1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334
|
; Correctness of a Factorial Program that Violates Actual JVM Stack Rules
; Problem: Define an M1 program to compute the factorial of a natural number n,
; by pushing all the factors onto the stack and then multiplying them in a second
; loop.
; Design Plan: I will go around a loop pushing n, n-1, ..., 1 onto the stack.
; Then I will go around another loop just doing IMULs. Note that I must go
; around the first loop n times and the second loop n-1 times. This program
; violates the bytecode verifier's run that the stack is a fixed size at every
; instruction.
; Verification of the program illustrates how to verify a two loop program
; where the loops are not nested. However, this program is very unusual
; because it essentially uses the stack as a list of values of arbitary length
; and its verification involves abandoning the push/top/pop abstraction and
; just manipulating lists. (Of course, we could redevelop list theory for
; push/top/pop, but it is counter to the spirit of stacks.) So this is not a
; good exemplar of two-loop verification.
; (0) Start ACL2
; (include-book "m1")
(in-package "M1")
; (1) Write your specification, i.e., define the expected inputs and the
; desired output, theta.
(defun ok-inputs (n)
(natp n))
(defun ! (n)
(if (zp n)
1
(* n (! (- n 1)))))
(defun theta (n)
(! n))
; (2) Write your algorithm. This will consist of a tail-recursive helper
; function and a wrapper, fn.
; With this algorithm we see something new: We have to have two loops and so we
; have to have two helpers, one which is mimicking pushing things onto the
; stack and the other which mimicks IMULing them away. Unlike our other
; helpers, these return the stack rather than some individual local.
(defun helper1 (n stack)
(if (zp n)
stack
(helper1 (- n 1) (push n stack))))
(defun helper2 (m stack)
(if (zp m)
stack
(helper2 (- m 1) (push (* (top (pop stack)) (top stack)) (pop (pop stack))))))
(defun fn (n)
(if (zp n)
1
(top (helper2 (- n 1)
(helper1 n nil)))))
; (3) Prove that the algorithm satisfies the spec, by proving first that the
; helper is appropriately related to theta and then that fn is theta on ok
; inputs.
; Important Note: When you verify your helper function, you must consider the
; most general case. For example, if helper is defined with formal parameters
; n, m, and a and fn calls helper initializing a to 0, your helper theorem must
; be about (helper n m a), not just about the special case (helper n m 0).
; -----------------------------------------------------------------
; Here begins a horrible development of list theory and the conversion of our
; stack stuff to lists! We could import a bunch of functions from ACL2 list
; books, but I'll just develop it all here. The end of this development is
; marked by another row of hyphens.
(in-theory (enable top pop push))
(defun ap (x y)
(if (endp x)
y
(cons (car x)
(ap (cdr x) y))))
(defun nats (n)
(if (zp n)
nil
(ap (nats (- n 1)) (list n))))
(defun prod (x)
(if (endp x)
1
(* (car x) (prod (cdr x)))))
(defun firstn (n x)
(if (or (zp n)
(endp x))
1
(cons (car x)
(firstn (- n 1) (cdr x)))))
(defun natp-list (x)
(if (endp x)
t
(and (natp (car x))
(natp-list (cdr x)))))
(defthm assoc-of-ap
(equal (ap (ap a b) c)
(ap a (ap b c))))
(defthm natp-list-ap
(equal (natp-list (ap a b))
(and (natp-list a)
(natp-list b))))
(defthm len-ap
(equal (len (ap a b))
(+ (len a) (len b))))
(defthm len-nats
(equal (len (nats n))
(nfix n)))
(defthm natp-list-nats
(natp-list (nats n)))
(defthm firstn-ap
(implies (natp n)
(equal (firstn n (ap a b))
(if (< n (len a))
(firstn n a)
(ap a (firstn (- n (len a)) b))))))
(defthm prod-ap
(equal (prod (ap a b))
(* (prod a) (prod b))))
(defthm prod-nats
(equal (prod (nats n))
(! (nfix n))))
(defthm nthcdr-ap
(implies (natp n)
(equal (nthcdr n (ap a b))
(if (< n (len a))
(ap (nthcdr n a) b)
(nthcdr (- n (len a)) b)))))
; -----------------------------------------------------------------
(defthm helper1-alt-def
(equal (helper1 n stack)
(ap (nats n) stack)))
(defthm helper2-alt-def
(implies (and (natp n)
(natp-list stack)
(< n (len stack)))
(equal (helper2 n stack)
(cons (prod (firstn (+ n 1) stack))
(nthcdr (+ n 1) stack)))))
(defthm helper2-helper1-is-theta
(implies (and (not (zp n))
(natp-list stack))
(equal (helper2 (- n 1) (helper1 n stack))
(push (! n) stack))))
(defthm fn-is-theta
(implies (ok-inputs n)
(equal (fn n) (theta n))))
; Disable these lemmas because they confuse the theorem prover when it is
; dealing with the code versus fn.
(in-theory (disable helper1-alt-def
helper2-alt-def
helper2-helper1-is-theta
fn-is-theta))
; (4) Write your M1 program with the intention of implementing your algorithm.
(defconst *pi*
'((ILOAD 0) ; 0
(IFEQ 21) ; 1
(ILOAD 0) ; 2
(ICONST 1) ; 3
(ISUB) ; 4
(ISTORE 1) ; 5
(ILOAD 0) ; 6 loop1
(IFEQ 7) ; 7
(ILOAD 0) ; 8
(ILOAD 0) ; 9
(ICONST 1) ; 10
(ISUB) ; 11
(ISTORE 0) ; 12
(GOTO -7) ; 13
(ILOAD 1) ; 14 loop2
(IFEQ 8) ; 15
(IMUL) ; 16
(ILOAD 1) ; 17
(ICONST 1) ; 18
(ISUB) ; 19
(ISTORE 1) ; 20
(GOTO -7) ; 21
(ICONST 1) ; 22
(HALT)) ; 23
)
; (5) Define the ACL2 function that clocks your program, starting with the
; loop clock and then using it to clock the whole program. The clock
; should take the program from pc 0 to a HALT statement. (Sometimes your
; clocks will require multiple inputs or other locals, but our example only
; requires the first local.)
(defun loop1-clk (n)
(if (zp n)
2
(clk+ 8
(loop1-clk (- n 1)))))
(defun loop2-clk (m)
(if (zp m)
2
(clk+ 8
(loop2-clk (- m 1)))))
(defun clk (n)
(if (zp n)
8
(clk+ 6
(clk+ (loop1-clk n)
(loop2-clk (- n 1))))))
; (6) Prove that the code implements your algorithm, starting with the lemma
; that the loop implements the helper. Each time you prove a lemma relating
; code to algorithm, disable the corresponding clock function so the theorem
; prover doesn't look any deeper into subsequent code.
; Important Note: Your lemma about the loop must consider the general case.
; For example, if the loop uses the locals n, m, and a, you must characterize
; its behavior for arbitrary, legal n, m, and a, not just a special case (e.g.,
; where a is 0).
(defthm loop1-is-helper1
(implies (ok-inputs n)
(equal (m1 (make-state 6
(list n m)
stack
*pi*)
(loop1-clk n))
(make-state 14
(list 0 m)
(helper1 n stack)
*pi*))))
(in-theory (disable loop1-clk))
(defthm loop2-is-helper2
(implies (and (natp m)
(natp-list stack)
(< m (len stack)))
(equal (m1 (make-state 14
(list n m)
stack
*pi*)
(loop2-clk m))
(make-state 23
(list n 0)
(helper2 m stack)
*pi*))))
(in-theory (disable loop2-clk))
(defthm program-is-fn
(implies (ok-inputs n)
(equal (m1 (make-state 0
(list n)
nil
*pi*)
(clk n))
(make-state 23
(if (zp n) (list 0) (list 0 0))
(push (fn n) nil)
*pi*)))
; This hint is necessary because we have to relieve the hypotheses on the two
; loop lemmas, e.g., that stack is a list of nats sufficiently long, and our
; way of doing that is to appeal to the list lemmas.
:hints (("Goal" :in-theory (enable helper1-alt-def helper2-alt-def))))
(in-theory (disable clk))
; (7) Put the two steps together to get correctness.
(in-theory (enable fn-is-theta))
(defthm program-correct
(implies (ok-inputs n)
(equal (m1 (make-state 0
(list n)
nil
*pi*)
(clk n))
(make-state 23
(if (zp n) (list 0) (list 0 0))
(push (theta n)
nil)
*pi*))))
; This corollary just shows we did what we set out to do:
(defthm total-correctness
(implies (and (natp n)
(equal sf (m1 (make-state 0
(list n)
nil
*pi*)
(clk n))))
(and (equal (next-inst sf) '(HALT))
(equal (top (stack sf)) (! n))))
:rule-classes nil)
; Think of the above theorem as saying: for all natural numbers n and m, there
; exists a clock (for example, the one constructed by (clk n)) such that
; running *pi* with (list n m) as input produces a state, sf, that is halted
; and which contains (* n m) on top of the stack. Note that the algorithm
; used by *pi* is not specified or derivable from this formula.
|