Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10130)
dcruette [Tue, 24 Dec 2019 21:48:19 +0000 (22:48 +0100)]
Update tls13_enc.c
Fix double + in hkdflabel declaration (FIXES #10675)
CLA: trivial
Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Kurt Roeckx <kurt@roeckx.be> Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/10700)
Matt Caswell [Mon, 23 Dec 2019 17:14:58 +0000 (17:14 +0000)]
Temporarily disable the explicit enable-asan build
The explicit enable-asan build fails in the memleak test for unknown
reasons. Therefore we disable it temporarily to get a green Travis.
Other builds that use -fsanitize=address in Travis seem to pass.
[extended tests]
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com> Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/10689)
Matt Caswell [Mon, 23 Dec 2019 14:52:03 +0000 (14:52 +0000)]
Temporarily disable some Travis external tests
The pyca-cryptography external test has been failing for a long time.
It looks like upstream needs to make some changes to adapt to 3.0.
[extended tests]
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com> Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/10689)
Matt Caswell [Mon, 23 Dec 2019 14:39:57 +0000 (14:39 +0000)]
Don't run test_conf in cross compiled builds
test_conf was failing in travis for mingw builds. We run these on linux
via wine. However due to line break differences the tests were failing.
We just skip these in a cross compiled build.
Reviewed-by: Nicola Tuveri <nic.tuv@gmail.com> Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/10689)
Bernd Edlinger [Sun, 22 Dec 2019 18:40:03 +0000 (19:40 +0100)]
Fix a race condition in the speed command
The timer alarm sets run = 0, while the benchmark
does run = 1 in the initialization code. That is
a race condition, if the timer goes off too early
the benchmark runs forever.
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10680)
Richard Levitte [Sun, 22 Dec 2019 22:52:30 +0000 (23:52 +0100)]
util/find-doc-nits: when loading "missing" files, check if documented
It may be that some "missing" manuals have been written since their
insertion in the "missing" files. Make sure to alert when such manual
references are found.
This works, because we collect all existing manual references into
%name_map first.
Fixes #10681
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10683)
Benjamin Kaduk [Mon, 23 Dec 2019 18:35:48 +0000 (10:35 -0800)]
Update the krb5 submodule
Bring us up to date with upstream's 1.17.1 release. Among other
things, it includes commit c2497d46b4bad473e164943d67b58cd1ae261c3a
which fixes several issues that affect running the test suite under
Travis CI. Hopefully those will work transitively for us as well.
[extended tests]
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10690)
Bernd Edlinger [Sun, 22 Dec 2019 10:48:54 +0000 (11:48 +0100)]
Add some missing cfi frame info in x25519-x86_64.pl
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/10676)
Shane Lontis [Sat, 21 Dec 2019 23:37:17 +0000 (09:37 +1000)]
Remove asn1 module dependency from RSASSA-PKCS1-v1_5 implementation.
Replace use of the asn1 module (X509_SIG, X509_ALGOR, ASN1_TYPE,
ASN1_OCTET_STRING, i2d_X509_SIG(), etc.) as well as OID lookups using
OBJ_nid2obj() with pre-generated DigestInfo encodings for MD2, MD5, MDC-2,
SHA-1, SHA-2 and SHA-3; the encoding is selected based on the NID. This is
similar to the approach used by the old FOM.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/9138)
Richard Levitte [Thu, 12 Dec 2019 18:50:41 +0000 (19:50 +0100)]
util/find-doc-nits: Better checking of missing documentation
The names collected in util/missing*.txt are not file names, but
symbol names, and to compare properly with script data, the section
name must be included.
All symbols found in util/lib*.num are library functions, so we know
that they are in manual section 3 and can simply add that info. The
same goes for all macros found in C headers.
Finally, we get rid of getdocced() and its associated hash table
%docced. We already have the appropriate information in %name_map.
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10621)
Bernd Edlinger [Wed, 18 Dec 2019 16:33:03 +0000 (17:33 +0100)]
Add some missing cfi frame info in aes-x86_64.pl
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/10650)
Bernd Edlinger [Wed, 18 Dec 2019 09:12:26 +0000 (10:12 +0100)]
Add some missing cfi frame info in camellia-x86_64.pl
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/10642)
Jussi Keranen [Wed, 11 Dec 2019 13:08:04 +0000 (15:08 +0200)]
Fix regression on x509 keyform argument
In OpenSSL pre 1.1.0, 'openssl x509 -keyform engine' was possible
and supported. In 1.1.0, type of keyform argument is OPT_FMT_PEMDER
which doesn't support engine. This changes type of keyform argument
to OPT_FMT_PDE which means PEM, DER or engine and updates the manpage
including keyform and CAkeyform.
This restores the pre 1.1.0 behavior.
This issue is very similar than https://github.com/openssl/openssl/issues/4366
CLA: trivial
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com> Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10609)
Rich Salz [Thu, 31 Oct 2019 03:35:08 +0000 (23:35 -0400)]
Use a function to generate do-not-edit comment
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10316)
Haohui Mai [Sat, 7 Dec 2019 08:44:16 +0000 (00:44 -0800)]
Make Windows build more robust
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10586)
Shane Lontis [Thu, 19 Dec 2019 07:50:50 +0000 (17:50 +1000)]
Fix travis timeout by excluding arm64 gcc -fsanitize=address build
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/10639)
Fangming.Fang [Fri, 31 May 2019 10:15:10 +0000 (10:15 +0000)]
Optimize AES-GCM implementation on aarch64
Comparing to current implementation, this change can get more
performance improved by tunning the loop-unrolling factor in
interleave implementation as well as by enabling high level parallelism.
Richard Levitte [Wed, 27 Nov 2019 16:58:01 +0000 (17:58 +0100)]
ASN1: Adapt ASN.1 output routines to use BIO_f_prefix()
We modify asn1_print_info() to print the full line. It pushes a
BIO_f_prefix() BIO to the given |bp| if it can't detect that it's
already present, then uses both the prefix and indent settings to get
formatting right.
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10531)
Bernd Edlinger [Tue, 17 Dec 2019 08:05:32 +0000 (09:05 +0100)]
Fix unwind info for some trivial functions
While stack unwinding works with gdb here, the
function _Unwind_Backtrace gives up when something outside
.cfi_startproc/.cfi_endproc is found in the call stack, like
OPENSSL_cleanse, OPENSSL_atomic_add, OPENSSL_rdtsc, CRYPTO_memcmp
and other trivial functions which don't save anything in the stack.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
(Merged from https://github.com/openssl/openssl/pull/10635)
Shane Lontis [Wed, 18 Dec 2019 04:46:01 +0000 (14:46 +1000)]
Cleanup legacy digest methods.
Macros have been added to generate the simple legacy methods.
Engines and EVP_MD_METH_get methods still require access to the old legacy methods,
so they needed to be added back in.
They may only be removed after engines are deprecated and removed.
Removed some unnecessary #includes and #ifndef guards (which are done in build.info instead).
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10602)
Saritha [Tue, 29 Oct 2019 06:40:55 +0000 (12:10 +0530)]
apps/speed.c: Fix eddsa sign and verify output with -multi option
Fixes #10261
CLA: trivial
Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10285)
Matt Caswell [Wed, 4 Dec 2019 10:21:52 +0000 (10:21 +0000)]
Update the HISTORY entry for RSA_get0_pss_params()
Make a note of when this function was first introduced
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10568)
Matt Caswell [Mon, 9 Dec 2019 11:51:48 +0000 (11:51 +0000)]
Ensure EVP_PKEY_set1_DH detects X9.42 keys
OpenSSL supports both PKCS#3 and X9.42 DH keys. By default we use PKCS#3
keys. The function `EVP_PKEY_set1_DH` was assuming that the supplied DH
key was a PKCS#3 key. It should detect what type of key it is and assign
the correct type as appropriate.
Fixes #10592
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10593)
kinichiro [Thu, 5 Dec 2019 11:00:50 +0000 (20:00 +0900)]
Return 1 when openssl req -addext kv is duplicated
CLA: trivial
Fixes #10273
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org> Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/10578)
Rich Salz [Wed, 11 Dec 2019 15:56:12 +0000 (10:56 -0500)]
Deprecated crypto-mdebug-backtrace
Reviewed-by: Paul Dale <paul.dale@oracle.com> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10572)
Rich Salz [Wed, 4 Dec 2019 18:15:08 +0000 (13:15 -0500)]
Deprecate most of debug-memory
Fixes #8322
The leak-checking (and backtrace option, on some platforms) provided
by crypto-mdebug and crypto-mdebug-backtrace have been mostly neutered;
only the "make malloc fail" capability remains. OpenSSL recommends using
the compiler's leak-detection instead.
The OPENSSL_DEBUG_MEMORY environment variable is no longer used.
CRYPTO_mem_ctrl(), CRYPTO_set_mem_debug(), CRYPTO_mem_leaks(),
CRYPTO_mem_leaks_fp() and CRYPTO_mem_leaks_cb() return a failure code.
CRYPTO_mem_debug_{malloc,realloc,free}() have been removed. All of the
above are now deprecated.
Merge (now really small) mem_dbg.c into mem.c
Reviewed-by: Paul Dale <paul.dale@oracle.com> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10572)
Large GOST ClientKeyExchange messages are sent by VipNet CSP, one of
Russian certified products implementing GOST TLS, when a server
certificate contains 512-bit keys.
This behaviour was present in 1.0.2 branch and needs to be restored.
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10376)
Richard Levitte [Wed, 11 Dec 2019 13:36:36 +0000 (14:36 +0100)]
Add better support for using deprecated symbols internally
OPENSSL_SUPPRESS_DEPRECATED only does half the job, in telling the
deprecation macros not to add the warning attribute. However, with
'no-deprecated', the symbols are still removed entirely, while we
might still want to use them internally.
The solution is to permit <openssl/opensslconf.h> macros to be
modified internally, such as undefining OPENSSL_NO_DEPRECATED in this
case.
However, with the way <openssl/opensslconf.h> includes
<openssl/macros.h>, that's easier said than done. That's solved by
generating <openssl/configuration.h> instead, and add a new
<openssl/opensslconf.h> that includes <openssl/configuration.h> as
well as <openssl/macros.h>, thus allowing to replace an inclusion of
<openssl/opensslconf.h> with this:
Note that undefining OPENSSL_NO_DEPRECATED must never be done by
applications, since the symbols must still be exported by the
library. Internal test programs are excempt of this rule, though.
Reviewed-by: Tim Hudson <tjh@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org> Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/10608)
Richard Levitte [Tue, 10 Dec 2019 19:38:09 +0000 (20:38 +0100)]
test/namemap_internal_test.c: use "cookie" instead of "foo"...
... in test_namemap()
Because tests may sometimes run in random order (subject of the
environment variable OPENSSL_TEST_RAND_ORDER being defined), and we're
dealing with the global namemap, each test must use names that are
globally unique for that test. Unfortunately, we used "foo" in two of
them, which might lead to surprising results.
Fixes #10401
Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10601)
kinichiro [Wed, 11 Dec 2019 12:12:53 +0000 (21:12 +0900)]
Check return value after loading config file
CLA: trivial
Reviewed-by: Paul Yang <kaishen.yy@antfin.com> Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/10607)
Rich Salz [Sat, 5 Oct 2019 22:14:30 +0000 (18:14 -0400)]
Various missing-link fixes
Also, turn missing L<foo(3)> into foo(3)
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10104)
Richard Levitte [Wed, 11 Dec 2019 12:43:24 +0000 (13:43 +0100)]
PROV: Move AES_GCM specialisation away from common cipher header
The AES_GCM specialisation was defined in the common cipher header
providers/implementations/include/prov/ciphercommon_gcm.h, when it
should in fact be in a local providers/implementations/ciphers/
header.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/10606)
Richard Levitte [Wed, 11 Dec 2019 12:11:34 +0000 (13:11 +0100)]
PROV: Move AES_CCM specialisation away from common cipher header
The AES_CCM specialisation was defined in the common cipher header
providers/implementations/include/prov/ciphercommon_ccm.h, when it
should in fact be in a local providers/implementations/ciphers/
header.
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/10606)
Fangming.Fang [Fri, 15 Nov 2019 07:47:22 +0000 (07:47 +0000)]
Enrich arm64 tests in Travis matrix
1, Remove simple test just with --strict-warnings enabled.
2, Share the three common envs with amd64.
3, Add matrix item running test in bionic(default xenial) for arm64.
4, Enable MSan test on arm64 for extended test.
5, Enable UBSan test on arm64 for extended test.
Reviewed-by: Paul Dale <paul.dale@oracle.com> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10519)
Rich Salz [Tue, 26 Nov 2019 14:16:41 +0000 (09:16 -0500)]
Fix docs for CRYPTO_secure_allocated
Fixes #9300
Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10523)
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10544)
XiaokangQian [Thu, 7 Nov 2019 02:36:45 +0000 (02:36 +0000)]
Optimize AES-ECB mode in OpenSSL for both aarch64 and aarch32
Aes-ecb mode can be optimized by inverleaving cipher operation on
several blocks and loop unrolling. Interleaving needs one ideal
unrolling factor, here we adopt the same factor with aes-cbc,
which is described as below:
If blocks number > 5, select 5 blocks as one iteration,every
loop, decrease the blocks number by 5.
If 3 < left blocks < 5 select 3 blocks as one iteration, every
loop, decrease the block number by 3.
If left blocks < 3, treat them as tail blocks.
Detailed implementation will have a little adjustment for squeezing
code space.
With this way, for small size such as 16 bytes, the performance is
similar as before, but for big size such as 16k bytes, the performance
improves a lot, even reaches to 100%, for some arches such as A57,
the improvement even exceeds 100%. The following table will list the
encryption performance data on aarch64, take a72 and a57 as examples.
Performance value takes the unit of cycles per byte, takes the format
as comparision of values. List them as below:
Andy Polyakov [Sun, 5 May 2019 16:30:55 +0000 (18:30 +0200)]
Configure: use ELFv2 ABI on some ppc64 big endian systems
If _CALL_ELF is defined to be 2, it's an ELFv2 system.
Conditionally switch to the v2 perlasm scheme.
Reviewed-by: Paul Dale <paul.dale@oracle.com> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8883)
Reviewed-by: Paul Dale <paul.dale@oracle.com> Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8883)
Richard Levitte [Tue, 3 Dec 2019 18:41:05 +0000 (19:41 +0100)]
Move providers/common/{ciphers,digests}/* to providers/implementations
The idea to have all these things in providers/common was viable as
long as the implementations was spread around their main providers.
This is, however, no longer the case, so we move the common blocks
closer to the source that use them.
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/10564)
Rich Salz [Mon, 18 Nov 2019 10:29:21 +0000 (05:29 -0500)]
Remove handling of outdated macro's
DECLARE_STACK_OF was renamed to DEFINE_STACK_OF in commit 8588571.
Expanded the only use of TYPEDEF_{D2I,I2D,D2I2D}_OF, so that they can
easily be removed in a future release
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
(Merged from https://github.com/openssl/openssl/pull/10464)
Richard Levitte [Mon, 25 Nov 2019 16:13:10 +0000 (17:13 +0100)]
test/memleaktest.c: Modify for use with address/leak sanitizer
Detects if leak sanitizing is on, and directs the exit code accordingly.
Note that this program is designed to fail when leaking, as that's
expected, so to make it easy for wrapper scripts, we also make it look
like it fails when sanitizing isn't on.
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/9294)
Richard Levitte [Tue, 2 Jul 2019 14:23:27 +0000 (16:23 +0200)]
Use leak sanitizer instead of internal mdebug to check for memory leaks
The leak sanitizer gives better reports (complete stack traces) and
works as a wrapper around the application instead of relying on
cooperative enabling and disabling calls (which are too easy to get
unbalanced).
Related to #8322
Reviewed-by: Paul Dale <paul.dale@oracle.com>
(Merged from https://github.com/openssl/openssl/pull/9294)
avas [Thu, 5 Dec 2019 14:14:44 +0000 (06:14 -0800)]
Move U64 macro from sha.h to sha512.c
Summary:
U64 is too common name for macro, being in public header sha.h it
conflicts with other projects (WAVM in my case). Moving macro from
public header to the only .c file using it.
CLA: trivial
Reviewed-by: Paul Dale <paul.dale@oracle.com> Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tim Hudson <tjh@openssl.org> Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/10579)
FdaSilvaYY [Wed, 9 May 2018 20:27:27 +0000 (22:27 +0200)]
apps/speed: replace list of #define by enums declarations.
it simplifies some pieces of code.
Improve internal assertions
Tag a few #endif with OPENSSL_NO_EC to mark its ending.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10078)
FdaSilvaYY [Tue, 15 Oct 2019 21:33:02 +0000 (23:33 +0200)]
apps/speed: factorize ec test curves declarations
remove 'test' prefix from variable names.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10078)
FdaSilvaYY [Sat, 19 Oct 2019 15:55:36 +0000 (17:55 +0200)]
apps/speed: improve sm2 failure code.
attach the new objects sooner, so error handling is simplified.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10078)
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10078)
FdaSilvaYY [Tue, 5 Jun 2018 17:44:42 +0000 (19:44 +0200)]
apps/speed: initialize key material only when its algo is selected.
Remove some duplicate key data declarations.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10078)
FdaSilvaYY [Sat, 19 Oct 2019 17:37:01 +0000 (19:37 +0200)]
apps/speed: remove a shared global variable
replace |save_count| by the right c[D_EVP(_xxx)] variable.
this may shared a value between various algorithm.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10078)
FdaSilvaYY [Sat, 19 Oct 2019 15:58:07 +0000 (17:58 +0200)]
apps/speed: discard useless protoypes as these methods are defines before being used.
Reviewed-by: Richard Levitte <levitte@openssl.org> Reviewed-by: Tomas Mraz <tmraz@fedoraproject.org>
(Merged from https://github.com/openssl/openssl/pull/10078)