What causes "text section not found" in golang/go?

Verify the `.text` section exists: `dumpbin /headers ` or use a PE viewer.

text section not found

Error message

text section not found

What it means

When analyzing a PE (Windows) binary, `peFile.text()` retrieves the `.text` section containing machine code. Before that it calls `imageBase()` which can itself fail. If `.text` section does not exist in the PE file, this error is returned. Standard Windows executables always have a `.text` section.

Source

Thrown at src/cmd/internal/objfile/pe.go:146

		// We didn't find the symbols, so look for the names used in 1.3 and earlier.
		// TODO: Remove code looking for the old symbols when we no longer care about 1.3.
		var err2 error
		if pclntab, err2 = loadPETable(f.pe, "pclntab", "epclntab"); err2 != nil {
			return 0, nil, err
		}
	}
	return textStart, pclntab, nil
}

func (f *peFile) text() (textStart uint64, text []byte, err error) {
	imageBase, err := f.imageBase()
	if err != nil {
		return 0, nil, err
	}

	sect := f.pe.Section(".text")
	if sect == nil {
		return 0, nil, fmt.Errorf("text section not found")
	}
	textStart = imageBase + uint64(sect.VirtualAddress)
	text, err = sect.Data()
	return
}

func findPESymbol(f *pe.File, name string) (*pe.Symbol, error) {
	for _, s := range f.Symbols {
		if s.Name != name {
			continue
		}
		if s.SectionNumber <= 0 {
			return nil, fmt.Errorf("symbol %s: invalid section number %d", name, s.SectionNumber)
		}
		if len(f.Sections) < int(s.SectionNumber) {
			return nil, fmt.Errorf("symbol %s: section number %d is larger than max %d", name, s.SectionNumber, len(f.Sections))
		}
		return s, nil

View on GitHub (pinned to b6b368adc5)

Solutions

Verify the `.text` section exists: `dumpbin /headers <file>` or use a PE viewer.
Rebuild the binary without packers/obfuscators that rename sections.
If analyzing a resource DLL or non-code PE, use the appropriate tool rather than `objfile`.
Unpack or restore original section names if the binary was processed by UPX or similar.

Example fix

// before — packed PE with renamed sections
f, err := objfile.Open("packed.exe")
_, _, err := f.Text()  // fails: no .text section

// after — unpack first or rebuild
$ upx -d packed.exe  # decompress
f, err := objfile.Open("packed.exe")
_, _, err := f.Text()  // succeeds

Defensive patterns

Strategy: try-catch

Validate before calling

// Check for .text section in PE
func hasPETextSection(path string) bool {
    f, err := pe.Open(path)
    if err != nil { return false }
    defer f.Close()
    for _, s := range f.Sections {
        if s.Name == ".text" { return true }
    }
    return false
}

Try / catch

_, text, err := f.Text()
if err != nil && strings.Contains(err.Error(), "text section not found") {
    return analyzeSymbolsOnly(f)
}

Prevention

Use 'dumpbin /headers <file>' to verify .text section exists.
Avoid using PE packers/obfuscators on binaries meant for analysis.
Unpack UPX-compressed binaries before running objfile tools.

When it happens

Trigger: Calling `objfile.Open` on a PE binary and then `File.Text()`, where the PE file has no `.text` section. This happens with stripped PE binaries, resource-only DLLs, or PE files with non-standard section naming.

Common situations: Analyzing a resource-only DLL that has no code section. Using `go tool objdump` or pprof on a PE binary that was processed by a packer or obfuscator that renames sections. Pointing at a PE file that is not a normal executable (e.g., a `.cpl` with non-standard layout).

Related errors

AI-assisted analysis of golang/go@b6b368adc5 (2026-08-12). Data as JSON: /api/errors/ca4284557d8a2ec7. Report an issue: GitHub.