cmd/compile: implement comparisons directly with memory

Allow the compiler to generate code like CMPQ 16(AX), $7

It's tricky because it's difficult to spill such a comparison during
flagalloc, because the same memory state might not be available at
the restore locations.

Solve this problem by decomposing the compare+load back into its parts
if it needs to be spilled.

The big win is that the write barrier test goes from:

MOVL	runtime.writeBarrier(SB), CX
TESTL	CX, CX
JNE	60

to

CMPL	runtime.writeBarrier(SB), $0
JNE	59

It's one instruction and one byte smaller.

Fixes #19485
Fixes #15245
Update #22460

Binaries are about 0.15% smaller.

Change-Id: I4fd8d1111b6b9924d52f9a0901ca1b2e5cce0836
Reviewed-on: https://go-review.googlesource.com/86035
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Ilya Tocar <ilya.tocar@intel.com>

1 files changed, 13 insertions, 0 deletions

diff --git a/src/cmd/compile/internal/ssa/func.go b/src/cmd/compile/internal/ssa/func.go
index 62550df0cc..7e8f68bf87 100644
--- a/src/cmd/compile/internal/ssa/func.go
+++ b/src/cmd/compile/internal/ssa/func.go
@@ -350,6 +350,19 @@ func (b *Block) NewValue2I(pos src.XPos, op Op, t *types.Type, auxint int64, arg
 	return v
 }
 
+// NewValue2IA returns a new value in the block with two arguments and both an auxint and aux values.
+func (b *Block) NewValue2IA(pos src.XPos, op Op, t *types.Type, auxint int64, aux interface{}, arg0, arg1 *Value) *Value {
+	v := b.Func.newValue(op, t, b, pos)
+	v.AuxInt = auxint
+	v.Aux = aux
+	v.Args = v.argstorage[:2]
+	v.argstorage[0] = arg0
+	v.argstorage[1] = arg1
+	arg0.Uses++
+	arg1.Uses++
+	return v
+}
+
 // NewValue3 returns a new value in the block with three arguments and zero aux values.
 func (b *Block) NewValue3(pos src.XPos, op Op, t *types.Type, arg0, arg1, arg2 *Value) *Value {
 	v := b.Func.newValue(op, t, b, pos)