Matt Caswell [Thu, 14 May 2015 12:48:47 +0000 (13:48 +0100)]
Move SSLv3_*method() functions
Move these functions into t1_clnt.c, t1_srvr.c and t1_meth.c and take
advantage of the existing tls1_get*_method() functions that all the other
methods are using. Since these now have to support SSLv3 anyway we might
as well use the same set of get functions for both TLS and SSLv3.
Matt Caswell [Tue, 31 Mar 2015 12:57:46 +0000 (13:57 +0100)]
Version negotiation rewrite cleanup
Following the version negotiation rewrite all of the previous code that was
dedicated to version negotiation can now be deleted - all six source files
of it!!
Matt Caswell [Mon, 30 Mar 2015 23:18:31 +0000 (00:18 +0100)]
Client side version negotiation rewrite
Continuing from the previous commit this changes the way we do client side
version negotiation. Similarly all of the s23* "up front" state machine code
has been avoided and again things now work much the same way as they already
did for DTLS, i.e. we just do most of the work in the
ssl3_get_server_hello() function.
Matt Caswell [Fri, 27 Mar 2015 23:01:51 +0000 (23:01 +0000)]
Server side version negotiation rewrite
This commit changes the way that we do server side protocol version
negotiation. Previously we had a whole set of code that had an "up front"
state machine dedicated to the negotiating the protocol version. This adds
significant complexity to the state machine. Historically the justification
for doing this was the support of SSLv2 which works quite differently to
SSLv3+. However, we have now removed support for SSLv2 so there is little
reason to maintain this complexity.
The one slight difficulty is that, although we no longer support SSLv2, we
do still support an SSLv3+ ClientHello in an SSLv2 backward compatible
ClientHello format. This is generally only used by legacy clients. This
commit adds support within the SSLv3 code for these legacy format
ClientHellos.
Server side version negotiation now works in much the same was as DTLS,
i.e. we introduce the concept of TLS_ANY_VERSION. If s->version is set to
that then when a ClientHello is received it will work out the most
appropriate version to respond with. Also, SSLv23_method and
SSLv23_server_method have been replaced with TLS_method and
TLS_server_method respectively. The old SSLv23* names still exist as
macros pointing at the new name, although they are deprecated.
Subsequent commits will look at client side version negotiation, as well of
removal of the old s23* code.
Richard Levitte [Thu, 14 May 2015 14:56:48 +0000 (16:56 +0200)]
Identify and move common internal libcrypto header files
There are header files in crypto/ that are used by a number of crypto/
submodules. Move those to crypto/include/internal and adapt the
affected source code and Makefiles.
Richard Levitte [Thu, 14 May 2015 13:55:59 +0000 (15:55 +0200)]
Adjust unixly mk1mf after introduction of tkey
Added depencies on the public variants of some keys in test to Makefile.
Added the newly introduced key files from test/ in the list of files
to copy in util/pl/unix.pl.
Richard Levitte [Thu, 14 May 2015 12:54:49 +0000 (14:54 +0200)]
Identify and move OpenSSL internal header files
There are header files in crypto/ that are used by the rest of
OpenSSL. Move those to include/internal and adapt the affected source
code, Makefiles and scripts.
Richard Levitte [Thu, 14 May 2015 06:44:06 +0000 (08:44 +0200)]
Move definition of INTxx_MIN et al to internal header
Having the INTxx_MIN et al macros defined in a public header is
unnecessary and risky. Also, it wasn't done for all platforms that
might need it.
So we move those numbers to an internal header file, do the math
ourselves and make sure to account for the integer representations we
know of.
This introduces include/internal, which is unproblematic since we
already use -I$(TOP)/include everywhere. This directory is different
from crypto/include/internal, as the former is more general internal
headers for all of OpenSSL, while the latter is for libcrypto only.
Rich Salz [Fri, 8 May 2015 16:23:56 +0000 (12:23 -0400)]
RT3841: memset() cipher_data when allocated
If an EVP implementation (such as an engine) fails out early, it's
possible to call EVP_CIPHER_CTX_cleanup() which will call
ctx->cipher->cleanup() before the cipher_data has been initialized
via ctx->cipher->init(). Guarantee it's all-bytes-zero as soon as
it is allocated.
Hanno Böck [Mon, 11 May 2015 10:33:37 +0000 (11:33 +0100)]
Call of memcmp with null pointers in obj_cmp()
The function obj_cmp() (file crypto/objects/obj_dat.c) can in some
situations call memcmp() with a null pointer and a zero length.
This is invalid behaviour. When compiling openssl with undefined
behaviour sanitizer (add -fsanitize=undefined to compile flags) this
can be seen. One example that triggers this behaviour is the pkcs7
command (but there are others, e.g. I've seen it with the timestamp
function):
apps/openssl pkcs7 -in test/testp7.pem
What happens is that obj_cmp takes objects of the type ASN1_OBJECT and
passes their ->data pointer to memcmp. Zero-sized ASN1_OBJECT
structures can have a null pointer as data.
RT#3816
Signed-off-by: Matt Caswell <matt@openssl.org> Reviewed-by: Rich Salz <rsalz@openssl.org>
Matt Caswell [Wed, 6 May 2015 20:31:16 +0000 (21:31 +0100)]
Don't allow a CCS when expecting a CertificateVerify
Currently we set change_cipher_spec_ok to 1 before calling
ssl3_get_cert_verify(). This is because this message is optional and if it
is not sent then the next thing we would expect to get is the CCS. However,
although it is optional, we do actually know whether we should be receiving
one in advance. If we have received a client cert then we should expect
a CertificateVerify message. By the time we get to this point we will
already have bombed out if we didn't get a Certificate when we should have
done, so it is safe just to check whether |peer| is NULL or not. If it is
we won't get a CertificateVerify, otherwise we will. Therefore we should
change the logic so that we only attempt to get the CertificateVerify if
we are expecting one, and not allow a CCS in this scenario.
Whilst this is good practice for TLS it is even more important for DTLS.
In DTLS messages can be lost. Therefore we may be in a situation where a
CertificateVerify message does not arrive even though one was sent. In that
case the next message the server will receive will be the CCS. This could
also happen if messages get re-ordered in-flight. In DTLS if
|change_cipher_spec_ok| is not set and a CCS is received it is ignored.
However if |change_cipher_spec_ok| *is* set then a CCS arrival will
immediately move the server into the next epoch. Any messages arriving for
the previous epoch will be ignored. This means that, in this scenario, the
handshake can never complete. The client will attempt to retransmit
missing messages, but the server will ignore them because they are the wrong
epoch. The server meanwhile will still be waiting for the CertificateVerify
which is never going to arrive.
Fix the heap corruption in libeay32!OBJ_add_object.
Original 'sizeof(ADDED_OBJ)' was replaced with 'sizeof(*ao)'. However,
they return different sizes. Therefore as the result heap gets corrupted
and at some point later debug version of malloc() detects the corruption.
Rich Salz [Fri, 8 May 2015 16:05:36 +0000 (12:05 -0400)]
Make COMP_CTX and COMP_METHOD opaque
Since COMP_METHOD is now defined in comp_lcl.h, it is no
longer possible to create new TLS compression methods without
using the OpenSSL source. Only ZLIB is supported by default.
Also, since the types are opaque, #ifdef guards to use "char *"
instead of the real type aren't necessary.
The changes are actually minor. Adding missing copyright to some
files makes the diff misleadingly big.
Digest cached records if not sending a certificate.
If server requests a certificate, but the client doesn't send one, cache
digested records. This is an optimisation and ensures the correct finished
mac is used when extended master secret is used with client authentication.
Richard Levitte [Wed, 6 May 2015 16:50:57 +0000 (18:50 +0200)]
Make -CAserial a type 's' option
The file name given to -CAserial might not exist yet. The
-CAcreateserial option decides if this is ok or not.
Previous to this change, -CAserial was a type '<' option, and in that
case, the existence of the file given as argument is tested quite
early, and is a failure if it doesn't. With the type 's' option, the
argument is just a string that the application can do whatever it
wants with.
Initialize potentially uninitialized local variables
Compiling OpenSSL code with MSVC and /W4 results in a number of warnings.
One category of warnings is particularly interesting - C4701 (potentially
uninitialized local variable 'name' used). This warning pretty much means
that there's a code path which results in uninitialized variables being used
or returned. Depending on compiler, its options, OS, values in registers
and/or stack, the results can be nondeterministic. Cases like this are very
hard to debug so it's rational to fix these issues.
This patch contains a set of trivial fixes for all the C4701 warnings (just
initializing variables to 0 or NULL or appropriate error code) to make sure
that deterministic values will be returned from all the execution paths.
RT#3835
Signed-off-by: Matt Caswell <matt@openssl.org>
Matt's note: All of these appear to be bogus warnings, i.e. there isn't
actually a code path where an unitialised variable could be used - its just
that the compiler hasn't been able to figure that out from the logic. So
this commit is just about silencing spurious warnings.
Rich Salz [Mon, 4 May 2015 22:00:15 +0000 (18:00 -0400)]
memset, memcpy, sizeof consistency fixes
Just as with the OPENSSL_malloc calls, consistently use sizeof(*ptr)
for memset and memcpy. Remove needless casts for those functions.
For memset, replace alternative forms of zero with 0.
Reviewed-by: Richard Levitte <levitte@openssl.org>
Matt Caswell [Thu, 23 Apr 2015 19:01:33 +0000 (20:01 +0100)]
Add Error state
Reusing an SSL object when it has encountered a fatal error can
have bad consequences. This is a bug in application code not libssl
but libssl should be more forgiving and not crash.
Matt Caswell [Mon, 4 May 2015 22:15:46 +0000 (23:15 +0100)]
Remove libcrypto to libssl dependency
Remove dependency on ssl_locl.h from v3_scts.c, and incidentally fix a build problem with
kerberos (the dependency meant v3_scts.c was trying to include krb5.h, but without having been
passed the relevanant -I flags to the compiler)
Reviewed-by: Dr. Stephen Henson <steve@openssl.org>
Rich Salz [Sun, 3 May 2015 12:45:27 +0000 (08:45 -0400)]
GH271: Warning on </dev/null to CA.pl
If CA.pl is reading from /dev/null, then "chop $FILE" gives a warning.
Sigh. Have to add "if $FILE". This just silences a build warning.
Thanks to GitHub user andrejs-igumenovs for help with this.
Reviewed-by: Richard Levitte <levitte@openssl.org>
Rich Salz [Sat, 2 May 2015 03:10:31 +0000 (23:10 -0400)]
Use safer sizeof variant in malloc
For a local variable:
TYPE *p;
Allocations like this are "risky":
p = OPENSSL_malloc(sizeof(TYPE));
if the type of p changes, and the malloc call isn't updated, you
could get memory corruption. Instead do this:
p = OPENSSL_malloc(sizeof(*p));
Also fixed a few memset() calls that I noticed while doing this.
Reviewed-by: Richard Levitte <levitte@openssl.org>
Richard Levitte [Mon, 4 May 2015 15:34:40 +0000 (17:34 +0200)]
RT2943: Check sizes if -iv and -K arguments
RT2943 only complains about the incorrect check of -K argument size,
we might as well do the same thing with the -iv argument.
Before this, we only checked that the given argument wouldn't give a
bitstring larger than EVP_MAX_KEY_LENGTH. we can be more precise and
check against the size of the actual cipher used.
Rich Salz [Fri, 1 May 2015 18:37:16 +0000 (14:37 -0400)]
free NULL cleanup -- coda
After the finale, the "real" final part. :) Do a recursive grep with
"-B1 -w [a-zA-Z0-9_]*_free" to see if any of the preceeding lines are
an "if NULL" check that can be removed.