Only field multiplication and squaring (but not reduction) show a
significant improvement. This is enabled on Power ISA >= 3.0.
On a Power 9 CPU an average 10% performance improvement is seen (ECHDE:
14%, ECDSA sign: 6%, ECDSA verify 10%), compared to existing code.
On an upcoming Power 10 CPU we see an average performance improvement
of 26% (ECHDE: 38%, ECDSA sign: 16%, ECDSA verify 25%), compared to
existing code.
Signed-off-by: Amitay Isaacs <amitay@ozlabs.org>
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15401)
This is only used if ECP_NISTP521_ASM is defined and this currently
only occurs on PPC64.
This simply chooses the C reference implementation, which will be the
default when custom code is available for certain CPUs.
Only the multiplication and squaring operations are handled, since the
upcoming assembly code only contains those. This scheme can be easily
extended to handle reduction too.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Signed-off-by: Amitay Isaacs <amitay@ozlabs.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15401)
This will allow clean addition of assembly versions of these operations.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15401)
Power has 2 numbering systems for vector registers:
* VR: Vector Registers are numbered from 0 to 31
* VSR: Vector-Scalar registers are numbers from 32 to 63
These refer to the same registers. Some instructions use VR numbering
for their operands, while others use VSR numbering.
When using Perl to provide a meaningful name for a register it makes
sense to use the same variable for both VR and VSR instructions. This
makes the code more readable.
However, providing a VSR number (i.e. >=32) to an instruction that
expects a VR number will cause an assembler error.
So, for instructions that require VR numbering, map VSR numbers
(i.e. >=32) to VR numbers. This also allows existing code that uses
VR numbering to remain unchanged.
Signed-off-by: Martin Schwenke <martin@meltin.net>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15401)
Fixes#15515
Another case of the order that tests run in causes a failure.
A new test was loading "legacy" into the default lib ctx. If it
ran first then everything fails. The test now has its own lib ctx.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15516)
EVP_EncryptInit.pod now follows the pattern used in EVP_DigestInit.pod.
i.e.
'=item' is used for methods
PARAMETERS and CONTROLS sections have been added.
The PARAMETERS list has been moved from provider-cipher.pod (this file just
has a link now).
Missing fields were updated.
The CONTROLS shows the mappings to OSSL_PARAM keys.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15416)
CLA: trivial
Signed-off-by: Sven Schwermer <sven.schwermer@disruptive-technologies.com>
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/15495)
This fixes a compiler warning on clang-1205.0.22.9 when compiling the
generated code as C++11:
ISO C++11 does not allow conversion from string literal to 'char *'
[-Wwritable-strings]
CLA: trivial
Signed-off-by: Sven Schwermer <sven.schwermer@disruptive-technologies.com>
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/15495)
Fixes#15392
PBE algorithms such as NID_pbe_WithSHA1And3_Key_TripleDES_CBC will
currently always fail to the EVP_CIPHER_fetch() call, so the fallback to
a legacy algorithm always happens. In this case the error stack should
ignore the fetch error.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15473)
The ctx->propq is strdup'ed, so it must be free'd too.
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15458)
Remove unused -DCONF_DEBUG and -DBN_CTX_DEBUG.
Rename REF_PRINT to REF_DEBUG for consistency, and add a new
tracing category and use it for printing reference counts.
Rename -DDEBUG_UNUSED to -DUNUSED_RESULT_DEBUG
Fix BN_DEBUG_RAND so it compiles and, when set, force DEBUG_RAND to
be set also.
Rename engine_debug_ref to be ENGINE_REF_PRINT also for consistency.
Fixes#15357
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15353)
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/15483)
Signed-off-by: Tom Cosgrove <tom.cosgrove@arm.com>
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/15486)
Otherwise, the state machine ends up being in a bad state:
```
SSL routines:write_state_machine:missing fatal:ssl/statem/statem.c:XXX:
```
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/15487)
Parameter "header" of ssl3_cbc_digest_record was fixed to a 13 bytes header
but used as a pointer. This caused a warning about out-of-bounds array access
with GCC 11.
Fixes#15462.
Signed-off-by: Juergen Christ <jchrist@linux.ibm.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15463)
These multi-prime tests were omitted when genrsa was deprecated but not
returned when it was restored.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15472)
These tests were omitted when genrsa was deprecated but not returned when
it was restored.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15472)
During key generation, any sized key can be asked for. Attempting to generate
a key with a security strength larger than the RNG strength now fails.
Fixes#15421
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15472)
The symbols renamed are:
RSAZ_amm52x20_x1_256
RSAZ_amm52x20_x2_256
rsaz_avx512ifma_eligible
RSAZ_mod_exp_avx512_x2
Additionally, RSAZ_exp52x20_x2_256 was made static
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/15445)
