When the providers change, the method cache needs to be flushed. This also
impacts the cache is full partial flushes and the algorithm flushing by ID.
A new function is introduced to clear all of the operation bits in all
providers in a library context.
Fixes#15032
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15042)
Reviewed-by: Richard Levitte <levitte@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/15130)
libimplementations.a was a nice idea, but had a few flaws:
1. The idea to have common code in libimplementations.a and FIPS
sensitive helper functions in libfips.a / libnonfips.a didn't
catch on, and we saw full implementation ending up in them instead
and not appearing in libimplementations.a at all.
2. Because more or less ALL algorithm implementations were included
in libimplementations.a (the idea being that the appropriate
objects from it would be selected automatically by the linker when
building the shared libraries), it's very hard to find only the
implementation source that should go into the FIPS module, with
the result that the FIPS checksum mechanism include source files
that it shouldn't
To mitigate, we drop libimplementations.a, but retain the idea of
collecting implementations in static libraries. With that, we not
have:
libfips.a
Includes all implementations that should become part of the FIPS
provider.
liblegacy.a
Includes all implementations that should become part of the legacy
provider.
libdefault.a
Includes all implementations that should become part of the
default and base providers.
With this, libnonfips.a becomes irrelevant and is dropped.
libcommon.a is retained to include common provider code that can be
used uniformly by all providers.
Fixes#15157
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15171)
It was discovered that eddsa.c exist in two places, here and in
crypto/ec/curve448/, which would result in a file name clash if they
ever end up in the same library.
To mitigate, we rename the copy in providers/implementations/signatures
to have '_sig' in the file name, and do the same with all other source
files in this directory, for consistency.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15171)
External ENCODER may not implement OSSL_FUNC_ENCODER_IMPORT_OBJECT,
so a check for NULL is needed.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/14919)
When reading the comment for ossl_dispatch_st it seems to indicate that
the function_id numbers are defined further down in the same file. But I
was not able to find them there, but instead in core_dispatch.h.
This commit suggests updating the comment to point to core_dispatch.h
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15156)
Add OSSL_STORE_PARAM_INPUT_TYPE and make it possible to be
set when OSSL_STORE_open_ex() or OSSL_STORE_attach() is called.
The input type format is enforced only in case the file
type file store is used.
By default we use FORMAT_UNDEF meaning the input type
is not enforced.
Fixes#14569
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15100)
The code path shouldn't occur in our code but could in an application.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/15128)
Removed error codes, and the mention of the functions.
This removal is already documented in the CHANGES doc.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15140)
Also remove redundant opt_name() and make names of opt_{i,u}ntmax() consistent.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15111)
The "bad DTLS" tests run into trouble due to the special behavior
for that "bad" version, and the SSL record tests need to set the
-legacy_server_connect flag to allow an SSLv2 ClientHello to work
against any TLS server (since SSLv2 ClientHello messages cannot
carry extensions as would be needed in order to negotiate the use
of the renegitiation_info extension).
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15127)
This option is only useful for the client, but it was previously
marked as only being applicable for servers.
Correct the entry to properly mark it as client-only, and update the
s_server/s_client manuals accordingly.
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15127)
Previously we would set SSL_OP_LEGACY_SERVER_CONNECT by default in
SSL_CTX_new(), to allow connections to legacy servers that did not
implement RFC 5746.
It has been more than a decade since RFC 5746 was published, so
there has been plenty of time for implmentation support to roll out.
Change the default behavior to be to require peers to support
secure renegotiation. Existing applications that already cleared
SSL_OP_LEGACY_SERVER_CONNECT will see no behavior change, as
re-clearing the flag is just a little bit of redundant work.
The old behavior is still available by explicitly setting the flag
in the application.
Also remove SSL_OP_LEGACY_SERVER_CONNECT from SSL_OP_ALL, for
similar reasons.
Document the behavior change in CHANGES.md, and update the
SSL_CTX_set_options() and SSL_CONF_cmd manuals to reflect the change
in default behavior.
Fixes: 14848
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15127)
Also add a negative test, and fix typo's.
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15090)
It was a mistake to allow relative paths for include files (just
like root shouldn't have "." in its PATH), but we probably can't
change it now. Add a new pragma "abspath" that someone can put
in the system-wide config file to require absolute paths.
Also update the config documentation to better explain how file
inclusion works.
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15090)
Use official (first) names for wrapping algorithms.
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/14182)
Setting an output length higher than 8191 was causing a buffer overflow.
This was reported by Acumen (FIPS lab).
The max output size has increased to ~2M and it now checks this during set_parameters.
The encoder related functions now pass in the maximum size of the output buffer so they
can correctly check their size. kmac_bytepad_encode_key() calls bytepad twice in
order to calculate and check the length before encoding.
Note that right_encode() is currently only used in one place but this
may change if other algorithms are supported (such as TupleHash).
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15106)
