encoding/asn1: fix off-by-one in parseBase128Int.
parseBase128Int compares |shifted| with four, seemingly to ensure the result
fits in an int32 on 32-bit platforms where int is 32-bit. However, there is an
off-by-one in this logic, so it actually allows five shifts, making the maximum
tag number or OID component 2^35-1.
Fix this so the maximum is 2^28-1 which should be plenty for OID components and
tag numbers while not overflowing on 32-bit platforms.
Change-Id: If825b30cc53a0fc08e68ea1a24d265e7eb1a13a4
Reviewed-on: https://go-review.googlesource.com/18225
Reviewed-by: Adam Langley <agl@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
diff --git a/src/encoding/asn1/asn1.go b/src/encoding/asn1/asn1.go
index 0070ea8..8bafefd 100644
--- a/src/encoding/asn1/asn1.go
+++ b/src/encoding/asn1/asn1.go
@@ -294,7 +294,7 @@
func parseBase128Int(bytes []byte, initOffset int) (ret, offset int, err error) {
offset = initOffset
for shifted := 0; offset < len(bytes); shifted++ {
- if shifted > 4 {
+ if shifted == 4 {
err = StructuralError{"base 128 integer too large"}
return
}
diff --git a/src/encoding/asn1/asn1_test.go b/src/encoding/asn1/asn1_test.go
index 509a2cb..e0e8331 100644
--- a/src/encoding/asn1/asn1_test.go
+++ b/src/encoding/asn1/asn1_test.go
@@ -380,6 +380,8 @@
{[]byte{0xa0, 0x84, 0x80, 0x00, 0x00, 0x00}, false, tagAndLength{}},
// Long length form may not be used for lengths that fit in short form.
{[]byte{0xa0, 0x81, 0x7f}, false, tagAndLength{}},
+ // Tag numbers which would overflow int32 are rejected. (The value below is 2^31.)
+ {[]byte{0x1f, 0x88, 0x80, 0x80, 0x80, 0x00, 0x00}, false, tagAndLength{}},
}
func TestParseTagAndLength(t *testing.T) {
