oweals/openssl.git
7 years agoBN_pseudo_rand is really BN_rand
Rich Salz [Wed, 21 Jun 2017 12:55:02 +0000 (13:55 +0100)]
BN_pseudo_rand is really BN_rand

And BN_pseudo_rand_range is really BN_rand_range.
Document that we might deprecate those functions.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3743)

7 years ago'make update' after objects/objects.txt update.
Andy Polyakov [Sat, 24 Jun 2017 19:26:44 +0000 (21:26 +0200)]
'make update' after objects/objects.txt update.

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3643)

7 years agoobjects/objects.txt: add SHA3 OIDs.
Andy Polyakov [Sun, 18 Jun 2017 21:30:24 +0000 (23:30 +0200)]
objects/objects.txt: add SHA3 OIDs.

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3643)

7 years agoCorrect documentation for UI_get0_result_string
Richard Levitte [Sat, 1 Jul 2017 16:28:50 +0000 (18:28 +0200)]
Correct documentation for UI_get0_result_string

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3824)

7 years agosha/keccak1600.c: internalize KeccakF1600 and simplify SHA3_absorb.
Andy Polyakov [Fri, 30 Jun 2017 11:35:59 +0000 (13:35 +0200)]
sha/keccak1600.c: internalize KeccakF1600 and simplify SHA3_absorb.

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
7 years agosha/asm/keccak1600-x86_64.pl: close gap with Keccak Code Package.
Andy Polyakov [Fri, 30 Jun 2017 09:45:34 +0000 (11:45 +0200)]
sha/asm/keccak1600-x86_64.pl: close gap with Keccak Code Package.

[Also typo and readability fixes. Ryzen result is added.]

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
7 years agosha/asm/keccak1600-s390x.pl: typo and readability, minor size optimization.
Andy Polyakov [Fri, 30 Jun 2017 09:42:58 +0000 (11:42 +0200)]
sha/asm/keccak1600-s390x.pl: typo and readability, minor size optimization.

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
7 years agox86_64 assembly pack: fill some blanks in Ryzen results.
Andy Polyakov [Fri, 30 Jun 2017 09:42:00 +0000 (11:42 +0200)]
x86_64 assembly pack: fill some blanks in Ryzen results.

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
7 years agoVarious doc fixes
Rich Salz [Sun, 2 Jul 2017 16:16:38 +0000 (12:16 -0400)]
Various doc fixes

Fix a =head1 section name
Fix a typo in POD label
Remove a spurious =back
Add a missing blank line
Avoid 'legacy' -- use 'deprecated' if still needed if we cannot just reword.
Always do strict checking
Do not warn about missing "RETURN VALUES" unless -s is set.
Change OpenSSL version 1.1 -> 1.1.0

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3830)

7 years agoFix potential crash in tls_construct_finished.
Bernd Edlinger [Wed, 14 Jun 2017 18:25:52 +0000 (20:25 +0200)]
Fix potential crash in tls_construct_finished.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3667)

7 years agoFix a crash in tls_construct_client_certificate.
Bernd Edlinger [Tue, 13 Jun 2017 05:25:43 +0000 (07:25 +0200)]
Fix a crash in tls_construct_client_certificate.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3667)

7 years agoFix TLSv1.3 exporter
Matt Caswell [Fri, 30 Jun 2017 09:45:11 +0000 (10:45 +0100)]
Fix TLSv1.3 exporter

We need to use the hashsize in generating the exportsecret not 0! Otherwise
we end up with random garbage for the secret.

It was pure chance that this passed the tests previously. It so happens
that, because we call SSL_export_keying_material() repeatedly for different
scenarios in the test, we end up in the tls13_export_keying_material() at
exactly the same position in the stack each time and therefore end up using
the same random garbage secret each time!

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3810)

7 years agoRemove the possibility to disable the UI module entirely
Richard Levitte [Sat, 1 Jul 2017 10:14:37 +0000 (12:14 +0200)]
Remove the possibility to disable the UI module entirely

Instead, make it possible to disable the console reader that's part of
the UI module.  This makes it possible to use the UI API and other UI
methods in environments where the console reader isn't useful.

To disable the console reader, configure with 'no-ui-console' /
'disable-ui-console'.

'no-ui' / 'disable-ui' is now an alias for  'no-ui-console' /
'disable-ui-console'.

Fixes #3806

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3820)

7 years agoMake sure OSSL_STORE_load() isn't caught in an endless loop
Richard Levitte [Sat, 1 Jul 2017 16:25:43 +0000 (18:25 +0200)]
Make sure OSSL_STORE_load() isn't caught in an endless loop

The post process callback might potentially say "no" to everything (by
constantly returning NULL) and thereby cause an endless loop.  Ensure
that we stop all processing when "eof" is reached.

Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/3823)

7 years agoFix copyright date for the ARIA evp file.
Pauli [Mon, 3 Jul 2017 02:14:59 +0000 (12:14 +1000)]
Fix copyright date for the ARIA evp file.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3831)

7 years agoUpdate fuzz corpora
Kurt Roeckx [Sun, 2 Jul 2017 15:40:51 +0000 (17:40 +0200)]
Update fuzz corpora

Reviewed-by: Rich Salz <rsalz@openssl.org>
GH: #3829

7 years agoSTORE 'file' scheme loader: DNS name in URI is case insensitive
Richard Levitte [Sun, 2 Jul 2017 09:54:40 +0000 (11:54 +0200)]
STORE 'file' scheme loader: DNS name in URI is case insensitive

...  so compare accordingly with "//localhost"

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3827)

7 years agoWhen apps_startup() fails, exit with a failure code and a message
Richard Levitte [Fri, 30 Jun 2017 18:47:45 +0000 (20:47 +0200)]
When apps_startup() fails, exit with a failure code and a message

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3816)

7 years agoCleanup some copyright stuff
Rich Salz [Thu, 15 Jun 2017 16:03:40 +0000 (12:03 -0400)]
Cleanup some copyright stuff

Remove some incorrect copyright references.
Move copyright to standard place
Add OpenSSL copyright where missing.
Remove copyrighted file that we don't use any more
Remove Itanium assembler for RC4 and MD5 (assembler versions of old and
weak algorithms for an old chip)
Standardize apps/rehash copyright comment; approved by Timo
Put dual-copyright notice on mkcert

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3691)

7 years agoFix build with no-threads no-ec
Dr. Stephen Henson [Fri, 30 Jun 2017 18:55:47 +0000 (19:55 +0100)]
Fix build with no-threads no-ec

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3817)

7 years agoFix atfork flag. Avoid double-negatives :)
Rich Salz [Fri, 30 Jun 2017 17:55:08 +0000 (13:55 -0400)]
Fix atfork flag.  Avoid double-negatives :)

Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3815)

7 years agoMerge Intel copyright notice into standard
Rich Salz [Thu, 15 Jun 2017 16:08:35 +0000 (12:08 -0400)]
Merge Intel copyright notice into standard

This is done with the kind permission of Intel.

Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3813)

7 years agoAdd ECHO to makefiles for real silence
Rich Salz [Thu, 29 Jun 2017 15:40:19 +0000 (11:40 -0400)]
Add ECHO to makefiles for real silence

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3801)

7 years agoDrop support for OPENSSL_NO_TLS1_3_METHOD
Matt Caswell [Thu, 29 Jun 2017 14:03:14 +0000 (15:03 +0100)]
Drop support for OPENSSL_NO_TLS1_3_METHOD

There are no public TLSv1_3_*method() functions so
OPENSSL_NO_TLS1_3_METHOD doesn't make any sense and should be removed.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3800)

7 years agoAdd fork handlers, based on pthread_atfork
Rich Salz [Thu, 22 Jun 2017 18:00:06 +0000 (14:00 -0400)]
Add fork handlers, based on pthread_atfork

Only for Unix platforms

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3754)

7 years agoSTORE: Make sure the loader to be registered is complete
Richard Levitte [Thu, 29 Jun 2017 19:47:54 +0000 (21:47 +0200)]
STORE: Make sure the loader to be registered is complete

Most of the loader function pointers are crucial, they must be defined
unconditionally.  Therefore, let's make sure OSSL_STORE_register_loader
refuses to register incomplete loaders

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3805)

7 years agoSTORE: simplify store_loader_cmp()
Richard Levitte [Thu, 29 Jun 2017 19:46:02 +0000 (21:46 +0200)]
STORE: simplify store_loader_cmp()

We have already made sure that the loader scheme isn't NULL, so
checking if they are NULL or not when comparing registered loaders
is redundant.  We still soft assert it, just to be entirely sure.

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3805)

7 years agoutil/mkdef.pl: Make symbol version processing Linux only
Richard Levitte [Thu, 29 Jun 2017 19:11:48 +0000 (21:11 +0200)]
util/mkdef.pl: Make symbol version processing Linux only

For Windows, we care which way it is, the resulting file is just a pile
of symbols.  For VMS, we really need to care about the numeric ordering,
and getting the symbols sorted by symbol version too didn't agree with
that.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3804)

7 years agoutil/mkdef.pl: Add UNIX as a platform
Richard Levitte [Thu, 29 Jun 2017 19:09:52 +0000 (21:09 +0200)]
util/mkdef.pl: Add UNIX as a platform

This allows us to guard Unix specific functions with
#ifndef / #ifdef OPENSSL_SYS_UNIX

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3804)

7 years agoAdd sha/asm/keccak1600-s390x.pl.
Andy Polyakov [Tue, 27 Jun 2017 19:45:18 +0000 (21:45 +0200)]
Add sha/asm/keccak1600-s390x.pl.

Reviewed-by: Richard Levitte <levitte@openssl.org>
7 years agosha/asm/keccak1600-x86_64.pl: add CFI directives.
Andy Polyakov [Mon, 26 Jun 2017 15:39:43 +0000 (17:39 +0200)]
sha/asm/keccak1600-x86_64.pl: add CFI directives.

Reviewed-by: Richard Levitte <levitte@openssl.org>
7 years agosha/asm/keccak1600-x86_64.pl: optimize by re-ordering instructions.
Andy Polyakov [Mon, 26 Jun 2017 15:29:24 +0000 (17:29 +0200)]
sha/asm/keccak1600-x86_64.pl: optimize by re-ordering instructions.

Reviewed-by: Richard Levitte <levitte@openssl.org>
7 years agosha/asm/keccak1600-x86_64.pl: remove redundant moves.
Andy Polyakov [Mon, 26 Jun 2017 15:28:13 +0000 (17:28 +0200)]
sha/asm/keccak1600-x86_64.pl: remove redundant moves.

Reviewed-by: Richard Levitte <levitte@openssl.org>
7 years agoAdd sha/asm/keccak1600-x86_64.pl.
Andy Polyakov [Mon, 26 Jun 2017 15:27:09 +0000 (17:27 +0200)]
Add sha/asm/keccak1600-x86_64.pl.

Reviewed-by: Richard Levitte <levitte@openssl.org>
7 years agoAdd internal functions to fetch PEM data from an opened BIO
Richard Levitte [Tue, 6 Dec 2016 13:41:33 +0000 (14:41 +0100)]
Add internal functions to fetch PEM data from an opened BIO

store_attach_pem_bio() creates a STORE_CTX with the 'file' scheme
loader backend in PEM reading mode on an already opened BIO.
store_detach_pem_bio() detaches the STORE_CTX from the BIO and
destroys it (without destroying the BIO).

These two functions can be used in place of STORE_open() and
STORE_close(), and are present as internal support for other OpenSSL
functions.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/2745)

7 years agoMake it possible to tell the file loader to use secure memory
Richard Levitte [Mon, 27 Feb 2017 13:30:00 +0000 (14:30 +0100)]
Make it possible to tell the file loader to use secure memory

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3483)

7 years agoFix double array increment in s_client mysql connect
Steven Danneman [Tue, 27 Jun 2017 22:53:11 +0000 (15:53 -0700)]
Fix double array increment in s_client mysql connect

The packet parsing code for the server version string was incrementing
the array index twice on every iteration. This meant that strings with
an even number of characters would pass, but strings with an odd number
(ex: 5.7.18-0ubuntu0.16.04.1) would cause the pos variable to get out
of sync.

This would cause a later failure with "MySQL packet is broken."

CLA: trivial

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3799)

7 years agoSTORE 'file' scheme loader: refactor the treatment of matches
Richard Levitte [Tue, 27 Jun 2017 21:08:54 +0000 (23:08 +0200)]
STORE 'file' scheme loader: refactor the treatment of matches

Sometimes, 'file_load' couldn't really distinguish if a file handler
matched the data and produced an error or if it didn't match the data
at all.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE: Add an entry in NEWS and CHANGES
Richard Levitte [Fri, 26 May 2017 16:36:26 +0000 (18:36 +0200)]
STORE: Add an entry in NEWS and CHANGES

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE test recipe: Remove comment refering to OpenConnect
Richard Levitte [Thu, 18 May 2017 14:07:15 +0000 (16:07 +0200)]
STORE test recipe: Remove comment refering to OpenConnect

These tests were inspired by OpenConnect and incorporated
by permission of David Woodhouse under CLA

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoAdd documentation for the storeutl app
Richard Levitte [Tue, 13 Dec 2016 12:47:13 +0000 (13:47 +0100)]
Add documentation for the storeutl app

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoAdd documentation for STORE functions
Richard Levitte [Tue, 13 Dec 2016 12:46:53 +0000 (13:46 +0100)]
Add documentation for STORE functions

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoengine app: print out information on STORE loaders and STORE FILE handlers
Richard Levitte [Sun, 11 Dec 2016 06:06:13 +0000 (07:06 +0100)]
engine app: print out information on STORE loaders and STORE FILE handlers

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE: add ENGINE information to loaders
Richard Levitte [Sun, 11 Dec 2016 06:02:06 +0000 (07:02 +0100)]
STORE: add ENGINE information to loaders

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoTest that storeutl with a directory path works as expected
Richard Levitte [Tue, 7 Feb 2017 15:30:31 +0000 (16:30 +0100)]
Test that storeutl with a directory path works as expected

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE 'file' scheme loader: Add directory listing capability
Richard Levitte [Tue, 7 Feb 2017 15:19:40 +0000 (16:19 +0100)]
STORE 'file' scheme loader: Add directory listing capability

This has it recognised when the given path is a directory.  In that
case, the file loader will give back a series of names, all as URI
formatted as possible given the incoming URI.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE 'file' scheme loader: Add handler for encrypted PKCS#8 data
Richard Levitte [Sat, 11 Feb 2017 00:18:29 +0000 (01:18 +0100)]
STORE 'file' scheme loader: Add handler for encrypted PKCS#8 data

Add a separate handler for encrypted PKCS#8 data.  This uses the new
restart functionality.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE 'file' scheme loader: refactor file_load to support decoding restart
Richard Levitte [Sat, 11 Feb 2017 00:17:50 +0000 (01:17 +0100)]
STORE 'file' scheme loader: refactor file_load to support decoding restart

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE: Add a OSSL_STORE_INFO type to help support file handler restarts
Richard Levitte [Thu, 8 Dec 2016 13:28:42 +0000 (14:28 +0100)]
STORE: Add a OSSL_STORE_INFO type to help support file handler restarts

Some containers might very simply decode into something new that
deserves to be considered as new (embedded) data.  With the help of a
special OSSL_STORE_INFO type, make that new data available to the
loader functions so they can start over.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE tests: add PKCS#12 tests
Richard Levitte [Mon, 5 Dec 2016 14:13:01 +0000 (15:13 +0100)]
STORE tests: add PKCS#12 tests

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE 'file' scheme loader: add support for the PKCS#12 container
Richard Levitte [Sat, 11 Feb 2017 00:16:07 +0000 (01:16 +0100)]
STORE 'file' scheme loader: add support for the PKCS#12 container

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoSTORE 'file' scheme loader: add support for containers
Richard Levitte [Mon, 5 Dec 2016 22:15:58 +0000 (23:15 +0100)]
STORE 'file' scheme loader: add support for containers

Containers are objects that are containers for a bunch of other
objects with types we recognise but aren't readable in a stream.  Such
containers are read and parsed, and their content is cached, to be
served one object at a time.

This extends the FILE_HANDLER type to include a function to destroy
the cache and a function to simulate the EOF check.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoAdd a test that checks the store utility
Richard Levitte [Sat, 19 Nov 2016 19:24:17 +0000 (20:24 +0100)]
Add a test that checks the store utility

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoAdd a simple store utility command
Richard Levitte [Sat, 19 Nov 2016 18:38:23 +0000 (19:38 +0100)]
Add a simple store utility command

This command can be used to view the contents of any supported type of
information fetched from a URI, and output them in PEM format.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoAdd a STORE loader for the "file" scheme
Richard Levitte [Mon, 28 Nov 2016 16:30:21 +0000 (17:30 +0100)]
Add a STORE loader for the "file" scheme

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoMake it possible to peek at BIO data through BIO_f_buffer()
Richard Levitte [Fri, 18 Nov 2016 17:18:37 +0000 (18:18 +0100)]
Make it possible to peek at BIO data through BIO_f_buffer()

This is needed for the upcoming "file" scheme STORE loader.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoMake asn1_d2i_read_bio accessible from STORE
Richard Levitte [Fri, 18 Nov 2016 17:17:20 +0000 (18:17 +0100)]
Make asn1_d2i_read_bio accessible from STORE

This is needed for the upcoming "file" scheme STORE loader.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoAdd the STORE module
Richard Levitte [Thu, 8 Dec 2016 14:51:31 +0000 (15:51 +0100)]
Add the STORE module

This STORE module adds the following functionality:

- A function OSSL_STORE_open(), OSSL_STORE_load() and OSSL_STORE_close()
  that accesses a URI and helps loading the supported objects (PKEYs,
  CERTs and CRLs for the moment) from it.
- An opaque type OSSL_STORE_INFO that holds information on each loaded
  object.
- A few functions to retrieve desired data from a OSSL_STORE_INFO
  reference.
- Functions to register and unregister loaders for different URI
  schemes.  This enables dynamic addition of loaders from applications
  or from engines.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoMake it possible to refer to ERR_R_UI_LIB
Richard Levitte [Tue, 6 Dec 2016 03:29:08 +0000 (04:29 +0100)]
Make it possible to refer to ERR_R_UI_LIB

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3542)

7 years agoUpdates following review of SSL_export_key_material() changes
Matt Caswell [Wed, 28 Jun 2017 16:18:27 +0000 (17:18 +0100)]
Updates following review of SSL_export_key_material() changes

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3782)

7 years agoUpdate the SSL_export_keying_material() documentation for TLSv1.3
Matt Caswell [Tue, 27 Jun 2017 15:32:40 +0000 (16:32 +0100)]
Update the SSL_export_keying_material() documentation for TLSv1.3

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3782)

7 years agoAdd an SSL_export_keying_material() test
Matt Caswell [Tue, 27 Jun 2017 15:28:25 +0000 (16:28 +0100)]
Add an SSL_export_keying_material() test

There aren't any test vectors for this, so all we do is test that both
sides of the communication create the same result for different protocol
versions.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3782)

7 years agoUpdate SSL_export_keying_material() for TLSv1.3
Matt Caswell [Tue, 27 Jun 2017 13:57:15 +0000 (14:57 +0100)]
Update SSL_export_keying_material() for TLSv1.3

Fixes #3680

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3782)

7 years agoFix sample code
Greg Zaverucha [Wed, 28 Jun 2017 00:38:25 +0000 (17:38 -0700)]
Fix sample code

Fix memory leak in sample encryption code and check return value of
fopen.

CLA: trivial

Signed-off-by: Greg Zaverucha <gregz@microsoft.com>
Reviewed-by: Paul Dale <paul.dale@oracle.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3790)

7 years agocrypto/mem.c: on Windows, use rand() instead of random()
Richard Levitte [Tue, 27 Jun 2017 09:25:03 +0000 (11:25 +0200)]
crypto/mem.c: on Windows, use rand() instead of random()

Windows doesn't provide random().  In this particular case, our
requirements on the quality of randomness isn't high, so we don't
need to care how good randomness rand() does or doesn't provide.

Fixes #3778

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3779)

7 years agossl_session_dup() missing ext.alpn_session
Todd Short [Mon, 26 Jun 2017 13:21:20 +0000 (09:21 -0400)]
ssl_session_dup() missing ext.alpn_session

Properly copy ext.alpn_session in ssl_session_dup()
Use OPENSSL_strndup() as that's used in ssl_asn1.c

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3770)

7 years agoAdd dependency on apps/progs.h for test/uitest.o
Richard Levitte [Wed, 28 Jun 2017 12:39:20 +0000 (14:39 +0200)]
Add dependency on apps/progs.h for test/uitest.o

uitest.o depends on apps.h which depends on progs.h, which is
dynamically generated, so we need to explicitely add a dependency
between uitest.o and progs.h for the latter to be generated in time.

Fixed #3793

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3794)

7 years agotest/uitest.c's pem_password_cb returned 1 instead of the password length
Richard Levitte [Wed, 28 Jun 2017 11:01:29 +0000 (13:01 +0200)]
test/uitest.c's pem_password_cb returned 1 instead of the password length

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3791)

7 years agoUI_UTIL_wrap_read_pem_callback: make sure to terminate the string received
Richard Levitte [Wed, 28 Jun 2017 09:17:24 +0000 (11:17 +0200)]
UI_UTIL_wrap_read_pem_callback: make sure to terminate the string received

The callback we're wrapping around may or may not return a
NUL-terminated string.  Let's ensure it is.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3791)

7 years agoDocument the added devcrypto engine in CHANGES
Richard Levitte [Wed, 21 Jun 2017 20:03:29 +0000 (22:03 +0200)]
Document the added devcrypto engine in CHANGES

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3744)

7 years agoComment on the lack of documentation for asymmetric ciphers
Richard Levitte [Thu, 11 May 2017 10:52:47 +0000 (12:52 +0200)]
Comment on the lack of documentation for asymmetric ciphers

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3744)

7 years agoAdapt for BSD cryptodev.h differences
Richard Levitte [Tue, 18 Apr 2017 06:51:51 +0000 (08:51 +0200)]
Adapt for BSD cryptodev.h differences

The BSD cryptodev.h doesn't have things like COP_FLAG_WRITE_IV and
COP_FLAG_UPDATE.  In that case, we need to implement that
functionality ourselves.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3744)

7 years agoAdd new /dev/crypto engine
Richard Levitte [Fri, 24 Mar 2017 15:19:00 +0000 (16:19 +0100)]
Add new /dev/crypto engine

Based on cryptodev-linux

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3744)

7 years agoAdd the common error ERR_R_OPERATION_FAIL
Richard Levitte [Tue, 27 Jun 2017 14:05:12 +0000 (16:05 +0200)]
Add the common error ERR_R_OPERATION_FAIL

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3744)

7 years agoUse randomness not entropy
Rich Salz [Mon, 26 Jun 2017 16:02:57 +0000 (12:02 -0400)]
Use randomness not entropy

Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/3773)

7 years agoFix return-value checks in OCSP_resp_get1_id()
Benjamin Kaduk [Mon, 26 Jun 2017 17:05:09 +0000 (12:05 -0500)]
Fix return-value checks in OCSP_resp_get1_id()

Commit db17e43d882ecde217e1dce4a2b8c76c3ed134bf added the function
but would improperly report success if the underlying dup operation
failed.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3775)

7 years agoutil/mkerr.pl: avoid getting an annoying warning about negative count
Richard Levitte [Tue, 27 Jun 2017 11:47:22 +0000 (13:47 +0200)]
util/mkerr.pl: avoid getting an annoying warning about negative count

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3781)

7 years agoutil/mkerr.pl: allow module names prefixed with OSSL_ or OPENSSL_
Richard Levitte [Tue, 27 Jun 2017 11:34:17 +0000 (13:34 +0200)]
util/mkerr.pl: allow module names prefixed with OSSL_ or OPENSSL_

To make sure that our symbols don't clash with other libraries, we
claim the namespaces OSSL and OPENSSL.  Because C doesn't provide
namespaces, the only solution is to have them as prefixes on symbols,
thus we allow OSSL_ and OPENSSL_ as prefixes.

These namespace prefixes are optional for the foreseeable future, and
will only be used for new modules as needed on a case by case basis,
until further notice.

For extra safety, there's an added requirement that module names -
apart from the namespace prefix - be at least 2 characters long.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3781)

7 years agoRename static global "lock" to "obj_lock"
Rich Salz [Mon, 26 Jun 2017 18:04:02 +0000 (14:04 -0400)]
Rename static global "lock" to "obj_lock"

Breaks djgpp, masks a common kernel function name.
Thanks to Gisle Vanem for pointing this out.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3776)

7 years agoChange to check last return value of BN_CTX_get
Paul Yang [Thu, 22 Jun 2017 10:52:29 +0000 (18:52 +0800)]
Change to check last return value of BN_CTX_get

To make it consistent in the code base

Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3749)

7 years agoFix return value checking for BIO_sock_init
Paul Yang [Sun, 25 Jun 2017 17:09:46 +0000 (01:09 +0800)]
Fix return value checking for BIO_sock_init

BIO_sock_init returns '-1' on error, not '0', so it's needed to check
explicitly istead of using '!'.

Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3766)

7 years agoFix inaccurate comments in bn_prime.c
Paul Yang [Sun, 25 Jun 2017 03:49:19 +0000 (11:49 +0800)]
Fix inaccurate comments in bn_prime.c

As well as a coding style nit is fixed.

Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3763)

7 years agotsget.in: remove call of WWW::Curl::Easy::global_cleanup
Richard Levitte [Sun, 25 Jun 2017 20:06:25 +0000 (22:06 +0200)]
tsget.in: remove call of WWW::Curl::Easy::global_cleanup

This function is undocumented, but similarly named functions (such as
'curl_global_cleanup') are documented as internals that should not be
called by scripts.

Fixes #3765

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3768)

7 years agoFix no-dsa build again
Benjamin Kaduk [Sun, 25 Jun 2017 02:46:36 +0000 (21:46 -0500)]
Fix no-dsa build again

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3762)

7 years agoAdd tests for deprecated sigalgs with TLS 1.3 ClientHellos
Benjamin Kaduk [Wed, 26 Apr 2017 20:17:57 +0000 (15:17 -0500)]
Add tests for deprecated sigalgs with TLS 1.3 ClientHellos

Test for each of DSA, SHA1, and SHA224.

Use the symbolic names for SignatureScheme comparisons just added.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3326)

7 years agoFix no-dsa build
Benjamin Kaduk [Thu, 27 Apr 2017 15:32:30 +0000 (10:32 -0500)]
Fix no-dsa build

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3326)

7 years agoAdd constants for TLS 1.3 SignatureScheme values
Benjamin Kaduk [Thu, 27 Apr 2017 14:46:07 +0000 (09:46 -0500)]
Add constants for TLS 1.3 SignatureScheme values

Put them into the TLSProxy::Message namespace along with the extension
type constants.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3326)

7 years agoDisallow DSA/SHA1/etc. for pure TLS 1.3 ClientHellos
Benjamin Kaduk [Mon, 24 Apr 2017 23:20:33 +0000 (18:20 -0500)]
Disallow DSA/SHA1/etc. for pure TLS 1.3 ClientHellos

In draft-ietf-tls-tls13-20 Appendix B we find that:

   This section describes protocol types and constants.  Values listed
   as _RESERVED were used in previous versions of TLS and are listed
   here for completeness.  TLS 1.3 implementations MUST NOT send them
   but might receive them from older TLS implementations.

Similarly, in section 4.2.3 we see:

   Legacy algorithms  Indicates algorithms which are being deprecated
      because they use algorithms with known weaknesses, specifically
      SHA-1 which is used in this context with either with RSA using
      RSASSA-PKCS1-v1_5 or ECDSA.  These values refer solely to
      signatures which appear in certificates (see Section 4.4.2.2) and
      are not defined for use in signed TLS handshake messages.
      Endpoints SHOULD NOT negotiate these algorithms but are permitted
      to do so solely for backward compatibility.  Clients offering
      these values MUST list them as the lowest priority (listed after
      all other algorithms in SignatureSchemeList).  TLS 1.3 servers
      MUST NOT offer a SHA-1 signed certificate unless no valid
      certificate chain can be produced without it (see
      Section 4.4.2.2).

However, we are currently sending the SHA2-based DSA signature schemes
and many SHA1-based schemes, which is in contradiction with the specification.

Because TLS 1.3 support will appear in OpenSSL 1.1, we are bound by
stability requirements to continue to offer the DSA signature schemes
and the deprecated hash algorithms.  at least until OpenSSL 1.2.
However, for pure TLS 1.3 clients that do not offer lower TLS versions,
we can be compliant.  Do so, and leave a note to revisit the issue when
we are permitted to break with sacred historical tradition.

Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3326)

7 years agosha/asm/keccak1600-mmx.pl: optimize for Atom and add comparison data.
Andy Polyakov [Wed, 21 Jun 2017 13:29:53 +0000 (15:29 +0200)]
sha/asm/keccak1600-mmx.pl: optimize for Atom and add comparison data.

Curiously enough out-of-order Silvermont benefited most from
optimization, 33%. [Originally mentioned "anomaly" turned to be
misreported frequency scaling problem. Correct results were
collected under older kernel.]

Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3739)

7 years agoAdd sha/asm/keccak1600-mmx.pl, x86 MMX module.
Andy Polyakov [Wed, 21 Jun 2017 13:25:52 +0000 (15:25 +0200)]
Add sha/asm/keccak1600-mmx.pl, x86 MMX module.

Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
(Merged from https://github.com/openssl/openssl/pull/3739)

7 years agoFix a memory leak in the new TAP filter BIO
Matt Caswell [Fri, 23 Jun 2017 17:18:21 +0000 (18:18 +0100)]
Fix a memory leak in the new TAP filter BIO

[extended tests]

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3760)

7 years agoFix travis clang-3.9 builds
Matt Caswell [Fri, 23 Jun 2017 12:58:49 +0000 (13:58 +0100)]
Fix travis clang-3.9 builds

Something environmental changed in travis so that it started preferring
the ubuntu clang-3.9 version instead of the llvm.org one. This breaks the
sanitiser based builds. This change forces travis to de-prioritise the
ubuntu clang packages.

[extended tests]

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3759)

7 years agoFix another EVP_DigestVerify() instance
Matt Caswell [Fri, 23 Jun 2017 10:40:47 +0000 (11:40 +0100)]
Fix another EVP_DigestVerify() instance

Following on from the previous commit this fixes another instance where
we need to treat a -ve return from EVP_DigestVerify() as a bad signature.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3756)

7 years agoTreat all failures from EVP_DigestVerify() as a bad signature
Matt Caswell [Fri, 23 Jun 2017 10:29:04 +0000 (11:29 +0100)]
Treat all failures from EVP_DigestVerify() as a bad signature

Prior to 72ceb6a we treated all failures from the call to
EVP_DigestVerifyFinal() as if it were a bad signature, and failures in
EVP_DigestUpdate() as an internal error. After that commit we replaced
this with the one-shot function EVP_DigestVerify() and treated a 0 return
as a bad signature and a negative return as an internal error. However,
some signature errors can be negative (e.g. according to the docs if the
form of the signature is wrong). Therefore we should treat all <=0
returns as a bad signature.

This fixes a boringssl test failure.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3756)

7 years agoFix the constant time 64 test
Matt Caswell [Fri, 23 Jun 2017 09:10:51 +0000 (10:10 +0100)]
Fix the constant time 64 test

We were adding more tests than we had data for due to use of
sizeof instead of OSSL_NELEM. I also changed the 8 bit tests
for consistency, although they were already working.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3755)

7 years agoRemove uses of the TEST_check macro.
Pauli [Thu, 22 Jun 2017 04:00:55 +0000 (14:00 +1000)]
Remove uses of the TEST_check macro.

This macro aborts the test which prevents later tests from executing.  It also
bypasses the test framework output functionality.

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3750)

7 years agoFix OBJ_create() to tolerate a NULL sn and ln
Matt Caswell [Thu, 22 Jun 2017 14:25:26 +0000 (15:25 +0100)]
Fix OBJ_create() to tolerate a NULL sn and ln

In 1.0.2 and before OBJ_create() allowed the sn or ln parameter to be NULL.
Commit 52832e47 changed that so that it crashed if they were NULL.

This was causing problems with the built-in config oid module. If a long
name was provided OBJ_create() is initially called with a NULL ln and
therefore causes a crash.

Fixes #3733

Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3753)

7 years agoFix tls1_generate_master_secret
Alexey Komnin [Thu, 22 Jun 2017 06:45:17 +0000 (09:45 +0300)]
Fix tls1_generate_master_secret

Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Rich Salz <rsalz@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3751)

7 years agoFix the names of ChaCha20-Poly1305 cipher suites in t1_trce.c.
David Benjamin [Thu, 22 Jun 2017 03:36:19 +0000 (23:36 -0400)]
Fix the names of ChaCha20-Poly1305 cipher suites in t1_trce.c.

Per RFC 7905, the cipher suite names end in "_SHA256". The original
implementation targeted the -03 draft, but there was a -04 draft right
before the RFC was published to make the names consistent.

Reviewed-by: Rich Salz <rsalz@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3748)

7 years agoTAP line filter BIO.
Pauli [Wed, 21 Jun 2017 23:35:08 +0000 (09:35 +1000)]
TAP line filter BIO.

This is an implementation of a BIO filter that produce TAP compatible output
for the test framework.  The current test indentation level is honoured.

The test output functions have been modified to not attempt to indent
their output and to not include the leading '#' character.

The filter is applied to bio_err only.  bio_out is left unchanged, although
tests using bio_out have been modified to use bio_err instead.

Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/3732)