cmd/compile: soften type matching when allocating stack slots

Currently we use pointer equality on types when deciding whether we can
reuse a stack slot. That's too strict, as we don't guarantee pointer
equality for the same type. In particular, it can vary based on whether
PtrTo has been called in the frontend or not.

Instead, use the type's LinkString, which is guaranteed to both be
unique for a type, and to not vary given two different type structures
describing the same type.

Update #65783

Change-Id: I64f55138475f04bfa30cfb819b786b7cc06aebe4
Reviewed-on: https://go-review.googlesource.com/c/go/+/565436
Reviewed-by: Keith Randall <khr@google.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Keith Randall <khr@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>

1 files changed, 29 insertions, 0 deletions

diff --git a/test/codegen/stack.go b/test/codegen/stack.go
index eebbbf1677..65c9868d67 100644
--- a/test/codegen/stack.go
+++ b/test/codegen/stack.go
@@ -113,3 +113,32 @@ func Defer() {
 	// amd64:`CALL\truntime\.deferprocStack`
 	defer func() {}()
 }
+
+// Check that stack slots are shared among values of the same
+// type, but not pointer-identical types. See issue 65783.
+
+func spillSlotReuse() {
+	// The return values of getp1 and getp2 need to be
+	// spilled around the calls to nopInt. Make sure that
+	// spill slot gets reused.
+
+	//arm64:`.*autotmp_2-8\(SP\)`
+	getp1()[nopInt()] = 0
+	//arm64:`.*autotmp_2-8\(SP\)`
+	getp2()[nopInt()] = 0
+}
+
+//go:noinline
+func nopInt() int {
+	return 0
+}
+
+//go:noinline
+func getp1() *[4]int {
+	return nil
+}
+
+//go:noinline
+func getp2() *[4]int {
+	return nil
+}