3 (declare (simple-string s))
4 (declare (optimize (speed 3) (safety 0) (debug 0)))
7 (dotimes (i (length s))
8 (when (eql (aref s i) #\1)
12 * On X86 I is represented as a tagged integer.
15 3: SLOT S!11[EDX] {SB-C::VECTOR-LENGTH 1 7} => t23[EAX]
16 4: MOVE t23[EAX] => t24[EBX]
18 --------------------------------------------------------------------------------
21 (declare (optimize (speed 3) (safety 0) (space 2) (debug 0)))
22 (declare (type (simple-array double-float 1) v))
24 (declare (type double-float s))
25 (dotimes (i (length v))
26 (setq s (+ s (aref v i))))
29 * Python does not combine + with AREF, so generates extra move and
32 * On X86 Python thinks that all FP registers are directly accessible
33 and emits costy MOVE ... => FR1.
35 --------------------------------------------------------------------------------
38 (declare (optimize (speed 3) (safety 0) (space 2))
40 (let ((v (make-list n)))
41 (setq v (make-array n))
44 * IR1 does not optimize away (MAKE-LIST N).
45 --------------------------------------------------------------------------------
48 (declare (optimize (speed 3) (safety 0) (space 2))
49 (type (simple-array base-char 1) v1 v2))
50 (dotimes (i (length v1))
51 (setf (aref v2 i) (aref v1 i))))
53 VOP DATA-VECTOR-SET/SIMPLE-STRING V2!14[EDI] t32[EAX] t30[S2]>t33[CL]
55 MOV #<TN t33[CL]>, #<TN t30[S2]>
56 MOV BYTE PTR [EDI+EAX+1], #<TN t33[CL]>
57 MOV #<TN t35[AL]>, #<TN t33[CL]>
58 MOV #<TN t34[S2]>, #<TN t35[AL]>
60 * The value of DATA-VECTOR-SET is not used, so there is no need in the
64 --------------------------------------------------------------------------------
67 (declare (optimize (speed 3) (safety 0) (debug 0)))
68 (declare (type (double-float 0d0 1d0) d))
69 (loop for i fixnum from 1 to 5
70 for x1 double-float = (sin d) ;;; !!!
71 do (loop for j fixnum from 1 to 4
72 sum x1 double-float)))
74 Without the marked declaration Python will use boxed representation for X1.
80 ;; use of X as DOUBLE-FLOAT
83 The initial binding is effectless, and without it X is of type
84 DOUBLE-FLOAT. Unhopefully, IR1 does not optimize away effectless
85 SETs/bindings, and IR2 does not perform type inference.
86 --------------------------------------------------------------------------------
87 #9 "Multi-path constant folding"
89 (if (= (cond ((irgh x) 0)
96 This code could be optimized to
102 --------------------------------------------------------------------------------
104 (inverted variant of #9)
107 (let ((y (sap-alien x c-string)))
111 It could be optimized to
113 (lambda (x) (list x x))
115 (if Y were used only once, the current compiler would optimize it)
116 --------------------------------------------------------------------------------
118 (typep (truly-the (simple-array * (*)) x) 'simple-vector)
121 --------------------------------------------------------------------------------
123 FAST-+/FIXNUM and similar should accept unboxed arguments in interests
124 of representation selection. Problem: inter-TN dependencies.
125 --------------------------------------------------------------------------------
127 The derived type of (/ (THE (DOUBLE-FLOAT (0D0)) X) (THE (DOUBLE-FLOAT
128 1D0) Y)) is (DOUBLE-FLOAT 0.0d0). While it might be reasonable, it is
129 better to derive (OR (MEMBER 0.0d0) (DOUBLE-FLOAT (0.0d0))).
130 --------------------------------------------------------------------------------
132 On the alpha, the system is reluctant to refer directly to a constant bignum,
133 preferring to load a large constant through a slow sequence of instructions,
134 then cons up a bignum for it:
137 (DECLARE (OPTIMIZE (SAFETY 1) (SPEED 3) (DEBUG 1))
138 (TYPE (INTEGER -10000 10000) A)
141 ((89 125 16) (ASH A (MIN 18 -706)))
142 (T (DPB -3 (BYTE 30 30) -1))))
143 --------------------------------------------------------------------------------
146 ((= i (the (integer 0 100) n)))
149 It is commonly expected for Python to derive (FIXNUMP I). (If ``='' is
150 replaced with ``>='', Python will do.)
151 --------------------------------------------------------------------------------
153 Type tests for (ARRAY BIT), (ARRAY T) and similar go through full
154 %TYPEP, even though it is relatively simple to establish the arrayness
155 of an object and also to obtain the element type of an array. As of
156 sbcl-0.8.12.30, this affects at least DUMP-OBJECT through
157 COMPOUND-OBJECT-P, and (LABELS MAYBE-EMIT-MAKE-LOAD-FORMS GROVEL)
158 through TYPEP UNBOXED-ARRAY, within the compiler itself.
159 --------------------------------------------------------------------------------
161 (lambda (x) (declare (null x)) (sxhash x)) goes through SYMBOL-HASH
162 rather than either constant-folding or manipulating NIL-VALUE or
164 --------------------------------------------------------------------------------
169 (declare (dynamic-extent dx))
172 DX is not allocated on stack.
173 --------------------------------------------------------------------------------
175 (defun-with-dx foo (x)
177 (let ((l (list nil nil)))
179 (setf (second l) (1- x))
182 (declare (dynamic-extent l))
185 Result of MAKE is not stack allocated, which means that
186 stack-allocation of structures is impossible.
187 --------------------------------------------------------------------------------
189 (defun-with-dx foo ()
190 (let ((dx (list (list 1 2) (list 3 4))))
191 (declare (dynamic-extent dx))
194 External list in DX is allocated on stack, but internal are not.
195 --------------------------------------------------------------------------------
197 IR2 does not perform unused code flushing.
198 --------------------------------------------------------------------------------
200 Python does not know that &REST lists are LISTs (and cannot derive it).
201 --------------------------------------------------------------------------------
203 a. Iterations on &REST lists, returning them as VALUES could be
204 rewritten with &MORE vectors.
205 b. Implement local unknown-values mv-call (useful for fast type checking).
206 --------------------------------------------------------------------------------
208 SBCL cannot derive upper bound for I and uses generic arithmetic here:
212 (dotimes (i (length l))
214 (map-foo (lambda (x) (if x (return t)))
219 (So the constraint propagator or a possible future SSA-convertor
220 should know the connection between an NLE and its CLEANUP.)
221 --------------------------------------------------------------------------------
223 (We always zeroize stack-allocated arrays of boxed elements. The
224 previous note here suggested that we could avoid that step on
225 platforms with conservative GC; it's not clear to me (NJF) that
226 doing so is a wise idea.)
228 x86 and x86-64 do not zeroize stack-allocated arrays of unboxed
229 elements; other platforms could copy what they do.
230 --------------------------------------------------------------------------------
232 a. Accessing raw slots in structure instances is more inefficient than
233 it could be; if we placed raw slots before the header word, we would
234 not need to do arithmetic at runtime to access them. (But beware:
235 this would complicate handling of the interior pointer).
237 b. (Also note that raw slots are currently disabled on HPPA)
238 --------------------------------------------------------------------------------
240 Python is overly zealous when converting high-level CL functions, such
241 as MIN/MAX, LOGBITP, and LOGTEST, to low-level CL functions. Reducing
242 Python's aggressiveness would make it easier to effect changes such as
245 * direct MIN/MAX on {SINGLE,DOUBLE}-FLOATs ({MIN,MAX}S{S,D})
248 * direct LOGBITP on word-sized integers and fixnums (BT + JC)
251 * branch-free MIN/MAX on word-sized integers and fixnums (floats could
252 be handled too, modulo safety considerations on the PPC)
255 * efficient LOGTESTs on word-sized integers and fixnums (TEST)
259 (The framework for this has been implemented as of 0.9.9.18; see the
260 vm-support-routine COMBINATION-IMPLEMENTATION-STYLE and its use in
261 src/compiler/ir1opt.lisp, IR1-OPTIMIZE-COMBINATION. The above
262 optimizations are left as an exercise for the reader.)
263 --------------------------------------------------------------------------------
268 FOO's IR1 representation is roughly:
275 However, if a full call is generated for < (and similarly for other
276 predicate functions), then the IF is unnecessary, since the return value
277 of (< x y) is already T or NIL.
278 --------------------------------------------------------------------------------
280 The typecheck generated for a declaration like (integer 0 45) on x86 looks
283 ; 12B: F6C203 TEST DL, 3
285 ; 130: 8BC2 MOV EAX, EDX
286 ; 132: 83F800 CMP EAX, 0
288 ; 137: 8BC2 MOV EAX, EDX
289 ; 139: 3DB4000000 CMP EAX, 180
292 A better code sequence for this would be:
300 Doing an unsigned comparison means that, similarly to %CHECK-BOUND, we can
301 combine the <0 and >=bound tests. This sort of test is generated often
302 in SBCL and any array-based code that's serious about type-checking its
304 --------------------------------------------------------------------------------
306 The code for a vector bounds check on x86 (similarly on x86-64) where
307 the vector is in EDX and the index in EAX looks like:
309 ; 49: L0: 8B5AFD MOV EBX, [EDX-3]
310 ; 4C: 39C3 CMP EBX, EAX
313 because %CHECK-BOUND is used for bounds-checking any array dimension.
314 A more efficient specialization (%CHECK-BOUND/VECTOR) would produce:
319 Which is slightly shorter and avoids using a register.
320 --------------------------------------------------------------------------------
322 Reports from the Java camp indicate that using an SSE2-based
323 floating-point backend on x86 when possible is highly preferable to
324 using the x86 FP stack. It would be nice if SBCL included an SSE2-based
325 floating point backend with a compile-time option to switch between the
327 --------------------------------------------------------------------------------
332 (declare (type (integer 0 45) x y))
335 results in the following error trapping code for type-checking the
338 ; 424: L0: 8B058CE31812 MOV EAX, [#x1218E38C] ; '(MOD 46)
339 ; 42A: 0F0B0A BREAK 10 ; error trap
341 ; 42E: 1F BYTE #X1F ; OBJECT-NOT-TYPE-ERROR
342 ; 42F: FECE01 BYTE #XFE, #XCE, #X01 ; EDI
343 ; 432: 0E BYTE #X0E ; EAX
344 ; 433: L1: 8B0590E31812 MOV EAX, [#x1218E390] ; '(MOD 46)
345 ; 439: 0F0B0A BREAK 10 ; error trap
347 ; 43D: 1F BYTE #X1F ; OBJECT-NOT-TYPE-ERROR
348 ; 43E: 8E BYTE #X8E ; EDX
349 ; 43F: 0E BYTE #X0E ; EAX
351 Notice that '(MOD 46) has two entries in the constant vector. Having
352 one would be preferable.
353 --------------------------------------------------------------------------------
358 (declare (type simple-vector a))
361 results in the following x86 code:
363 ; 115886E9: F7C703000000 TEST EDI, 3 ; no-arg-parsing entry point
365 ; 6F1: 8BC7 MOV EAX, EDI
366 ; 6F3: 83F800 CMP EAX, 0
368 ; 6F8: 8BC7 MOV EAX, EDI
369 ; 6FA: 3DF8FFFF7F CMP EAX, 2147483640
371 ; 701: L0: 8B057C865811 MOV EAX, [#x1158867C] ; '(MOD
373 ; 707: 0F0B0A BREAK 10 ; error trap
375 ; 70B: 1F BYTE #X1F ; OBJECT-NOT-TYPE-ERROR
376 ; 70C: FECE01 BYTE #XFE, #XCE, #X01 ; EDI
377 ; 70F: 0E BYTE #X0E ; EAX
378 ; 710: L1: 8B42FD MOV EAX, [EDX-3]
379 ; 713: 8BCF MOV ECX, EDI
380 ; 715: 39C8 CMP EAX, ECX
382 ; 719: 8B540A01 MOV EDX, [EDX+ECX+1]
384 ... plus the standard return sequence and some error blocks. The
385 `TEST EDI, 3' and associated comparisons are to ensure that `I' is a
386 positive fixnum. The associated comparisons are unnecessary, as the
387 %CHECK-BOUND VOP only requires its tested index to be a fixnum and takes
388 care of the negative fixnum case itself.
390 {HAIRY-,}DATA-VECTOR-REF are DEFKNOWN'd with EXPLICIT-CHECK, which would
391 seem to take care of this, but EXPLICIT-CHECK only seems to be used when
392 compiling calls to unknown functions or similar. Furthermore,
393 EXPLICIT-CHECK, as NJF understands it, doesn't have the right
394 semantics--it suppresses all type checking of arguments, whereas what we
395 really want is to ensure that the argument is a fixnum, but not check
397 --------------------------------------------------------------------------------
400 In #35, the CMP EAX, $foo instructions are all preceded by a MOV. They
401 appear to be unnecessary, but are necessary because in IR2, EDI is a
402 DESCRIPTOR-REG, whereas EAX is an ANY-REG--and the comparison VOPs only
403 accept ANY-REGs. Therefore, the MOVs are "necessary" to ensure that the
404 comparison VOP receives an TN of the appropriate storage class.
406 Obviously, it would be better if a) we only performed one MOV prior to
407 all three comparisons or b) eliminated the necessity of the MOV(s)
408 altogether. The former option is probably easier than the latter.
410 --------------------------------------------------------------------------------
413 Dynamic extent allocation doesn't currently work for one-element lists,
414 since there's a source transform from (LIST X) to (CONS X NIL).