Skip to content

Commit a35ec9a

Browse files
raskyrasky
committed
cmd/compile: implement CMOV on amd64
This builds upon the branchelim pass, activating it for amd64 and lowering CondSelect. Special care is made to FPU instructions for NaN handling. Benchmark results on Xeon E5630 (Westmere EP): name old time/op new time/op delta BinaryTree17-16 4.99s ± 9% 4.66s ± 2% ~ (p=0.095 n=5+5) Fannkuch11-16 4.93s ± 3% 5.04s ± 2% ~ (p=0.548 n=5+5) FmtFprintfEmpty-16 58.8ns ± 7% 61.4ns ±14% ~ (p=0.579 n=5+5) FmtFprintfString-16 114ns ± 2% 114ns ± 4% ~ (p=0.603 n=5+5) FmtFprintfInt-16 181ns ± 4% 125ns ± 3% -30.90% (p=0.008 n=5+5) FmtFprintfIntInt-16 263ns ± 2% 217ns ± 2% -17.34% (p=0.008 n=5+5) FmtFprintfPrefixedInt-16 230ns ± 1% 212ns ± 1% -7.99% (p=0.008 n=5+5) FmtFprintfFloat-16 411ns ± 3% 344ns ± 5% -16.43% (p=0.008 n=5+5) FmtManyArgs-16 828ns ± 4% 790ns ± 2% -4.59% (p=0.032 n=5+5) GobDecode-16 10.9ms ± 4% 10.8ms ± 5% ~ (p=0.548 n=5+5) GobEncode-16 9.52ms ± 5% 9.46ms ± 2% ~ (p=1.000 n=5+5) Gzip-16 334ms ± 2% 337ms ± 2% ~ (p=0.548 n=5+5) Gunzip-16 64.4ms ± 1% 65.0ms ± 1% +1.00% (p=0.008 n=5+5) HTTPClientServer-16 156µs ± 3% 155µs ± 3% ~ (p=0.690 n=5+5) JSONEncode-16 21.0ms ± 1% 21.8ms ± 0% +3.76% (p=0.016 n=5+4) JSONDecode-16 95.1ms ± 0% 95.7ms ± 1% ~ (p=0.151 n=5+5) Mandelbrot200-16 6.38ms ± 1% 6.42ms ± 1% ~ (p=0.095 n=5+5) GoParse-16 5.47ms ± 2% 5.36ms ± 1% -1.95% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 111ns ± 1% 111ns ± 1% ~ (p=0.635 n=5+4) RegexpMatchEasy0_1K-16 408ns ± 1% 411ns ± 2% ~ (p=0.087 n=5+5) RegexpMatchEasy1_32-16 103ns ± 1% 104ns ± 1% ~ (p=0.484 n=5+5) RegexpMatchEasy1_1K-16 659ns ± 2% 652ns ± 1% ~ (p=0.571 n=5+5) RegexpMatchMedium_32-16 176ns ± 2% 174ns ± 1% ~ (p=0.476 n=5+5) RegexpMatchMedium_1K-16 58.6µs ± 4% 57.7µs ± 4% ~ (p=0.548 n=5+5) RegexpMatchHard_32-16 3.07µs ± 3% 3.04µs ± 4% ~ (p=0.421 n=5+5) RegexpMatchHard_1K-16 89.2µs ± 1% 87.9µs ± 2% -1.52% (p=0.032 n=5+5) Revcomp-16 575ms ± 0% 587ms ± 2% +2.12% (p=0.032 n=4+5) Template-16 110ms ± 1% 107ms ± 3% -3.00% (p=0.032 n=5+5) TimeParse-16 463ns ± 0% 462ns ± 0% ~ (p=0.810 n=5+4) TimeFormat-16 538ns ± 0% 535ns ± 0% -0.63% (p=0.024 n=5+5) name old speed new speed delta GobDecode-16 70.7MB/s ± 4% 71.4MB/s ± 5% ~ (p=0.452 n=5+5) GobEncode-16 80.7MB/s ± 5% 81.2MB/s ± 2% ~ (p=1.000 n=5+5) Gzip-16 58.2MB/s ± 2% 57.7MB/s ± 2% ~ (p=0.452 n=5+5) Gunzip-16 302MB/s ± 1% 299MB/s ± 1% -0.99% (p=0.008 n=5+5) JSONEncode-16 92.4MB/s ± 1% 89.1MB/s ± 0% -3.63% (p=0.016 n=5+4) JSONDecode-16 20.4MB/s ± 0% 20.3MB/s ± 1% ~ (p=0.135 n=5+5) GoParse-16 10.6MB/s ± 2% 10.8MB/s ± 1% +2.00% (p=0.016 n=5+5) RegexpMatchEasy0_32-16 286MB/s ± 1% 285MB/s ± 3% ~ (p=1.000 n=5+5) RegexpMatchEasy0_1K-16 2.51GB/s ± 1% 2.49GB/s ± 2% ~ (p=0.095 n=5+5) RegexpMatchEasy1_32-16 309MB/s ± 1% 307MB/s ± 1% ~ (p=0.548 n=5+5) RegexpMatchEasy1_1K-16 1.55GB/s ± 2% 1.57GB/s ± 1% ~ (p=0.690 n=5+5) RegexpMatchMedium_32-16 5.68MB/s ± 2% 5.73MB/s ± 1% ~ (p=0.579 n=5+5) RegexpMatchMedium_1K-16 17.5MB/s ± 4% 17.8MB/s ± 4% ~ (p=0.500 n=5+5) RegexpMatchHard_32-16 10.4MB/s ± 3% 10.5MB/s ± 4% ~ (p=0.460 n=5+5) RegexpMatchHard_1K-16 11.5MB/s ± 1% 11.7MB/s ± 2% +1.57% (p=0.032 n=5+5) Revcomp-16 442MB/s ± 0% 433MB/s ± 2% -2.05% (p=0.032 n=4+5) Template-16 17.7MB/s ± 1% 18.2MB/s ± 3% +3.12% (p=0.032 n=5+5) Change-Id: I6972e8f35f2b31f9a42ac473a6bf419a18022558 Reviewed-on: https://go-review.googlesource.com/100935 Run-TryBot: Giovanni Bajo <[email protected]> TryBot-Result: Gobot Gobot <[email protected]> Reviewed-by: Keith Randall <[email protected]>
1 parent 4231110 commit a35ec9a

File tree

9 files changed

+17321
-12200
lines changed

9 files changed

+17321
-12200
lines changed

‎src/cmd/compile/internal/amd64/ssa.go‎

Lines changed: 77 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -398,7 +398,18 @@ func ssaGenValue(s *gc.SSAGenState, v *ssa.Value) {
398398
p.To.Type = obj.TYPE_REG
399399
p.To.Reg = r
400400

401-
case ssa.OpAMD64CMOVQEQ, ssa.OpAMD64CMOVLEQ:
401+
case ssa.OpAMD64CMOVQEQ, ssa.OpAMD64CMOVLEQ, ssa.OpAMD64CMOVWEQ,
402+
ssa.OpAMD64CMOVQLT, ssa.OpAMD64CMOVLLT, ssa.OpAMD64CMOVWLT,
403+
ssa.OpAMD64CMOVQNE, ssa.OpAMD64CMOVLNE, ssa.OpAMD64CMOVWNE,
404+
ssa.OpAMD64CMOVQGT, ssa.OpAMD64CMOVLGT, ssa.OpAMD64CMOVWGT,
405+
ssa.OpAMD64CMOVQLE, ssa.OpAMD64CMOVLLE, ssa.OpAMD64CMOVWLE,
406+
ssa.OpAMD64CMOVQGE, ssa.OpAMD64CMOVLGE, ssa.OpAMD64CMOVWGE,
407+
ssa.OpAMD64CMOVQHI, ssa.OpAMD64CMOVLHI, ssa.OpAMD64CMOVWHI,
408+
ssa.OpAMD64CMOVQLS, ssa.OpAMD64CMOVLLS, ssa.OpAMD64CMOVWLS,
409+
ssa.OpAMD64CMOVQCC, ssa.OpAMD64CMOVLCC, ssa.OpAMD64CMOVWCC,
410+
ssa.OpAMD64CMOVQCS, ssa.OpAMD64CMOVLCS, ssa.OpAMD64CMOVWCS,
411+
ssa.OpAMD64CMOVQGTF, ssa.OpAMD64CMOVLGTF, ssa.OpAMD64CMOVWGTF,
412+
ssa.OpAMD64CMOVQGEF, ssa.OpAMD64CMOVLGEF, ssa.OpAMD64CMOVWGEF:
402413
r := v.Reg()
403414
if r != v.Args[0].Reg() {
404415
v.Fatalf("input[0] and output not in same register %s", v.LongString())
@@ -409,6 +420,71 @@ func ssaGenValue(s *gc.SSAGenState, v *ssa.Value) {
409420
p.To.Type = obj.TYPE_REG
410421
p.To.Reg = r
411422

423+
case ssa.OpAMD64CMOVQNEF, ssa.OpAMD64CMOVLNEF, ssa.OpAMD64CMOVWNEF:
424+
r := v.Reg()
425+
if r != v.Args[0].Reg() {
426+
v.Fatalf("input[0] and output not in same register %s", v.LongString())
427+
}
428+
// Flag condition: ^ZERO || PARITY
429+
// Generate:
430+
// CMOV*NE SRC,DST
431+
// CMOV*PS SRC,DST
432+
p := s.Prog(v.Op.Asm())
433+
p.From.Type = obj.TYPE_REG
434+
p.From.Reg = v.Args[1].Reg()
435+
p.To.Type = obj.TYPE_REG
436+
p.To.Reg = r
437+
var q *obj.Prog
438+
if v.Op == ssa.OpAMD64CMOVQNEF {
439+
q = s.Prog(x86.ACMOVQPS)
440+
} else if v.Op == ssa.OpAMD64CMOVLNEF {
441+
q = s.Prog(x86.ACMOVLPS)
442+
} else {
443+
q = s.Prog(x86.ACMOVWPS)
444+
}
445+
q.From.Type = obj.TYPE_REG
446+
q.From.Reg = v.Args[1].Reg()
447+
q.To.Type = obj.TYPE_REG
448+
q.To.Reg = r
449+
450+
case ssa.OpAMD64CMOVQEQF, ssa.OpAMD64CMOVLEQF, ssa.OpAMD64CMOVWEQF:
451+
r := v.Reg()
452+
if r != v.Args[0].Reg() {
453+
v.Fatalf("input[0] and output not in same register %s", v.LongString())
454+
}
455+
456+
// Flag condition: ZERO && !PARITY
457+
// Generate:
458+
// MOV SRC,AX
459+
// CMOV*NE DST,AX
460+
// CMOV*PC AX,DST
461+
//
462+
// TODO(rasky): we could generate:
463+
// CMOV*NE DST,SRC
464+
// CMOV*PC SRC,DST
465+
// But this requires a way for regalloc to know that SRC might be
466+
// clobbered by this instruction.
467+
if v.Args[1].Reg() != x86.REG_AX {
468+
opregreg(s, moveByType(v.Type), x86.REG_AX, v.Args[1].Reg())
469+
}
470+
p := s.Prog(v.Op.Asm())
471+
p.From.Type = obj.TYPE_REG
472+
p.From.Reg = r
473+
p.To.Type = obj.TYPE_REG
474+
p.To.Reg = x86.REG_AX
475+
var q *obj.Prog
476+
if v.Op == ssa.OpAMD64CMOVQEQF {
477+
q = s.Prog(x86.ACMOVQPC)
478+
} else if v.Op == ssa.OpAMD64CMOVLEQF {
479+
q = s.Prog(x86.ACMOVLPC)
480+
} else {
481+
q = s.Prog(x86.ACMOVWPC)
482+
}
483+
q.From.Type = obj.TYPE_REG
484+
q.From.Reg = x86.REG_AX
485+
q.To.Type = obj.TYPE_REG
486+
q.To.Reg = r
487+
412488
case ssa.OpAMD64MULQconst, ssa.OpAMD64MULLconst:
413489
r := v.Reg()
414490
p := s.Prog(v.Op.Asm())

‎src/cmd/compile/internal/ssa/branchelim.go‎

Lines changed: 21 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -19,7 +19,10 @@ package ssa
1919
// rewrite Phis in the postdominator as CondSelects.
2020
func branchelim(f *Func) {
2121
// FIXME: add support for lowering CondSelects on more architectures
22-
if f.Config.arch != "arm64" {
22+
switch f.Config.arch {
23+
case "arm64", "amd64":
24+
// implemented
25+
default:
2326
return
2427
}
2528

@@ -32,10 +35,22 @@ func branchelim(f *Func) {
3235
}
3336
}
3437

35-
func canCondSelect(v *Value) bool {
38+
func canCondSelect(v *Value, arch string) bool {
3639
// For now, stick to simple scalars that fit in registers
37-
sz := v.Type.Size()
38-
return sz <= v.Block.Func.Config.RegSize && (v.Type.IsInteger() || v.Type.IsPtrShaped())
40+
switch {
41+
case v.Type.Size() > v.Block.Func.Config.RegSize:
42+
return false
43+
case v.Type.IsPtrShaped():
44+
return true
45+
case v.Type.IsInteger():
46+
if arch == "amd64" && v.Type.Size() < 2 {
47+
// amd64 doesn't support CMOV with byte registers
48+
return false
49+
}
50+
return true
51+
default:
52+
return false
53+
}
3954
}
4055

4156
func elimIf(f *Func, dom *Block) bool {
@@ -68,7 +83,7 @@ func elimIf(f *Func, dom *Block) bool {
6883
for _, v := range post.Values {
6984
if v.Op == OpPhi {
7085
hasphis = true
71-
if !canCondSelect(v) {
86+
if !canCondSelect(v, f.Config.arch) {
7287
return false
7388
}
7489
}
@@ -169,7 +184,7 @@ func elimIfElse(f *Func, b *Block) bool {
169184
for _, v := range post.Values {
170185
if v.Op == OpPhi {
171186
hasphis = true
172-
if !canCondSelect(v) {
187+
if !canCondSelect(v, f.Config.arch) {
173188
return false
174189
}
175190
}

0 commit comments

Comments
 (0)