cmd/compile: fold negation into multiplication

goos: linux
goarch: riscv64
pkg: cmd/compile/internal/test
cpu: Spacemit(R) X60
        │ /root/mul.base.log │          /root/mul.new.log          │
        │       sec/op       │   sec/op     vs base                │
MulNeg           6.426µ ± 0%   4.501µ ± 0%  -29.96% (p=0.000 n=10)
Mul2Neg          9.000µ ± 0%   6.431µ ± 0%  -28.54% (p=0.000 n=10)
Mul2             1.263µ ± 0%   1.263µ ± 0%        ~ (p=1.000 n=10)
MulNeg2          1.577µ ± 0%   1.577µ ± 0%        ~ (p=0.211 n=10)
geomean          3.276µ        2.756µ       -15.89%

goos: linux
goarch: amd64
pkg: cmd/compile/internal/test
cpu: AMD EPYC 7532 32-Core Processor
        │ /root/base  │              /root/new              │
        │   sec/op    │   sec/op     vs base                │
MulNeg    691.9n ± 1%   319.4n ± 0%  -53.83% (p=0.000 n=10)
Mul2Neg   630.0n ± 0%   629.6n ± 0%   -0.07% (p=0.000 n=10)
Mul2      438.1n ± 0%   438.1n ± 0%        ~ (p=0.728 n=10)
MulNeg2   439.3n ± 0%   439.4n ± 0%        ~ (p=0.656 n=10)
geomean   538.2n        443.6n       -17.58%

Change-Id: Ice8e6c8d1e8e3009ba8a0b1b689205174e199019
Reviewed-on: https://go-review.googlesource.com/c/go/+/720180
Reviewed-by: abner chenc <[email protected]>
Reviewed-by: Keith Randall <[email protected]>
Reviewed-by: Junyang Shao <[email protected]>
Reviewed-by: Joel Sing <[email protected]>
Reviewed-by: Keith Randall <[email protected]>
LUCI-TryBot-Result: Go LUCI <[email protected]>
Auto-Submit: Keith Randall <[email protected]>

Author: mengzhuo

src/cmd/compile/internal/ssa/_gen/LOONG64.rules

@@ -743,9 +743,6 @@
 
 (MULV  x (MOVVconst [c])) && canMulStrengthReduce(config, c) => {mulStrengthReduce(v, x, c)}
 
-(MULV (NEGV x) (MOVVconst [c])) => (MULV x (MOVVconst [-c]))
-(MULV (NEGV x) (NEGV y)) => (MULV x y)
-
 (ADDV x0 x1:(SLLVconst [c] y)) && x1.Uses == 1 && c > 0 && c <= 4 => (ADDshiftLLV x0 y [c])
 
 // fold constant in ADDshift op

src/cmd/compile/internal/ssa/_gen/generic.rules

@@ -195,6 +195,11 @@
 // Convert x * -1 to -x.
 (Mul(8|16|32|64)  (Const(8|16|32|64)  [-1]) x) => (Neg(8|16|32|64)  x)
 
+// Convert -x * c to x * -c
+(Mul(8|16|32|64) (Const(8|16|32|64) <t> [c]) (Neg(8|16|32|64) x)) => (Mul(8|16|32|64) x (Const(8|16|32|64) <t> [-c]))
+
+(Mul(8|16|32|64) (Neg(8|16|32|64) x) (Neg(8|16|32|64) y)) => (Mul(8|16|32|64) x y)
+
 // DeMorgan's Laws
 (And(8|16|32|64) <t> (Com(8|16|32|64) x) (Com(8|16|32|64) y)) => (Com(8|16|32|64) (Or(8|16|32|64) <t> x y))
 (Or(8|16|32|64) <t> (Com(8|16|32|64) x) (Com(8|16|32|64) y)) => (Com(8|16|32|64) (And(8|16|32|64) <t> x y))
@@ -2228,4 +2233,4 @@
 (Neq(64|32|16) (SignExt8to(64|32|16) (CvtBoolToUint8 x)) (Const(64|32|16) [0])) => x
 (Neq(64|32|16) (SignExt8to(64|32|16) (CvtBoolToUint8 x)) (Const(64|32|16) [1])) => (Not x)
 (Eq(64|32|16)  (SignExt8to(64|32|16) (CvtBoolToUint8 x)) (Const(64|32|16) [1])) => x
-(Eq(64|32|16)  (SignExt8to(64|32|16) (CvtBoolToUint8 x)) (Const(64|32|16) [0])) => (Not x)
+(Eq(64|32|16)  (SignExt8to(64|32|16) (CvtBoolToUint8 x)) (Const(64|32|16) [0])) => (Not x)

src/cmd/compile/internal/ssa/rewriteLOONG64.go

@@ -318,13 +318,19 @@ func MergeMuls5(a, n int) int {
 // Multiplications folded negation
 
 func FoldNegMul(a int) int {
-	// loong64:"SUBVU" "ALSLV [$]2" "ALSLV [$]1"
-	return (-a) * 11
+	// amd64:"IMUL3Q [$]-11" -"NEGQ"
+	// arm64:"MOVD [$]-11" "MUL" -"NEG"
+	// loong64:"ALSLV [$]2" "SUBVU" "ALSLV [$]4"
+	// riscv64:"MOV [$]-11" "MUL" -"NEG"
+	return -a * 11
 }
 
 func Fold2NegMul(a, b int) int {
+	// amd64:"IMULQ" -"NEGQ"
+	// arm64:"MUL" -"NEG"
 	// loong64:"MULV" -"SUBVU R[0-9], R0,"
-	return (-a) * (-b)
+	// riscv64:"MUL" -"NEG"
+	return -a * -b
 }
 
 // -------------- //