talk-2019-10-OOPSLA19/slides.tex

% vim: spell spelllang=en

\documentclass[11pt,xcolor={usenames,dvipsnames}]{beamer}
\usetheme{Warsaw}
\usepackage[utf8]{inputenc}
\usepackage[english]{babel}
\usepackage[T1]{fontenc}
\usepackage{amsmath}
\usepackage{amsfonts}
\usepackage{amssymb}
\usepackage{booktabs}
\usepackage{makecell}
\usepackage{ifthen}
\usepackage{colortbl}
\usepackage{tabularx}
\usepackage{pifont}

\usepackage{texlib/my_listings}
\usepackage{texlib/specific}
\usepackage{texlib/common}
\usepackage{texlib/todo}

\usepackage{inconsolata}
\lstset{basicstyle=\footnotesize\ttfamily}

\renewcommand\theadalign{c}
\renewcommand\theadfont{\scriptsize\bfseries}

\setbeamertemplate{navigation symbols}{}
\setbeamertemplate{headline}{}

\newcommand{\thenalert}[1]{\only<1>{#1}\only<2>{\alert{#1}}}
\newcommand{\slidecountline}{
    \ifthenelse{\theframenumber = 0}
        {}
        {\insertframenumber/\inserttotalframenumber}}
\newcommand{\sectionline}{
    \ifthenelse{\thesection = 0}
    {}
    {\Roman{section}~-- \insertsection}}

\newcommand{\cmark}{\color{OliveGreen}\ding{52}}
\newcommand{\xmark}{\color{BrickRed}\ding{56}}

\AtBeginSection[]{
    \begin{frame}
        \vfill
        \centering
        \begin{beamercolorbox}[sep=8pt,center,shadow=true,rounded=true]{title}
            \usebeamerfont{title}\insertsectionhead\par%
        \end{beamercolorbox}
        \vfill
    \end{frame}
}

\lstdefinelanguage{gdb}{
    morekeywords={gdb},
    sensitive=false,
}


\newcolumntype{b}{X}
\newcolumntype{s}{>{\hsize=.43\hsize}X}

\newcommand{\lstinl}
{\lstinline[language=C, keepspaces=true, basicstyle=\ttfamily]}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\title[\sectionline] {Growing the DWARF tougher:\\
    synthesis, validation and compilation}
\author[\slidecountline]{\textbf{Théophile Bastian} \\ \vspace{0.5em}
{{\footnotesize{}Based on work done with}\\
\textbf{Francesco Zappa Nardelli},
\textbf{Stephen Kell},
\textbf{Simon Ser}}}
\date{}
%\subject{}
%\logo{}
\institute{ENS Paris, INRIA}

\begin{document}

\begin{frame}
    \addtocounter{framenumber}{-1}
	\titlepage{}

    \vspace{-2em}
    \begin{center}
            \todo{}
        \begin{align*}
            \text{Slides: } &\text{\url{https://tobast.fr/m2/slides.pdf}} \\
            \text{Report: } &\text{\url{https://tobast.fr/m2/report.pdf}}
        \end{align*}
    \end{center}
\end{frame}

\begin{frame}{~}
    \addtocounter{framenumber}{-1}
    \tableofcontents[hideallsubsections]
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{DWARF and stack unwinding data}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\subsection{Introduction}

\begin{frame}[fragile]{We often use stack unwinding!}
    \begin{columns}[c]
        \begin{column}{0.70\textwidth}
            \begin{lstlisting}[language=gdb, numbers=none, escapechar=|]
Program received signal SIGSEGV.
0x54625 in fct_b at segfault.c:5
5           printf("%l\n", *b);

|\pause| (gdb) backtrace
#0   0x54625 in fct_b at segfault.c:5
#1   0x54663 in fct_a at segfault.c:10
#2   0x54674 in main at segfault.c:14

|\pause| (gdb) frame 1
#1   0x54663 in fct_a at segfault.c:10
10       fct_b((int*) a);

|\pause| (gdb) print a
$1 = 84
            \end{lstlisting}
            \vspace{-1em}
            \pause{}
            \begin{center}
                \textbf{\Large How does it work?!}
            \end{center}
        \end{column}
        \begin{column}{0.35\textwidth}
            \pause{}
            \includegraphics[width=0.95\linewidth]{img/call_stack}
        \end{column}
    \end{columns}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\subsection{Stack frames and unwinding}

\begin{frame}{Call stack and registers}
    \begin{columns}[c]
        \begin{column}{0.55\textwidth}
            \begin{center}
                \large\bf
                How do we get the grandparent RA\@?

                \medskip

                Isn't it as trivial as \texttt{pop()}?

                \vspace{2em}

                \only<2>{We only have \reg{rsp} and \reg{rip}.}

            \end{center}
        \end{column}
        \begin{column}{0.45\textwidth}
            \includegraphics[width=0.95\linewidth]{img/call_stack}
        \end{column}
    \end{columns}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\subsection{DWARF tables}

\newcolumntype{a}{>{\columncolor{RedOrange}}l}

\begin{frame}{DWARF unwinding data}
    \vspace{2em}
    \tt \footnotesize
    \begin{tabular}{
            >{\columncolor{YellowGreen}}l
            >{\columncolor{Thistle}}l
            l l l l l l
            >{\columncolor{Apricot}}l}
~LOC & CFA & rbx & rbp & r12 & r13 & r14 & r15 & ra \\
0084950 & rsp+8 & u & u & u & u & u & u & c-8 \\
0084952 & rsp+16 & u & u & u & u & u & c-16 & c-8 \\
0084954 & rsp+24 & u & u & u & u & c-24 & c-16 & c-8 \\
0084956 & rsp+32 & u & u & u & c-32 & c-24 & c-16 & c-8 \\
0084958 & rsp+40 & u & u & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084959 & rsp+48 & u & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
\rowcolor{Aquamarine} 008495a & rsp+56 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084962 & rsp+64 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084a19 & rsp+56 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084a1d & rsp+48 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084a1e & rsp+40 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084a20 & rsp+32 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084a22 & rsp+24 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084a24 & rsp+16 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084a26 & rsp+8 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
0084a30 & rsp+64 & c-56 & c-48 & c-40 & c-32 & c-24 & c-16 & c-8 \\
    \end{tabular}

    \pause{}

    \vspace{-3cm}
    \hfill\includegraphics[height=3cm, angle=45, origin=c]{img/dwarf_logo}
    \hspace{-1cm}
\end{frame}

\begin{frame}[t, fragile]{The real DWARF}
    \begin{lstlisting}[numbers=none, language=]
00009b30 48 009b34 FDE cie=0000 pc=0084950..0084b37
  DW_CFA_advance_loc: 2 to 0000000000084952
  DW_CFA_def_cfa_offset: 16
  DW_CFA_offset: r15 (r15) at cfa-16
  DW_CFA_advance_loc: 2 to 0000000000084954
  DW_CFA_def_cfa_offset: 24
  DW_CFA_offset: r14 (r14) at cfa-24
  DW_CFA_advance_loc: 2 to 0000000000084956
  DW_CFA_def_cfa_offset: 32
  DW_CFA_offset: r13 (r13) at cfa-32
  DW_CFA_advance_loc: 2 to 0000000000084958
  DW_CFA_def_cfa_offset: 40
  DW_CFA_offset: r12 (r12) at cfa-40
  DW_CFA_advance_loc: 1 to 0000000000084959
  [...]
    \end{lstlisting}

    \begin{itemize}
        \item[\textbf{$\longrightarrow$}] \textbf{\alert{constructed} on-demand
            by a \alert{Turing-complete bytecode}!}
    \end{itemize}

    \pause{}

    \vspace{-6.5cm}
    \begin{center}
    \bf \fontsize{8cm}{1cm}
        \colorbox{white}{\alert{Complex}} \\
        \colorbox{white}{\alert{\& slow!}}
    \end{center}
\end{frame}

\begin{frame}{Why does slow matter?}
    \begin{itemize}

        \item{} After all, we're talking about \alert{debugging procedures} ran
            by a \alert{human being} (slower than the machine).

            \ldots{}or are we?
    \end{itemize}

    \pause{}
    \begin{center}
        \textbf{\Large{}No!}
    \end{center}

    \begin{itemize}
        \pause{}\item{} Pretty much any \alert{program analysis tool}
        \pause{}\item{} \alert{Profiling} with polling profilers

        \pause{}\item{} \alert{Exception handling} in C++

    \end{itemize}

    \vspace{2em}

    \begin{center}
        \textbf{\Large{}Debug data is not only for debugging}
    \end{center}
\end{frame}

\begin{frame}{Difficult to generate}
    Major concern with DWARF\@: it is \alert{difficult to generate}
    (correctly).

    \begin{itemize}
        \item{} \alert{Hard to generate}: each compiler pass must keep it in
            sync
        \item{} Most of it is \alert{seldom used} (\eg{} unwinding data of
            dusty code), and thus \alert{seldom tested}
    \end{itemize}
    \vspace{1em}

    Yields to
    \begin{itemize}
        \item{} unreliable DWARF\@: can cause headaches when debugging
        \item{} or not generated at all (eg. OCaml until recently) \todo{Check
            this}
    \end{itemize}

    \vspace{1em}
    \begin{center}
        \Large\bf
        $\leadsto$ Complex, buggy, untested
    \end{center}
\end{frame}

\begin{frame}{A debugging hell: Linux kernel}
    ``Sorry, but last time was too f\dots painful. The whole (and
    only) point of unwinders is to make debugging easy
    when a bug occurs. But \alert{the dwarf unwinder had bugs}
    itself, or \alert{our dwarf information had bugs}, and in either
    case it actually turned several trivial bugs into a \alert{total
    undebuggable hell}.''

    \vspace{1em}
    \only<1>{\hfill ---~Linus Torvalds, Kernel mailing list, 2012}
    \pause{}

    ``If you can \alert{mathematically prove that the unwinder is
    correct} — even in the presence of bogus and actively
    incorrect unwinding information — and never ever
    follows a bad pointer, \alert{I’ll reconsider}.''

    \vspace{1em}
    \hfill ---~Linus Torvalds, Kernel mailing list, 2012

    \pause{}\vspace{1em}
    \begin{center}
        \Large\bf
        \alert{This is where we still are!}
    \end{center}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Unwinding data validation}

\begin{frame}{Main idea}
    \begin{itemize}
        \item If we follow \alert{one path of execution}, we can track the
            return address location
        \item If we read unwinding data \alert{at runtime}, we can \alert{check
            the RA consistency} at each step
    \end{itemize}
\end{frame}

\newcommand{\tblrowval}[4]{#1 & #2 & \only<2->{#3} & \only<2->{#4} \\}
\newcommand{\blknote}[1]
    {\begin{block}{}
            \centering\large
            #1
    \end{block}}
\newcommand{\blklnote}[1]
    {\begin{block}{}
            \large
            #1
    \end{block}}
\newcommand{\tblhl}{\rowcolor{Tan}}

\begin{frame}{Example}
    \begin{table}
        \ttfamily\large
        \begin{tabularx}{0.9\linewidth}{
                l
                b
                >{\columncolor{SkyBlue}}s
                >{\columncolor{SkyBlue}}s
            }
\tblrowval{\hspace{-2ex}<{\bf foo}>:}{}{\textbf{CFA}}{\textbf{ra}}
\rowonly<3>{\tblhl{}}  \tblrowval{push}{\%r15}{rsp+8}{c-8}
\rowonly<4>{\tblhl{}} \tblrowval{push}{\%r14}{rsp+16}{c-8}
\rowonly<5>{\tblhl{}} \tblrowval{mov}{\$0x3,\%eax}{rsp+24}{c-8}
\rowonly<6>{\tblhl{}} \tblrowval{push}{\%r13}{rsp+24}{c-8}
\tblrowval{push}{\%r12}{rsp+32}{c-8}
\tblrowval{push}{\%rbp}{rsp+40}{c-8}
\tblrowval{push}{\%rbx}{rsp+48}{c-8}
\tblrowval{sub}{\$0x68,\%rsp}{rsp+56}{c-8}
\rowonly<9>{\tblhl{}} \tblrowval{cmp}{\$0x1,\%edi}{rsp+160}{c-8}%
\only<-8>{%
\tblrowval{add}{\$0x68,\%rsp}{rsp+160}{c-8}
\tblrowval{pop}{\%rbx}{rsp+56}{c-8}
\tblrowval{pop}{\%rbp}{rsp+48}{c-8}}
        \end{tabularx}
    \end{table}

\only<-8> {
    \only<3>{\blknote{Upon function call, \alert{ra = *(\reg{rsp})} (ABI)}}
    \only<4>{\blknote{\texttt{push} decreases \reg{rsp} by 8: %
        \alert{ra = *(\reg{rsp} + 8)}}}
    \only<5>{\blknote{and again: %
        \alert{ra = *(\reg{rsp} + 16)}}}
    \only<6>{\blknote{This \texttt{mov} leaves \reg{rsp} untouched: %
        \alert{ra = *(\reg{rsp} + 16)}}}
    \only<7>{\blknote{The unwinding table can actually be seen as\\
            an \alert{abstract interpretation} of the code\ldots}}
    \only<8>{\blknote{\ldots and thus, for a given run, be
        \alert{re-computed for verification}}}
    }
\only<9-> {
    \vspace{-2em}
    \only<9>{\blklnote{If, within an execution,
            \begin{itemize}
                \item ra = \texttt{*(0xFFFF1098)}
                \item \reg{rsp} = \texttt{0xFFFF1000}
            \end{itemize}
            We can \alert{evaluate both expressions} and \alert{compare}
    }}
}
\end{frame}

\begin{frame}{Dynamic validation}
    \textbf{Abstract state}
    \begin{itemize}
        \item \alert{Stack} of actual \alert{addresses} where
            \alert{return addresses} are stored
    \end{itemize}

    \vspace{2em}\pause{}

    \textbf{Abstract instruction semantics}
    \begin{itemize}
        \item[\alert{\texttt{call}}] push \alert{\reg{rsp}} on the stack
        \item[\alert{\texttt{ret}}] pop from the stack
    \end{itemize}

    \vspace{2em}\pause{}

    \textbf{Validation of each instruction}
    \begin{itemize}
        \item Evaluate the return address provided by DWARF
        \item Compare it with the value at the top of the stack
    \end{itemize}
\end{frame}

\begin{frame}{In practice: \texttt{eh\_frame\_check}}
    Strategy implemented and working: \alert{\texttt{eh\_frame\_check}}

    \begin{itemize}
        \item \texttt{gdb} allows for Python instrumentation
            \vspace{1em}\pause{}
        \item Parse ELF and DWARF data (\texttt{pyelftools})
        \item Run the binary inside gdb
        \item Pause at each (assembly) step
        \item Jointly evaluate DWARF data and the abstract stack
        \item Report upon error
    \end{itemize}

    \vspace{1em}
    Works, but… \alert{Python is slow}!

    A few thousand of ASM instructions/second (good enough)
\end{frame}

\begin{frame}{A real bug!}
    \begin{columns}[c]
        \begin{column}{0.65\textwidth}
            \lstinputlisting[language=C]{src/llvm_bug.c}
        \end{column} \begin{column}{0.35\textwidth}
            \textbf{\texttt{CSmith}\\+ \texttt{Creduce}\\+
            \texttt{eh\_frame\_check}}

            \vspace{2em}$\leadsto$ \alert{\bf LLVM (3.8) bug!}
        \end{column}
    \end{columns}
\end{frame}

\renewcommand{\tblrowval}[5]{#1 & #2 & #3 & #4 & #5 \\}
\begin{frame}{A real bug!}

    \begin{columns}[c]
        \column{0.7\textwidth}
        \begin{align*}
            \onslide<2->{\textbf{Abstract state} \qquad &
                \left[\texttt{0xFFFF1000}\right]} \\
            \onslide<3->{\reg{rsp} \qquad & %
                ~\,\texttt{%
                    \only<3-4>{0xFFFF1000}%
                    \only<5-8>{0xFFFF0FF8}%
                    \only<9->{0xFFFF1000}%
                }
            }
        \end{align*}

        \column{0.3\textwidth}
        {\vspace{-4mm}\bf \fontsize{2cm}{5.5cm}\selectfont %
            \only<4>{\cmark}%
            \only<6>{\cmark}%
            \only<8>{\cmark}%
            \only<10->{\xmark}%
        }
    \end{columns}


    \vspace{1em}

    \begin{table}
        \ttfamily\large
        \begin{tabularx}{0.95\linewidth}{
                l
                l
                b
                >{\columncolor{SkyBlue}}s
                >{\columncolor{SkyBlue}}s
            }
\tblrowval{\hspace{-2ex}<{\bf foo}>:}{}{}{\textbf{CFA}}{\textbf{ra}}
%\rowonly<3>{\tblhl{}} \tblrowval{4004e0}{push}{\%rbx}{rsp+8}{c-8}
\rowonly<2-4>{\tblhl{}} \tblrowval{4004e0}{push}{\%rbx}{rsp+8}{c-8}
\rowonly<5-6>{\tblhl{}} \tblrowval{}{}{}{rsp+16}{c-8}
\tblrowval{}{[\ldots]}{}{}{}
\tblrowval{}{}{}{}{}
\rowonly<7-8>{\tblhl{}} \tblrowval{40061d}{pop}{\%rbx}{rsp+16}{c-8}
\rowonly<9->{\tblhl{}} \tblrowval{40061e}{retq}{}{rsp+16}{c-8}
        \end{tabularx}
    \end{table}

    \begin{center}
        \bf\Large %
        \onslide<11>{$\leadsto$ LLVM bug \#13161}
    \end{center}

\end{frame}

\begin{frame}{TODO}
    \todo{}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Unwinding data synthesis from assembly}

\begin{frame}{TODO}
    \todo{}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section{Unwinding data compilation}

\subsection{Compilation ahead-of-time}

\begin{frame}{Compilation overview}
    \begin{itemize}
        \item Compiled to \alert{C code}
        \item C code then \alert{compiled to native binary} (gcc)
            \begin{itemize}
                \item[$\leadsto$] gcc optimisations for free
            \end{itemize}
        \item Compiled as \alert{separate \texttt{.so} files}, called \ehelfs{}
            \bigskip{}
        \item Morally a \alert{monolithic switch} on IPs
        \item Each case contains assembly that computes a \alert{row of the
            table}
    \end{itemize}
\end{frame}


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{frame}{Compilation example: original C, DWARF}
    \lstinputlisting[language=C]{src/fib7/fib7.cfde}
\end{frame}

\begin{frame}[shrink]{Compilation example: generated C}
    \lstinputlisting[language=C]{src/fib7/fib7.eh_elf_basic.c}
\end{frame}


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{frame}{Compilation choices}
    \textbf{In order to keep the compiler \alert{simple} and \alert{easily
    testable}, the whole DWARF5 instruction set is not supported.}

    \begin{itemize}
        \item Focus on \alert{x86\_64}
        \item Focus on unwinding return address \\
            \vspace{0.3ex}
            $\leadsto$ \textit{Allows building a backtrace}
            \begin{itemize}
                \item \alert{suitable for perf, not for gdb}
                \item Only supports \alert{unwinding registers}: \reg{rip}, \reg{rsp},
                    \reg{rbp}, \reg{rbx}
                \item Supports the \alert{wide majority} ($> 99.9\%$) of instructions
                    used
                \item Among \alert{4000} randomly sampled filed, only \alert{24}
                    containing unsupported instructions
            \end{itemize}
    \end{itemize}
\end{frame}

\begin{frame}{Interface: libunwind}
    \begin{itemize}
        \item \alert{libunwind}: \textit{de facto} standard library for
            unwinding
        \item Relies on DWARF

            \bigskip{}

        \item \texttt{libunwind-eh\_elf}: alternative implementation using
            \ehelfs{}

        \item[$\leadsto$] \alert{alternative implementation} of libunwind,
            almost plug-and-play for existing projects!
            \begin{itemize}
                \item[$\leadsto$] It is \alert{easy} to use \ehelfs{}: just
                    link against the right library!
            \end{itemize}
    \end{itemize}
\end{frame}


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

\begin{frame}{Size optimisation: outlining}
    \begin{itemize}
        \item This \alert{works}, but \alert{takes space}: about \alert{7 times
            larger in size} than regular DWARF\@.

        \item DWARF optimisation strategy: \alert{alter previous row}. \\
            Causes slowness: we cannot do that.

        \item Remark: a lot of lines appear often.
            \begin{itemize}
                \item[$\leadsto$] \textbf{\emph{outline} them!}
            \end{itemize}

    \pause{}

        \item On libc, $20\,827$ rows $\rightarrow$ $302$ outlined ($1.5\,\%$)
        \item Turn the big switch into a binary search \alert{if/else tree}
    \end{itemize}

    \pause{}

    \bigskip{}
    \begin{center}
        $\leadsto$ only \textbf{2.5 times bigger than DWARF}
    \end{center}
\end{frame}

\begin{frame}{Example with outlining}
    \lstinputlisting[language=C]{src/fib7/fib7.eh_elf_outline.c}
\end{frame}


\begin{frame}[t]{A word on formalization}
    \begin{itemize}
        \item First task: \alert{writing semantics} for DWARF, written  as
            mapping to C code.
        \item DWARF5 specification: \alert{plain English}, no proper semantics
        \item Compiled code is in substance equivalent to semantics
        \item What remains to prove is mostly \alert{simple or classic
            optimisations}
    \end{itemize}

    \pause{}
    \vspace{-3cm}
    \begin{center}
        \includegraphics[width=0.8\linewidth, angle=10]{img/dw_spec.png}
    \end{center}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\subsection{Benchmarking}

\begin{frame}{Benchmarking requirements}
    \begin{enumerate}
        \item Thousands of samples (single unwind: $10\,\mu{}s$)
        \item Interesting enough program to unwind: nested functions, complex
            FDEs
        \item Mitigate caching: don't always unwind from the \emph{same} point
        \item Yet be fair: don't always unwind from totally different places
        \item Distribute evenly: if possible, also from within libraries
    \end{enumerate}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{frame}{perf instrumentation}
    \textbf{\alert{perf} is the state-of-the-art polling profiler for Linux.}
    \begin{itemize}
        \item{} used to get readings of the time spent in each function
        \item{} works by regularly stopping the program, unwinding its stack,
            then aggregating the gathered data
    \end{itemize}

    \pause{}\bigskip{}
    \textbf{Instrumenting perf matches all the requirements!}

    \begin{itemize}
        \item{} \alert{Plug \ehelfs{} into perf}: use \ehelfs{} instead of
            DWARF to unwind the stack
        \item{} Implement \alert{unwinding performance counters} inside perf
            \bigskip{}

        \item{} Use perf on \alert{hackbench}, a kernel stress-test program
            \begin{itemize}
                \item Small program
                \item Lots of calls
                \item Relies on libc, libpthread
            \end{itemize}
    \end{itemize}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\subsection{Results}

\begin{frame}{Time performance}
    \small
    \centering
    \begin{tabular}{l r r r r r}
        \toprule
        \thead{Unwinding method} & \thead{Frames \\ unwound}
            & \thead{Tot.\ time \\ ($\mu s$)}
            & \thead{Avg. \\ time / frame \\ ($ns$)}
            & \thead{Time \\ ratio} \\
        \midrule
        \alert{\ehelfs{}}
            & 23506 % Frames unwound
            & 14837 % Total time
            & 631 % Avg time
            & 1
            \\
        \prog{libunwind}, \alert{cached}
            & 27058 % Frames unwound
            & 441601 % Total time
            & 16320 % Avg time
            & \alert{25.9}
            \\
        \prog{libunwind}, \alert{uncached}
            & 27058 % Frames unwound
            & 671292 % Total time
            & 24809 % Avg time
            & \alert{39.3}
            \\
        \bottomrule
    \end{tabular}
\end{frame}

\begin{frame}{Space performance}
    \begin{center}
        \begin{tabular}{r r r r r r}
            \toprule
            \thead{Object}
                & \thead{\% of binary size}
                & \thead{Growth factor} \\
            \midrule
                libc
                    & 21.88 & 2.41 \\
                libpthread
                    & 43.71 & 2.19 \\
                ld
                    & 22.09 & 2.97 \\
                hackbench
                    & 93.87 & 4.99 \\
                    \midrule
                Total
                    & 22.81 & \alert{2.44} \\
            \bottomrule
        \end{tabular}
    \end{center}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\section*{}
\setcounter{section}{0}

\begin{frame}{What next?}
    \begin{itemize}
        \item Implement a release-ready, packageable, easy to use version of
            perf with \ehelfs{} and submit it for inclusion

        \item{} Measure \alert{C++ exceptions overhead} precisely in common
            software

        \item{} Implement \alert{\ehelfs{}} support for \alert{C++ runtime}
            exception handling, and other systems where unwinding is a
            performance bottleneck

        \medskip

        \item \alert{Outlining} was effective for
            compactness\ldots{} Try outlining DWARF bytecode\@?

    \end{itemize}
\end{frame}

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{frame}
    \vspace{5mm}
    \includegraphics[width=\linewidth]{img/keep_breathing}
    \vspace{-1cm}

    \begin{center}
        \large
        \begin{align*}
            \textbf{Slides: } &\text{\url{https://tobast.fr/m2/slides.pdf}} \\
            \textbf{Report: } &\text{\url{https://tobast.fr/m2/report.pdf}}
        \end{align*}
    \end{center}

\end{frame}

\end{document}