gasm-sdk/parser/parser_test.go

// Copyright (c) 2026 Petr Balvín <opensource@petrbalvin.org> (https://petrbalvin.org)
// SPDX-License-Identifier: BSD-3-Clause

package parser

import (
	"os"
	"path/filepath"
	"testing"

	"sourcedock.dev/petrbalvin/gasm-devkit/ast"
)

func mustParse(t *testing.T, path string) *ast.File {
	t.Helper()
	src, err := os.ReadFile(path)
	if err != nil {
		t.Fatalf("read %s: %v", path, err)
	}
	file, errs := Parse(path, string(src))
	if len(errs) > 0 {
		t.Fatalf("parse %s: %v", path, errs)
	}
	return file
}

func texts(f *ast.File) []*ast.Text {
	var out []*ast.Text
	for _, d := range f.Decls {
		if t, ok := d.(*ast.Text); ok {
			out = append(out, t)
		}
	}
	return out
}

func TestParseSample(t *testing.T) {
	f := mustParse(t, "../testdata/sample_amd64.s")

	// Includes, GLOBL/DATA and two TEXT functions.
	var includes, globls, datas int
	for _, d := range f.Decls {
		switch d.(type) {
		case *ast.Include:
			includes++
		case *ast.Globl:
			globls++
		case *ast.Data:
			datas++
		}
	}
	if includes != 1 {
		t.Errorf("includes = %d, want 1", includes)
	}
	if globls != 2 {
		t.Errorf("globls = %d, want 2", globls)
	}
	if datas != 4 {
		t.Errorf("datas = %d, want 4", datas)
	}

	txts := texts(f)
	if len(txts) != 2 {
		t.Fatalf("text functions = %d, want 2", len(txts))
	}

	fn := txts[0]
	if fn.Name.Name != "analyzeO1RangeAVX2" {
		t.Errorf("name = %q, want analyzeO1RangeAVX2", fn.Name.Name)
	}
	if fn.Name.Pseudo != "SB" {
		t.Errorf("pseudo = %q, want SB", fn.Name.Pseudo)
	}
	if len(fn.Flags) != 1 || fn.Flags[0] != "NOSPLIT" {
		t.Errorf("flags = %v, want [NOSPLIT]", fn.Flags)
	}
	if fn.Frame == nil || !fn.Frame.Imm.HasVal || fn.Frame.Imm.Val != 0 {
		t.Errorf("frame = %+v, want $0", fn.Frame)
	}
	if fn.Args == nil || fn.Args.Imm.Val != 65 {
		t.Errorf("args = %+v, want 65", fn.Args)
	}
	if fn.Doc == "" {
		t.Error("expected a doc comment on the first TEXT")
	}

	// The body must contain the two labels vec1 and vec1done.
	labels := map[string]bool{}
	for _, s := range fn.Body {
		if l, ok := s.(*ast.Label); ok {
			labels[l.Name.Text] = true
		}
	}
	for _, want := range []string{"vec1", "vec1done"} {
		if !labels[want] {
			t.Errorf("missing label %q", want)
		}
	}
}

func TestOperandStructure(t *testing.T) {
	f := mustParse(t, "../testdata/sample_amd64.s")
	fn := texts(f)[0]

	// Index instructions by mnemonic for targeted checks.
	byMnem := map[string]*ast.Instr{}
	for _, s := range fn.Body {
		if in, ok := s.(*ast.Instr); ok {
			byMnem[in.Mnemonic.Text] = in
		}
	}

	// MOVQ swin_base+0(FP), SI — the first MOVQ in the body.
	var mov *ast.Instr
	for _, s := range fn.Body {
		if in, ok := s.(*ast.Instr); ok && in.Mnemonic.Text == "MOVQ" {
			mov = in
			break
		}
	}
	if mov == nil {
		t.Fatal("MOVQ not found")
	}
	src := mov.Operands[0]
	if src.Kind != ast.OpAddr || src.Addr.Sym == nil {
		t.Fatalf("src operand = %+v, want symbol address", src)
	}
	if src.Addr.Sym.Name != "swin_base" || src.Addr.Sym.Pseudo != "FP" || src.Addr.Sym.Offset != 0 {
		t.Errorf("src symbol = %+v, want swin_base+0(FP)", src.Addr.Sym)
	}
	if mov.Operands[1].Addr.Sym.Name != "SI" {
		t.Errorf("dst = %+v, want SI", mov.Operands[1].Addr)
	}

	// LEAQ (SI)(BX*4), R9
	leaq := byMnem["LEAQ"]
	if leaq == nil {
		t.Fatal("LEAQ not found")
	}
	mem := leaq.Operands[0].Addr
	if mem.Base != "SI" || mem.Index != "BX" || mem.Scale != 4 {
		t.Errorf("LEAQ addr = %+v, want base SI index BX scale 4", mem)
	}

	// ANDQ $-8, R10
	andq := byMnem["ANDQ"]
	if andq == nil {
		t.Fatal("ANDQ not found")
	}
	imm := andq.Operands[0]
	if imm.Kind != ast.OpImmediate || !imm.Imm.Neg || imm.Imm.Val != 8 {
		t.Errorf("ANDQ imm = %+v, want -8", imm.Imm)
	}
}

func TestAVX512Operands(t *testing.T) {
	f := mustParse(t, "../testdata/sample_amd64.s")
	fn := texts(f)[1]
	byMnem := map[string]*ast.Instr{}
	for _, s := range fn.Body {
		if in, ok := s.(*ast.Instr); ok {
			byMnem[in.Mnemonic.Text] = in
		}
	}

	// VALIGND $15, Z9, Z0, Z1 — four operands.
	val := byMnem["VALIGND"]
	if val == nil {
		t.Fatal("VALIGND not found")
	}
	if len(val.Operands) != 4 {
		t.Errorf("VALIGND operands = %d, want 4", len(val.Operands))
	}
	if val.Operands[0].Kind != ast.OpImmediate || val.Operands[0].Imm.Val != 15 {
		t.Errorf("VALIGND first operand = %+v, want $15", val.Operands[0])
	}

	// VMOVDQU32 Z0, 4(SI)(AX*1)
	vmov := byMnem["VMOVDQU32"]
	if vmov == nil {
		t.Fatal("VMOVDQU32 not found")
	}
	dst := vmov.Operands[len(vmov.Operands)-1].Addr
	if dst.Offset != 4 || dst.Base != "SI" || dst.Index != "AX" || dst.Scale != 1 {
		t.Errorf("VMOVDQU32 dst = %+v, want 4(SI)(AX*1)", dst)
	}

	// KTESTW K1, K1 — mask registers parse as bare names.
	kt := byMnem["KTESTW"]
	if kt == nil || len(kt.Operands) != 2 {
		t.Fatalf("KTESTW = %+v, want two operands", kt)
	}
}

func TestDataWidthAndStatic(t *testing.T) {
	f := mustParse(t, "../testdata/sample_amd64.s")
	var datas []*ast.Data
	for _, d := range f.Decls {
		if dd, ok := d.(*ast.Data); ok {
			datas = append(datas, dd)
		}
	}
	if datas[0].Width != 4 {
		t.Errorf("first DATA width = %d, want 4", datas[0].Width)
	}
	if datas[0].Name.Pseudo != "SB" || datas[0].Name.Offset != 0 {
		t.Errorf("first DATA name = %+v, want +0(SB)", datas[0].Name)
	}
	if datas[0].Value.Kind != ast.OpImmediate || datas[0].Value.Imm.Val != 1 {
		t.Errorf("first DATA value = %+v, want $1", datas[0].Value)
	}
	// The mask24<> entries are static.
	if !datas[2].Name.Static {
		t.Errorf("mask24 DATA should be static, got %+v", datas[2].Name)
	}
}

// TestParseRealGoLibraries parses every .s file in the sibling go-libraries
// repository when it is checked out, asserting a clean, error-free parse.  It
// is skipped when the repository is not present.
func TestParseRealGoLibraries(t *testing.T) {
	matches, _ := filepath.Glob("../../go-libraries/go-*/*.s")
	if len(matches) == 0 {
		t.Skip("go-libraries repository not present next to gasm-devkit")
	}
	for _, path := range matches {
		src, err := os.ReadFile(path)
		if err != nil {
			t.Fatalf("read %s: %v", path, err)
		}
		file, errs := Parse(path, string(src))
		if len(errs) > 0 {
			t.Errorf("parse %s: %v", path, errs)
			continue
		}
		if len(texts(file)) == 0 {
			t.Errorf("parse %s: no TEXT functions found", path)
		}
		t.Logf("%s: %d decls, %d functions", filepath.Base(path), len(file.Decls), len(texts(file)))
	}
}