source: Deliverables/D1.2/CompilerProofOutline/outline.tex @ 1763

Last change on this file since 1763 was 1763, checked in by mulligan, 8 years ago

to avoid conflicts

File size: 52.4 KB
Line 
1\documentclass[a4paper, 10pt]{article}
2
3\usepackage{a4wide}
4\usepackage{amsfonts}
5\usepackage{amsmath}
6\usepackage{amssymb}
7\usepackage[english]{babel}
8\usepackage{color}
9\usepackage{diagrams}
10\usepackage{graphicx}
11\usepackage[colorlinks]{hyperref}
12\usepackage[utf8x]{inputenc}
13\usepackage{listings}
14\usepackage{microtype}
15\usepackage{skull}
16\usepackage{stmaryrd}
17\usepackage{array}
18\newcolumntype{b}{@{}>{{}}}
19\newcolumntype{B}{@{}>{{}}c<{{}}@{}}
20\newcolumntype{h}[1]{@{\hspace{#1}}}
21\newcolumntype{L}{>{$}l<{$}}
22\newcolumntype{C}{>{$}c<{$}}
23\newcolumntype{R}{>{$}r<{$}}
24\newcolumntype{S}{>{$(}r<{)$}}
25\newcolumntype{n}{@{}}
26\usepackage{wasysym}
27
28
29\lstdefinelanguage{matita-ocaml} {
30  mathescape=true
31}
32\lstset{
33  language=matita-ocaml,basicstyle=\tt,columns=flexible,breaklines=false,
34  showspaces=false, showstringspaces=false, extendedchars=false,
35  inputencoding=utf8x, tabsize=2
36}
37
38\title{Proof outline for the correctness of the CerCo compiler}
39\date{\today}
40\author{The CerCo team}
41
42\begin{document}
43
44\maketitle
45
46\section{Introduction}
47\label{sect.introduction}
48
49In the last project review of the CerCo project, the project reviewers
50recommended us to quickly outline a paper-and-pencil correctness proof
51for each of the stages of the CerCo compiler in order to allow for an
52estimation of the complexity and time required to complete the formalization
53of the proof. This has been possible starting from month 18 when we have
54completed the formalization in Matita of the datastructures and code of
55the compiler.
56
57In this document we provide a very high-level, pen-and-paper
58sketch of what we view as the best path to completing the correctness proof
59for the compiler. In particular, for every translation between two intermediate languages, in both the front- and back-ends, we identify the key translation steps, and identify some invariants that we view as being important for the correctness proof.  We sketch the overall correctness results, and also briefly describe the parts of the proof that have already
60been completed at the end of the First Period.
61
62In the last section we finally present an estimation of the effort required
63for the certification in Matita of the compiler and we draw conclusions.
64
65\section{Front-end: Clight to RTLabs}
66
67The front-end of the CerCo compiler consists of several stages:
68
69\begin{center}
70\begin{minipage}{.8\linewidth}
71\begin{tabbing}
72\quad \= $\downarrow$ \quad \= \kill
73\textsf{Clight}\\
74\> $\downarrow$ \> cast removal\\
75\> $\downarrow$ \> add runtime functions\footnote{Following the last project
76meeting we intend to move this transformation to the back-end}\\
77\> $\downarrow$ \> cost labelling\\
78\> $\downarrow$ \> loop optimizations\footnote{\label{lab:opt2}To be ported from the untrusted compiler and certified only in case of early completion of the certification of the other passes.} (an endo-transformation)\\
79\> $\downarrow$ \> partial redundancy elimination$^{\mbox{\scriptsize \ref{lab:opt2}}}$ (an endo-transformation)\\
80\> $\downarrow$ \> stack variable allocation and control structure
81 simplification\\
82\textsf{Cminor}\\
83\> $\downarrow$ \> generate global variable initialisation code\\
84\> $\downarrow$ \> transform to RTL graph\\
85\textsf{RTLabs}\\
86\> $\downarrow$ \> \\
87\>\,\vdots
88\end{tabbing}
89\end{minipage}
90\end{center}
91
92Here, by `endo-transformation', we mean a mapping from language back to itself:
93the loop optimization step maps the Clight language to itself.
94
95%Our overall statements of correctness with respect to costs will
96%require a correctly labelled program
97There are three layers in most of the proofs proposed:
98\begin{enumerate}
99\item invariants closely tied to the syntax and transformations using
100  dependent types (such as the presence of variable names in environments),
101\item a forward simulation proof relating each small-step of the
102  source to zero or more steps of the target, and
103\item proofs about syntactic properties of the cost labelling.
104\end{enumerate}
105The first will support both functional correctness and allow us to
106show the totality of some of the compiler stages (that is, those
107stages of the compiler cannot fail).  The second provides the main
108functional correctness result, including the preservation of cost
109labels in the traces, and the last will be crucial for applying
110correctness results about the costings from the back-end by showing
111that they appear in enough places so that we can assign all of the
112execution costs to them.
113
114We will also prove that a suitably labelled RTLabs trace can be turned
115into a \emph{structured trace} which splits the execution trace into
116cost-label to cost-label chunks with nested function calls.  This
117structure was identified during work on the correctness of the
118back-end cost analysis as retaining important information about the
119structure of the execution that is difficult to reconstruct later in
120the compiler.
121
122\subsection{Clight cast removal}
123
124This transformation removes some casts inserted by the parser to make
125arithmetic promotion explicit but which are superfluous (such as
126\lstinline[language=C]'c = (short)((int)a + (int)b);' where
127\lstinline'a' and \lstinline'b' are \lstinline[language=C]'short').
128This is necessary for producing good code for our target architecture.
129
130It only affects Clight expressions, recursively detecting casts that
131can be safely eliminated.  The semantics provides a big-step
132definition for expression, so we should be able to show a lock-step
133forward simulation between otherwise identical states using a lemma
134showing that cast elimination does not change the evaluation of
135expressions.  This lemma will follow from a structural induction on
136the source expression.  We have already proved a few of the underlying
137arithmetic results necessary to validate the approach.
138
139\subsection{Clight cost labelling}
140
141This adds cost labels before and after selected statements and
142expressions, and the execution traces ought to be equivalent modulo
143the new cost labels.  Hence it requires a simple forward simulation
144with a limited amount of stuttering whereever a new cost label is
145introduced.  A bound can be given for the amount of stuttering allowed
146based on the statement or continuation to be evaluated next.
147
148We also intend to show three syntactic properties about the cost
149labelling:
150\begin{enumerate}
151\item every function starts with a cost label,
152\item every branching instruction is followed by a cost label (note that
153  exiting a loop is treated as a branch), and
154\item the head of every loop (and any \lstinline'goto' destination) is
155  a cost label.
156\end{enumerate}
157These can be shown by structural induction on the source term.
158
159\subsection{Clight to Cminor translation}
160
161This translation is the first to introduce some invariants, with the
162proofs closely tied to the implementation by dependent typing.  These
163are largely complete and show that the generated code enjoys:
164\begin{itemize}
165\item some minimal type safety shown by explicit checks on the
166  Cminor types during the transformation (a little more work remains
167  to be done here, but follows the same form);
168\item that variables named in the parameter and local variable
169  environments are distinct from one another, again by an explicit
170  check;
171\item that variables used in the generated code are present in the
172  resulting environment (either by checking their presence in the
173  source environment, or from a list of freshly generated temporary variables);
174  and
175\item that all \lstinline[language=C]'goto' labels are present (by
176  checking them against a list of source labels and proving that all
177  source labels are preserved).
178\end{itemize}
179
180The simulation will be similar to the relevant stages of CompCert
181(Clight to Csharpminor and Csharpminor to Cminor --- in the event that
182the direct proof is unwieldy we could introduce an intermediate
183language corresponding to Csharpminor).  During early experimentation
184with porting CompCert definitions to the Matita proof assistant we
185found little difficulty reproving the results for the memory model, so
186we plan to port the memory injection properties and use them to relate
187Clight in-memory variables with either the value of the local variable or a
188stack slot, depending on how it was classified.
189
190This should be sufficient to show the equivalence of (big-step)
191expression evaluation.  The simulation can then be shown by relating
192corresponding blocks of statement and continuations with their Cminor
193counterparts and proving that a few steps reaches the next matching
194state.
195
196The syntactic properties required for cost labels remain similar and a
197structural induction on the function bodies should be sufficient to
198show that they are preserved.
199
200\subsection{Cminor global initialisation code}
201
202This short phase replaces the global variable initialisation data with
203code that executes when the program starts.  Each piece of
204initialisation data in the source is matched by a new statement
205storing that data.  As each global variable is allocated a distinct
206memory block, the program state after the initialisation statements
207will be the same as the original program's state at the start of
208execution, and will proceed in the same manner afterwards.
209
210% Actually, the above is wrong...
211% ... this ought to be in a fresh main function with a fresh cost label
212
213\subsection{Cminor to RTLabs translation}
214
215In this part of the compiler we transform the program's functions into
216control flow graphs.  It is closely related to CompCert's Cminorsel to
217RTL transformation, albeit with target-independent operations.
218
219We already enforce several invariants with dependent types: some type
220safety, mostly shown using the type information from Cminor; and
221that the graph is closed (by showing that each successor was recently
222added, or corresponds to a \lstinline[language=C]'goto' label which
223are all added before the end).  Note that this relies on a
224monotonicity property; CompCert maintains a similar property in a
225similar way while building RTL graphs.  We will also add a result
226showing that all of the pseudo-register names are distinct for use by
227later stages using the same method as Cminor.
228
229The simulation will relate Cminor states to RTLabs states which are about to
230execute the code corresponding to the Cminor statement or continuation.
231Each Cminor statement becomes zero or more RTLabs statements, with a
232decreasing measure based on the statement and continuations similar to
233CompCert's.  We may also follow CompCert in using a relational
234specification of this stage so as to abstract away from the functional
235(and highly dependently typed) definition.
236
237The first two labelling properties remain as before; we will show that
238cost labels are preserved, so the function entry point will be a cost
239label, and successors to any statement that are cost labels map still
240map to cost labels, preserving the condition on branches.  We replace
241the property for loops with the notion that we will always reach a
242cost label or the end of the function after following a bounded number of
243successors.  This can be easily seen in Cminor using the requirement
244for cost labels at the head of loops and after gotos.  It remains to
245show that this is preserved by the translation to RTLabs.  % how?
246
247\subsection{RTLabs structured trace generation}
248
249This proof-only step incorporates the function call structure and cost
250labelling properties into the execution trace.  As the function calls
251are nested within the trace, we need to distinguish between
252terminating and non-terminating function calls.  Thus we use the
253excluded middle (specialised to a function termination property) to do
254this.
255
256Structured traces for terminating functions are built by following the
257flat trace, breaking it into chunks between cost labels and
258recursively processing function calls.  The main difficulties here are
259the non-structurally recursive nature of the function (instead we use
260the size of the termination proof as a measure) and using the RTLabs
261cost labelling properties to show that the constraints of the
262structured traces are observed.  We also show that the lower stack
263frames are preserved during function calls in order to prove that
264after returning from a function call we resume execution of the
265correct code.  This part of the work has already been constructed, but
266still requires a simple proof to show that flattening the structured
267trace recreates the original flat trace.
268
269The non-terminating case follows the trace like the terminating
270version to build up chunks of trace from cost-label to cost-label
271(which, by the finite distance to a cost label property shown before,
272can be represented by an inductive type).  These chunks are chained
273together in a coinductive data structure that can represent
274non-terminating traces.  The excluded middle is used to decide whether
275function calls terminate, in which case the function described above
276constructs an inductive terminating structured trace which is nested
277in the caller's trace.  Otherwise, another coinductive constructor is
278used to embed the non-terminating trace of the callee, generated by
279corecursion.  This part of the trace transformation is currently under
280construction, and will also need a flattening result to show that it
281is correct.
282
283
284\section{Backend: RTLabs to machine code}
285\label{sect.backend.rtlabs.machine.code}
286
287The compiler backend consists of the following intermediate languages, and stages of translation:
288
289\begin{center}
290\begin{minipage}{.8\linewidth}
291\begin{tabbing}
292\quad \=\,\vdots\= \\
293\> $\downarrow$ \>\\
294\> $\downarrow$ \quad \= \kill
295\textsf{RTLabs}\\
296\> $\downarrow$ \> copy propagation\footnote{\label{lab:opt}To be ported from the untrusted compiler and certified only in case of early completion of the certification of the other passes.} (an endo-transformation) \\
297\> $\downarrow$ \> instruction selection\\
298\> $\downarrow$ \> change of memory models in compiler\\
299\textsf{RTL}\\
300\> $\downarrow$ \> constant propagation$^{\mbox{\scriptsize \ref{lab:opt}}}$ (an endo-transformation) \\
301\> $\downarrow$ \> calling convention made explicit \\
302\> $\downarrow$ \> layout of activation records \\
303\textsf{ERTL}\\
304\> $\downarrow$ \> register allocation and spilling\\
305\> $\downarrow$ \> dead code elimination\\
306\textsf{LTL}\\
307\> $\downarrow$ \> function linearisation\\
308\> $\downarrow$ \> branch compression (an endo-transformation) \\
309\textsf{LIN}\\
310\> $\downarrow$ \> relabeling\\
311\textsf{ASM}\\
312\> $\downarrow$ \> pseudoinstruction expansion\\
313\textsf{MCS-51 machine code}\\
314\end{tabbing}
315\end{minipage}
316\end{center}
317
318\subsection{The RTLabs to RTL translation}
319\label{subsect.rtlabs.rtl.translation}
320
321The RTLabs to RTL translation pass marks the frontier between the two memory models used in the CerCo project.
322As a result, we require some method of translating between the values that the two memory models permit.
323Suppose we have such a translation, $\sigma$.
324Then the translation between values of the two memory models may be pictured with:
325
326\begin{displaymath}
327\mathtt{Value} ::= \bot \mid \mathtt{int(size)} \mid \mathtt{float} \mid \mathtt{null} \mid \mathtt{ptr} \quad\stackrel{\sigma}{\longrightarrow}\quad \mathtt{BEValue} ::= \bot \mid \mathtt{byte} \mid \mathtt{null}_i \mid \mathtt{ptr}_i
328\end{displaymath}
329
330In the front-end, we have both integer and float values, where integer values are `sized', along with null values and pointers. Some frontenv values are
331representables in a byte, but some others require more bits.
332
333In the back-end model all values are meant to be represented in a single byte.
334Values can thefore be undefined, be one byte long integers or be indexed
335fragments of a pointer, null or not. Floats values are no longer present, as floating point arithmetic is not supported by the CerCo compiler.
336
337The $\sigma$ map implements a one-to-many relation: a single front-end value
338is mapped to a sequence of back-end values when its size is more then one byte.
339
340We further require a map, $\sigma$, which maps the front-end \texttt{Memory} and the back-end's notion of \texttt{BEMemory}. Both kinds of memory can be
341thought as an instance of a generic \texttt{Mem} data type parameterized over
342the kind of values stored in memory.
343
344\begin{displaymath}
345\mathtt{Mem}\ \alpha = \mathtt{Block} \rightarrow (\mathbb{Z} \rightarrow \alpha)
346\end{displaymath}
347
348Here, \texttt{Block} consists of a \texttt{Region} paired with an identifier.
349
350\begin{displaymath}
351\mathtt{Block} ::= \mathtt{Region} \times \mathtt{ID}
352\end{displaymath}
353
354We now have what we need for defining what is meant by the `memory' in the backend memory model.
355Namely, we instantiate the previously defined \texttt{Mem} type with the type of back-end memory values.
356
357\begin{displaymath}
358\mathtt{BEMem} = \mathtt{Mem} \mathtt{BEValue}
359\end{displaymath}
360
361Memory addresses consist of a pair of back-end memory values:
362
363\begin{displaymath}
364\mathtt{Address} = \mathtt{BEValue} \times  \mathtt{BEValue} \\
365\end{displaymath}
366
367The back- and front-end memory models differ in how they represent sized integeer values in memory.
368In particular, the front-end stores integer values as a header, with size information, followed by a string of `continuation' blocks, marking out the full representation of the value in memory.
369In contrast, the layout of sized integer values in the back-end memory model consists of a series of byte-sized `chunks':
370
371\begin{center}
372\begin{picture}(0, 25)
373\put(-125,0){\framebox(25,25)[c]{\texttt{v,4}}}
374\put(-100,0){\framebox(25,25)[c]{\texttt{cont}}}
375\put(-75,0){\framebox(25,25)[c]{\texttt{cont}}}
376\put(-50,0){\framebox(25,25)[c]{\texttt{cont}}}
377\put(-15,10){\vector(1, 0){30}}
378\put(25,0){\framebox(25,25)[c]{\texttt{\texttt{v$_1$}}}}
379\put(50,0){\framebox(25,25)[c]{\texttt{\texttt{v$_2$}}}}
380\put(75,0){\framebox(25,25)[c]{\texttt{\texttt{v$_3$}}}}
381\put(100,0){\framebox(25,25)[c]{\texttt{\texttt{v$_4$}}}}
382\end{picture}
383\end{center}
384
385Chunks for pointers are pairs made of the original pointer and the index of the chunk.
386Therefore, when assembling the chunks together, we can always recognize if all chunks refer to the same value or if the operation is meaningless.
387
388The differing memory representations of values in the two memory models imply the need for a series of lemmas on the actions of \texttt{load} and \texttt{store} to ensure correctness.
389The first lemma required has the following statement:
390\begin{displaymath}
391\mathtt{load}\ s\ a\ M = \mathtt{Some}\ v \rightarrow \forall i \leq s.\ \mathtt{load}\ s\ (a + i)\ \sigma(M) = \mathtt{Some}\ v_i
392\end{displaymath}
393That is, if we are successful in reading a value of size $s$ from memory at address $a$ in front-end memory, then we should successfully be able to read a value from memory in the back-end memory at \emph{any} address from address $a$ up to and including address $a + i$, where $i \leq s$.
394
395Next, we must show that \texttt{store} properly commutes with the $\sigma$-map between memory spaces:
396\begin{displaymath}
397\sigma(\mathtt{store}\ a\ v\ M) = \mathtt{store}\ \sigma(v)\ \sigma(a)\ \sigma(M)
398\end{displaymath}
399That is, if we store a value \texttt{v} in the front-end memory \texttt{M} at address \texttt{a} and transform the resulting memory with $\sigma$, then this is equivalent to storing a transformed value $\mathtt{\sigma(v)}$ at address $\mathtt{\sigma(a)}$ into the back-end memory $\mathtt{\sigma(M)}$.
400
401Finally, we must prove that \texttt{load}, \texttt{store} and $\sigma$ all properly commute.
402Writing \texttt{load}$^*$ for multiple consecutive iterations of \texttt{load}, we must prove:
403\begin{displaymath}
404\texttt{load}^* (\mathtt{store}\ \sigma(a')\ \sigma(v)\ \sigma(M))\ \sigma(a)\ \sigma(M) = \mathtt{load}^*\ \sigma(s)\ \sigma(a)\ \sigma(M)
405\end{displaymath}
406That is, suppose we store a transformed value $\mathtt{\sigma(v)}$ into a back-end memory $\mathtt{\sigma(M)}$ at address $\mathtt{\sigma(a')}$, using \texttt{store}, and then load the correct number of bytes (for the size of $\mathtt{\sigma(v)}$ at address $\sigma(a)$.
407Then, this should be equivalent to loading the correct number of bytes from address $\sigma(a)$ in an unaltered version of $\mathtt{\sigma(M)}$, \emph{providing} that the memory regions occupied by the two sequences of bytes at the two addresses do not overlap.
408This will entail more proof obligations, demonstrating that the $\sigma$-map between memory spaces respects memory regions.
409
410RTLabs states come in three flavours:
411\begin{displaymath}
412\begin{array}{rll}
413\mathtt{State} & ::=  & (\mathtt{State} : \mathtt{Frame}^* \times \mathtt{Frame} \\
414               & \mid & \mathtt{Call} : \mathtt{Frame}^* \times \mathtt{Args} \times \mathtt{Return} \times \mathtt{Fun} \\
415               & \mid & \mathtt{Return} : \mathtt{Frame}^* \times \mathtt{Value} \times \mathtt{Return}) \times \mathtt{Mem}
416\end{array}
417\end{displaymath}
418\texttt{State} is the default state in which RTLabs programs are almost always in.
419The \texttt{Call} state is only entered when a call instruction is being executed, and then we immediately return to being in \texttt{State}.
420Similarly, \texttt{Return} is only entered when a return instruction is being executed, before returning immediately to \texttt{State}.
421All RTLabs states are accompanied by a memory, \texttt{Mem}, with \texttt{Call} and \texttt{Return} keeping track of arguments, return addresses and the results of functions.
422\texttt{State} keeps track of a list of stack frames.
423
424RTL states differ from their RTLabs counterparts, in including a program counter \texttt{PC}, stack-pointer \texttt{SP}, internal stack pointer \texttt{ISP}, a carry flag \texttt{CARRY} and a set of registers \texttt{REGS}:
425\begin{displaymath}
426\mathtt{State} ::= \mathtt{Frame}^* \times \mathtt{PC} \times \mathtt{SP} \times \mathtt{ISP} \times \mathtt{CARRY} \times \mathtt{REGS}
427\end{displaymath}
428The internal stack pointer \texttt{ISP}, and its relationship with the stack pointer \texttt{SP}, needs some comment.
429Due to the design of the MCS-51, and its minuscule stack, it was decided that the compiler would implement an emulated stack in external memory.
430As a result, we have two stack pointers in our state: \texttt{ISP}, which is the real, hardware stack, and \texttt{SP}, which is the stack pointer of the emulated stack in memory.
431The emulated stack is used for pushing and popping stack frames when calling or returning from function calls, however this is done using the hardware stack, indexed by \texttt{ISP} as an intermediary.
432Instructions like \texttt{LCALL} and \texttt{ACALL} are hardwired by the processor's design to push on to the hardware stack, therefore after a call has been made, and after a call returns, the compiler emits code to move parameters and return addresses to-and-from the hardware stack.
433As a result, for most of the execution of the processor, the hardware stack is empty, or contains a single item ready to be moved into external memory.
434
435Once more, we require a map from RTLabs states to RTL states.
436Call it $\sigma$:
437\begin{displaymath}
438\mathtt{State} \stackrel{\sigma}{\longrightarrow} \mathtt{State}
439\end{displaymath}
440
441Translating an RTLabs state to an RTL state proceeds by cases on the particular type of state we are trying to translate, either a \texttt{State}, \texttt{Call} or a \texttt{Return}.
442For \texttt{State} we perform a further case analysis of the top stack frame, which decomposes into a tuple holding the current program counter value, the current stack pointer and the value of the registers:
443\begin{displaymath}
444\sigma(\mathtt{State} (\mathtt{Frame}^* \times \mathtt{\langle PC, REGS, SP \rangle})) \longrightarrow ((\sigma(\mathtt{Frame}^*), \sigma(\mathtt{PC}), \sigma(\mathtt{SP}), 0, 0, \sigma(\mathtt{REGS})), \sigma(\mathtt{Mem}))
445\end{displaymath}
446Translation then proceeds by translating the remaining stack frames, as well as the contents of the top stack frame, and setting all untranslated values to zero, their default value.
447
448Translating \texttt{Call} and \texttt{Return} states is more involved, as a commutation between a single step of execution and the translation process must hold:
449\begin{displaymath}
450\sigma(\mathtt{Return}(-)) \longrightarrow \sigma \circ \text{return one step}
451\end{displaymath}
452
453\begin{displaymath}
454\sigma(\mathtt{Call}(-)) \longrightarrow \sigma \circ \text{call one step}
455\end{displaymath}
456
457Here \emph{return one step} and \emph{call one step} refer to a pair of commuting diagrams relating the one-step execution of a call and return state and translation of both.
458We provide the one step commuting diagrams in Figure~\ref{fig.commuting.diagrams}.
459
460\begin{figure}
461\begin{displaymath}
462\begin{diagram}
463s & \rTo^{\text{one step of execution}} & s'   \\
464  & \rdTo                             & \dTo \\
465  &                                   & \llbracket s'' \rrbracket
466\end{diagram}
467\end{displaymath}
468
469\begin{displaymath}
470\begin{diagram}
471s & \rTo^{\text{one step of execution}} & s'   \\
472  & \rdTo                             & \dTo \\
473  &                                   & \llbracket s'' \rrbracket
474\end{diagram}
475\end{displaymath}
476\caption{The one-step commuting diagrams for \texttt{Call} and \texttt{Return} state translations}
477\label{fig.commuting.diagrams}
478\end{figure}
479
480We outline how states are executed:
481\begin{displaymath}
482\begin{array}{rcl}
483\mathtt{Call(id,\ args,\ dst,\ pc),\ State(Frame^*, Frame)} & \longrightarrow & \mathtt{Call(M(args), dst)}, \\
484                                                           &                 & \mathtt{PUSH(Frame[PC := after\_return])}
485\end{array}
486\end{displaymath}
487Suppose we are given a \texttt{State} with a list of stack frames, with the top frame being \texttt{Frame}.
488Suppose also that the program counter in \texttt{Frame} points to a \texttt{Call} instruction, complete with arguments and destination address.
489Then this is executed by entering into a \texttt{Call} state were the arguments are looked up in memory, and the destination address is filled in, with the current stack frame being pushed on top of the stack with the return address substituted for the program counter.
490
491Now, what happens next depends on whether we are executing an internal or an external function.
492In the case where the call is to an external function, we have:
493\begin{displaymath}
494\begin{array}{rcl}
495\mathtt{Call(M(args), dst)},                       & \stackrel{\mathtt{ret\_val = f(M(args))}}{\longrightarrow} & \mathtt{Return(ret\_val,\ dst,\ PUSH(...))} \\
496\mathtt{PUSH(current\_frame[PC := after\_return])} &                                                            & 
497\end{array}
498\end{displaymath}
499
500
501then:
502
503\begin{displaymath}
504\begin{array}{rcl}
505\mathtt{Return(ret\_val,\ dst,\ PUSH(...))} & \longrightarrow & \mathtt{pc = POP\_STACK(regs[dst := M(ret\_val)],\ pc)}
506\end{array}
507\end{displaymath}
508
509In the case where the call is to an internal function, we have:
510
511\begin{displaymath}
512\begin{array}{rcl}
513\mathtt{CALL}(\mathtt{id}, \mathtt{args}, \mathtt{dst}, \mathtt{pc}) & \longrightarrow & \mathtt{CALL\_ID}(\mathtt{id}, \sigma'(\mathtt{args}), \sigma(\mathtt{dst}), \mathtt{pc}) \\
514\mathtt{RETURN}                                                      & \longrightarrow & \mathtt{RETURN} \\
515\end{array} 
516\end{displaymath}
517
518\begin{displaymath}
519\begin{array}{rcl}
520\mathtt{Call(M(args), dst)}                        & \longrightarrow & \mathtt{sp = alloc}, regs = \emptyset[- := PARAMS] \\
521\mathtt{PUSH(current\_frame[PC := after\_return])} &                 & \mathtt{State}(regs,\ sp,\ pc_\emptyset,\ dst)
522\end{array}
523\end{displaymath}
524
525then:
526
527\begin{displaymath}
528\begin{array}{rcl}
529\mathtt{sp = alloc}, regs = \emptyset[- := PARAMS] & \longrightarrow & \mathtt{free(sp)} \\
530\mathtt{State}(regs,\ sp,\ pc_\emptyset,\ dst)     &                 & \mathtt{Return(M(ret\_val), dst, frames)}
531\end{array}
532\end{displaymath}
533
534and finally:
535
536\begin{displaymath}
537\begin{array}{rcl}
538\mathtt{free(sp)}                         & \longrightarrow & \mathtt{pc = POP\_STACK(regs[dst := M(ret\_val)],\ pc)} \\
539\mathtt{Return(M(ret\_val), dst, frames)} &                 & 
540\end{array}
541\end{displaymath}
542
543\subsection{The RTL to ERTL translation}
544\label{subsect.rtl.ertl.translation}
545
546\begin{displaymath}
547\begin{diagram}
548& & \llbracket \mathtt{CALL\_ID}(\mathtt{id}, \mathtt{args}, \mathtt{dst}, \mathtt{pc})\rrbracket & & \\
549& \ldTo^{\text{external}} & & \rdTo^{\text{internal}} & \\
550\skull & & & & \mathtt{regs} = [\mathtt{params}/-] \\
551& & & & \mathtt{sp} = \mathtt{ALLOC} \\
552& & & & \mathtt{PUSH}(\mathtt{carry}, \mathtt{regs}, \mathtt{dst}, \mathtt{return\_addr}), \mathtt{pc}_{0}, \mathtt{regs}, \mathtt{sp} \\
553\end{diagram}
554\end{displaymath}
555
556\begin{align*}
557\llbracket \mathtt{RETURN} \rrbracket \\
558\mathtt{return\_addr} & := \mathtt{top}(\mathtt{stack}) \\
559v*                    & := m(\mathtt{rv\_regs}) \\
560\mathtt{dst}, \mathtt{sp}, \mathtt{carry}, \mathtt{regs} & := \mathtt{pop} \\
561\mathtt{regs}[v* / \mathtt{dst}] \\
562\end{align*}
563
564\begin{displaymath}
565\begin{diagram}
566s    & \rTo^1 & s' & \rTo^1 & s'' \\
567\dTo &        &    & \rdTo  & \dTo \\
568\llbracket s \rrbracket & \rTo(1,3)^1 & & & \llbracket s'' \rrbracket \\ 
569\mathtt{CALL} \\
570\end{diagram}
571\end{displaymath}
572
573\begin{displaymath}
574\begin{diagram}
575s    & \rTo^1 & s' & \rTo^1 & s'' \\
576\dTo &        &    & \rdTo  & \dTo \\
577\  & \rTo(1,3) & & & \ \\
578\mathtt{RETURN} \\
579\end{diagram}
580\end{displaymath}
581
582\begin{displaymath}
583\mathtt{b\_graph\_translate}: (\mathtt{label} \rightarrow \mathtt{blist'})
584\rightarrow \mathtt{graph} \rightarrow \mathtt{graph}
585\end{displaymath}
586
587\begin{align*}
588\mathtt{theorem} &\ \mathtt{b\_graph\_translate\_ok}: \\
589& \forall  f.\forall G_{i}.\mathtt{let}\ G_{\sigma} := \mathtt{b\_graph\_translate}\ f\ G_{i}\ \mathtt{in} \\
590&       \forall l \in G_{i}.\mathtt{subgraph}\ (f\ l)\ l\ (\mathtt{next}\ l\ G_{i})\ G_{\sigma}
591\end{align*}
592
593\begin{align*}
594\mathtt{lemma} &\ \mathtt{execute\_1\_step\_ok}: \\
595&       \forall s.  \mathtt{let}\ s' := s\ \sigma\ \mathtt{in} \\
596&       \mathtt{let}\ l := pc\ s\ \mathtt{in} \\
597&       s \stackrel{1}{\rightarrow} s^{*} \Rightarrow \exists n. s' \stackrel{n}{\rightarrow} s'^{*} \wedge s'^{*} = s'\ \sigma
598\end{align*}
599
600\begin{align*}
601\mathrm{RTL\ status} & \ \ \mathrm{ERTL\ status} \\
602\mathtt{sp} & = \mathtt{spl} / \mathtt{sph} \\
603\mathtt{graph} &  \mathtt{graph} + \mathtt{prologue}(s) + \mathtt{epilogue}(s) \\
604& \mathrm{where}\ s = \mathrm{callee\ saved} + \nu \mathrm{RA} \\
605\end{align*}
606
607\begin{displaymath}
608\begin{diagram}
609\mathtt{CALL} & \rTo^1 & \mathtt{inside\ function} \\
610\dTo & & \dTo \\
611\underbrace{\ldots}_{\llbracket \mathtt{CALL} \rrbracket} & \rTo &
612\underbrace{\ldots}_{\mathtt{prologue}} \\
613\end{diagram}
614\end{displaymath}
615
616\begin{displaymath}
617\begin{diagram}
618\mathtt{RETURN} & \rTo^1 & \mathtt{.} \\
619\dTo & & \dTo \\
620\underbrace{\ldots}_{\mathtt{epilogue}} & \rTo &
621\underbrace{\ldots} \\
622\end{diagram}
623\end{displaymath}
624
625\begin{align*}
626\mathtt{prologue}(s) = & \mathtt{create\_new\_frame}; \\
627                       & \mathtt{pop\ ra}; \\
628                       & \mathtt{save\ callee\_saved}; \\
629                                                                                         & \mathtt{get\_params} \\
630                                                                                         & \ \ \mathtt{reg\_params}: \mathtt{move} \\
631                                                                                         & \ \ \mathtt{stack\_params}: \mathtt{push}/\mathtt{pop}/\mathtt{move} \\
632\end{align*}
633
634\begin{align*}
635\mathtt{epilogue}(s) = & \mathtt{save\ return\ to\ tmp\ real\ regs}; \\
636                                                                                         & \mathtt{restore\_registers}; \\
637                       & \mathtt{push\ ra}; \\
638                       & \mathtt{delete\_frame}; \\
639                       & \mathtt{save return} \\
640\end{align*}
641
642\begin{displaymath}
643\mathtt{CALL}\ id \mapsto \mathtt{set\_params};\ \mathtt{CALL}\ id;\ \mathtt{fetch\_result}
644\end{displaymath}
645
646\subsection{The ERTL to LTL translation}
647\label{subsect.ertl.ltl.translation}
648\newcommand{\declsf}[1]{\expandafter\newcommand\expandafter{\csname #1\endcsname}{\mathop{\mathsf{#1}}\nolimits}}
649\declsf{Livebefore}
650\declsf{Liveafter}
651\declsf{Defined}
652\declsf{Used}
653\declsf{Eliminable}
654\declsf{StatementSem}
655For the liveness analysis, we aim at a map
656$\ell \in \mathtt{label} \mapsto $ live registers at $\ell$.
657We define the following operators on ERTL statements.
658$$
659\begin{array}{lL>{(ex. $}L<{)$}}
660\Defined(s) & registers defined at $s$ & r_1\leftarrow r_2+r_3 \mapsto \{r_1,C\}, \mathtt{CALL}~id\mapsto \text{caller-save}
661\\
662\Used(s) & registers used at $s$ & r_1\leftarrow r_2+r_3 \mapsto \{r_2,r_3\}, \mathtt{CALL}~id\mapsto \text{parameters}
663\end{array}
664$$
665Given $LA:\mathtt{label}\to\mathtt{lattice}$ (where $\mathtt{lattice}$
666is the type of sets of registers\footnote{More precisely, it is thethe lattice
667of pairs of sets of pseudo-registers and sets of hardware registers,
668with pointwise operations.}, we also have have the following
669predicates:
670$$
671\begin{array}{lL}
672\Eliminable_{LA}(\ell) & iff $s(\ell)$ has side-effects only on $r\notin LA(\ell)$
673\\&
674(ex.\ $\ell : r_1\leftarrow r_2+r_3 \mapsto (\{r_1,C\}\cap LA(\ell)\neq\emptyset,
675  \mathtt{CALL}id\mapsto \text{never}$)
676\\
677\Livebefore_{LA}(\ell) &$:=
678  \begin{cases}
679    LA(\ell) &\text{if $\Eliminable_{LA}(\ell)$,}\\
680    (LA(\ell)\setminus \Defined(s(\ell)))\cup \Used(s(\ell) &\text{otherwise}.
681  \end{cases}$
682\end{array}
683$$
684In particular, $\Livebefore$ has type $(\mathtt{label}\to\mathtt{lattice})\to
685\mathtt{label}\to\mathtt{lattice}$.
686
687The equation on which we build the fixpoint is then
688$$\Liveafter(\ell) \doteq \bigcup_{\ell' >_1 \ell} \Livebefore_{\Liveafter}(\ell')$$
689where $\ell' >_1 \ell$ denotes that $\ell'$ is an immediate successor of $\ell$
690in the graph. We do not require the fixpoint to be the least one, so the hypothesis
691on $\Liveafter$ that we require is
692$$\Liveafter(\ell) \supseteq \bigcup_{\ell' >_1 \ell} \Livebefore(\ell')$$
693(for shortness we drop the subscript from $\Livebefore$).
694\subsection{The LTL to LIN translation}
695\label{subsect.ltl.lin.translation}
696Ad detailed elsewhere in the reports, due to the parameterized representation of
697the back-end languages, the pass described here is actually much more generic
698than the translation from LTL to LIN. It consists in a linearization pass
699that maps any graph-based back-end language to its corresponding linear form,
700preserving its semantics. In the rest of the section, however, we will keep
701the names LTL and LIN for the two partial instantiations of the parameterized
702language.
703
704We require a map, $\sigma$, from LTL statuses, where program counters are represented as labels in a graph data structure, to LIN statuses, where program counters are natural numbers:
705\begin{displaymath}
706\mathtt{pc : label} \stackrel{\sigma}{\longrightarrow} \mathbb{N}
707\end{displaymath}
708
709The LTL to LIN translation pass also linearises the graph data structure into a list of instructions.
710Pseudocode for the linearisation process is as follows:
711
712\begin{lstlisting}
713let rec linearise graph visited required generated todo :=
714  match todo with
715  | l::todo ->
716    if l $\in$ visited then
717      let generated := generated $\cup\ \{$ Goto l $\}$ in
718      let required := required $\cup$ l in
719        linearise graph visited required generated todo
720    else
721      -- Get the instruction at label `l' in the graph
722      let lookup := graph(l) in
723      let generated := generated $\cup\ \{$ lookup $\}$ in
724      -- Find the successor of the instruction at label `l' in the graph
725      let successor := succ(l, graph) in
726      let todo := successor::todo in
727        linearise graph visited required generated todo
728  | []      -> (required, generated)
729\end{lstlisting}
730
731It is easy to see that this linearisation process eventually terminates.
732In particular, the size of the visited label set is monotonically increasing, and is bounded above by the size of the graph that we are linearising.
733
734The initial call to \texttt{linearise} sees the \texttt{visited}, \texttt{required} and \texttt{generated} sets set to the empty set, and \texttt{todo} initialized with the singleton list consisting of the entry point of the graph.
735We envisage needing to prove the following invariants on the linearisation function above:
736
737\begin{enumerate}
738\item
739$\mathtt{visited} \approx \mathtt{generated}$, where $\approx$ is \emph{multiset} equality, as \texttt{generated} is a set of instructions where instructions may mention labels multiple times, and \texttt{visited} is a set of labels,
740\item
741$\forall \mathtt{l} \in \mathtt{generated}.\ \mathtt{succ(l,\ graph)} \subseteq \mathtt{required} \cup \mathtt{todo}$,
742\item
743$\mathtt{required} \subseteq \mathtt{visited}$,
744\item
745$\mathtt{visited} \cap \mathtt{todo} = \emptyset$.
746\end{enumerate}
747
748The invariants collectively imply the following properties, crucial to correctness, about the linearisation process:
749
750\begin{enumerate}
751\item
752Every graph node is visited at most once,
753\item
754Every instruction that is generated is generated due to some graph node being visited,
755\item
756The successor instruction of every instruction that has been visited already will eventually be visited too.
757\end{enumerate}
758
759Note, because the LTL to LIN transformation is the first time the code of
760a function is linearised in the back-end, we must discover a notion of `well formed function code' suitable for linearised forms.
761In particular, we see the notion of well formedness (yet to be formally defined) resting on the following conditions:
762
763\begin{enumerate}
764\item
765For every jump to a label in a linearised function code, the target label exists at some point in the function code,
766\item
767Each label is unique, appearing only once in the function code,
768\item
769The final instruction of a function code must be a return or an unconditional
770jump.
771\end{enumerate}
772
773We assume that these properties will be easy consequences of the invariants on the linearisation function defined above.
774
775The final condition above is potentially a little opaque, so we explain further.
776The only instructions that can reasonably appear in final position at the end of a function code are returns or backward jumps, as any other instruction would cause execution to `fall out' of the end of the program (for example, when a function invoked with \texttt{CALL} returns, it returns to the next instruction past the \texttt{CALL} that invoked it).
777
778\subsection{The LIN to ASM and ASM to MCS-51 machine code translations}
779\label{subsect.lin.asm.translation}
780
781The LIN to ASM translation step is trivial, being almost the identity function.
782The only non-trivial feature of the LIN to ASM translation is that all labels are `named apart' so that there is no chance of freshly generated labels from different namespaces clashing with labels from another namespace.
783
784The ASM to MCS-51 machine code translation step, and the required statements of correctness, are found in an unpublished manuscript attached to this document.
785This is the most complex translation because of the huge number of cases
786to be addressed and because of the complexity of the two semantics.
787Moreover, in the assembly code we have conditional and unconditional jumps
788to arbitrary locations in the code, which are not supported by the MCS-51
789instruction set. The latter has several kind of jumps characterized by a
790different instruction size and execution time, but limited in range. For
791instance, conditional jumps to locations whose destination is more than
792$2^7$ bytes away from the jump instruction location are not supported at
793all and need to be emulated with a code transformation. The problem, which
794is known in the litterature as branch displacement and that applies also
795to modern architectures, is known to be hard and is often NP. As far as we
796know, we will provide the first formally verified proof of correctness for
797an assembler that implements branch displacement. We are also providing
798the first verified proof of correctness of a mildly optimizing branch
799displacement algorithm that, at the moment, is almost finished, but not
800described in the companion paper. This proof by itself took about 6 men
801months.
802
803\section{Correctness of cost prediction}
804Roughly speaking,
805the proof of correctness of cost prediction shows that the cost of executing
806a labelled object code program is the same as the sum over all labels in the
807program execution trace of the cost statically associated to the label and
808computed on the object code itself.
809
810In presence of object level function calls, the previous statement is, however,
811incorrect. The reason is twofold. First of all, a function call may diverge.
812To the last labels that comes before the call, however, we also associate
813the cost of the instructions that follow the call. Therefore, in the
814sum over all labels, when we meet a label we pre-pay for the instructions
815after function calls, assuming all calls to be terminating. This choice is
816driven by considerations on the source code. Functions can be called also
817inside expressions and it would be too disruptive to put labels inside
818expressions to capture the cost of instructions that follow a call. Moreover,
819adding a label after each call would produce a much higher number of proof
820obligations in the certification of source programs using Frama-C. The
821proof obligations, moreover, would be guarded by termination of all functions
822involved, that also generates lots of additional complex proof obligations
823that have little to do with execution costs. With our approach, instead, we
824put less burden on the user, at the price of proving a weaker statement:
825the estimated and actual costs will be the same if and only if the high level
826program is converging. For prefixes of diverging programs we can provide
827a similar result where the equality is replaced by an inequality (loss of
828precision).
829
830Assuming totality of functions is however not sufficient yet at the object
831level. Even if a function returns, there is no guarantee that it will transfer
832control back to the calling point. For instance, the function could have
833manipulated the return address from its stack frame. Moreover, an object level
834program can forge any address and transfer control to it, with no guarantee
835on the execution behaviour and labelling properties of the called program.
836
837To solve the problem, we introduced the notion of \emph{structured trace}
838that come in two flavours: structured traces for total programs (an inductive
839type) and structured traces for diverging programs (a co-inductive type based
840on the previous one). Roughly speaking, a structured trace represents the
841execution of a well behaved program that is subject to several constraints
842like
843\begin{enumerate}
844 \item All function calls return control just after the calling point
845 \item The execution of all function bodies start with a label and end with
846   a RET (even the ones reached by invoking a function pointer)
847 \item All instructions are covered by a label (required by correctness of
848   the labelling approach)
849 \item The target of all conditional jumps must be labelled (a sufficient
850   but not necessary condition for precision of the labelling approach)
851 \item \label{prop5} Two structured traces with the same structure yield the same
852   cost traces.
853\end{enumerate}
854
855Correctness of cost predictions is proved only for structured execution traces,
856i.e. well behaved programs. The forward simulation proof for all back-end
857passes will actually be a proof of preservation of the structure of
858the structured traces that, because of property \ref{prop5}, will imply
859correctness of the cost prediction for the back-end. The Clight to RTLabs
860will also include a proof that associates to each converging execution its
861converging structured trace and to each diverging execution its diverging
862structured trace.
863
864There are also other two issues that invalidate the naive statement of
865correctness of cost prediciton given above. The algorithm that statically
866computes the cost of blocks is correct only when the object code is \emph{well
867formed} and the program counter is \emph{reachable}.
868A well formed object code is such that
869the program counter will never overflow after the execution step of
870the processor. An overflow that occurs during fetching but is overwritten
871during execution is, however, correct and necessary to accept correct
872programs that are as large as the program memory. Temporary overflows add
873complications to the proof. A reachable address is an address that can be
874obtained by fetching (not executing!) a finite number of times from the
875beginning of the code memory without ever overflowing. The complication is that
876the static prediction traverses the code memory assuming that the memory will
877be read sequentially from the beginning and that all jumps jump only to
878reachable addresses. When this property is violated, the way the code memory
879is interpreted is uncorrect and the cost computed is totally meaningless.
880The reachability relation is closed by fetching for well formed programs.
881The property that calls to function pointers only target reachable and
882well labelled locations, however, is not statically predictable and it is
883enforced in the structured trace.
884
885The proof of correctness of cost predictions has been quite complex. Setting
886up the good invariants (structured traces, well formed programs, reachability)
887and completing the proof has required more than 3 men months while the initally
888estimated effort was much lower. In the paper-and-pencil proof for IMP, the
889corresponding proof was obvious and only took two lines.
890
891The proof itself is quite involved. We
892basically need to show as an important lemma that the sum of the execution
893costs over a structured trace, where the costs are summed in execution order,
894is equivalent to the sum of the execution costs in the order of pre-payment.
895The two orders are quite different and the proof is by mutual recursion over
896the definition of the converging structured traces, which is a family of three
897mutual inductive types. The fact that this property only holds for converging
898function calls in hidden in the definition of the structured traces.
899Then we need to show that the order of pre-payment
900corresponds to the order induced by the cost traces extracted from the
901structured trace. Finally, we need to show that the statically computed cost
902for one block corresponds to the cost dinamically computed in pre-payment
903order.
904
905\section{Overall results}
906
907Functional correctness of the compiled code can be shown by composing
908the simulations to show that the target behaviour matches the
909behaviour of the source program, if the source program does not `go
910wrong'.  More precisely, we show that there is a forward simulation
911between the source trace and a (flattened structured) trace of the
912output, and conclude equivalence because the target's semantics are
913in the form of an executable function, and hence
914deterministic.
915
916Combining this with the correctness of the assignment of costs to cost
917labels at the ASM level for a structured trace, we can show that the
918cost of executing any compiled function (including the main function)
919is equal to the sum of all the values for cost labels encountered in
920the \emph{source code's} trace of the function.
921
922\section{Estimated effort}
923Based on the rough analysis performed so far we can estimate the total
924effort for the certification of the compiler. We obtain this estimation by
925combining, for each pass: 1) the number of lines of code to be certified;
9262) the ratio of number of lines of proof to number of lines of code from
927the CompCert project~\cite{compcert} for the CompCert pass that is closest to
928ours; 3) an estimation of the complexity of the pass according to the
929analysis above.
930
931\begin{tabular}{lrlrr}
932Pass origin & Code lines & CompCert ratio & Estimated effort & Estimated effort \\
933            &            &                & (based on CompCert) & \\
934\hline
935Common &  4864 & 4.25 \permil & 20.67 & 17.0 \\
936Cminor &  1057 & 5.23 \permil & 5.53  &  6.0 \\
937Clight &  1856 & 5.23 \permil & 9.71  & 10.0 \\ 
938RTLabs &  1252 & 1.17 \permil & 1.48  &  5.0 \\
939RTL    &   469 & 4.17 \permil & 1.95  &  2.0 \\
940ERTL   &   789 & 3.01 \permil & 2.38  & 2.5 \\
941LTL    &    92 & 5.94 \permil & 0.55  & 0.5 \\
942LIN    &   354 & 6.54 \permil & 2.31  &   1.0 \\
943ASM    &   984 & 4.80 \permil & 4.72  &  10.0 \\
944\hline
945Total common    &  4864 & 4.25 \permil & 20.67 & 17.0 \\
946Total front-end &  2913 & 5.23 \permil & 15.24 & 16.0 \\
947Total back-end  &  6853 & 4.17 \permil & 13.39 & 21.0 \\
948\hline
949Total           & 14630 & 3.75 \permil & 49.30 & 54.0 \\
950\end{tabular}
951
952We provide now some additional informations on the methodology used in the
953computation. The passes in Cerco and CompCert front-end closely match each
954other. However, there is no clear correspondence between the two back-ends.
955For instance, we enforce the calling convention immediately after instruction
956selection, whereas in CompCert this is performed in a later phase. Or we
957linearize the code at the very end, whereas CompCert performs linearization
958as soon as possible. Therefore, the first part of the exercise has consisted
959in shuffling and partitioning the CompCert code in order to assign to each
960CerCo pass the CompCert code that performs the same transformation.
961
962After this preliminary step, using the data given in~\cite{compcert} (which
963are relative to an early version of CompCert) we computed the ratio between
964men months and lines of code in CompCert for each CerCo pass. This is shown
965in the third column of Table~\ref{wildguess}. For those CerCo passes that
966have no correspondence in CompCert (like the optimizing assembler) or where
967we have insufficient data, we have used the average of the ratios computed
968above.
969
970The first column of the table shows the number of lines of code for each
971pass in CerCo. The third column is obtained multiplying the first with the
972CompCert ratio. It provides an estimate of the effort required (in men months)
973if the complexity of the proofs for CerCo and Compcert would be the same.
974
975The two proof styles, however, are on purpose completely different. Where
976CompCert uses non executable semantics, describing the various semantics with
977inductive types, we have preferred executable semantics. Therefore, CompCert
978proofs by induction and inversion become proof by functional inversion,
979performed using the Russel methodology (now called Program in Coq, but whose
980behaviour differs from Matita's one). Moreover, CompCert code is written using
981only types that belong to the Hindley-Milner fragment, whereas we have
982heavily exploited dependent types all over the code. The dependent type
983discipline offers many advantages from the point of view of clarity of the
984invariants involved and early detection of errors and it naturally combines
985well with the Russel approach which is based on dependent types. However, it
986is also well known to introduce technical problems all over the code, like
987the need to explicitly prove type equalities to be able to manipulate
988expressions in certain ways. In many situations, the difficulties encountered
989with manipulating dependent types are better addressed by improving the Matita
990system, according to the formalization driven system development. For this
991reason, and assuming a pessimistic point of view on our performance, the
992fourth columns presents the final estimation of the effort required, that also
993takes in account the complexity of the proof suggested by the informal proofs
994sketched in the previous section.
995
996\section{Conclusions}
997The overall exercise, whose details have been only minimally reported here,
998has been very useful. It has allowed to spot in an early moment some criticities
999of the proof that have required major changes in the proof plan. It has also
1000shown that the last passes of the compilation (e.g. assembly) and cost
1001prediction on the object code are much more involved than more high level
1002passes.
1003
1004The final estimation for the effort is surely affected by a low degree of
1005confidence. It is however sufficient to conclude that the effort required
1006is in line with the man power that was scheduled for the second half of the
1007second period and for the third period. Compared to the number of men months
1008declared in Annex I of the contract, we will need more men months. However,
1009both at UNIBO and UEDIN there have been major differences in hiring with
1010respect to the Annex. Therefore both sites have now an higher number of men
1011months available, with the trade-off of a lower level of maturity of the
1012people employed.
1013
1014The reviewers suggested that we use this estimation to compare two possible
1015scenarios: a) proceed as planned, porting all the CompCert proofs to Matita
1016or b) port D3.1 and D4.1 to Coq and re-use the CompCert proofs.
1017We remark here again that the back-end of the two compilers, from the
1018memory model on, are sensibly different: we are not re-proving correctness
1019of the same piece of code. Moreover, the proof techniques are different for
1020the front-end too. Switching to the CompCert formalization would imply
1021the abandon of the untrusted compiler, the abandon of the experiment with
1022a different proof technique, the abandon of the quest for an open source
1023proof, and the abandon of the co-development of the formalization and the
1024Matita proof assistant. In the Commitment Letter~\cite{letter} delivered
1025to the Officer in May we clarified our personal perspective on the project
1026goals and objectives. We do not re-describe here the point of view presented
1027in the letter that we can condense in ``we value diversity''.
1028
1029Clearly, if the execise would have suggested the infeasability in terms of
1030effort of concluding the formalization or getting close to that, we would have
1031abandoned our path and embraced the reviewer's suggestion. However, we
1032have been comforted in the analysis we did in autumn and further progress done
1033during the winter does not show yet any major delay with respect to the
1034proof schedule. We are thus planning to continue the certification according
1035to the more detailed proof plan that came out from the exercise reported in
1036this manuscript.
1037
1038\end{document}
Note: See TracBrowser for help on using the repository browser.