Merge branch 'master' of tildegit.org:lucic71/dissertation into HEAD

At some point I should delete other unused flags too.

TSW/Makefile

@@ -0,0 +1,13 @@
+PROJECT=main
+
+build:
+	pdflatex ${PROJECT}.tex
+	bibtex ${PROJECT}
+	pdflatex ${PROJECT}.tex
+	pdflatex ${PROJECT}.tex
+
+show:
+	evince ${PROJECT}.pdf
+
+clean:
+	rm -f *.aux *.blg *.out *.bbl *.log *.toc *.pdf

TSW/abstract.tex

@@ -0,0 +1,15 @@
+\begin{abstract} 
+\noindent
+The ISO C Standard added the undefined behavior notion as a mean to
+portability. State-of-the-art compilers such as gcc and LLVM use it to
+issue aggressive optimizations that break the semantics of the code.
+\todo{or the intention of the progammer} We argue that the performance
+impact of undefined behavior based optimizations in operating systems,
+such as OpenBSD, is low. Furthermore they introduce more unobservable
+and undocumented effects than performance advantages. To test our
+hypothesis we aim to disable the above mentioned \todo{should i write UB
+based optimiatoins instead?} optimizations and test the generated system
+against real-life testbenches \todo{too generic}.
+\end{abstract}
+
+\noindent\keywords{compiler, optimization, operating system}\\

TSW/bib.bib

@@ -0,0 +1,53 @@
+@inproceedings{wang2012undefined,
+  title={Undefined behavior: what happened to my code?},
+  author={Wang, Xi and Chen, Haogang and Cheung, Alvin and Jia, Zhihao
+and Zeldovich, Nickolai and Kaashoek, M Frans},
+  booktitle={Proceedings of the Asia-Pacific Workshop on Systems},
+  pages={1--7},
+  year={2012}
+}
+
+@misc{checks_2008,
+  title={CERT/CC Vulnerability note vu162289},
+  url={https://www.kb.cert.org/vuls/id/162289/},
+  journal={VU162289 - C compilers may silently discard some wraparound
+checks},
+  year={2008},
+  month={Apr}
+} 
+
+@inproceedings{wang2013towards,
+  title={Towards optimization-safe systems: Analyzing the impact of
+undefined behavior},
+  author={Wang, Xi and Zeldovich, Nickolai and Kaashoek, M Frans and
+Solar-Lezama, Armando},
+  booktitle={Proceedings of the Twenty-Fourth ACM Symposium on Operating
+Systems Principles},
+  pages={260--275},
+  year={2013}
+}
+
+@article{lomont2003fast,
+  title={Fast inverse square root},
+  author={Lomont, Chris},
+  journal={Tech-315 nical Report},
+  volume={32},
+  year={2003}
+}
+
+@misc{google_2015, title={BORINGCC},
+url={https://groups.google.com/g/boring-crypto/c/48qa1kWignU/m/o8GGp2K1DAAJ},
+journal={Google}, publisher={Google}, year={2015}, month={Dec}} 
+
+@inproceedings{ertl2015every,
+  title={What every compiler writer should know about programmers or
+Optimization based on undefined behaviour hurts performance},
+  author={Ertl, M Anton},
+  booktitle={Kolloquium Programmiersprachen und Grundlagen der
+Programmierung (KPS 2015)},
+  year={2015}
+}
+
+@misc{regehr_2014, url={https://blog.regehr.org/archives/1180},
+journal={Embedded in Academia}, author={Regehr, John}, year={2014},
+month={Aug}} 

TSW/bibliography.bib

@@ -0,0 +1,215 @@
+@inproceedings{tanaka,
+  title={Approaches to making software porting more productive},
+  author={Tanaka, Toshikiyo and Hakuta, M and Iwata, N and Ohminami, M},
+  booktitle={Proceedings of the 12th TRON Project international Symposium},
+  pages={73--85},
+  year={1995},
+  organization={IEEE}
+}
+
+@article{hakuta,
+  title={A study of software portability evaluation},
+  author={Hakuta, Mitsuari and Ohminami, Masato},
+  journal={Journal of Systems and Software},
+  volume={38},
+  number={2},
+  pages={145--154},
+  year={1997},
+  publisher={Elsevier}
+}
+
+@inproceedings{kanai,
+  title={A cost model for software conversion based on program characteristics and a converter effect},
+  author={Kanai, A and Furuyama, T and Takahashi, M},
+  booktitle={1992 Proceedings. The Sixteenth Annual International Computer Software and Applications Conference},
+  pages={63--64},
+  year={1992},
+  organization={IEEE Computer Society}
+}
+
+@incollection{mooney2004developing,
+  title={Developing portable software},
+  author={Mooney, James D},
+  booktitle={Information Technology},
+  pages={55--84},
+  year={2004},
+  publisher={Springer}
+}
+
+@article{capretz,
+  title={Bringing the human factor to software engineering},
+  author={Capretz, Luiz Fernando},
+  journal={IEEE software},
+  volume={31},
+  number={2},
+  pages={104--104},
+  year={2014},
+  publisher={IEEE}
+}
+
+@article{ejiogu,
+  title={A simple measure of software complexity},
+  author={Ejiogu, Lem O},
+  journal={ACM SIGPLAN Notices},
+  volume={20},
+  number={3},
+  pages={16--31},
+  year={1985},
+  publisher={ACM New York, NY, USA}
+}
+
+@article{mooney1990strategies,
+  title={Strategies for supporting application portability},
+  author={Mooney, James D.},
+  journal={Computer},
+  volume={23},
+  number={11},
+  pages={59--70},
+  year={1990},
+  publisher={IEEE}
+}
+
+@inproceedings{cho2011case,
+  title={Case Study on Installing a Porting Process for Embedded Operating System in a Small Team},
+  author={Cho, DongSeok and Bae, DooHwan},
+  booktitle={2011 Fifth International Conference on Secure Software Integration and Reliability Improvement-Companion},
+  pages={19--25},
+  year={2011},
+  organization={IEEE}
+}
+
+@misc{porquet2015,
+  title={Porting {Linux} to a new processor architecture, part 1: The basics},
+  howpublished="\url{https://lwn.net/Articles/654783/}",
+  journal={[LWN.net]},
+  author={Porquet, Joël},
+  year={2015},
+  month={Aug}
+} 
+
+@article{bodenstab1984unix,
+  title={The {UNIX} system: {UNIX} operating system porting experiences},
+  author={Bodenstab, DE and Houghton, Thomas F and Kelleman, Keith A and Ronkin, George and Schan, Edward P},
+  journal={AT\&T Bell Laboratories Technical Journal},
+  volume={63},
+  number={8},
+  pages={1769--1790},
+  year={1984},
+  publisher={Nokia Bell Labs}
+}
+
+@misc{osdevcrossport,
+  title={Cross-Porting Software},
+  howpublished="\url{https://wiki.osdev.org/Cross-Porting\_Software}",
+  journal={[osdev.org]},
+  year={2019},
+  month={Sept}
+} 
+
+@article{jolitz1990porting,
+  title={Porting {UNIX} to the 386: A practical approach},
+  author={Jolitz, William Frederick and Jolitz, Lynne Greer},
+  journal={Dr. Dobb's Journal},
+  volume={16},
+  number={1},
+  pages={16--46},
+  year={1990},
+  publisher={CMP Media, Inc.}
+}
+
+@article{frakes1995sixteen,
+  title={Sixteen questions about software reuse},
+  author={Frakes, William B and Fox, Christopher J},
+  journal={Communications of the ACM},
+  volume={38},
+  number={6},
+  pages={75--ff},
+  year={1995},
+  publisher={ACM New York, NY, USA}
+}
+
+@article{tanenbaum1978guidelines,
+  title={Guidelines for software portability},
+  author={Tanenbaum, Andrew S and Klint, Paul and Bohm, Wim},
+  journal={Software: Practice and Experience},
+  volume={8},
+  number={6},
+  pages={681--698},
+  year={1978},
+  publisher={Wiley Online Library}
+}
+
+@article{johnson1978unix,
+  title={{UNIX} time-sharing system: Portability of C programs and the {UNIX} system},
+  author={Johnson, Steven C and Ritchie, Dennis M},
+  journal={The Bell System Technical Journal},
+  volume={57},
+  number={6},
+  pages={2021--2048},
+  year={1978},
+  publisher={Nokia Bell Labs}
+}
+
+@article{xia2017measuring,
+  title={Measuring program comprehension: A large-scale field study with professionals},
+  author={Xia, Xin and Bao, Lingfeng and Lo, David and Xing, Zhenchang and Hassan, Ahmed E and Li, Shanping},
+  journal={IEEE Transactions on Software Engineering},
+  volume={44},
+  number={10},
+  pages={951--976},
+  year={2017},
+  publisher={IEEE}
+}
+
+@article{morgan1994controlling,
+  title={Controlling software development costs},
+  author={Morgan, Malcolm J},
+  journal={Industrial Management \& Data Systems},
+  year={1994},
+  publisher={MCB UP Ltd}
+}
+
+@article{boehm2000software,
+  title={Software development cost estimation approaches—A survey},
+  author={Boehm, Barry and Abts, Chris and Chulani, Sunita},
+  journal={Annals of software engineering},
+  volume={10},
+  number={1},
+  pages={177--205},
+  year={2000},
+  publisher={Springer}
+}
+
+@article{walli1995posix,
+  title={The POSIX family of standards},
+  author={Walli, Stephen R},
+  journal={StandardView},
+  volume={3},
+  number={1},
+  pages={11--17},
+  year={1995},
+  publisher={ACM New York, NY, USA}
+}
+
+@inproceedings{spencer1992ifdef,
+  title={\# ifdef considered harmful, or portability experience with C News},
+  author={Spencer, Henry and Collyer, Geoff},
+  booktitle={USENIX Summer 1992 Technical Conference (USENIX Summer 1992 Technical Conference)},
+  year={1992}
+}
+
+@misc{callahanopenbsd,
+  title={I ported the new Hare compiler to {OpenBSD}},
+  howpublished="\url={https://briancallahan.net/blog/20220427.html}",
+  journal={[Brian Robert Callahan]},
+  author={Callahan, B. R.},
+  year={2022},
+  month={April}
+} 
+
+@article{lyonunixportability,
+author = {Lyon, Tom},
+year = {1977},
+month = {08},
+title = {Inter-{UNIX} Portability}
+}

TSW/intro.tex

@@ -0,0 +1,64 @@
+\section{Context and motivation}
+
+The C89 standard provides a loose definition of undefined
+behavior. This permits compiler implementations to
+abuse the definition and use it as a mean to agressive optimizations
+that break the semantics of the code. Code that works with a lower level
+of optimization is broken when the optimization level is elavated.
+Furthermore code that works on previous versions of the compiler is
+suddenly broken in newer versions because the standard imposes no
+requirements on undefined behavior.
+
+This has created serious security problems throughout the
+years~\cite{wang2012undefined,checks_2008}. A number of initiatives to
+solve this problem were started from different
+parties~\cite{google_2015,regehr_2014,wang2013towards} however the
+problem still persists. The primary open source developer groups have
+seized the unsteady definition of undefined behavior to justify
+dangerous slient code transformations that break the intention of the
+programmer.
+
+This philosophy is very dangerous in terms of programming expresivity
+and intentionality. The compiler has very little context of what the
+developer wants to accomplish with a specific piece of code. For example
+the compiler cannot make the distinction between a read from an
+uninitialized memory region that might produce unspecified results and a
+read from a memory mapped device that cannot be written in order to
+initialize it. Or it cannot distinguish between an erroneous floating
+pointer access to an integer variable and a smart method of computing an
+arithmetic function~\cite{lomont2003fast}. The general principle is that
+the developer has the responsability to decide what the code should to,
+the job of the compiler is to translate the code into machine readable
+instructions and to apply optimizations only when there is no risk of
+losing developer intentionality.
+
+The argument of the people that defend this kind of optimizations is
+that C code that contains undefined behavior has no meaning and the
+compiler is free to do various types of modifications on it. In Control
+Theory terms, such a system is described by a low degree of
+controllability and observability. Which is paradoxical in the
+philosophy described above where the compiler forcefully takes the
+responsability, from the developer, of generating relevant code. The
+implications of this are that no meaningful engineering can be done in
+this framework where processes inside a compiler cannot be understood
+and analyzed.
+
+Another argument that defends the aggresive optimizations view is that
+code generated by these compilers runs faster on artificial benchmarks.
+\todo{do some more research here} This does not necessarly hold for real
+life benchmarks that differ in complexity from the artificial ones and
+that make use of non-trivial code constructs.  Ertl~\cite{ertl2015every}
+makes an interesting observation regarding the performance of undefined
+behavior based optimizations. He notes that source level changes buy
+greater speedup factors than UB based optimizations for certain classes
+of programs. While his research in this field is valuable, the
+limitation of his work is that he draws conclusions based on SPECint
+benchmarks.
+
+The contribution of this work is that we try to analyze the speedup
+factors for real life programs, such as operating systems, in particular
+OpenBSD. \todo{motivate the choice of openbsd} By doing this we want to
+provide a tradeoff analysis between the performance gained using UB
+based optimizations and the risks of issuing them.
+
+This paper is structured as follows. \todo{later}

TSW/johd.sty

@@ -0,0 +1,36 @@
+\setlength{\paperwidth}{21cm}   % A4
+\setlength{\paperheight}{29.7cm}% A4
+\setlength\topmargin{-0.5cm}    
+\setlength\oddsidemargin{0cm}   
+\setlength\textheight{24.7cm} 
+\setlength\textwidth{16.0cm}
+\setlength\columnsep{0.6cm}  
+\newlength\titlebox 
+\setlength\titlebox{5cm}
+\setlength\headheight{5pt}   
+\setlength\headsep{0pt}
+\pagestyle{plain}
+\usepackage{color}
+\usepackage[natbibapa]{apacite}
+\usepackage{xurl}
+\usepackage[colorlinks,citecolor=blue,urlcolor=blue, linkcolor=blue, bookmarks=false,hypertexnames=true]{hyperref}
+\usepackage{url}
+%\usepackage{libertine}
+\usepackage{float}
+\usepackage{graphicx}
+\usepackage{doi} % hyperlink URLs
+\renewcommand{\doi}{DOI:~}
+
+\newcommand\outauthor{
+	\begin{tabular}[t]{c}
+		\bf\@author
+	\end{tabular}}
+	
+%Add keyword command
+\providecommand{\keywords}[1]
+{\small\textbf{Keywords:} #1
+}
+
+\providecommand{\authorroles}[1]
+{\small\textbf{Author roles:} #1
+}

TSW/main.tex

@@ -0,0 +1,64 @@
+\documentclass{article}
+\usepackage[english]{babel}
+\usepackage[utf8]{inputenc}
+\usepackage{johd}
+\usepackage{color}
+
+\newcommand{\todo}[1]{}
+\renewcommand{\todo}[1]{{\color{red} TODO: {#1}}}
+
+\title{Semantic Preservation of Undefined Behavior Based Optimizations}
+
+\author{Popescu Lucian-Ioan \\
+        \small CS Department, Politehnica Univesity of Bucharest,
+Romania \\
+        \small lucian.popescu187@gmail.com
+}
+
+\date{}
+
+\begin{document}
+
+\maketitle
+
+\input{abstract}
+\input{intro}
+
+\section{Dataset description}
+Here you can provide, if applicable, information about the dataset(s) whose creation, collection, management, access, processing or analysis have been discussed in this paper, following this schema:
+\paragraph{Object name} Typically the name of the file or file set in the repository.
+\paragraph{Format names and versions} E.g., ASCII, CSV, Autocad, EPS, JPEG, Excel, SQL, etc.
+\paragraph{Creation dates} The start and end dates of when the data was created (YYYY-MM-DD).
+\paragraph{Dataset creators} Please list anyone who helped to create the dataset (who may or may not be an author of the data paper), including their roles (using and affiliations).
+\paragraph{Language} Languages used in the dataset (i.e., for variable names etc.).
+\paragraph{License} The open license under which the data has been deposited (e.g., CC0). 
+\paragraph{Repository name} The name of the repository to which the data is uploaded. E.g., Figshare, Dataverse, etc. 
+\paragraph{Publication date} If already known, the date in which the dataset was published in the repository (YYYY-MM-DD).
+
+\section{Method}
+Describe the methods used in the study.
+
+\section{Results and discussion}
+Describe and discuss the results of the study.
+
+\section{Implications/Applications}
+Provide information about the implications of this research and/or how it can be applied.
+
+\section*{Acknowledgements}
+Please add any relevant acknowledgements to anyone else that assisted with the project in which the data was created but did not work directly on the data itself.
+
+\section*{Funding Statement}
+If the research resulted from funded research please list the funder and grant number here.
+
+\section*{Competing interests} 
+If any of the authors have any competing interests then these must be declared. If there are no competing interests to declare then the following statement should be present: The author(s) has/have no competing interests to declare.
+
+
+\bibliographystyle{johd}
+\bibliography{bib}
+
+\section*{Supplementary Files (optional)}
+Any supplementary/additional files that should link to the main publication must be listed, with a corresponding number, title and option description. Ideally the supplementary files are also cited in the main text.
+Note: supplementary files will not be typeset so they must be provided in their final form. They will be assigned a DOI and linked to from the publication.
+
+\end{document}

TSW/relwork.tex

@@ -0,0 +1 @@
+http://www.complang.tuwien.ac.at/kps2015/proceedings/KPS_2015_submission_29.pdf

Tech/sys/arch/amd64/amd64/locore0.md

@@ -0,0 +1,13 @@
+# locore0.S
+
+I should have expected problems with the compiler, but not so early. The
+first problem that I encountered here is that for 64 bit mode intel
+there were no .code32 and .code16 opcodes. I don't know why they didn't
+add this support. Maybe because no-one tried to use the compiler for
+such a task.
+
+However, I patched the compiler and the .S file is ok with the opcode.
+We will see on the generated output if the machine code is correct.
+Until then I still have to patch the compiler because there is no popfl
+opcode. I don't even know what it does, but the problem needs to be
+solved.

Tech/tcc/gnu.md

@@ -0,0 +1,15 @@
+# GNU
+
+TCC tries to keep a lot of compatibility with GCC for whatever reason. I
+wonder what is the reason for that? I don't think that the author was
+dumb enough to think that if the project kisses the ass of GCC it will
+stay popular and relevant.
+
+Just grep GCC in /tinycc and you will see how much dependency has the
+project on GNU. Even the problem I solved earlier with .code32 in 64 bit
+mode had to do with GAS, the GNU assembler. The makefile is written in
+GNU make dialect, in /tinycc/configure gcc is the default compiler, etc,
+etc.
+
+I'm starting to think that maybe tcc is not the best solution for what
+I'm trying to do.

sys/arch/amd64/conf/Makefile.amd64

@@ -77,7 +77,7 @@ CWARNFLAGS+=	-Wno-address-of-packed-member -Wno-constant-conversion \
 DEBUG?=		-g
 COPTIMIZE?=	-O2
 CFLAGS=		${DEBUG} ${CWARNFLAGS} ${CMACHFLAGS} ${COPTIMIZE} ${COPTS} ${PIPE}
-AFLAGS=		-D_LOCORE -x assembler-with-cpp ${CWARNFLAGS} ${CMACHFLAGS}
+AFLAGS=		-D_LOCORE ${CWARNFLAGS} ${CMACHFLAGS}
 LINKFLAGS=	-T ld.script -X --warn-common -nopie
 
 HOSTCC?=	${CC}