In case of a restart, v[0] and v[1] were incorrectly initialised.
This was interpreted by ssl3_get_client_key_exchange as an RSA decryption
failure (don't ask me why) and caused it to create a _random_ master key
instead (even weirder), which obviously led to incorrect input to
ssl3_generate_master_secret and thus caused "block cipher pad is
wrong" error messages from ssl3_enc for the client's Finished message.
Arrgh.
Changes between 0.9.4 and 0.9.5 [xx XXX 1999]
+ *) Bugfix: ssl23_get_client_hello did not work properly when called in
+ state SSL23_ST_SR_CLNT_HELLO_B, i.e. when the first 7 bytes of
+ a SSLv2-compatible client hello for SSLv3 or TLSv1 could be read,
+ but a retry condition occured while trying to read the rest.
+ [Bodo Moeller]
+
*) The PKCS7_ENC_CONTENT_new() function was setting the content type as
NID_pkcs7_encrypted by default: this was wrong since this should almost
always be NID_pkcs7_data. Also modified the PKCS7_set_type() to handle
int type=0,use_sslv2_strong=0;
int v[2];
- /* read the initial header */
- v[0]=v[1]=0;
if (s->state == SSL23_ST_SR_CLNT_HELLO_A)
{
+ /* read the initial header */
+ v[0]=v[1]=0;
+
if (!ssl3_setup_buffers(s)) goto err;
n=ssl23_read_bytes(s,7);
type=1;
if (s->options & SSL_OP_NON_EXPORT_FIRST)
+ /* not only confusing, but broken! */
{
STACK_OF(SSL_CIPHER) *sk;
SSL_CIPHER *c;
/* we have a SSLv3/TLSv1 in a SSLv2 header */
type=2;
p=s->packet;
+ v[0] = p[3];
+ v[1] = p[4];
n=((p[0]&0x7f)<<8)|p[1];
if (n > (1024*4))
{
#define SSL_OP_PKCS1_CHECK_1 0x08000000L
#define SSL_OP_PKCS1_CHECK_2 0x10000000L
#define SSL_OP_NETSCAPE_CA_DN_BUG 0x20000000L
+/* SSL_OP_NON_EXPORT_FIRST looks utterly broken .. */
#define SSL_OP_NON_EXPORT_FIRST 0x40000000L
#define SSL_OP_NETSCAPE_DEMO_CIPHER_CHANGE_BUG 0x80000000L
#define SSL_OP_ALL 0x000FFFFFL