cmd/compile: generate subfic on ppc64

This merges an lis + subf into subfic, and for 32b constants
lwa + subf into oris + ori + subf.

The carry bit is no longer used in code generation, therefore
I think we can clobber it as needed.  Note, lowered borrow/carry
arithmetic is self-contained and thus is not affected.

A few extra rules are added to ensure early transformations to
SUBFCconst don't trip up earlier rules, fold constant operations,
or otherwise simplify lowering.  Likewise, tests are added to
ensure all rules are hit.  Generic constant folding catches
trivial cases, however some lowering rules insert arithmetic
which can introduce new opportunities (e.g BitLen or Slicemask).

I couldn't find a specific benchmark to demonstrate noteworthy
improvements, but this is generating subfic in many of the default
bent test binaries, so we are at least saving a little code space.

Change-Id: Iad7c6e5767eaa9dc24dc1c989bd1c8cfe1982012
Reviewed-on: https://go-review.googlesource.com/c/go/+/249461
Run-TryBot: Lynn Boger <laboger@linux.vnet.ibm.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Carlos Eduardo Seo <cseo@linux.vnet.ibm.com>

3 files changed, 77 insertions, 0 deletions

diff --git a/test/codegen/arithmetic.go b/test/codegen/arithmetic.go
index afd4d66bd9..0bdb66a376 100644
--- a/test/codegen/arithmetic.go
+++ b/test/codegen/arithmetic.go
@@ -42,6 +42,48 @@ func SubMem(arr []int, b, c, d int) int {
 	return arr[0] - arr[1]
 }
 
+func SubFromConst(a int) int {
+	// ppc64le: `SUBC\tR[0-9]+,\s[$]40,\sR`
+	// ppc64: `SUBC\tR[0-9]+,\s[$]40,\sR`
+	b := 40 - a
+	return b
+}
+
+func SubFromConstNeg(a int) int {
+	// ppc64le: `ADD\t[$]40,\sR[0-9]+,\sR`
+	// ppc64: `ADD\t[$]40,\sR[0-9]+,\sR`
+	c := 40 - (-a)
+	return c
+}
+
+func SubSubFromConst(a int) int {
+	// ppc64le: `ADD\t[$]20,\sR[0-9]+,\sR`
+	// ppc64: `ADD\t[$]20,\sR[0-9]+,\sR`
+	c := 40 - (20 - a)
+	return c
+}
+
+func AddSubFromConst(a int) int {
+	// ppc64le: `SUBC\tR[0-9]+,\s[$]60,\sR`
+	// ppc64: `SUBC\tR[0-9]+,\s[$]60,\sR`
+	c := 40 + (20 - a)
+	return c
+}
+
+func NegSubFromConst(a int) int {
+	// ppc64le: `ADD\t[$]-20,\sR[0-9]+,\sR`
+	// ppc64: `ADD\t[$]-20,\sR[0-9]+,\sR`
+	c := -(20 - a)
+	return c
+}
+
+func NegAddFromConstNeg(a int) int {
+	// ppc64le: `SUBC\tR[0-9]+,\s[$]40,\sR`
+	// ppc64: `SUBC\tR[0-9]+,\s[$]40,\sR`
+	c := -(-40 + a)
+	return c
+}
+
 // -------------------- //
 //    Multiplication    //
 // -------------------- //
diff --git a/test/codegen/mathbits.go b/test/codegen/mathbits.go
index 942605de55..4c35f26997 100644
--- a/test/codegen/mathbits.go
+++ b/test/codegen/mathbits.go
@@ -76,9 +76,17 @@ func Len64(n uint64) int {
 	// arm:"CLZ" arm64:"CLZ"
 	// mips:"CLZ"
 	// wasm:"I64Clz"
+	// ppc64le:"SUBC","CNTLZD"
+	// ppc64:"SUBC","CNTLZD"
 	return bits.Len64(n)
 }
 
+func SubFromLen64(n uint64) int {
+	// ppc64le:"CNTLZD",-"SUBC"
+	// ppc64:"CNTLZD",-"SUBC"
+	return 64 - bits.Len64(n)
+}
+
 func Len32(n uint32) int {
 	// amd64:"BSRQ","LEAQ",-"CMOVQEQ"
 	// s390x:"FLOGR"
@@ -291,6 +299,12 @@ func TrailingZeros64(n uint64) int {
 	return bits.TrailingZeros64(n)
 }
 
+func TrailingZeros64Subtract(n uint64) int {
+	// ppc64le/power8:"NEG","SUBC","ANDN","POPCNTD"
+	// ppc64le/power9:"SUBC","CNTTZD"
+	return bits.TrailingZeros64(1 - n)
+}
+
 func TrailingZeros32(n uint32) int {
 	// amd64:"BTSQ\\t\\$32","BSFQ"
 	// arm:"CLZ"
diff --git a/test/codegen/slices.go b/test/codegen/slices.go
index 40e857f9f6..38e8a62f4b 100644
--- a/test/codegen/slices.go
+++ b/test/codegen/slices.go
@@ -347,3 +347,24 @@ func InitNotSmallSliceLiteral() []int {
 		42,
 	}
 }
+
+// --------------------------------------- //
+//   Test PPC64 SUBFCconst folding rules   //
+//   triggered by slice operations.        //
+// --------------------------------------- //
+
+func SliceWithConstCompare(a []int, b int) []int {
+	var c []int = []int{1, 2, 3, 4, 5}
+	if b+len(a) < len(c) {
+		// ppc64le:-"NEG"
+		// ppc64:-"NEG"
+		return c[b:]
+	}
+	return a
+}
+
+func SliceWithSubtractBound(a []int, b int) []int {
+	// ppc64le:"SUBC",-"NEG"
+	// ppc64:"SUBC",-"NEG"
+	return a[(3 - b):]
+}