From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Akira Yokosawa <akiyks@gmail.com>
Cc: perfbook@vger.kernel.org
Subject: Re: [PATCH] defer/rcuapi: Merge Tables 9.3 and 9.4
Date: Sun, 24 Jul 2016 17:06:03 -0700 [thread overview]
Message-ID: <20160725000603.GY7094@linux.vnet.ibm.com> (raw)
In-Reply-To: <fa1dcdab-7d30-3066-4286-74038c355735@gmail.com>
On Sun, Jul 24, 2016 at 11:39:54AM +0900, Akira Yokosawa wrote:
> >From 1e4c0d9ae90cf0f5dcd44a27651614119aebef62 Mon Sep 17 00:00:00 2001
> From: Akira Yokosawa <akiyks@gmail.com>
> Date: Sun, 24 Jul 2016 10:39:37 +0900
> Subject: [PATCH] defer/rcuapi: Merge Tables 9.3 and 9.4
>
> By using "sidewaystable" environment, Tables 9.3 and 9.4 can be
> merged into a wider single table.
>
> Signed-off-by: Akira Yokosawa <akiyks@gmail.com>
Nice! Applied and pushed, thank you!
Thanx, Paul
> ---
> defer/rcuapi.tex | 112 ++++++++++++++++++++-----------------------------------
> perfbook.tex | 1 +
> 2 files changed, 41 insertions(+), 72 deletions(-)
>
> diff --git a/defer/rcuapi.tex b/defer/rcuapi.tex
> index 183bb8b..6f69115 100644
> --- a/defer/rcuapi.tex
> +++ b/defer/rcuapi.tex
> @@ -16,159 +16,127 @@ presents concluding remarks.
> \subsubsection{RCU has a Family of Wait-to-Finish APIs}
> \label{sec:defer:RCU has a Family of Wait-to-Finish APIs}
>
> -\begin{table*}[p]
> +\begin{sidewaystable*}[htbp]
> \centering
> -\scriptsize\tymin=1.1in
> -\begin{tabulary}{5.9in}{L|L|L|L|L}
> +\scriptsize\tymin=1.4in
> +\begin{tabulary}{7.8in}{L|L|L|L|L|L}
> Attribute &
> RCU Classic &
> RCU BH &
> RCU Sched &
> - Realtime RCU \\
> + Realtime RCU &
> + SRCU \\
> \hline
> \hline
> Purpose &
> Original &
> Prevent DDoS attacks &
> Wait for preempt-disable regions, hardirqs, \& NMIs &
> - Realtime response \\
> + Realtime response &
> + Sleeping readers \\
> \hline
> Availability &
> 2.5.43 &
> 2.6.9 &
> 2.6.12 &
> - 2.6.26 \\
> + 2.6.26 &
> + 2.6.19 \\
> \hline
> Read-side primitives &
> - \begin{minipage}[t]{1.0in}{ \raggedright
> + \begin{minipage}[t]{1.4in}{ \raggedright
> \co{rcu_read_lock()}~! \\
> \co{rcu_read_unlock()}~! }\end{minipage} &
> - \begin{minipage}[t]{1.0in}{ \raggedright
> + \begin{minipage}[t]{1.4in}{ \raggedright
> \co{rcu_read_lock_bh()} \\
> \co{rcu_read_unlock_bh()} }\end{minipage} &
> - \begin{minipage}[t]{1.0in}{ \raggedright
> + \begin{minipage}[t]{1.4in}{ \raggedright
> \co{preempt_disable()} \\
> \co{preempt_enable()} \\
> (and friends) }\end{minipage} &
> - \begin{minipage}[t]{1.0in}{ \raggedright
> + \begin{minipage}[t]{1.4in}{ \raggedright
> \co{rcu_read_lock()} \\
> - \co{rcu_read_unlock()} }\end{minipage} \\
> + \co{rcu_read_unlock()} }\end{minipage} &
> + \begin{minipage}[t]{1.4in}{ \raggedright
> + \co{srcu_read_lock()} \\
> + \co{srcu_read_unlock()} }\end{minipage} \\
> \hline
> { Update-side primitives (synchronous) } &
> { \co{synchronize_rcu()} \co{synchronize_net()} } &
> \co{synchronize_rcu_bh()} &
> \co{synchronize_sched()} &
> - { \co{synchronize_rcu()} \co{synchronize_net()} } \\
> + { \co{synchronize_rcu()} \co{synchronize_net()} } &
> + \co{synchronize_srcu()} \\
> \hline
> { Update-side primitives (asynchronous/callback) } &
> \co{call_rcu()} ! &
> \co{call_rcu_bh()} &
> \co{call_rcu_sched()} &
> - \co{call_rcu()} \\
> + \co{call_rcu()} &
> + \co{call_srcu()} \\
> \hline
> { Update-side primitives (wait for callbacks) } &
> \co{rcu_barrier()} &
> \co{rcu_barrier_bh()} &
> \co{rcu_barrier_sched()} &
> - \co{rcu_barrier()} \\
> + \co{rcu_barrier()} &
> + N/A \\
> \hline
> Type-safe memory &
> \co{SLAB_DESTROY_BY_RCU} &
> &
> &
> - \co{SLAB_DESTROY_BY_RCU} \\
> + \co{SLAB_DESTROY_BY_RCU} &
> + \\
> \hline
> Read side constraints &
> No blocking &
> No bottom-half (BH) enabling &
> No blocking &
> - Only preemption and lock acquisition \\
> + Only preemption and lock acquisition &
> + No \co{synchronize_srcu()} wtih same \co{srcu_struct} \\
> \hline
> Read side overhead &
> Preempt disable/enable (free on non-PREEMPT) &
> BH disable/enable &
> Preempt disable/enable (free on non-PREEMPT) &
> - Simple instructions, irq disable/enable \\
> + Simple instructions, irq disable/enable &
> + Simple instructions, preempt disable/enable, memory barriers \\
> \hline
> Asynchronous update-side overhead &
> sub-microsecond &
> sub-microsecond &
> sub-microsecond &
> - sub-microsecond \\
> + sub-microsecond &
> + N/A \\
> \hline
> Grace-period latency &
> 10s of milliseconds &
> 10s of milliseconds &
> 10s of milliseconds &
> - 10s of milliseconds \\
> + 10s of milliseconds &
> + 10s of milliseconds \\
> \hline
> Non-\co{PREEMPT_RT} implementation &
> RCU Classic &
> RCU BH &
> RCU Classic &
> - Preemptible RCU \\
> + Preemptible RCU &
> + SRCU \\
> \hline
> \co{PREEMPT_RT} implementation &
> Preemptible RCU &
> Realtime RCU &
> Forced Schedule on all CPUs &
> - Realtime RCU \\
> + Realtime RCU &
> + SRCU \\
> \end{tabulary}
> \caption{RCU Wait-to-Finish APIs}
> \label{tab:defer:RCU Wait-to-Finish APIs}
> -\end{table*}
> -
> -\begin{table*}[p]
> -\centering
> -\scriptsize\tymin=1.1in\tymax=1.5in
> -\begin{tabulary}{4.5in}{L|L}
> -Attribute &
> - SRCU \\
> -\hline
> -\hline
> -Purpose &
> - Sleeping readers \\
> -\hline
> -Availability &
> - 2.6.19 \\
> -\hline
> -Read-side primitives &
> - \begin{minipage}[t]{1.5in}{ \raggedright
> - \co{srcu_read_lock()} \\
> - \co{srcu_read_unlock()} }\end{minipage} \\
> -\hline
> -Update-side primitives (synchronous) &
> - \co{synchronize_srcu()} \\
> -\hline
> -Update-side primitives (asynchronous/callback) &
> - \co{call_srcu()} \\
> -\hline
> -Update-side primitives (wait for callbacks) &
> - N/A \\
> -\hline
> -Type-safe memory &
> - \\
> -\hline
> -Read side constraints &
> - No \co{synchronize_srcu()} wtih same \co{srcu_struct} \\
> -\hline
> -Read side overhead &
> - Simple instructions, preempt disable/enable, memory barriers \\
> -\hline
> -Asynchronous update-side overhead &
> - N/A \\
> -\hline
> -Grace-period latency &
> - 10s of milliseconds \\
> -\end{tabulary}
> -\caption{Sleepable RCU Wait-to-Finish APIs}
> -\label{tab:defer:Sleepable RCU Wait-to-Finish APIs}
> -\end{table*}
> +\end{sidewaystable*}
>
> The most straightforward answer to ``what is RCU'' is that RCU is
> an API used in the Linux kernel, as summarized by
> -Tables~\ref{tab:defer:RCU Wait-to-Finish APIs} and
> -\ref{tab:defer:Sleepable RCU Wait-to-Finish APIs},
> +Table~\ref{tab:defer:RCU Wait-to-Finish APIs},
> which shows the wait-for-RCU-readers portions of the non-sleepable and
> sleepable APIs, respectively,
> and by
> @@ -386,7 +354,7 @@ elsewhere~\cite{PaulEMcKenney2007PreemptibleRCU}.
> } \QuickQuizEnd
>
> The ``SRCU'' column in
> -Table~\ref{tab:defer:Sleepable RCU Wait-to-Finish APIs}
> +Table~\ref{tab:defer:RCU Wait-to-Finish APIs}
> displays a specialized RCU API that permits
> general sleeping in RCU read-side critical
> sections~\cite{PaulEMcKenney2006c}.
> diff --git a/perfbook.tex b/perfbook.tex
> index 493c05a..9216485 100644
> --- a/perfbook.tex
> +++ b/perfbook.tex
> @@ -15,6 +15,7 @@
> \usepackage{subfigure}
> % \usepackage{breakurl}
> \usepackage{graphicx}
> +\usepackage{rotating}
> \usepackage{enumerate}
> \usepackage{ifthen}
> \usepackage{xcolor}
> --
> 1.9.1
>
prev parent reply other threads:[~2016-07-25 0:05 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-24 2:39 [PATCH] defer/rcuapi: Merge Tables 9.3 and 9.4 Akira Yokosawa
2016-07-25 0:06 ` Paul E. McKenney [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20160725000603.GY7094@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=akiyks@gmail.com \
--cc=perfbook@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.