More specifically, don't sort the stack when searching when it isn't sorted.
This avoids a race condition.
Fixes#20135
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20782)
Signed-off-by: Liu-ErMeng <liuermeng2@huawei.com>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20797)
Added a function to allocate a buffer and copy a maching param.
Added a function to allocate a buffer and concatenate all matching params.
Fixes#20717
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20724)
default_context_do_init may be never called and CRYPTO_THREAD_init_local
inside it may be never called too. But corresponding
CRYPTO_THREAD_cleanup_local is always called at cleanup stage. This lead
to undefined behavior.
So, add flag to check that default_context_do_init will be called
successfully or not.
Fix: #20697
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20801)
Fixes#20710
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20745)
According to docs, EVP_PKEY_get_int_param should return 1 on Success, and
0 on Failure. So, fix checking of this return value at check_curve
CLA: trivial
Reviewed-by: Todd Short <todd.short@me.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20770)
Original author: Nevine Ebeid (Amazon)
Fixes: CVE-2023-1255
The buffer overread happens on decrypts of 4 mod 5 sizes.
Unless the memory just after the buffer is unmapped this is harmless.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
(Merged from https://github.com/openssl/openssl/pull/20759)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/20736)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/20736)
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/20298)
We no longer need to cast function pointers to PTR_SIZE_INT.
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20748)
Bit-fiddling pointers is technically implementation defined behavior
in the C specification so the following code is not supported in all
platforms:
PTR_SIZE_INT mask;
void * a, b, c;
int boolean_flag;
mask = 0 - boolean_flag;
/* Not guaranteed to be a valid ptr to a or b on all platforms */
a = (void *)
((((PTR_SIZE_INT) b & ~mask) | (((PTR_SIZE_INT)) c & mask)));
Using a ternary conditional operator is supported on all platforms
(i.e. `a = boolean_flag ? b : c;`).
On most modern compilers/CPUs, this will be faster, since it will
get converted to a CMOV instruction.
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20748)
This keeps the interface consistent with other HPKE API's.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20583)
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/20281)
This reverts commit b1892d21f8.
Except for the moving derive_kdk to a separate function.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20281)
Function EVP_PKEY_get_bn_param() uses temporary buffer (on stack or
heap allocated) to store serialized bignum, but after deserializing it
into BIGNUM*, the buffer is not erased and may contain sensitive data.
This change makes sure the buffer is erased if it was successfully
filled before. Unfortunately, it does not distinguish between public and
private key components, and will always erase the buffer.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20639)
Calling OPENSSL_init_crypto from inside a RUN_ONCE seems like a bad idea.
This is especially bad if OPENSSL_init_crypto can recursively end up
attempting to call the RUN_ONCE that we're already inside.
The initialisation in OPENSSL_init_crypto is already "run once" protected.
There is no need to protect it "twice".
Fixes#20653
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20662)
(cherry picked from commit a9745427cd)
Discovered during coverage testing.
Remove unneccesary check when using ossl_dh_get0_params() and
ossl_dsa_get0_params(). These point to addresses and can not fail
for any existing calls.
Make dsa keygen tests only available in the FIPS module - as they are
not used in the default provider.
Change ossl_ffc_set_digest() to return void as it cannot fail.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20359)
Fixes a "Incompatible function pointer types" error.
In prior versions, this was a warning, but is now an error.
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20654)
Related to #20268
DSA_sign() assumes that the signature passed in is related to DSA_size().
If q is negative then DSA_size() actually fails and returns 0.
A test that tries to allocate the signature buffer using DSA_size() and then
pass it to DSA_sign() will then either.
(1) Have a signature buffer of NULL. In this case it was leaking data
returned via i2d_DSA_SIG.
(2) Cause a seg fault because we created a buffer that was not large
enough to hold the signature. As it already checked zero we also now
check for negative values also.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20553)
CLA: trivial
There is an incorrect null pointer check and this ccommit resolves it.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20646)
Similiar to the issue found in PR #20553 for DSA_sign().
ECDSA_sign() leaked memory if the signature was NULL
when i2d_ECDSA_SIG was called.
Note that this does not affect the higher level EVP
functions as they correctly handle NULL.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20554)
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20629)
Change-Id: Ia94e528a2d55934435de6a2949784c52eb38d82f
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20621)
Add support for the RFC7250 certificate-type extensions.
Alows the use of only private keys for connection (i.e. certs not needed).
Add APIs
Add unit tests
Add documentation
Add s_client/s_server support
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Viktor Dukhovni <viktor@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18185)
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/20585)
CLA: trivial
The the provider, context duplication method for signature, key
exchange, asymmetric cipher, and key encapsulation is optional. But if
they are missing, we will get a segmentation fault in `EVP_PKEY_CTX_dup`
because they are called without null pointer checking.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20581)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/20204)
Fixes a regression from 3.0 from the obj creation refactoring.
Fixes#20555
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20556)
The changes from the following commit should also apply to
Visual Studio 2010
2d46a44ff2 (r104867505)
Fixes build errors: undefined symbol InterlockedOr64
on Windows 2003, Visual Studio 2010 for x86 target.
CLA: trivial
Signed-off-by: Georgi Valkov <gvalkov@gmail.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20557)
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/20571)
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/20436)
Commit ac21c1780a VMS knows POSIX threads too!
removed ossl_crypto_mem_barrier for POSIX systems.
Remove it for Win32 and other architectures as well.
Resolves issue #19506 Unable to build under bcc32c environment (Embarcadero
clang compiler).
Signed-off-by: Čestmír Kalina <ckalina@redhat.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20538)
For example if would be helpful if we got more useful information if the
caller forgot to set the peer address.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20514)
As the sleep() call is interruptible, it is not even a good idea to call
it in a loop if the caller uses some ridiculously large value as an
infinity just waiting for an interrupt.
Fixes#20524
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
(Merged from https://github.com/openssl/openssl/pull/20533)
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20519)
(cherry picked from commit d4765408c7)
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20519)
(cherry picked from commit 110dac5783)
Signed-off-by: Čestmír Kalina <ckalina@redhat.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/12256)
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)
RISC-V currently only offers a GMULT() callback for accelerated
processing. Let's implement the missing piece to have GHASH()
available as well. Like GMULT(), we provide a variant for
systems with the Zbkb extension (including brev8).
The integration follows the existing pattern for GMULT()
in RISC-V. We keep the C implementation as we need to decide
if we can call an optimized routine at run-time.
The C implementation is the fall-back in case we don't have
any extensions available that can be used to accelerate
the calculation.
Tested with all combinations of possible extensions
on QEMU (limiting the available instructions accordingly).
No regressions observed.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
The existing GCM calculation provides some potential
for further optimizations. Let's use the demo code
from the RISC-V cryptography extension groups
(https://github.com/riscv/riscv-crypto), which represents
the extension architect's intended use of the clmul instruction.
The GCM calculation depends on bit and byte reversal.
Therefore, we use the corresponding instructions to do that
(if available at run-time).
The resulting computation becomes quite compact and passes
all tests.
Note, that a side-effect of this change is a reduced register
usage in .gmult(), which opens the door for an efficient .ghash()
implementation.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
A recent commit introduced a Perl module for common code.
This patch changes the GCM code to use this module, removes duplicated code,
and moves the instruction encoding functions into the module.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
On systems where Devel::StackTrace is available, we can use this module
to create more usable error messages. Further, don't print error
messages in case of official register aliases, but simply accept them.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
Move helper functions and instruction encoding functions
into a riscv.pm Perl module to avoid pointless code duplication.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
In RISC-V we have multiple extensions, that can be
used to accelerate processing.
The known extensions are defined in riscv_arch.def.
From that file test functions of the following
form are generated: RISCV_HAS_$ext().
In recent commits new ways to define the availability
of these test macros have been defined. E.g.:
#define RV32I_ZKND_ZKNE_CAPABLE \
(RISCV_HAS_ZKND() && RISCV_HAS_ZKNE())
[...]
#define RV64I_ZKND_ZKNE_CAPABLE \
(RISCV_HAS_ZKND() && RISCV_HAS_ZKNE())
This leaves us with two different APIs to test capabilities.
Further, creating the same macros for RV32 and RV64 results
in duplicated code (see example above).
This inconsistent situation makes it hard to integrate
further code. So let's clean this up with the following steps:
* Replace RV32I_* and RV64I_* macros by RICSV_HAS_* macros
* Move all test macros into riscv_arch.h
* Use "AND" and "OR" to combine tests with more than one extension
* Rename include files for accelerated processing (remove extension
postfix).
We end up with compile time tests for RV32/RV64 and run-time tests
for available extensions. Adding new routines (e.g. for vector crypto
instructions) should be straightforward.
Testing showed no regressions.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20078)
Fixes#19092
The code looks like it was written to work with PBES1.
As it had no tests, this would of then broken when PBES2
was introduced at a later point.
Also added libctx and propq support.
This affects the shroudedkeybag object.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20134)
The EVP layer should not rely on the underlying low level code to
handle catching incorrect reuse of contexts.
Add a flag to mark a context as finalised as needed and then catch and
immediately error on Update/Final operations if called improperly.
Signed-off-by: Simo Sorce <simo@redhat.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/20375)
If the ctx was *really* needed we'll probably fail later with an error
anyway, so no point in failing immediately.
Document that this behavior is dependent on the provider used to
implement the signature/verification.
Signed-off-by: Simo Sorce <simo@redhat.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/20375)
Probe first to see if we have a PKCS8 file to improve decoder performance.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/20416)
The assignment of the result of EVP_get_digestbynid() did not happen
which made the fallback not actually perform the fallback.
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20447)
"adrl" is a pseudo-instruction used to calculate an address relative
to PC. It's not recognized by clang resulting in a compilation error.
I've stumbled upon it when trying to integrate the bsaes-armv7 assmebly
logic into FreeBSD kernel, which uses clang as it's default compiler.
Note that this affect the build only if BSAES_ASM_EXTENDED_KEY is
defined, which is not the default option in OpenSSL.
The solution here is to replace it with an add instruction.
This mimics what has already been done in !BSAES_ASM_EXTENDED_KEY logic.
Because of that I've marked this as trivial CLA.
CLA: trivial
Signed-off-by: Kornel Dulęba <mindal@semihalf.com>
Reviewed-by: Tom Cosgrove <tom.cosgrove@arm.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20458)
This file was only recently introduced and the missing header slipped through
the review process.
Fixes#20461
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20470)
The keccak XOF used for KMAC can be simplified by using klmd. This speeds up
XOF processing in cases where more than one result block is needed.
Signed-off-by: Juergen Christ <jchrist@linux.ibm.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20431)
Especially, it's important to use BIO_ADDR_sockaddr_size() instead of
taking sizeof(addr), as BIO_ADDR is a union of several sockaddr variants
with different sizes, and some sendto() implementations are very picky
that the size is correct for the indicated sockaddr family.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20449)
include/internal/thread_arch.h didn't indicate this, now it does.
This also removes ossl_crypto_mem_barrier(), because we isn't used
anywhere, and doesn't build with compilers that don't support the GNU
extension __asm__.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/20440)
We can easily find out the keytype which should significantly improve
the performance of this function because we don't have to try every loaded
decoder.
Partial fix for #20399
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20412)
Make the SIGILL-based code easier to read, and don't use it on Apple Silicon.
Also fix "error: 'HWCAP(2)_*' macro redefined" warnings on FreeBSD.
Fixes#20188
Change-Id: I5618bbe9444cc40cb5705c6ccbdc331c16bab794
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20305)
Check the return from DSA_set0_key and generate an error on failure.
Technically a false positive since the function always returns success.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/20409)
This is an unused internal function. It should be added back
in if there is a need to set this value to something other than
NID_mgf1 (which is not likely).
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20321)
Coverage testing showed that ossl_ffc_params_fromdata() was not setting
OSSL_PKEY_PARAM_FFC_DIGEST_PROPS.
Adding a negative test also showed that ossl_ffc_params_copy() did not
do a shallow copy of the digest or digest property.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20385)
Similiar checks to the DSA code have been added for ECDSA also.
This should not be a problem when using named groups.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20384)
Fixes#20268
Values such as q=1 or priv=0 caused infinite loops when calling
DSA_sign() without these changes.
There are other cases where bad domain parameters may have caused
infinite loops where the retry counter has been added. The simpler case
of priv=0 also hits this case. q=1 caused an infinite loop in the setup.
The max retry value has been set to an arbitrary value of 8 (it is
unlikely to ever do a single retry for valid values).
The minimum q bits was set to an arbitrary value of 128 (160 is still
used for legacy reasons when using 512 bit keys).
Thanks @guidovranken for detecting this, and @davidben for his
insightful analysis.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20384)
Since GCC commit e154242724b084380e3221df7c08fcdbd8460674 the flag "-many"
is sometimes not passed to the assembler. Use .machine "any" just like
ecp_nistz256-ppc64 to prevent compile errors when built with some
configurations of GCC.
CLA: trivial
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20378)
According to the documentation and my analysis tool
ASN1_item_i2d() can return a negative value on error,
but this is not checked. Fix it by changing the error check condition.
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20251)
The provider functions OSSL_FUNC_keymgmt_import_types() and
OSSL_FUNC_keymgmt_export_types() do not get the provider context passed.
This makes it difficult for providers to implement these functions unless
its a static implementation returning a truly constant OSSL_PARAM array.
Some providers may have a need to return an OSSL_PARAM array that is
dependent on the provider configuration, or anything else that is contained
in its provider context.
Add extended variants of these functions that get the provider context passed.
The functions should still return a static and constant OSSL_PARAM array, but
may use the provider context to select the array to return dependent on its
context. The returned array must be constant at least until the provider is
unloaded.
Providers can implement only the original functions, or only the extended
functions, or both. Implementing at least one of those functions is required
if also the respective OSSL_FUNC_keymgmt_import() or OSSL_FUNC_keymgmt_export()
function is implemented. If an extended function is available, it is called by
evp_keymgmt_import_types() or evp_keymgmt_export_types(), otherwise the original
function is called.
This makes the code backward compatible. Existing providers will only implement
the original functions, so these functions will continued to be called.
Newer providers can choose to implement the extended functions, and thus can
benefit from the provider context being passed to the implementation.
Signed-off-by: Ingo Franzki <ifranzki@linux.ibm.com>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20255)
Also add checks on failing cert/CRL up_ref calls; improve coding style.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/19199)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/19199)
CLA: trivial
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20351)
When calling BIO-recvmmsg() and using a dgram pair we were failing to
raise an error in the case that a problem occurs. This means that the
reason behind a failure cannot be detected and all problems are treated
as fatal even if they may not be.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20030)
This results in spurious errors appearing on the queue in normal
operation, e.g. calling SSL_tick() with a QUIC connection will succeed,
but an error will end up on the queue anyway.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20030)
There is no point in calling OPENSSL_init_crypto() unless we are actually
going to be using the default libctx.
Fixes#20315
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20341)
BN_priv_rand_range_ex() and BN_add() both return a 0 on failure and a 1
on success. In case of failure, the algorithm should fail. However, the
branch that it goes through on failure is "goto end", not "goto err".
Therefore, the algorithm will return 1 which indicates success instead
of 0 for failure, leading to potential problems for the callers.
Fix it by changing the goto to "goto err" instead of "goto end".
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20279)
Conditioning it on $disabled{shared} isn't right, it will still end up
in the static variant of the library. It's better to use SHARED_SOURCE
for these sorts of things.
Fixes#20238
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20240)
Also add corresponding tests and to this end update credentials
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/20160)
_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)
These aren't currently checked when they are called in cipher_aes_gcm_hw_armv8.inc,
but they are declared as returning as size_t the number of bytes they have processed,
and the aes_gcm_*_*_kernel (unroll by 4) versions of these do return the correct
values.
Change-Id: Ic3eaf139e36e29e8779b5bd8b867c08fde37a337
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20191)
This function returns an errorcode <= 0, but only < 0 is checked. Other
callers that check the return value perform this check correctly. Fix it
by changing the check to <= 0.
CLA: trivial
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20186)
BIO_dup_state() returns an error code <= 0 according to my analysis tool
and the documentation. Currently only == 0 is checked. Fix it by
changing the check condition.
CLA: trivial
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20194)
CMS_SharedInfo_encode() can also return a negative error value, but this
is not checked in the current check, only the zero error return value is
covered. A previous PR [1] fixed the other caller's check of
CMS_SharedInfo_encode in this file, but it seems like this place was
missed. Fix it by changing the check to <= 0.
[1] a752fc4da5
CLA: trivial
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20181)
default_check() can return a zero value to indicate an internal error in
one condition for the PRE_CTRL_STR_TO_PARAMS state. This state can be
reached from the default_fixup_args() function which does not check for
a zero value. All other callers of default_check() in that file do check
for a zero return value. Fix it by changing the check to <= 0.
CLA: trivial
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20175)
S390x has to ability to offload modular exponentiation and CRT operations to
Crypto Express Adapters. This possible performance optimization was not yet
used by OpenSSL. Add support for offloading and implement an optimized
version of RSA and DH with it.
The environment variable OPENSSL_s390xcap now recognizes the token "nocex" to
prevent offloading.
Signed-off-by: Juergen Christ <jchrist@linux.ibm.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20113)
BIO_set_md() can return an error value <= 0 according to my analysis
tool and the documentation. But only an error value == 0 is currently
checked. Fix it by changing the check condition.
CLA: trivial
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20195)
Fixes#19975
for CMS_decrypt_set1_pkey_and_peer() in the obvious way,
and a related potential crash in CMS_decrypt_set1_password().
The point is that the input might have an unexpected content type,
so a guard is needed at both places after `ec` is obtained.
Note that in CMS_decrypt_set1_pkey_and_peer() there was
no such ec != NULL guard for
```
if (ris != NULL)
debug = ec->debug;
```
maybe because it is implied here by ris != NULL.
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19981)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20205)
Avoid including QUIC related stuff in the FIPS sources.
Also avoid including libssl headers in ssl3_cbc.c.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19658)
The unused and untested internal function ossl_a2ucompare() has been
removed.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/20177)
BIO_set_accept_name() can return error values -1 and 0 according to
my analysis tool and the documentation. Documentation says a value of 1
indicates success. Currently, only an error value != 0 is checked which
erroneously interprets a -1 error return value as success.
Fix it by changing the check condition.
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20206)
Unlike gcc, the clang assembler has issues with the maximum value of the literal
in the `ldr REG, #VALUE` pseudo-instruction (where the assembler places the
value into a literal pool and generates a PC-relative load from that pool) when
used with Neon registers.
Specifically, while dN refers to 64-bit Neon registers, and qN refers to 128-bit
Neon registers, clang assembly only supports a maximum of 32-bit loads to
either with this instruction.
Therefore restrict accordingly to avoid breakage when building with clang.
clang appears to support the correct maximums with the scalar registers xN etc.
This will prevent the kind of breakage we saw when #19914 was merged (which has
since been fixed by #20202) - assembly authors will need to manually apply the
literal load, as is done in #20202.
None of the Arm assembler code uses this pseudo-instruction anyway, as it
doesn't seem to avoid duplication of constants.
Change-Id: If52f6ce22c10feb1cc334d996ff71b1efed3218e
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/20222)
These calls invoke EVP_DigestInit() which can fail for digests
with implicit fetches. Subsequent EVP_DigestUpdate() from BIO_write()
or EVP_DigestFinal() from BIO_read() will segfault on NULL
dereference. This can be triggered by an attacker providing
PKCS7 data digested with MD4 for example if the legacy provider
is not loaded.
If BIO_set_md() fails the md BIO cannot be used.
CVE-2023-0401
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Fixes CVE-2023-0217
When attempting to do a BN_Copy of params->p there was no NULL check.
Since BN_copy does not check for NULL this is a NULL reference.
As an aside BN_cmp() does do a NULL check, so there are other checks
that fail because a NULL is passed. A more general check for NULL params
has been added for both FFC public and private key validation instead.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
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: Tomas Mraz <tomas@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: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
This function assumes that if the "gen" is an OtherName, then the "base"
is a rfc822Name constraint. This assumption is not true in all cases.
If the end-entity certificate contains an OtherName SAN of any type besides
SmtpUtf8Mailbox and the CA certificate contains a name constraint of
OtherName (of any type), then "nc_email_eai" will be invoked, with the
OTHERNAME "base" being incorrectly interpreted as a ASN1_IA5STRING.
Reported by Corey Bonnell from Digicert.
CVE-2022-4203
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20202)
Signed-off-by: Xu Yizhou <xuyizhou1@huawei.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/19914)
That function is a wrapper around evp_cipher_param_to_asn1_ex() which
can return 0 as an error value via its ret <= 0 check [1].
Furthermore, all other callers of this function check against <= 0
instead of < 0 and this is also in line with what the documentation
tells us. Fix the incomplete check by changing it to <= 0 as well.
CLA: trivial
[1] 114d99b46b/crypto/evp/evp_lib.c (L164-L165)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20180)
Fix issue #20063.
CLA: trivial
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20163)
If no-module or no-shared is used, the symbols from
libcrypto should not be duplicated in legacy.a
Also the BIGNUM functions are currently not needed
in legacy.a at all.
Fixes#20124
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20137)
X509V3_add1_i2d() can return both -1 and 0 as an error code. This check
only checked for 0. Change it into <= 0 to also catch the -1 error code.
CLA: trivial
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20157)
Gives a performance enhancement of 16-38%, similar to the M1.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20141)
The check is unnecessary as the condition is already checked
before the switch statement.
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20121)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18704)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18704)
EVP_CIPHER_asn1_to_param() returns a value <= 0 in case of an error, and
a value greater than 0 in case of success. Two callsites only check for
< 0 instead of <= 0. The other callsites perform this check correctly.
Change the two callsites to <= 0. Additionally correctly handle a zero
return value from EVP_CIPHER_get_asn1_iv as success.
Fixes: #20116
CLA: trivial
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/201213)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/19271)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/19271)
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: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/20076)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/20052)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/19391)
To this end, tweak the internal handling of ctx->total_timeout.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/19391)
