That caused several memory leaks in case of error.
Also when the CMS object that is created by CMS_EncryptedData_encrypt
is not used in the normal way, but instead just deleted
by CMS_ContentInfo_free some memory was lost.
Fixes#21985
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/22008)
That seems to be only an issue for RSA-PSS with parameters.
Spotted by code review, so it looks like there is no test coverage for this.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/22007)
Additionally use OPENSSL_clear_free on the private
key data in case of error.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/22007)
This affects some Poly1305 assembler functions
which are only used for certain CPU types.
Remove those functions for Windows targets,
as a simple interim solution.
Fixes#21522
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/21808)
(cherry picked from commit 7b8e27bc2e02238986d89ef0ece067ec1b48e165)
Add the copy of the omitted ASN1 public key method and
other members.
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/21125)
If |q| >= |p| then the q value is obviously wrong as q
is supposed to be a prime divisor of p-1.
We check if p is overly large so this added test implies that
q is not large either when performing subsequent tests using that
q value.
Otherwise if it is too large these additional checks of the q value
such as the primality test can then trigger DoS by doing overly long
computations.
Fixes CVE-2023-3817
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/21551)
The pre-existing error cases where DH_check returned zero
are not related to the dh params in any way, but are only
triggered by out-of-memory errors, therefore having *ret
set to zero feels right, but since the new error case is
triggered by too large p values that is something different.
On the other hand some callers of this function might not
be prepared to handle the return value correctly but only
rely on *ret. Therefore we set some error bits in *ret as
additional safety measure.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/21533)
The DH_check() function checks numerous aspects of the key or parameters
that have been supplied. Some of those checks use the supplied modulus
value even if it is excessively large.
There is already a maximum DH modulus size (10,000 bits) over which
OpenSSL will not generate or derive keys. DH_check() will however still
perform various tests for validity on such a large modulus. We introduce a
new maximum (32,768) over which DH_check() will just fail.
An application that calls DH_check() and supplies a key or parameters
obtained from an untrusted source could be vulnerable to a Denial of
Service attack.
The function DH_check() is itself called by a number of other OpenSSL
functions. An application calling any of those other functions may
similarly be affected. The other functions affected by this are
DH_check_ex() and EVP_PKEY_param_check().
CVE-2023-3446
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/21452)
OBJ_obj2txt() would translate any size OBJECT IDENTIFIER to canonical
numeric text form. For gigantic sub-identifiers, this would take a very
long time, the time complexity being O(n^2) where n is the size of that
sub-identifier.
To mitigate this, a restriction on the size that OBJ_obj2txt() will
translate to canonical numeric text form is added, based on RFC 2578
(STD 58), which says this:
> 3.5. OBJECT IDENTIFIER values
>
> An OBJECT IDENTIFIER value is an ordered list of non-negative numbers.
> For the SMIv2, each number in the list is referred to as a sub-identifier,
> there are at most 128 sub-identifiers in a value, and each sub-identifier
> has a maximum value of 2^32-1 (4294967295 decimal).
Fixesotc/security#96
Fixes CVE-2023-2650
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
The invocation of ossl_policy_level_add_node in tree_calculate_user_set
did not have any error handling. Add it to prevent a memory leak for the
allocated extra policy data.
Also add error handling to sk_X509_POLICY_NODE_push to ensure that if
a new node was allocated, but could not be added to the stack, it is
freed correctly.
Fix error handling if tree->user_policies cannot be allocated by
returning 0, indicating failure, rather than 1.
Signed-off-by: Clemens Lang <cllang@redhat.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/21066)
ossl_policy_level_add_node() first adds the new node to the level->nodes
stack, and then attempts to add extra data if extra_data is true. If
memory allocation or adding the extra data to tree->extra_data fails,
the allocated node (that has already been added to the level->nodes
stack) is freed using ossl_policy_node_free(), which leads to
a potential use after free.
Additionally, the tree's node count and the parent's child count would
not be updated, despite the new node being added.
Fix this by either performing the function's purpose completely, or not
at all by reverting the changes on error.
Signed-off-by: Clemens Lang <cllang@redhat.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/21066)
This is an alternative to #20893
Additionally this fixes also a possible issue in UI_UTIL_read_pw:
When UI_new returns NULL, the result code would still be zero
as if UI_UTIL_read_pw succeeded, but the password buffer is left
uninitialized, with subsequent possible stack corruption or worse.
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20957)
(cherry picked from commit a64c48cff88e032cf9513578493c4536df725a22)
The negative integer check is done to prevent potential overflow.
Fixes#20719.
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20862)
(cherry picked from commit 1258a8e4361320cd3cfaf9ede692492ce01034c8)
This is about a timing leak in the topmost limb
of the internal result of RSA_private_decrypt,
before the padding check.
There are in fact at least three bugs together that
caused the timing leak:
First and probably most important is the fact that
the blinding did not use the constant time code path
at all when the RSA object was used for a private
decrypt, due to the fact that the Montgomery context
rsa->_method_mod_n was not set up early enough in
rsa_ossl_private_decrypt, when BN_BLINDING_create_param
needed it, and that was persisted as blinding->m_ctx,
although the RSA object creates the Montgomery context
just a bit later.
Then the infamous bn_correct_top was used on the
secret value right after the blinding was removed.
And finally the function BN_bn2binpad did not use
the constant-time code path since the BN_FLG_CONSTTIME
was not set on the secret value.
In order to address the first problem, this patch
makes sure that the rsa->_method_mod_n is initialized
right before the blinding context.
And to fix the second problem, we add a new utility
function bn_correct_top_consttime, a const-time
variant of bn_correct_top.
Together with the fact, that BN_bn2binpad is already
constant time if the flag BN_FLG_CONSTTIME is set,
this should eliminate the timing oracle completely.
In addition the no-asm variant may also have
branches that depend on secret values, because the last
invocation of bn_sub_words in bn_from_montgomery_word
had branches when the function is compiled by certain
gcc compiler versions, due to the clumsy coding style.
So additionally this patch stream-lined the no-asm
C-code in order to avoid branches where possible and
improve the resulting code quality.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20284)
This reverts commit 43d8f88511991533f53680a751e9326999a6a31f.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20284)
Even though we check the leaf cert to confirm it is valid, we
later ignored the invalid flag and did not notice that the leaf
cert was bad.
Fixes: CVE-2023-0465
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20588)
A security vulnerability has been identified in all supported versions
of OpenSSL related to the verification of X.509 certificate chains
that include policy constraints. Attackers may be able to exploit this
vulnerability by creating a malicious certificate chain that triggers
exponential use of computational resources, leading to a denial-of-service
(DoS) attack on affected systems.
Fixes CVE-2023-0464
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/20569)
The mention of the GPL shouldn't have been there.
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20517)
(cherry picked from commit c879f8ac56170a5cf929fab8067beb2a5902be2b)
Fix the gcc build warning from conf_sap.c:
variable flags set but not used [-Wunused-but-set-variable]
variable appname set but not used [-Wunused-but-set-variable]
variable filename set but not used [-Wunused-but-set-variable]
CLA: trivial
Signed-off-by: Gang Chen <gang.c.chen@intel.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20347)
_umul128() is x86_64 (x64) only, while __umulh() works everywhere, but
doesn't generate optimal code on x64
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20244)
(cherry picked from commit 075652f224479dad2e64b92e791b296177af8705)
If the aux->asn1_cb() call fails in BIO_new_NDEF then the "out" BIO will
be part of an invalid BIO chain. This causes a "use after free" when the
BIO is eventually freed.
Based on an original patch by Viktor Dukhovni and an idea from Theo
Buehler.
Thanks to Octavio Galland for reporting this issue.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
In the event of a failure in PEM_read_bio_ex() we free the buffers we
allocated for the header and data buffers. However we were not clearing
the ptrs stored in *header and *data. Since, on success, the caller is
responsible for freeing these ptrs this can potentially lead to a double
free if the caller frees them even on failure.
Thanks to Dawei Wang for reporting this issue.
Based on a proposed patch by Kurt Roeckx.
CVE-2022-4450
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
A timing based side channel exists in the OpenSSL RSA Decryption
implementation which could be sufficient to recover a plaintext across
a network in a Bleichenbacher style attack. To achieve a successful
decryption an attacker would have to be able to send a very large number
of trial messages for decryption. The vulnerability affects all RSA
padding modes: PKCS#1 v1.5, RSA-OEAP and RSASVE.
Patch written by Dmitry Belyavsky and Hubert Kario
CVE-2022-4304
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Using floating point is not supported in UEFI and can cause build
problems, for example due to SSE being disabled and x64 calling
convention passing floats in SSE registers.
Avoid those problems by not compiling the formating code for floating
point numbers.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19738)
The by_dir certificate/CRL lookup code uses an OPENSSL_STACK to track
how many sequentially numbered CRL files have been loaded for a given
X509_NAME hash which is being requested. This avoids loading already
loaded CRL files and repeated stat() calls.
This OPENSSL_STACK is searched using sk_find, however this mutates
the OPENSSL_STACK unless it is known to be sorted. This operation
therefore requires a write lock, which was not taken.
Fix this issue by sorting the OPENSSL_STACK whenever it is mutated. This
guarantees no mutation will occur during sk_find. This is chosen over
taking a write lock during sk_find as retrieving a CRL by X509_NAME is
assumed to be a hotter path than the case where a new CRL is installed.
Also optimise the code by avoiding creating the structure to track the
last CRL file sequence number in the circumstance where it would match
the initial value, namely where no CRL with the given hash is installed.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20127)
Otherwise the alloca can cause an exception.
Issue reported by Jiayi Lin.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19735)
If the cipher being used in ossl_cms_EncryptedContent_init_bio() has no
associated OID then we should report an error rather than continuing on
regardless. Continuing on still ends up failing - but later on and with a
more cryptic error message.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19920)
Some things that may go wrong in asn1_bio_write() are serious errors
that should be reported as -1, rather than 0 (which just means "we wrote
no data").
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19920)
If the BIO unexpectedly fails to flush then SMIME_crlf_copy() was not
correctly reporting the error. We modify it to properly propagate the
error condition.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19920)
If an error occurs during a flush on a BIO_f_cipher() then in some cases
we could get into an infinite loop. We add a check to make sure we are
making progress during flush and exit if not.
This issue was reported by Octavio Galland who also demonstrated an
infinite loop in CMS encryption as a result of this bug.
The security team has assessed this issue as not a CVE. This occurs on
*encryption* only which is typically processing trusted data. We are not
aware of a way to trigger this with untrusted data.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19920)
If PKCS8_pkey_set0 fails, the memory in rk need to be clear freed
otherwise it is owned by the PKCS8_PRIV_KEY_INFO.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19694)
SYS$GETTIM_PREC is a very new function, only available on OpenVMS v8.4.
OpenSSL binaries built on OpenVMS v8.4 become unusable on older OpenVM
versions, but building for the older CRTL version will make the high
precision time functions unavailable.
Tests have shown that on Alpha and Itanium, the time update granularity
between SYS$GETTIM and SYS$GETTIM_PREC is marginal, so the former plus
a sequence number turns out to be better to guarantee a unique nonce.
Fixes#18727
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18730)
We need to reencode X509_req_info_st if member data updated.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18879)
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19007)
(cherry picked from commit c92c3dfb99485eb2cfb840e92bd0ece8cdd72d0c)
As of clang-14 the strict aliasing is causing code to magically disappear.
By explicitly inlining the code, the aliasing problem evaporates.
Fixes#18225
Backport of #18258 to 1.1.1.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18948)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18926)
