324,000,000} & \\
> >> \end{tabular}
> >> \caption{Performance of Synchronization Mechanisms on 16-CPU 2.8GHz Intel X5550 (Nehalem) System}
> >> \label{tab:cpu:Performance of Synchronization Mechanisms on 16-CPU 2.8GHz Intel X5550 (Nehalem) System}
> >
> > How about the following instead, to get rid of the decimal point as well?
> >
> > Thanx, Paul
> >
> > ------------------------------------------------------------------------
> >
> > diff --git a/cpu/overheads.tex b/cpu/overheads.tex
> > index 82853a8e1832..7d33c4800edb 100644
> > --- a/cpu/overheads.tex
> > +++ b/cpu/overheads.tex
> > @@ -126,9 +126,13 @@ This simplified sequence is just the beginning of a discipline called
> > \hline
> > CAS cache miss & 306.0 & 510.0 \\
> > \hline
> > - Comms Fabric & 3,000 & 5,000 \\
> > + Comms Fabric & 3,000\textcolor{white}{.0}
> > + & 5,000\textcolor{white}{.0}
> > + \\
> > \hline
> > - Global Comms & 130,000,000 & 216,000,000 \\
> > + Global Comms & 130,000,000\textcolor{white}{.0}
> > + & 216,000,000\textcolor{white}{.0}
> > + \\
> > \end{tabular}
> > \caption{Performance of Synchronization Mechanisms on 4-CPU 1.8GHz AMD Opteron 844 System}
> > \label{tab:cpu:Performance of Synchronization Mechanisms on 4-CPU 1.8GHz AMD Opteron 844 System}
> > diff --git a/perfbook.tex b/perfbook.tex
> > index 938c5ee3fd46..5c74adb204af 100644
> > --- a/perfbook.tex
> > +++ b/perfbook.tex
> > @@ -17,6 +17,7 @@
> > \usepackage{graphicx}
> > \usepackage{enumerate}
> > \usepackage{ifthen}
> > +\usepackage{xcolor}
> > \usepackage{listings}
> > \lstset{basicstyle=\ttfamily}
> > % \usepackage[strings]{underscore}
> >
> >
>
> Ah, this is way simpler and easy to maintain.
>
> Feel free to add
>
> Reported-by: Akira Yokosawa <akiyks@xxxxxxxxx>
>
> Or do you want me to resend a revised patch?
I committed with your Reported-by an pushed it, so we are set.
Thanx, Paul
--
To unsubscribe from this list: send the line "unsubscribe perfbook" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
