1 //===---------------------------------------------------------------------===//
2 // Random ideas for the X86 backend: SSE-specific stuff.
3 //===---------------------------------------------------------------------===//
5 - Consider eliminating the unaligned SSE load intrinsics, replacing them with
6 unaligned LLVM load instructions.
8 //===---------------------------------------------------------------------===//
10 Expand libm rounding functions inline: Significant speedups possible.
11 http://gcc.gnu.org/ml/gcc-patches/2006-10/msg00909.html
13 //===---------------------------------------------------------------------===//
15 When compiled with unsafemath enabled, "main" should enable SSE DAZ mode and
18 //===---------------------------------------------------------------------===//
20 Think about doing i64 math in SSE regs.
22 //===---------------------------------------------------------------------===//
24 This testcase should have no SSE instructions in it, and only one load from
27 double %test3(bool %B) {
28 %C = select bool %B, double 123.412, double 523.01123123
32 Currently, the select is being lowered, which prevents the dag combiner from
33 turning 'select (load CPI1), (load CPI2)' -> 'load (select CPI1, CPI2)'
35 The pattern isel got this one right.
37 //===---------------------------------------------------------------------===//
39 SSE doesn't have [mem] op= reg instructions. If we have an SSE instruction
44 and the register allocator decides to spill X, it is cheaper to emit this as:
55 ..and this uses one fewer register (so this should be done at load folding
56 time, not at spiller time). *Note* however that this can only be done
57 if Y is dead. Here's a testcase:
59 %.str_3 = external global [15 x sbyte] ; <[15 x sbyte]*> [#uses=0]
60 implementation ; Functions:
61 declare void %printf(int, ...)
65 no_exit.i7: ; preds = %no_exit.i7, %build_tree.exit
66 %tmp.0.1.0.i9 = phi double [ 0.000000e+00, %build_tree.exit ], [ %tmp.34.i18, %no_exit.i7 ] ; <double> [#uses=1]
67 %tmp.0.0.0.i10 = phi double [ 0.000000e+00, %build_tree.exit ], [ %tmp.28.i16, %no_exit.i7 ] ; <double> [#uses=1]
68 %tmp.28.i16 = add double %tmp.0.0.0.i10, 0.000000e+00
69 %tmp.34.i18 = add double %tmp.0.1.0.i9, 0.000000e+00
70 br bool false, label %Compute_Tree.exit23, label %no_exit.i7
71 Compute_Tree.exit23: ; preds = %no_exit.i7
72 tail call void (int, ...)* %printf( int 0 )
73 store double %tmp.34.i18, double* null
82 *** movsd %XMM2, QWORD PTR [%ESP + 8]
83 *** addsd %XMM2, %XMM1
84 *** movsd QWORD PTR [%ESP + 8], %XMM2
85 jmp .BBmain_1 # no_exit.i7
87 This is a bugpoint reduced testcase, which is why the testcase doesn't make
88 much sense (e.g. its an infinite loop). :)
90 //===---------------------------------------------------------------------===//
92 SSE should implement 'select_cc' using 'emulated conditional moves' that use
93 pcmp/pand/pandn/por to do a selection instead of a conditional branch:
95 double %X(double %Y, double %Z, double %A, double %B) {
96 %C = setlt double %A, %B
97 %z = add double %Z, 0.0 ;; select operand is not a load
98 %D = select bool %C, double %Y, double %z
107 addsd 24(%esp), %xmm0
108 movsd 32(%esp), %xmm1
109 movsd 16(%esp), %xmm2
110 ucomisd 40(%esp), %xmm1
120 //===---------------------------------------------------------------------===//
122 It's not clear whether we should use pxor or xorps / xorpd to clear XMM
123 registers. The choice may depend on subtarget information. We should do some
124 more experiments on different x86 machines.
126 //===---------------------------------------------------------------------===//
128 Currently the x86 codegen isn't very good at mixing SSE and FPStack
131 unsigned int foo(double x) { return x; }
135 movsd 24(%esp), %xmm0
143 This will be solved when we go to a dynamic programming based isel.
145 //===---------------------------------------------------------------------===//
147 Lower memcpy / memset to a series of SSE 128 bit move instructions when it's
150 //===---------------------------------------------------------------------===//
152 Teach the coalescer to commute 2-addr instructions, allowing us to eliminate
153 the reg-reg copy in this example:
155 float foo(int *x, float *y, unsigned c) {
158 for (i = 0; i < c; i++) {
159 float xx = (float)x[i];
168 cvtsi2ss %XMM0, DWORD PTR [%EDX + 4*%ESI]
169 mulss %XMM0, DWORD PTR [%EAX + 4*%ESI]
173 **** movaps %XMM1, %XMM0
174 jb LBB_foo_3 # no_exit
176 //===---------------------------------------------------------------------===//
179 if (copysign(1.0, x) == copysign(1.0, y))
184 //===---------------------------------------------------------------------===//
186 Use movhps to update upper 64-bits of a v4sf value. Also movlps on lower half
189 //===---------------------------------------------------------------------===//
191 Better codegen for vector_shuffles like this { x, 0, 0, 0 } or { x, 0, x, 0}.
192 Perhaps use pxor / xorp* to clear a XMM register first?
194 //===---------------------------------------------------------------------===//
196 How to decide when to use the "floating point version" of logical ops? Here are
199 movaps LCPI5_5, %xmm2
202 mulps 8656(%ecx), %xmm3
203 addps 8672(%ecx), %xmm3
209 movaps LCPI5_5, %xmm1
212 mulps 8656(%ecx), %xmm3
213 addps 8672(%ecx), %xmm3
217 movaps %xmm3, 112(%esp)
220 Due to some minor source change, the later case ended up using orps and movaps
221 instead of por and movdqa. Does it matter?
223 //===---------------------------------------------------------------------===//
225 X86RegisterInfo::copyRegToReg() returns X86::MOVAPSrr for VR128. Is it possible
226 to choose between movaps, movapd, and movdqa based on types of source and
229 How about andps, andpd, and pand? Do we really care about the type of the packed
230 elements? If not, why not always use the "ps" variants which are likely to be
233 //===---------------------------------------------------------------------===//
235 External test Nurbs exposed some problems. Look for
236 __ZN15Nurbs_SSE_Cubic17TessellateSurfaceE, bb cond_next140. This is what icc
239 movaps (%edx), %xmm2 #59.21
240 movaps (%edx), %xmm5 #60.21
241 movaps (%edx), %xmm4 #61.21
242 movaps (%edx), %xmm3 #62.21
243 movl 40(%ecx), %ebp #69.49
244 shufps $0, %xmm2, %xmm5 #60.21
245 movl 100(%esp), %ebx #69.20
246 movl (%ebx), %edi #69.20
247 imull %ebp, %edi #69.49
248 addl (%eax), %edi #70.33
249 shufps $85, %xmm2, %xmm4 #61.21
250 shufps $170, %xmm2, %xmm3 #62.21
251 shufps $255, %xmm2, %xmm2 #63.21
252 lea (%ebp,%ebp,2), %ebx #69.49
254 lea -3(%edi,%ebx), %ebx #70.33
256 addl 32(%ecx), %ebx #68.37
257 testb $15, %bl #91.13
258 jne L_B1.24 # Prob 5% #91.13
260 This is the llvm code after instruction scheduling:
262 cond_next140 (0xa910740, LLVM BB @0xa90beb0):
263 %reg1078 = MOV32ri -3
264 %reg1079 = ADD32rm %reg1078, %reg1068, 1, %NOREG, 0
265 %reg1037 = MOV32rm %reg1024, 1, %NOREG, 40
266 %reg1080 = IMUL32rr %reg1079, %reg1037
267 %reg1081 = MOV32rm %reg1058, 1, %NOREG, 0
268 %reg1038 = LEA32r %reg1081, 1, %reg1080, -3
269 %reg1036 = MOV32rm %reg1024, 1, %NOREG, 32
270 %reg1082 = SHL32ri %reg1038, 4
271 %reg1039 = ADD32rr %reg1036, %reg1082
272 %reg1083 = MOVAPSrm %reg1059, 1, %NOREG, 0
273 %reg1034 = SHUFPSrr %reg1083, %reg1083, 170
274 %reg1032 = SHUFPSrr %reg1083, %reg1083, 0
275 %reg1035 = SHUFPSrr %reg1083, %reg1083, 255
276 %reg1033 = SHUFPSrr %reg1083, %reg1083, 85
277 %reg1040 = MOV32rr %reg1039
278 %reg1084 = AND32ri8 %reg1039, 15
280 JE mbb<cond_next204,0xa914d30>
282 Still ok. After register allocation:
284 cond_next140 (0xa910740, LLVM BB @0xa90beb0):
286 %EDX = MOV32rm <fi#3>, 1, %NOREG, 0
287 ADD32rm %EAX<def&use>, %EDX, 1, %NOREG, 0
288 %EDX = MOV32rm <fi#7>, 1, %NOREG, 0
289 %EDX = MOV32rm %EDX, 1, %NOREG, 40
290 IMUL32rr %EAX<def&use>, %EDX
291 %ESI = MOV32rm <fi#5>, 1, %NOREG, 0
292 %ESI = MOV32rm %ESI, 1, %NOREG, 0
293 MOV32mr <fi#4>, 1, %NOREG, 0, %ESI
294 %EAX = LEA32r %ESI, 1, %EAX, -3
295 %ESI = MOV32rm <fi#7>, 1, %NOREG, 0
296 %ESI = MOV32rm %ESI, 1, %NOREG, 32
298 SHL32ri %EDI<def&use>, 4
299 ADD32rr %EDI<def&use>, %ESI
300 %XMM0 = MOVAPSrm %ECX, 1, %NOREG, 0
301 %XMM1 = MOVAPSrr %XMM0
302 SHUFPSrr %XMM1<def&use>, %XMM1, 170
303 %XMM2 = MOVAPSrr %XMM0
304 SHUFPSrr %XMM2<def&use>, %XMM2, 0
305 %XMM3 = MOVAPSrr %XMM0
306 SHUFPSrr %XMM3<def&use>, %XMM3, 255
307 SHUFPSrr %XMM0<def&use>, %XMM0, 85
309 AND32ri8 %EBX<def&use>, 15
311 JE mbb<cond_next204,0xa914d30>
313 This looks really bad. The problem is shufps is a destructive opcode. Since it
314 appears as operand two in more than one shufps ops. It resulted in a number of
315 copies. Note icc also suffers from the same problem. Either the instruction
316 selector should select pshufd or The register allocator can made the two-address
317 to three-address transformation.
319 It also exposes some other problems. See MOV32ri -3 and the spills.
321 //===---------------------------------------------------------------------===//
323 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25500
325 LLVM is producing bad code.
327 LBB_main_4: # cond_true44
338 jne LBB_main_4 # cond_true44
340 There are two problems. 1) No need to two loop induction variables. We can
341 compare against 262144 * 16. 2) Known register coalescer issue. We should
342 be able eliminate one of the movaps:
344 addps %xmm2, %xmm1 <=== Commute!
347 movaps %xmm1, %xmm1 <=== Eliminate!
354 jne LBB_main_4 # cond_true44
356 //===---------------------------------------------------------------------===//
360 __m128 test(float a) {
361 return _mm_set_ps(0.0, 0.0, 0.0, a*a);
372 Because mulss doesn't modify the top 3 elements, the top elements of
373 xmm1 are already zero'd. We could compile this to:
379 //===---------------------------------------------------------------------===//
381 Here's a sick and twisted idea. Consider code like this:
383 __m128 test(__m128 a) {
384 float b = *(float*)&A;
386 return _mm_set_ps(0.0, 0.0, 0.0, b);
389 This might compile to this code:
391 movaps c(%esp), %xmm1
396 Now consider if the ... code caused xmm1 to get spilled. This might produce
399 movaps c(%esp), %xmm1
400 movaps %xmm1, c2(%esp)
404 movaps c2(%esp), %xmm1
408 However, since the reload is only used by these instructions, we could
409 "fold" it into the uses, producing something like this:
411 movaps c(%esp), %xmm1
412 movaps %xmm1, c2(%esp)
415 movss c2(%esp), %xmm0
418 ... saving two instructions.
420 The basic idea is that a reload from a spill slot, can, if only one 4-byte
421 chunk is used, bring in 3 zeros the the one element instead of 4 elements.
422 This can be used to simplify a variety of shuffle operations, where the
423 elements are fixed zeros.
425 //===---------------------------------------------------------------------===//
429 #include <emmintrin.h>
430 void test(__m128d *r, __m128d *A, double B) {
431 *r = _mm_loadl_pd(*A, &B);
437 movsd 24(%esp), %xmm0
449 movl 4(%esp), %edx #3.6
450 movl 8(%esp), %eax #3.6
451 movapd (%eax), %xmm0 #4.22
452 movlpd 12(%esp), %xmm0 #4.8
453 movapd %xmm0, (%edx) #4.3
456 So icc is smart enough to know that B is in memory so it doesn't load it and
457 store it back to stack.
459 //===---------------------------------------------------------------------===//
461 __m128d test1( __m128d A, __m128d B) {
462 return _mm_shuffle_pd(A, B, 0x3);
467 shufpd $3, %xmm1, %xmm0
469 Perhaps it's better to use unpckhpd instead?
471 unpckhpd %xmm1, %xmm0
473 Don't know if unpckhpd is faster. But it is shorter.
475 //===---------------------------------------------------------------------===//
477 This code generates ugly code, probably due to costs being off or something:
479 void %test(float* %P, <4 x float>* %P2 ) {
480 %xFloat0.688 = load float* %P
481 %loadVector37.712 = load <4 x float>* %P2
482 %inFloat3.713 = insertelement <4 x float> %loadVector37.712, float 0.000000e+00, uint 3
483 store <4 x float> %inFloat3.713, <4 x float>* %P2
491 movd %xmm0, %eax ;; EAX = 0!
494 pinsrw $6, %eax, %xmm0
495 shrl $16, %eax ;; EAX = 0 again!
496 pinsrw $7, %eax, %xmm0
500 It would be better to generate:
506 pinsrw $6, %eax, %xmm0
507 pinsrw $7, %eax, %xmm0
511 or use pxor (to make a zero vector) and shuffle (to insert it).
513 //===---------------------------------------------------------------------===//
515 Some useful information in the Apple Altivec / SSE Migration Guide:
517 http://developer.apple.com/documentation/Performance/Conceptual/
518 Accelerate_sse_migration/index.html
520 e.g. SSE select using and, andnot, or. Various SSE compare translations.
522 //===---------------------------------------------------------------------===//
524 Add hooks to commute some CMPP operations.
526 //===---------------------------------------------------------------------===//
528 Apply the same transformation that merged four float into a single 128-bit load
529 to loads from constant pool.
531 //===---------------------------------------------------------------------===//
533 Floating point max / min are commutable when -enable-unsafe-fp-path is
534 specified. We should turn int_x86_sse_max_ss and X86ISD::FMIN etc. into other
535 nodes which are selected to max / min instructions that are marked commutable.
537 //===---------------------------------------------------------------------===//
539 We should compile this:
540 #include <xmmintrin.h>
546 void swizzle (const void *a, vector4_t * b, vector4_t * c) {
547 b->v = _mm_loadl_pi (b->v, (__m64 *) a);
548 c->v = _mm_loadl_pi (c->v, ((__m64 *) a) + 1);
559 movlps 8(%eax), %xmm0
573 movlps 8(%ecx), %xmm0
577 //===---------------------------------------------------------------------===//
581 #include <emmintrin.h>
582 __m128i test(long long i) { return _mm_cvtsi64x_si128(i); }
584 Should turn into a single 'movq %rdi, %xmm0' instruction. Instead, we
585 get this (on x86-64):
595 target triple = "x86_64-apple-darwin8"
596 define <2 x i64> @test(i64 %i) {
598 %tmp10 = insertelement <2 x i64> undef, i64 %i, i32 0
599 %tmp11 = insertelement <2 x i64> %tmp10, i64 0, i32 1
603 //===---------------------------------------------------------------------===//
605 These functions should produce the same code:
607 #include <emmintrin.h>
609 typedef long long __m128i __attribute__ ((__vector_size__ (16)));
611 int foo(__m128i* val) {
612 return __builtin_ia32_vec_ext_v4si(*val, 1);
614 int bar(__m128i* val) {
622 We currently produce (with -m64):
625 pshufd $1, (%rdi), %xmm0
632 //===---------------------------------------------------------------------===//
634 We should materialize vector constants like "all ones" and "signbit" with
637 cmpeqps xmm1, xmm1 ; xmm1 = all-ones
640 cmpeqps xmm1, xmm1 ; xmm1 = all-ones
641 psrlq xmm1, 31 ; xmm1 = all 100000000000...
643 instead of using a load from the constant pool. The later is important for
644 ABS/NEG/copysign etc.
646 //===---------------------------------------------------------------------===//
648 "converting 64-bit constant pool entry to 32-bit not necessarily beneficial"
649 http://llvm.org/PR1264
653 define double @foo(double %x) {
654 %y = mul double %x, 5.000000e-01
658 llc -march=x86-64 currently produces a 32-bit constant pool entry and this code:
660 cvtss2sd .LCPI1_0(%rip), %xmm1
663 instead of just using a 64-bit constant pool entry with this:
665 mulsd .LCPI1_0(%rip), %xmm0
667 This is due to the code in ExpandConstantFP in LegalizeDAG.cpp. It notices that
668 x86-64 indeed has an instruction to load a 32-bit float from memory and convert
669 it into a 64-bit float in a register, however it doesn't notice that this isn't
670 beneficial because it prevents the load from being folded into the multiply.
672 //===---------------------------------------------------------------------===//
676 LBB4_6: # bb47.preheader
683 xorps %xmm0, %xmm0 # (1)
685 movaps %xmm0, 16(%edx)
690 LBB4_8: # bb42.return_crit_edge
694 The instruction at (1) can be moved out of the main body of the loop.
696 //===---------------------------------------------------------------------===//