Fix i2d_X509_AUX: pp can be NULL.

Reported by David Benjamin

Reviewed-by: Emilia Käsper <emilia@openssl.org>

GH875: Document -no_check_time
Date:   Tue Mar 15 15:19:44 2016 +0100

This commit updates the documentation of cms, ocsp, s_client,
s_server, and verify to reflect the new "-no_check_time"
option introduced in commit d35ff2c0ade0a12e84aaa2e9841b4983a2f3cf45
on 2015-07-31.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

Remove confusing comment.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

Issue #719:

If no serverinfo extension is found in some cases, do not abort the handshake,
but simply omit/skip that extension.

Check for already-registered serverinfo callbacks during serverinfo
registration.

Update SSL_CTX_use_serverinfo() documentation to mention the need to reload the
same serverinfo per certificate, for servers with multiple server certificates.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

Document the addition of OPENSSL_LOCAL_CONFIG_DIR

Reviewed-by: Stephen Henson <steve@openssl.org>

Add the possibility for local build file templates

Use the environment variable OPENSSL_LOCAL_CONFIG_DIR to find build
file templates as well.

Reviewed-by: Stephen Henson <steve@openssl.org>

Add the possibility to have an extra local OpenSSL configs directory

The environment variable OPENSSL_LOCAL_CONFIG_DIR is used to indicate
that there's a local directory with extra configuration files.

Reviewed-by: Stephen Henson <steve@openssl.org>

Remove personal configs from version control

As per a team decision back in 2014.

Reviewed-by: Stephen Henson <steve@openssl.org>

Secure memory fixes

Fix some of the variables to be (s)size_t, so that more than 1GB of
secure memory can be allocated. The arena has to be a power of 2, and
2GB fails because it ends up being a negative 32-bit signed number.

The |too_late| flag is not strictly necessary; it is easy to figure
out if something is secure memory by looking at the arena. As before,
secure memory allocations will not fail, but now they can be freed
correctly. Once initialized, secure memory can still be used, even if
allocations occured before initialization.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

ssl/record/rec_layer_s3.c: fix typo from previous commit.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Handle multi-line "written by/for" comments.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Android build fixes.

Reviewed-by: Rich Salz <rsalz@openssl.org>

Remove obsolete defined(__INTEL__) condition.

This macro was defined by no-longer-supported __MWERKS__ compiler.

Reviewed-by: Richard Levitte <levitte@openssl.org>

chacha/asm/chacha-x86.pl: make it compile on legacy systems.

Usage of $ymm variable is a bit misleading here, it doesn't refer
to %ymm register bank, but rather to VEX instruction encoding,
which AMD XOP code path depends on.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Fix spelling in pod files

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

make update

Reviewed-by: Rich Salz <rsalz@openssl.org>

Add the adjusted perl scripts to the set of "update" scripts

Reviewed-by: Rich Salz <rsalz@openssl.org>

Adjust a last few generators to new license boilerplate and C code style

Reviewed-by: Rich Salz <rsalz@openssl.org>

Skip blank lines if old copyright comment was removed, and only then

Reviewed-by: Rich Salz <rsalz@openssl.org>

Fix some X509_STORE macros

Some X509_STORE macros do not work since the type was made opaque.

Reviewed-by: Viktor Dukhovni <viktor@openssl.org>

Add checks to X509_NAME_oneline()

Sanity check field lengths and sums to avoid potential overflows and reject
excessively large X509_NAME structures.

Issue reported by Guido Vranken.

Reviewed-by: Matt Caswell <matt@openssl.org>

Sanity check buffer length.

Reject zero length buffers passed to X509_NAME_onelne().

Issue reported by Guido Vranken.

Reviewed-by: Matt Caswell <matt@openssl.org>

Add size limit to X509_NAME structure.

This adds an explicit limit to the size of an X509_NAME structure. Some
part of OpenSSL (e.g. TLS) already effectively limit the size due to
restrictions on certificate size.

Reviewed-by: Matt Caswell <matt@openssl.org>

Remove BIO_dummy, it's old cruft

Reviewed-by: Rich Salz <rsalz@openssl.org>

Change 'struct bio_st' in all public header where applicable

Reviewed-by: Rich Salz <rsalz@openssl.org>

Add asn1_mac

Reviewed-by: Richard Levitte <levitte@openssl.org>

Remove some dead code from EC_GROUP_check()

EC_GROUP_check() was obtaining a temporary BIGNUM from the BN_CTX, but
then not using it.

Reviewed-by: Rich Salz <rsalz@openssl.org>

Don't use an uninitialised variable in srp application

The srp application created an uninitialised DB_ATTR object and then
passed it to the load_index function which attempted to read it. A
DB_ATTR object only contains a single field called "unique_subject".
AFAICT this attribute is unused in the SRP case, and therefore it would be
better to pass a NULL DB_ATTR to load_index (which handles that case
gracefully).

Reviewed-by: Rich Salz <rsalz@openssl.org>

Avoid a NULL ptr deref if group is not set

We should only copy parameters and keys if the group is set. Otherwise
they don't really make any sense. Previously we copied the private key
regardless of whether the group was set...but if it wasn't a NULL ptr
deref could occur. It's unclear whether we could ever get into that
situation, but since we were already checking it for the public key we
should be consistent.

Reviewed-by: Rich Salz <rsalz@openssl.org>

Check for a NULL return value from a call to X509_STORE_CTX_new()

Reviewed-by: Rich Salz <rsalz@openssl.org>

Check for failed malloc in BIO_ADDR_new

BIO_ADDR_new() calls OPENSSL_zalloc() which can fail - but the return
value is not checked.

Reviewed-by: Rich Salz <rsalz@openssl.org>

The x509_name_canon function doesn't check for an error return

i2d_name_canon can return a negative number on error. We should check it
before continuing.

Reviewed-by: Rich Salz <rsalz@openssl.org>

Make header signature of CRYPTO_mem_leaks BIO instead of struct bio_st

Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/1012)

Avoid overflow issues in X509_cmp.

The length is a long, so returning the difference does not quite work.

Thanks to Torbjörn Granlund for noticing.

Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>

Misc tweaks for EBCDIC based on feedback received

Reviewed-by: Andy Polyakov <appro@openssl.org>

Fix EBCDIC problem in conf_def.h

The non-ascii version of this set of macros ensures that the "a" variable
is inside the expected range. This logic wasn't quite right for the
EBCDIC version.

Reviewed-by: Andy Polyakov <appro@openssl.org>

Fix the tests to work with EBCDIC

Most of the tests already pass with EBCIDC but a few were trying to
write into read only memory.

Reviewed-by: Andy Polyakov <appro@openssl.org>

Fix a build error with strict-warnings and CHARSET_EBCDIC

Reviewed-by: Andy Polyakov <appro@openssl.org>

Add the ability to test EBCDIC builds

This adds the define CHARSET_EBCDIC_TEST which enables testing of EBCDIC
code on an ASCII system.

Reviewed-by: Andy Polyakov <appro@openssl.org>

Fix building with -DCHARSET_EBCDIC

Building with -DCHARSET_EBCDIC and using --strict-warnings resulted in
lots of miscellaneous errors. This fixes it.

Reviewed-by: Andy Polyakov <appro@openssl.org>

VMS: only explicitely translate names in library C files.

When compiling all other C files, rely on the compiler to
automatically pick up the name translation information from the header
files __DECC_INCLUDE_{PRO,EPI}LOGUE.H.

Reviewed-by: Andy Polyakov <appro@openssl.org>

VMS: It seems DEC C doesn't handle certain header files quite right

With DEC C on VMS, you can use __DECC_INCLUDE_PROLOGUE.H and
__DECC_INCLUDE_EPILOGUE.H to include some DEC C specific features or
pragmas without having to touch the other header files.

It seems, however, that the current version of the compiler requires
the file names to be upcased, or it doesn't handle them quite right.

Reviewed-by: Andy Polyakov <appro@openssl.org>

Add aliases for des-ede-ecb and des-ede3-ecb ciphers.

Currently we can get all block ciphers with
EVP_get_cipherbyname("<alg_name>-<block-mode-name>")
for example, by names "aes-128-ecb" or "des-ede-cbc".
I found a problem with des-ede-ecb and des-ede3-ecb ciphers as
they can be accessed only with names:
EVP_get_cipherbyname("des-ede")
EVP_get_cipherbyname("des-ede3")
It breaks the general concept.

In this patch I add aliases which allow to use names:
EVP_get_cipherbyname("des-ede-ecb")
EVP_get_cipherbyname("des-ede3-ecb")
in addition to the currently used names.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

Fixed scripts order for generate_crypto_objects target

Script obj_dat.pl depends on file obj_mac.h generated by script objects.pl

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

crypto/ppccap.c: fix missing declaration warning.

Reviewed-by: Richard Levitte <levitte@openssl.org>

crypto/ppccap.c: permit build with no-chacha and no-poly1305.

RT#4508

Reviewed-by: Richard Levitte <levitte@openssl.org>

Remove some dead code

Commit e1d9f1ab39eea left some dead code behind. This removes it.

Reviewed-by: Stephen Henson <steve@openssl.org>

A call to RSA_set0_key had the arguments in the wrong order

Reviewed-by: Richard Levitte <levitte@openssl.org>

Client side CKE processing can double free on error

The tls_client_key_exchange_post_work() frees the pms on error. It also
calls ssl_generate_master_secret() which also free the pms. If an error
occurs after ssl_generate_master_secret() has been called then a double
free can occur.

Reviewed-by: Andy Polyakov <appro@openssl.org>

Don't free the BIGNUM passed to BN_mpi2bn

Commit 91fb42dd fixed a leak but introduced a problem where a parameter
is erroneously freed instead.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

Fix a leak in i2b_PVK

Commit 8e588e28 fixed a leak but introduced a new one.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

make update

Reviewed-by: Matt Caswell <matt@openssl.org>

apps/progs.pl: don't make digests disablable by default

Some digest algorithms can't be disabled, don't pretend they can.

Reviewed-by: Matt Caswell <matt@openssl.org>

BIO_free should call method->destroy before free'ing member fields

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/1007)

Add checks on CRYPTO_new_ex_data return value...
with some adaptation to new multi-threading API.

Once reference, lock, meth and flag fields are setup,
DSA_free/DH_free can be called directly.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/996)

Add checks on CRYPTO_new_ex_data return value

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/996)

Fix an error code spelling.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/952)

various spelling fixes

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/952)

Add getters for X509_STORE and X509_OBJECT members

OpenSSL 1.1.0-pre5 has made some additional structs opaque. Python's ssl
module requires access to some of the struct members. Three new getters
are added:

int X509_OBJECT_get_type(X509_OBJECT *a);
STACK_OF(X509_OBJECT) *X509_STORE_get0_objects(X509_STORE *v);
X509_VERIFY_PARAM *X509_STORE_get0_param(X509_STORE *ctx);

Signed-off-by: Christian Heimes <cheimes@redhat.com>
Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>

make update

Recycling an unused slot.

Reviewed-by: Rich Salz <rsalz@openssl.org>

Implement X509_STORE_CTX_set_current_cert() accessor

Reviewed-by: Rich Salz <rsalz@openssl.org>

Fix BIO_set_nbio_accept()

The code that implements this control would work when enabling nbio,
but the disabling code needed fixing.

Reviewed-by: Matt Caswell <matt@openssl.org>

Don't leak memory on error path in dane_ctx_enable()

The function dane_ctx_enable() allocated some memory that it did not
free in an error path.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Free an ASN1_OBJECT in an error path

The r2i_certpol() function allocates an ASN1_OBJECT but can fail to free
it in an error path.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak an ASN1_OCTET_STRING on error in rsa_cms_encrypt

The rsa_cms_encrypt() function allocates an ASN1_OCTET_STRING but can
then fail to free it in an error condition.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Free memory on error in PKCS7_dataFinal()

The PKCS7_dataFinal() function allocates a memory buffer but then fails
to free it on an error condition.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak memory on error in PKCS12_key_gen_uni

The PKCS12_key_gen_uni() had one error path which did not free memory
correctly.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak memory on error in i2b_PVK

The i2b_PVK function leaked a number of different memory allocations on
error paths (and even some non-error paths).

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak memory on error in b2i_rsa

The b2i_rsa() function uses a number of temporary local variables which
get leaked on an error path.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak resource on error in OCSP_url_svcloc_new

On error we could leak a ACCESS_DESCRIPTION and an ASN1_IA5STRING. Both
should be freed in the error path.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Check that we were actually allocated BIGNUMs in dsa_builtin_paramgen2

Calls to BN_CTX_get() can fail so we should check that they were
successful.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak EVP_MD_CTX on error path

The cms_SignerInfo_content_sign() function allocated an EVP_MD_CTX but
then failed to free it on an error path.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak memory on error in cms_RecipientInfo_pwri_crypt

The cms_RecipientInfo_pwri_crypt() allocated an EVP_CIPHER_CTX but then
failed to free it in some error paths. By allocating it a bit later that
can be avoided.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak memory on error in BN_generate_prime_ex

In BN_generate_prime_ex() we do some sanity checks first and return
with an error if they fail. We should do that *before* allocating any
resources to avoid a memory leak.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Free a BIGNUM on error in BN_mpi2bn

In the BN_mpi2bn() function, a failure of a call to BN_bin2bn() could
result in the leak of a previously allocated BIGNUM value.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Don't leak memory on failure to create a mem BIO

During construction of a mem BIO we allocate some resources. If this
allocation fails we can end up leaking everything we have allocated so
far.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Close the accept socket on error

When setting an accepted socket for non-blocking, if the operation fails
make sure we close the accepted socket.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Make BIO_sock_error return a proper error code when getsockopt fails

BIO_sock_error() returned 1 when getsockopt() fails when it should
return the error code for that failure.

Additionally, the optlen parameter to getsockopt() has to point at
the size of the area that the optval parameter points at rather than
zero.  Some systems may forgive it being zero, but others don't.

Reviewed-by: Matt Caswell <matt@openssl.org>

Reject inappropriate private key encryption ciphers.

The traditional private key encryption algorithm doesn't function
properly if the IV length of the cipher is zero. These ciphers
(e.g. ECB mode) are not suitable for private key encryption
anyway.

Reviewed-by: Emilia Käsper <emilia@openssl.org>

Fix set0 reuse test

We must test for new object == current object, not !=.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Future proof build_chain() in x509_vfy.c

Coverity reports a potential NULL deref when "2 0 0" DANE trust-anchors
from DNS are configured via SSL_dane_tlsa_add() and X509_STORE_CTX_init()
is called with a NULL stack of untrusted certificates.

Since ssl_verify_cert_chain() always provideds a non-NULL stack of
untrusted certs, and no other code path enables DANE, the problem
can only happen in applications that use SSL_CTX_set_cert_verify_callback()
to implement their own wrappers around X509_verify_cert() passing
only the leaf certificate to the latter.

Regardless of the "improbability" of the problem, we do need to
ensure that build_chain() handles this case correctly.

Reviewed-by: Matt Caswell <matt@openssl.org>

Documentation the changed {RSA,DSA,DH}_set0_* functionality change

Reviewed-by: Matt Caswell <matt@openssl.org>

RSA, DSA, DH: Allow some given input to be NULL on already initialised keys

The diverse {RSA,DSA,DH}_set0_* functions are made to allow some
parameters to be NULL IF the corresponding numbers in the given key
structure have already been previously initialised.  Specifically,
this allows the addition of private components to be added to a key
that already has the public half, approximately like this:

    RSA_get0_key(rsa, NULL, &e, NULL);
    RSA_get0_factors(rsa, &p, &q);
    /* calculate new d */
    RSA_set0_key(rsa, NULL, NULL, d);

Reviewed-by: Matt Caswell <matt@openssl.org>

GH975 Add ex_data functions for X509_STORE

Add X509_STORE_{set,get}_ex_data() function and
X509_STORE_get_ex_new_index() macro.

X509_STORE has ex_data and the documentation also mentions them but they
are not actually implemented.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>

Fix no-gost no-srp no-ec no-dh.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Fix enable-ssl-trace no-nextprotoneg.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Fix no-engine no-ui.

Reviewed-by: Richard Levitte <levitte@openssl.org>

split_send_fragment should always be less than or equal to max_send_fragment

A bug meant that SSL_CTRL_SET_MAX_SEND_FRAGMENT was not adjusting
split_send_fragment properly.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Fix BIO_CTRL_DUP for an SSL BIO

The variables in the BIO weren't being duplicated properly.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>

Fix passwd seg fault

Passing the -stdin arg to the passwd command line app *and* supply a
password on the command line causes a seg fault.

Reviewed-by: Richard Levitte <levitte@openssl.org>

SPARCv9 assembly pack: fine-tune run-time switch.

Reviewed-by: Tim Hudson <tjh@openssl.org>

camellia/build.info: fix typo

RT#4520

Reviewed-by: Tim Hudson <tjh@openssl.org>

Remove dead-code infinite loop

Commit d32f5d8733df9938727710d4194e92813c421ef1 added a 'goto end;' statement
at the end of the code block for the 'end' label.  Fortunately, it was after a
return statement, so no infinite loop occurred, but it is still dead code.

Remove the extra goto statement as cleanup.

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>

Add ex_data functions for X509_STORE

Add X509_STORE_{set,get}_ex_data() function and
X509_STORE_get_ex_new_index() macro.

X509_STORE has ex_data and the documentation also mentions them but they
are not actually implemented.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>

Ensure we check i2d_X509 return val

The i2d_X509() function can return a negative value on error. Therefore
we should make sure we check it.

Issue reported by Yuan Jochen Kang.

Reviewed-by: Emilia Käsper <emilia@openssl.org>

*.der files are binary.

Reviewed-by: Richard Levitte <levitte@openssl.org>

Remove the never-functional no-sct

It was added as part of 2df84dd3299ff25fa078ca7ffbdeaac65b361feb
but has never actually been used for anything; presumably it was
a typo for one of SCTP or CT.

This removes the last '??' entry from INSTALL.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

Remove some OPENSSL_NO_SHA that snuck in

Commit d064e6ab52ac8e7b80b2a5d11b31bca583b769c7 removed all the
OPENSSL_NO_SHA guards, but commit
a50ad1daaa68c109ea1a14225a7aba8660526101 regenerated some due to the
sha entries in the %md_disabler table in apps/progs.pl.

Update %md_disabler to reflect that sha is not disableable, and
remove OPENSSL_NO_SHA for good.

Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>

crypto/poly1305/asm: chase overflow bit on x86 and ARM platforms.

Even though no test could be found to trigger this, paper-n-pencil
estimate suggests that x86 and ARM inner loop lazy reductions can
loose a bit in H4>>*5+H0 step.

Reviewed-by: Emilia Käsper <emilia@openssl.org>

make update

And recycle some disused slots.

Reviewed-by: Rich Salz <rsalz@openssl.org>