- Dec 13, 2018
-
-
Mansour Ahmadi authored
Fixes #7657 Reviewed-by:
Nicola Tuveri <nic.tuv@gmail.com> Reviewed-by:
Matt Caswell <matt@openssl.org> (Merged from https://github.com/openssl/openssl/pull/7877)
-
Mansour Ahmadi authored
Fixes #7650 Reviewed-by:
Kurt Roeckx <kurt@roeckx.be> Reviewed-by:
-
Mansour Ahmadi authored
Fixes #7117 Reviewed-by:
-
Rich Salz authored
Reviewed-by:
Richard Levitte <levitte@openssl.org> Reviewed-by:
Paul Yang <yang.yang@baishancloud.com> (Merged from https://github.com/openssl/openssl/pull/7423)
-
- Dec 12, 2018
-
-
Richard Levitte authored
It turned out that .S files aren't to be treated as lightly as I thought. They need to go through a preprocessing step, which .s files don't need to. Corrects #7703 Reviewed-by:
Tim Hudson <tjh@openssl.org> (Merged from https://github.com/openssl/openssl/pull/7889)
-
Dr. Matthias St. Pierre authored
Fixes #7883 Reviewed-by:
-
Jakub Jelen authored
CLA: trivial Signed-off-by:
Jakub Jelen <jjelen@redhat.com> Reviewed-by:
-
Richard Levitte authored
Reviewed-by:
-
Tobias Stoeckmann authored
There was a trailing :w at a line, which didn't make sense in context of the sentence/styling. Removed it, because I think it's a leftover vi command. CLA: trivial Signed-off-by:
Tobias Stoeckmann <tobias@stoeckmann.org> Reviewed-by:
Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com> Reviewed-by:
-
- Dec 11, 2018
-
-
Todd Short authored
Based originally on github.com/dfoxfranke/libaes_siv This creates an SIV128 mode that uses EVP interfaces for the CBC, CTR and CMAC code to reduce complexity at the cost of perfomance. The expected use is for short inputs, not TLS-sized records. Add multiple AAD input capacity in the EVP tests. Reviewed-by:
Paul Dale <paul.dale@oracle.com> Reviewed-by:
Bernd Edlinger <bernd.edlinger@hotmail.de> (Merged from https://github.com/openssl/openssl/pull/3540)
-
Matt Caswell authored
Reviewed-by:
-
Matt Caswell authored
Check that s is less than the order before attempting to verify the signature as per RFC8032 5.2.7 Fixes #7706 Reviewed-by:
-
nxtstep authored
CLA: trivial Reviewed-by:
-
Richard Levitte authored
Reviewed-by:
-
Richard Levitte authored
This was forgotten in a few places Fixes #7862 Reviewed-by:
-
- Dec 10, 2018
-
-
Richard Levitte authored
Fixes #7861 Reviewed-by:
-
Richard Levitte authored
The regexp to parse the incoming version number was flawed, and since we allow ourselves to add missing APIs in PATCH releases, the compatibility settings still need to include the PATCH part of the version number. Reviewed-by:
-
Eneas U de Queiroz authored
This is useful to determine the kernel driver running each algorithm. Signed-off-by:
Eneas U de Queiroz <cote2004-github@yahoo.com> Reviewed-by:
-
Eneas U de Queiroz authored
USE_SOFTDRIVERS: whether to use software (not accelerated) drivers CIPHERS: list of ciphers to enable DIGESTS: list of digests to enable Signed-off-by:
-
Eneas U de Queiroz authored
Since each ioctl causes a context switch, slowing things down, if EVP_MD_CTX_FLAG_ONESHOT is set, then: - call the ioctl in digest_update, saving the result; and - just copy the result in digest_final, instead of using another ioctl. Signed-off-by:
-
Eneas U de Queiroz authored
Digest must be able to do partial-state copy to be used. Signed-off-by:
-
Eneas U de Queiroz authored
Make CTR mode behave like a stream cipher. Signed-off-by:
-
Eneas U de Queiroz authored
The engine needs a custom cipher context copy function to open a new /dev/crypto session. Signed-off-by:
-
Eneas U de Queiroz authored
Close the session in digest_cleanup instead of digest_final. A failure in closing the session does not mean a previous successful digest final has failed as well. Signed-off-by:
-
Eneas U de Queiroz authored
If the source ctx has not been initialized, don't initialize the copy either. Signed-off-by:
-
Eneas U de Queiroz authored
Return failure when the digest_ctx is null in digest_update and digest_final, and when md is null in digest_final. Signed-off-by:
-
Eneas U de Queiroz authored
Call functions to prepare methods after confirming that /dev/crytpo was sucessfully open and that the destroy function has been set. Signed-off-by:
-
Eneas U de Queiroz authored
Signed-off-by:
-
Matt Caswell authored
Fixes #7727 Reviewed-by:
-
Matt Caswell authored
For the same reasons as in the previous commit we must preserve errno across dlopen calls. Some implementations (e.g. solaris) do not preserve errno even on a successful dlopen call. Fixes #6953 Reviewed-by:
-
Matt Caswell authored
This function can end up being called during ERR_get_error() if we are initialising. ERR_get_error() must preserve errno since it gets called via SSL_get_error(). If that function returns SSL_ERROR_SYSCALL then you are supposed to inspect errno. Reviewed-by:
-
Richard Levitte authored
Reviewed-by:
-
Richard Levitte authored
Reviewed-by:
-
Viktor Dukhovni authored
Engine names and ids are typically static strings. If an application actually dynamically allocated these, the application owns the storage, and should dispose of it via the original handle, rather than the "const char *" returned by the engine. In any case, this resolves the test code issue without resort to "unconst" macros/casts. Reviewed-by: -
Viktor Dukhovni authored
Reviewed-by: -
Christos Zoulas authored
Reviewed-by:
Viktor Dukhovni <viktor@openssl.org>
-
Christos Zoulas authored
strings. Reviewed-by:
-
Christos Zoulas authored
Reviewed-by:
-
Christos Zoulas authored
Reviewed-by:
-
Christos Zoulas authored
Reviewed-by:
-
