Based on the value, it would with work properly or produce an error. Most likely seems to have been the former.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14638)
Some private libcrypto symbols are also included in legacy.so.
Unfortunately this included some files with "RUN_ONCE" functions and
global data. This doesn't get properly cleaned up when OpenSSL exits.
Therefore we are more selective about the symbols we include in legacy.so.
Fixes#13560
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14646)
with AVX512_IFMA + AVX512_VL instructions, primarily for RSA CRT private key
operations. It uses 256-bit registers to avoid CPU frequency scaling issues.
The performance speedup for RSA2k signature on ICL is ~2x.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13750)
Fixes#14559
The intitial implementation of the gets() function tried using the next bio's gets() function.
For a file BIO this returned incorrect data for binary data containing 0x00.
Just buffering all data during gets() did not work however since some
applications open and close the bio multiple times when dealing with pem
files containing multiple entries.. This does not work
when reading from stdin unless the data if buffered one byte at a time.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14599)
When encountering a badly coded item, the DER printer (ASN1_print_dump())
sets a flag to ensure that an additional hex dump of the offending content
is printed as part of the output. Unfortunately, this flag is never reset,
which means that all following items are printed with the extra hex dump,
whether they are faulty or not.
Resetting the flag after hex dumping ensures that only the faulty contents
are printed with the additional hex dump.
Fixes#14626
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14627)
We also rename it to d2i_PrivateKey_legacy(), to match d2i_PrivateKey_decoder()
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14314)
To make this cleaner, decoder_ms2key.c is split into decoder_msblob2key.c
and decoder_pvk2key.c.
This required a great deal of refactoring of crypto/pem/pvkfmt.c, to
make cleaner internal functions that our decoder implementations can
use.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14314)
This required refactoring a number of functions from the diverse
EVP_PKEY_ASN1_METHOD implementations to become shared backend
functions. It also meant modifying a few of them to return pointers
to our internal RSA / DSA/ DH / EC_KEY, ... structures instead of
manipulating an EVP_PKEY pointer directly, letting the caller do the
latter.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14314)
This makes it possible to use d2i_<TYPE>_PUBKEY instead of the generic
d2i_PUBKEY()
This required adding a number of new d2i_<TYPE>_PUBKEY functions.
These are all kept internal.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14314)
Also do not shortcut the pkey == NULL case
to allow EVP_PKEY_get_params() to raise an error.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14606)
If we attempt to init a provider but that init fails, then we should
still deregister any thread handlers. The provider may have failed after
these were registered.
Fixes#13338
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14576)
Since the Unicode 4.0.0 standard, the valid code point range is U+0000
to U+10FFFF. Make code points outside this range invalid when converting
from/to UTF-8.
Signed-off-by: Beat Bolli <dev@drbeat.li>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14185)
There is no need to make the suggested changes in the 3.0 timescale.
These are just suggested improvements for the future.
Fixes#14375
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14556)
The TODO described a case where a legacy derive operation is called, but
the peer key is provider based. In practice this will almost never be a
problem. We should never end up in our own legacy EVP_PKEY_METHOD
implementations if no ENGINE has been configured. If an ENGINE has been
configured then we we will be using a third party EVP_PKEY_METHOD
implementation and public APIs will be used to obtain the key data from the
peer key so there will be no "reaching inside" the pkey.
There is a theoretical case where a third party ENGINE wraps our own
internal EVP_PKEY_METHODs using EVP_PKEY_meth_find() or
EVP_PKEY_meth_get0(). For these cases we just ensure all our
EVP_PKEY_METHODs never reach "inside" the implementation of a peer key. We
can never assume that it is a legacy key.
Fixes#14399
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14555)
They aren't relevant:
. Digest Sign isn't supported in the FIPS provider.
. Remove legacy NID use.
Fixes#14394Fixes#14395
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14565)
All 3 files that included crypto/siphash.h also included siphash_local.h,
and no other files included siphash_local.h independently. They probably
should be just one header file.
Fixes#14360
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14558)
The comments are either about legacy stuff that is going to be
removed in later releases or about a safety check that can
be kept.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14554)
Using ossl_assert makes the build fail with --strict-warnings
because the ossl_assert is declared with warn_unused_result.
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14571)
We need to keep the check for prov == NULL in ossl_provider_libctx
but it is not needed in core_get_libctx as there it can happen only when
there is a serious coding error in a third party provider and returning
NULL as libctx would be seriously wrong as that has a special meaning.
The second TODO is valid but not something that is relevant
for 3.0. Change it into a normal comment.
Fixes#14377
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14535)
Without this, it is necessary to query an algorithm before setting the default
property query. With this, the value will be created and the default will
work.
Fixes#14516
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14542)
CLA: trivial
On line 136, a period is added. I think this is what was intended.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14540)
Fixes#14480
An internal flag that is set during param gen was not being tested, so
the wrong type was used to select the dsa domain param validation method.
In the default provider - if no gen_type is set then by default the fips186_4 gentype
will be selected when pbits >=2048 otherwise it selects fips186_2.
The fips provider ignores the gen_type and always uses fips186_4.
Before this change dsa used fips186_2 by default in the default
provider.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14508)
The first TODO 3.0 is not really a TODO, just a comment.
The second one is something that is needed for compatibility
with existing applications. There is no major reason in
trying to change this behavior right now.
Fixes#14400
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14534)
The TODO is describing something that would be nice to fix. In fact the
problem exists even in 1.1.1. It would be nice to fix it, but it does
not need to be done in the 3.0 timeframe.
Fixes#14376
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14533)
There is nothing to be done here for the time being. If at some point
we make the async code libctx aware then we might need to make a change
but there are no plans to do that at the moment.
Fixes#14402
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14532)
Fixes#14390
The only caller of this function tests EVP_KEYMGMT_is_a() beforehand
which will fail if the RSA key types do not match. So the test is not
necessary. The assert has been removed when it does the test.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14524)
Some functions that lock things are void, so we just return early.
Also make ossl_namemap_empty return 0 on error. Updated the docs, and added
some code to ossl_namemap_stored() to handle the failure, and updated the
tests to allow for failure.
Fixes: #14230
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14238)
Calling OPENSSL_init_crypto(0, NULL) is a no-op and will
not properly initialize thread local handling.
Only the calls that are needed to initialize thread locals
are kept, the rest of the no-op calls are removed.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14497)
To avoid recursive lock issues, a copy is taken of the provider list and
the callbacks are made without holding the store lock.
Fixes#14251
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14489)
Fixes#13914
The "SRWLock" synchronization primitive is available in Windows Vista
and later. CRYPTO_THREAD functions now use SRWLock functions when the
target operating system supports them.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14381)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14429)
Providers (particularly the FIPS provider) needs access to BIOs from libcrypto.
Libcrypto is allowed to change the internal format of the BIO structure and it
is still expected to work with providers that were already built. This means
that the libcrypto BIO must be distinct from and not castable to the provider
side OSSL_CORE_BIO.
Unfortunately, this requirement was broken in both directions. This fixes
things by forcing the two to be different and any casts break loudly.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14419)
Fixes#13185Fixes#13352
Removed the existing code in file_store that was trying to figure out the
input type.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14407)
This allows BIO_tell() and BIO_seek() to work for BIO's that do
not support these methods. The main use case for this is file/fd BIO's
that use stdin.
This works for stdin taken from input redirection (command < file),
and stdin via pipe (cat file | command).
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14407)
Use the modern defaults as now set in the pkcs12 app. This also
allows modifying the application to not override the default values
when calling the API.
Fixes#14034
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/14450)
A trivial PR to remove some commonly repeated words. It looks like this is
not the first PR to do this.
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/14420)
OTC have decided that the EVP_PKEY_get0* functions should have a const
return type. This is a breaking change to emphasise that these values
should be considered as immutable.
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14319)
Most of these were already deprecated but a few have been missed. This
commit corrects that.
Fixes#14303Fixes#14317
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14319)
If someone calls an EVP_PKEY_get0*() function then we create a legacy
key and cache it in the EVP_PKEY - but it doesn't become an "origin" and
it doesn't ever get updated. This will be documented as a restriction of
the EVP_PKEY_get0*() function with provided keys.
Fixes#14020
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14319)
Make sure we were sucessful in creating an EVP_PKEY
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14319)
This fixes also failure behavior of OSSL_HTTP_REQ_CTX_sendreq_d2i(), OCSP_sendreq_nbio(), etc.
Fixes#14322
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14356)
provider_init() makes changes in the provider structure, and needs a
bit of protection to ensure that doesn't happen concurrently with race
conditions.
This also demands a bit of protection of the flags, since they are
bits and presumably occupy the same byte in memory.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14354)
We can try to do that although for legacy keys the keymgmt
will not be set. This function will disappear with legacy support
removed.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14404)
The OSSL_PARAM_set_BN() pads to data_size so there is no
need for OSSL_PARAM_set_BN_pad().
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14404)
The chacha and poly1305 algorithms are not FIPS approved so
they should stay out of FIPS module.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14404)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14405)
Fixes#13995
For small keys the MR test on the modulus can return
BN_PRIMETEST_COMPOSITE_WITH_FACTOR status although the modulus
is correct.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14389)
This allows making the signature operations return different
settable params when the context is initialized with
EVP_DigestSign/VerifyInit.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14338)
Also improve diagnostics on inconsistent cert request input in apps/cmp.c,
add trace output for transactionIDs on new sessions,
and update the documentation in openssl-cmp.pod.in.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14018)
Now handle [http[s]://][userinfo@]host[:port][/path][?query][#frag]
by optionally providing any userinfo, query, and frag components.
All usages of this function, which are client-only,
silently ignore userinfo and frag components,
while the query component is taken as part of the path.
Update and extend the unit tests and all affected documentation.
Document and deprecat OCSP_parse_url().
Fixes an issue that came up when discussing FR #14001.
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14009)
This introduces the encoder output type "blob", to be used for
anything that outputs an unstructured blob of data.
Fixes#14258
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/14291)
Return an error instead of trying to malloc a negative number.
The other usage in this file already had a similar check, and the caller
should have put an entry on the error stack already.
Note that we only check the initial calls to obtain the encoded length,
and assume that the follow-up call to actually encode to the allocated
storage will succeed if the first one did.
Fixes: #14177
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14308)
Partial fix for #12964
This adds ossl_ names for the following symbols:
ec_*, ecx_*, ecdh_*, ecdsa_*, sm2_*
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14231)
Partial fix for #12964
This adds ossl_ names for the following symbols:
dsa_check_pairwise, dsa_check_params, dsa_check_priv_key, dsa_check_pub_key, dsa_check_pub_key_partial,
dsa_do_sign_int, dsa_ffc_params_fromdata,
dsa_generate_ffc_parameters, dsa_generate_public_key,
dsa_get0_params, dsa_key_fromdata, dsa_new_with_ctx, dsa_pkey_method, dsa_sign_int
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14231)
If input offset_sec is sufficiently large (> INT32_MAX * SECS_PER_DAY, which is possible for a long on 64-bit platforms), then the first assignment contains an overflow.
I think leaving offset_hms as an int is still safe.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14252)
We don't want to hold a read lock when calling a user supplied callback.
That callback could do anything so the risk of a deadlock is high.
Instead we collect all the names first inside the read lock, and then
subsequently call the user callback outside the read lock.
Fixes#14225
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14250)
OSSL_PARAM_BLD_push_utf8_string() was still setting the length in
bytes of the UTF8 string to include the terminating NUL byte, while
recent changes excludes that byte from the length. It's still made to
add a NUL byte at the end of the string no matter what.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14035)
OSSL_PARAM_allocate_from_text() was still setting the length in bytes
of the UTF8 string to include the terminating NUL byte, while recent
changes excludes that byte from the length.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14035)
Errors raised from a provider that is subsequently unloaded from memory
may have references to strings representing the file and function that
are no longer present because the provider is no longer in memory. This
can cause crashes. To avoid this we duplicate the file and func strings.
Fixes#13623
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14213)
Without this, a provider has no way to know that an application
has finished with the array it returned earlier. A non-caching provider
requires this information.
Fixes#12974
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/12974)
The functions are not needed and require returning octet ptr parameters
from providers that would like to support them which complicates provider
implementations.
Fixes#12985
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14279)
When a SubjectPublicKeyInfo (SPKI) is decoded into an X509_PUBKEY
structure, the corresponding EVP_PKEY is automatically added as well.
This used to only support our built-in keytypes, and only in legacy
form.
This is now refactored by making The ASN1 implementation of the
X509_PUBKEY an EXTERN_ASN1, resulting in a more manual implementation
of the basic support routines. Specifically, the d2i routine will do
what was done in the callback before, and try to interpret the input
as an EVP_PKEY, first in legacy form, and then using OSSL_DECODER.
Fixes#13893
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14281)
Add an argument to PROVIDER_try_load() that permits a provider to be
loaded without changing the fallback status. This is useful when an
additional provider needs to be loaded without perturbing any other setup.
E.g. adding mock providers as part of unit testing.
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13652)
It would check the keytype and optype before determining if it even
supported the ctrl command number. This turned out to be disruptive,
so we make it check that it supports the request ctrl command number
first.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13913)
legacy_ctrl_to_param() and legacy_ctrl_str_to_param() are now
replaced with calls to evp_pkey_ctx_ctrl_to_param() and
evp_pkey_ctx_ctrl_str_to_param().
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13913)
EVP_PKEY_get_group_name() now simply calls EVP_PKEY_get_utf8_string_param().
EVP_PKEY_CTX_set_group_name() now simply calls EVP_PKEY_CTX_set_params().
EVP_PKEY_get_bn_param(), EVP_PKEY_get_octet_string_param(),
EVP_PKEY_get_utf8_string_param() and EVP_PKEY_get_int_param() can now
handle legacy EVP_PKEYs by calling evp_pkey_get_params_to_ctrl().
EVP_PKEY_CTX_get_params() can now handle a legacy backed EVP_PKEY_CTX
by calling evp_pkey_ctx_get_params_to_ctrl().
Note: EVP_PKEY_CTX_set_params() doesn't call the translator yet.
Should it ever?
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13913)
In the interest of calling these functions on legacy EVP_PKEY
contexts, only check the settable / gettable params for provider side
keys, leaving to the translated EVP_PKEY_CTX_ctrl() call check the
ctrl commands on its own.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13913)
The idea is to make it as transparent as possible to call things like
EVP_PKEY_CTX_ctrl() with a provider backed EVP_PKEY_CTX, or things
like EVP_PKEY_get_bn_param() with a legacy EVP_PKEY.
All these sorts of calls demand that we translate between ctrl
commands and OSSL_PARAM keys, and treat the arguments appropriately.
This implementation has it being as data driven as possible, thereby
centralizing everything into one table of translation data, which
supports both directions.
Fixes#13528
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13913)
This will help with transitioning diverse functions to be able to use the
ctrl<->OSSL_PARAM translators.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13913)
This does what was previously done by looking at pctx->pmeth->pkey_id,
but handles both legacy and provider side contexts, and is supposed to
become a replacement for the old way.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/13913)
Fixes#14263
If the new decoder code fails, it now falls back to the old legacy code
and tries that also.
Tested manually using gost engine master.
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/14266)
Adding the EVP_PKEY_param_check_quick() reminded me that there are also
partial checks for public keys as part of SP800-56A for FFC (DH named safe
prime groups) and ECC. The code was mainly already there and just needed
to be plumbed into the validate methods.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14206)
Any decoder can now also declare the name of the data structure for
the object it decoded in the OSSL_PARAM array they pass back to the
decoding process. The decoding process will use that as another
criterion to select the next decoder in the chain to consider.
Together with declaring the data type, this becomes a means to refine
how the decoded data is treated along the chain.
Fixes#13539
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14233)
BIO_get_ktls_send() and BIO_get_ktls_recv() are documented as
returning either 0 or 1. However, they were actually returning the
internal value of the associated BIO flag for the true case instead of
1.
Also trim redundant ternary operators.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14023)
The incorrect code is in #ifdef branch that is normally
not compiled in.
Signed-off-by: Zhang Jinde <zjd5536@163.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/12968)
The dh->nid was not being set if the loaded p,g matched an inbuilt named
group for "DH".
NOTE: The "DHX" related path already worked since it calls DH_set0_pqg()
(which does the name group check).
This bug was detected when new tests were added for dh5114 groups, combined
with the no-cache tests i.e. loading+import+export set the nid,
but just loading did not.
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14207)
The pem2der decoder can infer certain information about the endoded der
data based on the PEM headers. This information should be passed to the
next decoders in the chain to ensure we end up loading the correct type of
thing.
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14191)
When the string "ABCDEFGH" is passed, what's considered its data, this?
{ 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H' }
or this?
{ 'A', 'B', 'C', 'D', 'E', 'F', 'G', 'H', '\0' }
If it's passed as a pass phrase, should the terminating NUL byte be
considered part of the pass phrase, or not?
Our treatment of OSSL_PARAMs with the data type OSSL_PARAM_UTF8_STRING
set the length of the string to include the terminating NUL byte,
which is quite confusing. What should the recipient of such a string
believe?
Instead of perpetuating this confusion, we change the assumption to
set the OSSL_PARAM to the length of the string, not including the
terminating NUL byte, thereby giving it the same value as a strlen()
call would give.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14168)
Partial fix for #12964
This adds ossl_ names for the following symbols:
blake2b512_init,blake2b_final,blake2b_init,blake2b_init_key,
blake2b_param_init,blake2b_param_set_digest_length,blake2b_param_set_key_length,
blake2b_param_set_personal,blake2b_param_set_salt,blake2b_update,
blake2s256_init,blake2s_final,blake2s_init,blake2s_init_key,
blake2s_param_init,blake2s_param_set_digest_length,blake2s_param_set_key_length,
blake2s_param_set_personal,blake2s_param_set_salt,blake2s_update,
digest_default_get_params,digest_default_gettable_params
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14211)
Added primality check on p and q in the ossl_ffc_params_simple_validate().
Checking for p and q sizes in the default provider is made more
lenient.
Added two testcases for invalid parameters.
Fixes#13950
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14148)
Fixes: Variable "sk_untrusted" going out of scope leaks the storage it points to.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14187)
Additional renames done in encoder and decoder implementation
to follow the style.
Fixes#13622
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14155)
Fixes#14183. Fix the condition to detect legacy engines, so the
`props` are considered even when libctx == NULL.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14188)
This also fixes the public function RSA_padding_check_SSLv23.
Commit 6555a89 changed the padding check logic in RSA_padding_check_SSLv23
so that padding is rejected if the nul delimiter byte is not immediately
preceded by at least 8 bytes containing 0x03. Prior to that commit the
padding is rejected if it *is* preceded by at least 8 bytes containing 0x03.
Presumably this change was made to be consistent with what it says in
appendix E.3 of RFC 5246. Unfortunately that RFC is in error, and the
original behaviour was correct. This is fixed in later errata issued for
that RFC.
This has no impact on libssl for modern versions of OpenSSL because
there is no protocol support for SSLv2 in these versions. However
applications that call RSA_paddin_check_SSLv23 directly, or use the
RSA_SSLV23_PADDING mode may still be impacted. The effect of the original
error is that an RSA message encrypted by an SSLv2 only client will fail to
be decrypted properly by a TLS capable server, or a message encrypted by a
TLS capable client will fail to decrypt on an SSLv2 only server. Most
significantly an RSA message encrypted by a TLS capable client will be
successfully decrypted by a TLS capable server. This last case should fail
due to a rollback being detected.
Thanks to D. Katz and Joel Luellwitz (both from Trustwave) for reporting
this issue.
CVE-2021-23839
Reviewed-by: Paul Dale <pauli@openssl.org>
The OpenSSL public API function X509_issuer_and_serial_hash() attempts
to create a unique hash value based on the issuer and serial number data
contained within an X509 certificate. However it fails to correctly
handle any errors that may occur while parsing the issuer field (which
might occur if the issuer field is maliciously constructed). This may
subsequently result in a NULL pointer deref and a crash leading to a
potential denial of service attack.
The function X509_issuer_and_serial_hash() is never directly called by
OpenSSL itself so applications are only vulnerable if they use this
function directly and they use it on certificates that may have been
obtained from untrusted sources.
CVE-2021-23841
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
The low level DH API has two functions for checking parameters:
DH_check_ex() and DH_check_params_ex(). The former does a "full" check,
while the latter does a "quick" check. Most importantly it skips the
check for a safe prime. We're ok without using safe primes here because
we're doing ephemeral DH.
Now that libssl is fully using the EVP API, we need a way to specify that
we want a quick check instead of a full check. Therefore we introduce
EVP_PKEY_param_check_quick() and use it.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14146)
Fix CID 1472833 by removing a codepath that attempts to allocate a
stack if not already allocated, when the stack was already allocated
unconditionally a few lines previously.
Interestingly enough, this additional allocation path (and the comment
describing the need for it) were added in commit
69664d6af0, also prompted by Coverity(!).
It seems that the intervening (and much more recent) commit
d53b437f99 that allowed sk_X509_dup()
to accept a NULL argument allowed the earlier initialization path
to unconditionally allocate a stack, rendering this later allocation fully
redundant.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/14161)
Fixes CID #1472393
Previously this switch handled CMS & PCKS7 controls (e.g ANS1_PKEY_CTRL_PKCS7_SIGN)
which fell thru to the dead code to set the X509_ALG.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14163)
Using ERR_LIB_* causes the error output to say 'reason(n)' instead of
the name of the sub-library in question.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14152)
The OTC decided that all low level APIs should be deprecated. This extends
to SRP, even though at the current time there is no "EVP" interface to it.
This could be added in a future release.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14132)
RFC 8805 Geofeed files can be authenticated with RPKI
CLA: trivial
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14050)
Pauli
Matt Caswell
Shane Lontis
Andrey Matyukov
Richard Levitte
Tomas Mraz
Jon Spillett
Dr. David von Oheimb
Beat Bolli
div2016bit
Rich Salz
Vincent Drake
Pedro Monreal
Dmitry Belyavskiy
Tobias Nießen
Benjamin Kaduk
jwalch
Daniel Bevenius
Pauli
John Baldwin
Zhang Jinde
Sahana Prasad
Petr Gotthard
Job Snijders