[release-branch.go1.19] archive/tar: limit size of headers

Set a 1MiB limit on special file blocks (PAX headers, GNU long names,
GNU link names), to avoid reading arbitrarily large amounts of data
into memory.

Thanks to Adam Korczynski (ADA Logics) and OSS-Fuzz for reporting
this issue.

Fixes CVE-2022-2879
Updates #54853
Fixes #55926

Change-Id: I85136d6ff1e0af101a112190e027987ab4335680
Reviewed-on: https://team-review.git.corp.google.com/c/golang/go-private/+/1565555
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
Run-TryBot: Roland Shoemaker <bracewell@google.com>
Reviewed-by: Roland Shoemaker <bracewell@google.com>
(cherry picked from commit 6ee768cef6b82adf7a90dcf367a1699ef694f3b2)
Reviewed-on: https://team-review.git.corp.google.com/c/golang/go-private/+/1591053
Reviewed-by: Julie Qiu <julieqiu@google.com>
Reviewed-by: Damien Neil <dneil@google.com>
Reviewed-on: https://go-review.googlesource.com/c/go/+/438498
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Reviewed-by: Carlos Amedee <carlos@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
Run-TryBot: Carlos Amedee <carlos@golang.org>

src/archive/tar/format.go

@@ -143,6 +143,10 @@ const (
 	blockSize  = 512 // Size of each block in a tar stream
 	nameSize   = 100 // Max length of the name field in USTAR format
 	prefixSize = 155 // Max length of the prefix field in USTAR format
+
+	// Max length of a special file (PAX header, GNU long name or link).
+	// This matches the limit used by libarchive.
+	maxSpecialFileSize = 1 << 20
 )
 
 // blockPadding computes the number of bytes needed to pad offset up to the

src/archive/tar/reader.go

@@ -103,7 +103,7 @@ func (tr *Reader) next() (*Header, error) {
 			continue // This is a meta header affecting the next header
 		case TypeGNULongName, TypeGNULongLink:
 			format.mayOnlyBe(FormatGNU)
-			realname, err := io.ReadAll(tr)
+			realname, err := readSpecialFile(tr)
 			if err != nil {
 				return nil, err
 			}
@@ -293,7 +293,7 @@ func mergePAX(hdr *Header, paxHdrs map[string]string) (err error) {
 // parsePAX parses PAX headers.
 // If an extended header (type 'x') is invalid, ErrHeader is returned
 func parsePAX(r io.Reader) (map[string]string, error) {
-	buf, err := io.ReadAll(r)
+	buf, err := readSpecialFile(r)
 	if err != nil {
 		return nil, err
 	}
@@ -828,6 +828,16 @@ func tryReadFull(r io.Reader, b []byte) (n int, err error) {
 	return n, err
 }
 
+// readSpecialFile is like io.ReadAll except it returns
+// ErrFieldTooLong if more than maxSpecialFileSize is read.
+func readSpecialFile(r io.Reader) ([]byte, error) {
+	buf, err := io.ReadAll(io.LimitReader(r, maxSpecialFileSize+1))
+	if len(buf) > maxSpecialFileSize {
+		return nil, ErrFieldTooLong
+	}
+	return buf, err
+}
+
 // discard skips n bytes in r, reporting an error if unable to do so.
 func discard(r io.Reader, n int64) error {
 	// If possible, Seek to the last byte before the end of the data section.

src/archive/tar/reader_test.go

@@ -6,6 +6,7 @@ package tar
 
 import (
 	"bytes"
+	"compress/bzip2"
 	"crypto/md5"
 	"errors"
 	"fmt"
@@ -243,6 +244,9 @@ func TestReader(t *testing.T) {
 	}, {
 		file: "testdata/pax-bad-hdr-file.tar",
 		err:  ErrHeader,
+	}, {
+		file: "testdata/pax-bad-hdr-large.tar.bz2",
+		err:  ErrFieldTooLong,
 	}, {
 		file: "testdata/pax-bad-mtime-file.tar",
 		err:  ErrHeader,
@@ -625,9 +629,14 @@ func TestReader(t *testing.T) {
 			}
 			defer f.Close()
 
+			var fr io.Reader = f
+			if strings.HasSuffix(v.file, ".bz2") {
+				fr = bzip2.NewReader(fr)
+			}
+
 			// Capture all headers and checksums.
 			var (
-				tr      = NewReader(f)
+				tr      = NewReader(fr)
 				hdrs    []*Header
 				chksums []string
 				rdbuf   = make([]byte, 8)

src/archive/tar/writer.go

@@ -199,6 +199,9 @@ func (tw *Writer) writePAXHeader(hdr *Header, paxHdrs map[string]string) error {
 			flag = TypeXHeader
 		}
 		data := buf.String()
+		if len(data) > maxSpecialFileSize {
+			return ErrFieldTooLong
+		}
 		if err := tw.writeRawFile(name, data, flag, FormatPAX); err != nil || isGlobal {
 			return err // Global headers return here
 		}

src/archive/tar/writer_test.go

@@ -1004,6 +1004,33 @@ func TestIssue12594(t *testing.T) {
 	}
 }
 
+func TestWriteLongHeader(t *testing.T) {
+	for _, test := range []struct {
+		name string
+		h    *Header
+	}{{
+		name: "name too long",
+		h:    &Header{Name: strings.Repeat("a", maxSpecialFileSize)},
+	}, {
+		name: "linkname too long",
+		h:    &Header{Linkname: strings.Repeat("a", maxSpecialFileSize)},
+	}, {
+		name: "uname too long",
+		h:    &Header{Uname: strings.Repeat("a", maxSpecialFileSize)},
+	}, {
+		name: "gname too long",
+		h:    &Header{Gname: strings.Repeat("a", maxSpecialFileSize)},
+	}, {
+		name: "PAX header too long",
+		h:    &Header{PAXRecords: map[string]string{"GOLANG.x": strings.Repeat("a", maxSpecialFileSize)}},
+	}} {
+		w := NewWriter(io.Discard)
+		if err := w.WriteHeader(test.h); err != ErrFieldTooLong {
+			t.Errorf("%v: w.WriteHeader() = %v, want ErrFieldTooLong", test.name, err)
+		}
+	}
+}
+
 // testNonEmptyWriter wraps an io.Writer and ensures that
 // Write is never called with an empty buffer.
 type testNonEmptyWriter struct{ io.Writer }