Prevent crashes on error by making sure the info is freed after OSSL_CMP_CTX_free(),
which may call OSSL_HTTP_close() and thus indirectly reference the info.
Moreover, should not attempt to reference the cmp_ctx variable when NULL.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/18702)
The existing help text says:
> -badsig Corrupt last byte of loaded OSCP response signature (for test)
but this should be OCSP. This is the only occurrence within the project
of this typo.
CLA: trivial
Signed-off-by: Alexander Scheel <alex.scheel@hashicorp.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18684)
This happens if use_ssl is not set but an SSL_CTX is provided.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18674)
The -provider and -propquery options did not work on dsaparam and
gendsa. Fix this and add tests that check that operations that are not
supported by the FIPS provider work when run with
| -provider default -propquery '?fips!=yes'
See also https://bugzilla.redhat.com/show_bug.cgi?id=2094956, where this
was initially reported.
Signed-off-by: Clemens Lang <cllang@redhat.com>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18576)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18657)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18657)
Fixed#18489
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Hugo Landau <hlandau@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18550)
As the potential failure of the OPENSSL_strdup(),
it should be better to check the return value and
return error if fails.
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Reviewed-by: Todd Short <todd.short@me.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18595)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/18424)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/18424)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18368)
To avoid the issue of overflowing the buffer start while
building up the help string prefix this rewrite of the
string building logic does multiple smaller writes to
opt_printf_stderr. While this is slower it completely
avoids the buffer overflow issue and does not place
any (unchecked) length constraints on the name of passed
options. Instead such long options are gracefully
wrapped onto the next line.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/12265)
This can't currently happen due to sizeof(start) being way larger than MAX_OPT_HELP_WIDTH,
but wasn't checked for previously. With this patch there still remains one (static) OOB,
when the length of the option name and the valtype2param string for that argument overflow
the buffer in opt_print. This is kinda unlikely, unless someone intentionally crafts a
long option name, in which case this would become some trivial stack buffer overrun with
possibility to overwrite pointer to the OPTIONS structure (a long o->name is critical here).
I sincerely hope we trust our built-in documentation to not exploit ourselves.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/12265)
Fixes: openssl#18047.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18154)
The option -proxy of openssl s_client works fine. The option
-starttls also works fine. However, try putting both of them
on command line. It breaks, these options don't work together.
The problem is that -proxy option is implemented using starttls_proto
(the option parsing code sets it to PROTO_CONNECT) and -starttls option
overwrites the same variable again based on argument value.
The suggested fix is to independently handle -proxy option before
-starttls so the s_client can connect through HTTP proxy server and
then use STARTTLS command.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17925)
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Todd Short <todd.short@me.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18220)
In create_cert_store(), X509_STORE_new() is called and there is a
dereference of it in following function X509_STORE_add_lookup()
without check, which could lead to NULL pointer dereference.
Fix this by adding a NULL check of X509_STORE_new()
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18057)
The function OCSP_basic_add1_status() will return NULL on malloc failure.
However the return value is not checked before being passed to
OCSP_SINGLERESP_add1_ext_i2d(), and there is a wild field pointer,
which could lead to wild pointer dereference.
Fix this by adding return value check
Reviewed-by: Kurt Roeckx <kurt@roeckx.be>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/18081)
CLI changes: New parameter -digest to CLI command openssl cms, to
provide pre-computed digest for use with -sign.
API changes: New function CMS_final_digest(), like CMS_final() but
uses a pre-computed digest instead of computing it from the data.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Todd Short <todd.short@me.com>
(Merged from https://github.com/openssl/openssl/pull/15348)
Supports Linux, MacOS and FreeBSD
Disabled by default, enabled via `enabled-tfo`
Some tests
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/8692)
If TCP is being used, protocol = 0 is passed to init_client(), then
protocol == IPPROTO_TCP fails when attempting to set BIO_SOCK_NODELAY.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17838)
Also check return value of functions that call BIO_new() internally
such as dup_bio_out().
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17421)
Fix a bug in `openssl s_server -WWW` where it would attempt to invoke
`SSL_sendfile` if `-ktls -sendfile` was passed on the command line, even
if KTLS has not actually been enabled, for example because it is not
supported by the host. Since `SSL_sendfile` is only supported when KTLS
is actually being used, this resulted in a failure to serve requests.
Fixes#17503.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17788)
`openssl req -x509` has code allowing it to generate certificates from CSRs
as a replacement for `openssl x509`, but a bug prevents it from working
properly. -CA and -CAkey can now be passed to generate a CA-signed
certificate as documented in openssl-req(1).
Regression testing has been added to `openssl req`.
Fixes#17736.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17782)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17741)
As the potential failure of the BIO_new(), it should be better to check the return value and return error if fails in order to avoid the dereference of NULL pointer.
And because 'bio_s_msg' is checked before being used everytime, which has no need to add the check.
But 'bio_s_out' is not.
And since the check 'if (bio_s_out == NULL)' is redundant, it can be removed to make the code succincter.
Also the 'sbio' and so forth should be checked like the other places in the same file.
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17710)
Just assert 'bn' to be non-NULL is not enough.
The check for 'itmp' is still needed.
If 'bn' is 0, the 'itmp' is assigned by OPENSSL_strdup().
Since OPENSSL_strdup() may fail because of the lack of memory,
the 'itmp' will be NULL and be an valid parameter hashed in
TXT_DB_get_by_index(), returning a wrong result.
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17677)
Since the OPENSSL_strdup() may return NULL if allocation
fails, the 'port' could be NULL.
And then it will be used in do_server(), which can accept
NULL as an valid parameter.
That means that the system could run with a wrong parameter.
Therefore it should be better to check it, like the other
memory allocation.
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17673)
Fixes#13944
Moved "opt_printf_stderr" out of apps.c to avoid duplicate definition in tests.
Added function "asn1_string_to_time_t" including tests.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17645)
As the potential failure of the memory allocation,
it should be better to check the return value of
OPENSSL_strdup() and return error if fails.
Also, we need to restore the 'ep' to be NULL if fails.
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17676)
Fixes openssl#17629 by including internal/numbers.h which defines SIZE_MAX
CLA: trivial
Fixes#17629
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17632)
Including e_os.h with a path from a header file doesn't work well on
certain exotic platform. It simply fails to build.
Since we don't seem to be able to stop ourselves, the better move is
to move e_os.h to an include directory that's part of the inclusion
path given to the compiler.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17641)
Add copyright to files that were missing it.
Update license from OpenSSL to Apache as needed.
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17606)
We use opt_int_arg() to prevent malformed processing of
hexadecimal input to some command line options.
Updated help to mention the change.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17585)
As with 11d7d903, use SOURCE_DATE_EPOCH for the copyright year if it is
defined, to avoid reproducibility problems.
CLA: trivial
Signed-off-by: Ross Burton <ross.burton@arm.com>
Change-Id: I1bea19070411a69155c43de7082350fb2c499da3
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17601)
The option SSL_OP_NO_EXTENDED_MASTER_SECRET was added in #3910.
And it is valid for versions below (D)TLS 1.2.
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17597)
Allow dhparam to run quietly in scripts, etc.
For other commands that took a -verbose flag already, also support -quiet.
For genpkey which only supported -quiet, add the -verbose flag.
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17336)
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17336)
Any propquery passed via the -propquery option to the ciphers command was
being ignored.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17595)
Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17359)
as the code uses BIO_gets, and it always null terminates the
strings it reads, when it reads a record 2^14 byte long, it actually
returns 2^14-1 bytes to the calling application, in general it returns
size-1 bytes to the caller
This makes the code sub-optimal (as every 2^14 record will need two
BIO_gets() calls) and makes it impossible to use -rev option to test
all plaintext lengths (like in openssl#15706)
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17538)
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17528)
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Bernd Edlinger <bernd.edlinger@hotmail.de>
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17528)
Add null checks to avoid dereferencing a pointer that could be null.
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/17488)
use goto instead of returning directly while error handling
Signed-off-by: Peiwei Hu <jlu.hpw@foxmail.com>
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17404)
See discussion in #17088, where the real solution was postponed to 4.0.
This preliminarily fixes the issue that the HTTP(S) proxy environment vars
were neglected when determining whether a proxy should be used for HTTPS.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17310)
The use of -proxy prevented s_client from correctly sending the target
hostname as SNI data.
Fixes#17232
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17248)
Make app_http_tls_cb() tidy up on disconnect the SSL BIO it pushes on connect.
Make OSSL_HTTP_close() respect this.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17318)
Ignore -server with -rspin and exclude all of -use_mock_srv/-port/-rspin.
On the other hand, -server is required if no -use_mock_srv/-port/-rspin is given.
Ignore -tls_used with -use_mock_srv and -rspin; it is not supported with -port.
If -server is not given, ignore -proxy, -no_proxy, and -tls_used.
Also slightly improve the documentation of the two mock server variants.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17254)
When s_server responds to a file data with the -WWW parameter, it
always gets a path named "GET". In this case, we need to skip the
"GET /" character to get the correct file path.
Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Paul Yang <kaishen.yy@antfin.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17231)
Check is done to prevent wrong memory access by EVP_PKEY_get0_asn1()
Also fix wrong coding style in the s_cb.c file.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17176)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17110)
In the x509 app when doing the ip/email/host validation of a
certificate we return nonzero exit code if the validation fails.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17022)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17031)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/17030)
In a recent upstream change
(43da9a14f0)
the parameter sigsize become a read/write input in
EVP_PKEY_sign(), and after signing, sigsize will be overwritten with
the actual size and used in the verify step. As the speed program
calls EVP_PKEY_sign() on the same context repeatedly, sigsize value is
no longer the initial available size, and may fail in later buffer
size checks.
This fix adds a new buflen member in struct loopargs (which is only
used within apps/speed.c), to save available buffer size and
to be used as sigsize input in EVP_PKEY_sign() calls.
Sigsize still contains the signature size for the verify step.
Signed-off-by: Xiaofei Bai <xiaofei.bai@arm.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16991)
Since TLS v1.3 eschews renegotiation entirely it’s misleading to have
these apps say it’s “not supported” when in fact the TLS version is
new enough not to need renegotiation at all.
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16937)
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16918)
For machines where sizeof(size_t) == sizeof(int) there is a possible overflow
which could cause a crash.
For machines where sizeof(size_t) > sizeof(int), the existing checks adequately
detect the situation.
Fixes#16899
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16904)
This prevents a compile-time warning on newer gcc.
Also fix the related warning message.
Fixes#16814
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16821)
Commit 0007ff257c added a protocol version check to psk_server_cb but
failed to take account of DTLS causing DTLS based psk connections to
fail.
Fixes#16707
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
(Merged from https://github.com/openssl/openssl/pull/16838)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16783)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16783)
This avoids a crash in e.g. `openssl chacha20` as reported by
Steffen Nurpmeso on openssl-users.
Resolves: #16713
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16714)
From openssl-3.0.0-alpha15, KTLS is turned off by default, even if
KTLS feature in compilation, which makes it difficult to use KTLS
through s_server/s_client, so a parameter option 'ktls' is added
to enable KTLS through cmdline.
Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Paul Yang <kaishen.yy@antfin.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16609)
From openssl-3.0.0-alpha15, KTLS is turned off by default, even if
KTLS feature in compilation, which makes it difficult to use KTLS
through s_server/s_client, so a parameter option 'ktls' is added
to enable KTLS through cmdline.
At the same time, SSL_sendfile() depends on KTLS feature to work
properly, make parameters sendfile depend on parameters ktls.
Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Paul Yang <kaishen.yy@antfin.com>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16609)
Also fix indentation in c_rehash.in to 4 spaces, where a mixture of 4 and 8
spaces was used before, in addition to tabs.
CLA: trivial
Reviewed-by: Tomas Mraz <tomas@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16459)
Most of all, this reduces the following:
built on: built on: Wed Sep 8 19:41:55 2021 UTC
to:
built on: Wed Sep 8 19:41:55 2021 UTC
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16563)
(cherry picked from commit c1dc3536a8)
Command 'openssl ciphers -convert <name>' always returns failure,
this patch set the correct return value.
Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Paul Yang <kaishen.yy@antfin.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16383)
If there are no suitable encoders it might mean the key is in
an engine and thus it is a legacy key. Try legacy encoding
routines to encode the public key. We do not attempt encoding
a private key as it would be in most cases impossible anyway.
Fixes#16256
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16289)
Current implementation of warn_binary introduces a regression
when the content is passed in /dev/stdin as an explicit file name
and reads the file to be processed twice otherwise.
I suggest to reimplement this functionality after 3.0 if necessary.
Fixes#16359
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
(Merged from https://github.com/openssl/openssl/pull/16367)
The change to a more configuration based approach to enable FIPS mode
operation highlights a shortcoming in the default should do something
approach we've taken for bad configuration files.
Currently, a bad configuration file will be automatically loaded and
once the badness is detected, it will silently stop processing the
configuration and continue normal operations. This is good for remote
servers, allowing changes to be made without bricking things. It's bad
when a user thinks they've configured what they want but got something
wrong and it still appears to work.
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16171)
Fixes#16197
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16199)
This was introduced in 814b5133e for MSVC. C++Builder doesn't need it.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Matthias St. Pierre <Matthias.St.Pierre@ncp-e.com>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/16192)
This is a follow up of 15729bef38. Even
when the host does not support IPv6 at all, BIO_lookup_ex may now
return IN6ADDR_ANY in addition to INADDR_ANY, as the second element of
the ai_next field.
After eee8a40aa5, the do_server function
prefers the IPv6 address and fails on the BIO_socket call. This adds
a fallback code to retry with the IPv4 address returned as the first
element to avoid the error.
The failure had been partially avoided in the previous code with
AI_ADDRCONFIG, because getaddrinfo returns only IPv4 address if no
IPv6 address is associated with external interface. However, it would
be still a problem if the external interface has an IPv6 address
assigned, while the loopback interface doesn't.
Signed-off-by: Daiki Ueno <dueno@redhat.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16074)
Fall back to stdio functions if not available.
Fixes a daily run-checker failure (no-posix-io)
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16073)
Remove the executable attributes of some C code files and key files,
change the file mode from 0755 to 0644.
Signed-off-by: Tianjia Zhang <tianjia.zhang@linux.alibaba.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16045)
By default they are. However, if a provider, provider path or a property query has been specified
they are not. Likewise, if a library context or a property query has been
specified by the command, they are not.
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16022)
Issue #15951 describes a scenario which causes s_server to fail when using
a PSK. In the originally described issue this only impacted master and not
1.1.1. However, in fact this issue does also impact 1.1.1 - but only if you
additionally supply the option "-no_ticket" to the s_server command line.
The difference between the behaviour in master and 1.1.1 is due to 9c13b49,
which changed PSK_MAX_IDENTITY_LEN from 128 to 256. It just so happens that
a default OpenSSL TLSv1.3 ticket length happens to fall between those 2
values. Tickets are presented in TLSv1.3 as a PSK "identity". Passing
"no_ticket" doesn't actually stop TLSv1.3 tickets completely, it just
forces the use of "session ids as a ticket" instead. This significantly
reduces the ticket size to below 128 in 1.1.1.
The problem was due to s_server setting a TLSv1.2 PSK callback and a
TLSv1.3 PSK callback. For backwards compat reasons the TLSv1.2 PSK
callbacks also work in TLSv1.3 but are not preferred. In the described
scenario we use a PSK to create the initial connection. Subsequent to that
we attempt a resumption using a TLSv1.3 ticket (psk). If the psk length is
below PSK_MAX_IDENTITY_LEN then we first call the TLSv1.2 PSK callback.
Subsequently we call the TLSv1.3 PSK callback. Unfortunately s_server's
TLSv1.2 PSK callback accepts the identity regardless, even though it is an
unexpected value, and hence the binder subsequently fails to verify.
The fix is to bail early in the TLSv1.2 callback if we detect we are being
called from a TLSv1.3 connection.
Fixes#15951
Reviewed-by: Ben Kaduk <kaduk@mit.edu>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16008)
This fixes a regression introduced by commit 1d6c867. When exporting a set
of certificates to a PKCS12 file we shouldn't add the first one twice. Also
we restore historic behaviour with respect to the canames option where we
have no ee certificate with key.
Fixes#15983
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/16001)
When there is other PEM data in between certs the OSSL_STORE_load
returns NULL and reports error. Avoid printing that error unless
there was nothing read at all.
Fixes#15945
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15949)
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15790)
We used the original string, which meant fetching for, for example,
'rsa:2048'. That was, of course, doomed to fail.
Reviewed-by: Matt Caswell <matt@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15912)
The command ``openssl x509 -noout -modulus -in cert.pem`` used to segfaults
sometimes because an uninitialized variable was passed to
``BN_lebin2bn``. The bug triggered an assertion in bn_expand_internal().
Fixes: https://github.com/openssl/openssl/issues/15899
Signed-off-by: Christian Heimes <christian@python.org>
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15900)
This avoids use of uninitialised memory.
Follow on to #15900
Reviewed-by: Shane Lontis <shane.lontis@oracle.com>
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15908)
Fixes#15839
Reviewed-by: David von Oheimb <david.von.oheimb@siemens.com>
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15861)
Also ensure that opt_intmax() and opt_uintmax() does the right thing
if sizeof([u]intmax_t) is smaller than sizeof(ossl_[u]intmax_t).
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15825)
Since the service is echo-like (see TCP port 7 from RFC 862 or
gnutls-serv --echo), make it easier to find by mentioning "echo" in
the description of it in the help message an man page
Also fixes the man page inconsistency ("sends it back to the server")
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Dmitry Belyavskiy <beldmit@gmail.com>
(Merged from https://github.com/openssl/openssl/pull/15739)
Reviewed-by: Matt Caswell <matt@openssl.org>
Reviewed-by: Tim Hudson <tjh@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15415)
A Solaris compiler complains:
"apps/s_client.c", line 2994: statement not reached
It takes a bit of scrutiny to see that this is true, on all platforms.
Reviewed-by: Paul Dale <pauli@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15719)
Fixes#5430
Added the configuration file option "date_opt" to the openssl applications ca,
crl and x509.
Added ASN1_TIME_print_ex which supports the new datetime format using the
flag ASN1_DTFLGS_ISO8601
Reviewed-by: Paul Dale <pauli@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/14384)
Fixes#15071
It always tries loading the cert as DH which previously did not produce
an error. The errors are not suppressed for these operations.
The output now matches previous versions of OpenSSL.
Reviewed-by: Paul Dale <pauli@openssl.org>
Reviewed-by: Tomas Mraz <tomas@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/15670)
