compiler: add support for new unsafe slice/string functions

This adds support for unsafe.SliceData, unsafe.String, and
unsafe.SringData that were introduced in Go 1.20.

5 files changed, 116 insertions, 18 deletions

diff --git a/compiler/asserts.go b/compiler/asserts.go
index ba4824943..0fb112e0b 100644
--- a/compiler/asserts.go
+++ b/compiler/asserts.go
@@ -89,10 +89,11 @@ func (b *builder) createSliceToArrayPointerCheck(sliceLen llvm.Value, arrayLen i
 	b.createRuntimeAssert(isLess, "slicetoarray", "sliceToArrayPointerPanic")
 }
 
-// createUnsafeSliceCheck inserts a runtime check used for unsafe.Slice. This
-// function must panic if the ptr/len parameters are invalid.
-func (b *builder) createUnsafeSliceCheck(ptr, len llvm.Value, elementType llvm.Type, lenType *types.Basic) {
-	// From the documentation of unsafe.Slice:
+// createUnsafeSliceStringCheck inserts a runtime check used for unsafe.Slice
+// and unsafe.String. This function must panic if the ptr/len parameters are
+// invalid.
+func (b *builder) createUnsafeSliceStringCheck(name string, ptr, len llvm.Value, elementType llvm.Type, lenType *types.Basic) {
+	// From the documentation of unsafe.Slice and unsafe.String:
 	//   > At run time, if len is negative, or if ptr is nil and len is not
 	//   > zero, a run-time panic occurs.
 	// However, in practice, it is also necessary to check that the length is
@@ -117,7 +118,7 @@ func (b *builder) createUnsafeSliceCheck(ptr, len llvm.Value, elementType llvm.T
 	lenIsNotZero := b.CreateICmp(llvm.IntNE, len, zero, "")
 	assert := b.CreateAnd(ptrIsNil, lenIsNotZero, "")
 	assert = b.CreateOr(assert, lenOutOfBounds, "")
-	b.createRuntimeAssert(assert, "unsafe.Slice", "unsafeSlicePanic")
+	b.createRuntimeAssert(assert, name, "unsafeSlicePanic")
 }
 
 // createChanBoundsCheck creates a bounds check before creating a new channel to
diff --git a/compiler/compiler.go b/compiler/compiler.go
index 4abfdd559..aad5e0d9a 100644
--- a/compiler/compiler.go
+++ b/compiler/compiler.go
@@ -1646,20 +1646,20 @@ func (b *builder) createBuiltin(argTypes []types.Type, argValues []llvm.Value, c
 	case "Sizeof": // unsafe.Sizeof
 		size := b.targetData.TypeAllocSize(argValues[0].Type())
 		return llvm.ConstInt(b.uintptrType, size, false), nil
-	case "Slice": // unsafe.Slice
-		// This creates a slice from a pointer and a length.
+	case "Slice", "String": // unsafe.Slice, unsafe.String
+		// This creates a slice or string from a pointer and a length.
 		// Note that the exception mentioned in the documentation (if the
 		// pointer and length are nil, the slice is also nil) is trivially
 		// already the case.
 		ptr := argValues[0]
 		len := argValues[1]
-		slice := llvm.Undef(b.ctx.StructType([]llvm.Type{
-			ptr.Type(),
-			b.uintptrType,
-			b.uintptrType,
-		}, false))
-		elementType := b.getLLVMType(argTypes[0].Underlying().(*types.Pointer).Elem())
-		b.createUnsafeSliceCheck(ptr, len, elementType, argTypes[1].Underlying().(*types.Basic))
+		var elementType llvm.Type
+		if callName == "Slice" {
+			elementType = b.getLLVMType(argTypes[0].Underlying().(*types.Pointer).Elem())
+		} else {
+			elementType = b.ctx.Int8Type()
+		}
+		b.createUnsafeSliceStringCheck("unsafe."+callName, ptr, len, elementType, argTypes[1].Underlying().(*types.Basic))
 		if len.Type().IntTypeWidth() < b.uintptrType.IntTypeWidth() {
 			// Too small, zero-extend len.
 			len = b.CreateZExt(len, b.uintptrType, "")
@@ -1667,10 +1667,24 @@ func (b *builder) createBuiltin(argTypes []types.Type, argValues []llvm.Value, c
 			// Too big, truncate len.
 			len = b.CreateTrunc(len, b.uintptrType, "")
 		}
-		slice = b.CreateInsertValue(slice, ptr, 0, "")
-		slice = b.CreateInsertValue(slice, len, 1, "")
-		slice = b.CreateInsertValue(slice, len, 2, "")
-		return slice, nil
+		if callName == "Slice" {
+			slice := llvm.Undef(b.ctx.StructType([]llvm.Type{
+				ptr.Type(),
+				b.uintptrType,
+				b.uintptrType,
+			}, false))
+			slice = b.CreateInsertValue(slice, ptr, 0, "")
+			slice = b.CreateInsertValue(slice, len, 1, "")
+			slice = b.CreateInsertValue(slice, len, 2, "")
+			return slice, nil
+		} else {
+			str := llvm.Undef(b.getLLVMRuntimeType("_string"))
+			str = b.CreateInsertValue(str, argValues[0], 0, "")
+			str = b.CreateInsertValue(str, len, 1, "")
+			return str, nil
+		}
+	case "SliceData", "StringData": // unsafe.SliceData, unsafe.StringData
+		return b.CreateExtractValue(argValues[0], 0, "slice.data"), nil
 	default:
 		return llvm.Value{}, b.makeError(pos, "todo: builtin: "+callName)
 	}
diff --git a/compiler/compiler_test.go b/compiler/compiler_test.go
index a5cd9c6bf..f8221a08e 100644
--- a/compiler/compiler_test.go
+++ b/compiler/compiler_test.go
@@ -9,6 +9,7 @@ import (
 	"testing"
 
 	"github.com/tinygo-org/tinygo/compileopts"
+	"github.com/tinygo-org/tinygo/goenv"
 	"github.com/tinygo-org/tinygo/loader"
 	"tinygo.org/x/go-llvm"
 )
@@ -27,6 +28,12 @@ type testCase struct {
 func TestCompiler(t *testing.T) {
 	t.Parallel()
 
+	// Determine Go minor version (e.g. 16 in go1.16.3).
+	_, goMinor, err := goenv.GetGorootVersion(goenv.Get("GOROOT"))
+	if err != nil {
+		t.Fatal("could not read Go version:", err)
+	}
+
 	// Determine which tests to run, depending on the Go and LLVM versions.
 	tests := []testCase{
 		{"basic.go", "", ""},
@@ -43,6 +50,9 @@ func TestCompiler(t *testing.T) {
 		{"channel.go", "", ""},
 		{"gc.go", "", ""},
 	}
+	if goMinor >= 20 {
+		tests = append(tests, testCase{"go1.20.go", "", ""})
+	}
 
 	for _, tc := range tests {
 		name := tc.file
diff --git a/compiler/testdata/go1.20.go b/compiler/testdata/go1.20.go
new file mode 100644
index 000000000..be65363ec
--- /dev/null
+++ b/compiler/testdata/go1.20.go
@@ -0,0 +1,15 @@
+package main
+
+import "unsafe"
+
+func unsafeSliceData(s []int) *int {
+	return unsafe.SliceData(s)
+}
+
+func unsafeString(ptr *byte, len int16) string {
+	return unsafe.String(ptr, len)
+}
+
+func unsafeStringData(s string) *byte {
+	return unsafe.StringData(s)
+}
diff --git a/compiler/testdata/go1.20.ll b/compiler/testdata/go1.20.ll
new file mode 100644
index 000000000..fc66b47c5
--- /dev/null
+++ b/compiler/testdata/go1.20.ll
@@ -0,0 +1,58 @@
+; ModuleID = 'go1.20.go'
+source_filename = "go1.20.go"
+target datalayout = "e-m:e-p:32:32-p10:8:8-p20:8:8-i64:64-n32:64-S128-ni:1:10:20"
+target triple = "wasm32-unknown-wasi"
+
+%runtime._string = type { ptr, i32 }
+
+declare noalias nonnull ptr @runtime.alloc(i32, ptr, ptr) #0
+
+declare void @runtime.trackPointer(ptr nocapture readonly, ptr) #0
+
+; Function Attrs: nounwind
+define hidden void @main.init(ptr %context) unnamed_addr #1 {
+entry:
+  ret void
+}
+
+; Function Attrs: nounwind
+define hidden ptr @main.unsafeSliceData(ptr %s.data, i32 %s.len, i32 %s.cap, ptr %context) unnamed_addr #1 {
+entry:
+  call void @runtime.trackPointer(ptr %s.data, ptr undef) #2
+  ret ptr %s.data
+}
+
+; Function Attrs: nounwind
+define hidden %runtime._string @main.unsafeString(ptr dereferenceable_or_null(1) %ptr, i16 %len, ptr %context) unnamed_addr #1 {
+entry:
+  %0 = icmp slt i16 %len, 0
+  %1 = icmp eq ptr %ptr, null
+  %2 = icmp ne i16 %len, 0
+  %3 = and i1 %1, %2
+  %4 = or i1 %3, %0
+  br i1 %4, label %unsafe.String.throw, label %unsafe.String.next
+
+unsafe.String.next:                               ; preds = %entry
+  %5 = zext i16 %len to i32
+  %6 = insertvalue %runtime._string undef, ptr %ptr, 0
+  %7 = insertvalue %runtime._string %6, i32 %5, 1
+  call void @runtime.trackPointer(ptr %ptr, ptr undef) #2
+  ret %runtime._string %7
+
+unsafe.String.throw:                              ; preds = %entry
+  call void @runtime.unsafeSlicePanic(ptr undef) #2
+  unreachable
+}
+
+declare void @runtime.unsafeSlicePanic(ptr) #0
+
+; Function Attrs: nounwind
+define hidden ptr @main.unsafeStringData(ptr %s.data, i32 %s.len, ptr %context) unnamed_addr #1 {
+entry:
+  call void @runtime.trackPointer(ptr %s.data, ptr undef) #2
+  ret ptr %s.data
+}
+
+attributes #0 = { "target-features"="+bulk-memory,+nontrapping-fptoint,+sign-ext" }
+attributes #1 = { nounwind "target-features"="+bulk-memory,+nontrapping-fptoint,+sign-ext" }
+attributes #2 = { nounwind }