cmd/asm: add test for verification of instruction encodings

Not much testing yet, but the test now exists.

Another step toward #13822.

Change-Id: Idb2b06bf53a6113c83008150b4c0b631bb195279
Reviewed-on: https://go-review.googlesource.com/18844
Reviewed-by: Rob Pike <r@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>

3 files changed, 131 insertions, 19 deletions

diff --git a/src/cmd/asm/internal/asm/endtoend_test.go b/src/cmd/asm/internal/asm/endtoend_test.go
index bba82b5fca..620bb97417 100644
--- a/src/cmd/asm/internal/asm/endtoend_test.go
+++ b/src/cmd/asm/internal/asm/endtoend_test.go
@@ -10,6 +10,7 @@ import (
 	"io/ioutil"
 	"os"
 	"path/filepath"
+	"sort"
 	"strconv"
 	"strings"
 	"testing"
@@ -22,9 +23,9 @@ import (
 // Output is generated by, in effect, turning on -S and comparing the
 // result against a golden file.
 
-func testEndToEnd(t *testing.T, goarch string) {
+func testEndToEnd(t *testing.T, goarch, file string) {
 	lex.InitHist()
-	input := filepath.Join("testdata", goarch+".s")
+	input := filepath.Join("testdata", file+".s")
 	architecture, ctxt := setArch(goarch)
 	lexer := lex.NewLexer(input, ctxt)
 	parser := NewParser(ctxt, architecture, lexer)
@@ -33,23 +34,31 @@ func testEndToEnd(t *testing.T, goarch string) {
 	testOut = new(bytes.Buffer) // The assembler writes test output to this buffer.
 	ctxt.Bso = obj.Binitw(os.Stdout)
 	defer ctxt.Bso.Flush()
-	ctxt.Diag = t.Errorf
+	failed := false
+	ctxt.Diag = func(format string, args ...interface{}) {
+		failed = true
+		t.Errorf(format, args...)
+	}
 	obj.Binitw(ioutil.Discard)
 	pList.Firstpc, ok = parser.Parse()
-	if !ok || t.Failed() {
-		t.Fatalf("asm: %s assembly failed", goarch)
+	if !ok || failed {
+		t.Errorf("asm: %s assembly failed", goarch)
+		return
 	}
 	output := strings.Split(testOut.String(), "\n")
 
 	// Reconstruct expected output by independently "parsing" the input.
 	data, err := ioutil.ReadFile(input)
 	if err != nil {
-		t.Fatal(err)
+		t.Error(err)
+		return
 	}
 	lineno := 0
 	seq := 0
+	hexByLine := map[string]string{}
+	lines := strings.SplitAfter(string(data), "\n")
 Diff:
-	for _, line := range strings.SplitAfter(string(data), "\n") {
+	for _, line := range lines {
 		lineno++
 
 		// The general form of a test input line is:
@@ -62,14 +71,31 @@ Diff:
 		}
 		seq++
 
+		var hexes string
 		switch len(parts) {
 		default:
 			t.Errorf("%s:%d: unable to understand comments: %s", input, lineno, line)
 		case 1:
 			// no comment
 		case 2:
-			// one comment, printed form
+			// might be printed form or hex
+			note := strings.TrimSpace(parts[1])
+			if isHexes(note) {
+				hexes = note
+			} else {
+				printed = note
+			}
+		case 3:
+			// printed form, then hex
 			printed = strings.TrimSpace(parts[1])
+			hexes = strings.TrimSpace(parts[2])
+			if !isHexes(hexes) {
+				t.Errorf("%s:%d: malformed hex instruction encoding: %s", input, lineno, line)
+			}
+		}
+
+		if hexes != "" {
+			hexByLine[fmt.Sprintf("%s:%d", input, lineno)] = hexes
 		}
 
 		// Canonicalize spacing in printed form.
@@ -142,28 +168,114 @@ Diff:
 		t.Errorf("unexpected output: %q", output[0])
 		output = output[1:]
 	}
+
+	// Checked printing.
+	// Now check machine code layout.
+
+	top := pList.Firstpc
+	var text *obj.LSym
+	ok = true
+	ctxt.Diag = func(format string, args ...interface{}) {
+		t.Errorf(format, args...)
+		ok = false
+	}
+	obj.Flushplist(ctxt)
+
+	for p := top; p != nil; p = p.Link {
+		if p.As == obj.ATEXT {
+			text = p.From.Sym
+		}
+		hexes := hexByLine[p.Line()]
+		if hexes == "" {
+			continue
+		}
+		delete(hexByLine, p.Line())
+		if text == nil {
+			t.Errorf("%s: instruction outside TEXT", p)
+		}
+		size := int64(len(text.P)) - p.Pc
+		if p.Link != nil {
+			size = p.Link.Pc - p.Pc
+		} else if p.Isize != 0 {
+			size = int64(p.Isize)
+		}
+		var code []byte
+		if p.Pc < int64(len(text.P)) {
+			code = text.P[p.Pc:]
+			if size < int64(len(code)) {
+				code = code[:size]
+			}
+		}
+		codeHex := fmt.Sprintf("%x", code)
+		if codeHex == "" {
+			codeHex = "empty"
+		}
+		ok := false
+		for _, hex := range strings.Split(hexes, " or ") {
+			if codeHex == hex {
+				ok = true
+				break
+			}
+		}
+		if !ok {
+			t.Errorf("%s: have encoding %s, want %s", p, codeHex, hexes)
+		}
+	}
+
+	if len(hexByLine) > 0 {
+		var missing []string
+		for key := range hexByLine {
+			missing = append(missing, key)
+		}
+		sort.Strings(missing)
+		for _, line := range missing {
+			t.Errorf("%s: did not find instruction encoding", line)
+		}
+	}
+
 }
 
-func TestPPC64EndToEnd(t *testing.T) {
-	testEndToEnd(t, "ppc64")
+func isHexes(s string) bool {
+	if s == "" {
+		return false
+	}
+	if s == "empty" {
+		return true
+	}
+	for _, f := range strings.Split(s, " or ") {
+		if f == "" || len(f)%2 != 0 || strings.TrimLeft(f, "0123456789abcdef") != "" {
+			return false
+		}
+	}
+	return true
+}
+
+func Test386EndToEnd(t *testing.T) {
+	testEndToEnd(t, "386", "386")
 }
 
 func TestARMEndToEnd(t *testing.T) {
-	testEndToEnd(t, "arm")
+	defer os.Setenv("GOARM", os.Getenv("GOARM"))
+
+	for _, goarm := range []string{"5", "6", "7"} {
+		os.Setenv("GOARM", goarm)
+		t.Logf("GOARM=%v", os.Getenv("GOARM"))
+		testEndToEnd(t, "arm", "arm")
+	}
 }
 
 func TestARM64EndToEnd(t *testing.T) {
-	testEndToEnd(t, "arm64")
+	testEndToEnd(t, "arm64", "arm64")
 }
 
 func TestAMD64EndToEnd(t *testing.T) {
-	testEndToEnd(t, "amd64")
+	testEndToEnd(t, "amd64", "amd64")
 }
 
-func Test386EndToEnd(t *testing.T) {
-	testEndToEnd(t, "386")
+func TestMIPS64EndToEnd(t *testing.T) {
+	testEndToEnd(t, "mips64", "mips64")
 }
 
-func TestMIPS64EndToEnd(t *testing.T) {
-	testEndToEnd(t, "mips64")
+func TestPPC64EndToEnd(t *testing.T) {
+	testEndToEnd(t, "ppc64", "ppc64")
 }
diff --git a/src/cmd/asm/internal/asm/testdata/amd64.s b/src/cmd/asm/internal/asm/testdata/amd64.s
index 35af32a5cd..5512df0034 100644
--- a/src/cmd/asm/internal/asm/testdata/amd64.s
+++ b/src/cmd/asm/internal/asm/testdata/amd64.s
@@ -122,4 +122,4 @@ loop:
 	LOOP	loop // LOOP
 
 // LTYPE0 nonnon	{ outcode($1, &$2); }
-	RET
+	RET // c3
diff --git a/src/cmd/asm/internal/asm/testdata/arm.s b/src/cmd/asm/internal/asm/testdata/arm.s
index 0102cd8414..8062750250 100644
--- a/src/cmd/asm/internal/asm/testdata/arm.s
+++ b/src/cmd/asm/internal/asm/testdata/arm.s
@@ -189,7 +189,7 @@ TEXT	foo(SB), 7, $0
 //		outcode($1, $2, &$3, 0, &$5);
 //	}
 	ADDD.S	F1, F2
-	MOVF.S	$0.5, F2 // MOVF.S $(0.5), F2
+	MOVF	$0.5, F2 // MOVF $(0.5), F2
 
 //	LTYPEK cond frcon ',' LFREG ',' freg
 //	{