| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
| |
OpenSSL stopped building it last year and removed it this year.
Based on OpenSSL commit c436e05bdc7f49985a750df64122c960240b3ae1.
Also cranked major version in libcrypto, libssl and libtls.
"fine with me" bcook@ miod@
|
| |
|
|
|
|
|
|
| |
changed in an incompatible way regarding partial writes.
OK miod@ deraadt@
|
|
|
|
| |
ok beck@
|
|
|
|
|
|
|
| |
in libtls. This gives tls_write() a similar short write semantics
as write(2). So implementing daemons with libevent buffers will
be easier and workarounds in syslogd and httpd can be removed.
OK tedu@ beck@ reyk@
|
|
|
|
| |
ok miod@ jsing@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This makes using libtls easier to include by including dependent headers,
making something like this work as expected:
#include <iostream>
#include <tls.h>
int main()
{
std::cout << "tls_init: " << tls_init() << "\n";
}
This also makes building a standalone libtls-portable simpler.
ok doug@, jsing@
|
|
|
|
| |
ok mpi@
|
| |
|
|
|
|
|
|
| |
RFC 5280 says " " must not be used as a dNSName.
ok jsing@ jca@
|
|
|
|
|
|
|
| |
ensure that outlen is set to zero so that tls_read() has read(2) like
semantics for EOF.
Spotted by doug@
|
|
|
|
|
|
|
|
|
|
|
| |
from the other side and only return TLS_READ_AGAIN/TLS_WRITE_AGAIN if we
failed to send a close notify on a non-blocking socket.
Otherwise be more forceful and always shutdown/close the socket regardless
of other failures. Also do not consider ENOTCONN or ECONNRESET to be a
shutdown failure, since there are various situations where this can occur.
ok doug@ guenther@
|
| |
|
|
|
|
|
|
| |
close the connection. Also correctly handle the error on failure.
Diff from cookieandscream via github.
|
|
|
|
|
|
| |
Diff from Tim van der Molen.
ok jmc@
|
|
|
|
|
|
| |
TLS_READ_AGAIN and TLS_WRITE_AGAIN.
Based on a diff from Tim van der Molen.
|
| |
|
|
|
|
|
|
| |
accepted via an existing pair of file descriptors.
Based on a diff from Jan Klemkow.
|
|
|
|
|
|
|
| |
for the server, rather than on the context for the connection. This makes
more sense than the current behaviour does.
Issue reported by Tim van der Molen.
|
|
|
|
| |
repeated use of tls_connect. ok jsing
|
|
|
|
|
|
| |
them guaranteed to not conflict per POSIX.
ok espie@ guenther@
|
| |
|
|
|
|
|
|
|
| |
tls_config_insecure_noverifyname(), so that it is more accurate and keeps
inline with the distinction between DNS hostname and server name.
Requested by tedu@ during s2k15.
|
|
|
|
| |
configuration.
|
|
|
|
|
|
|
| |
be those that are TLSv1.2 with AEAD and PFS. Provide a "compat" mode that
allows the previous default ciphers to be selected.
Discussed with tedu@ during s2k15.
|
| |
|
| |
|
|
|
|
| |
ok jsing
|
| |
|
|
|
|
| |
tls_config_set_protocols().
|
| |
|
|
|
|
|
|
|
|
|
| |
to be converted into a libtls protocols value. This allows for things like:
"tlsv1.0,tlsv1.1" (TLSv1.0 and TLSv1.1)
"all,!tlsv1.0" (all protocols except TLSv1.0)
Discussed with tedu@ and reyk@
|
|
|
|
| |
Found by reyk@
|
|
|
|
|
|
|
|
| |
that includes all currently supported protocols (TLSv1.0, TLSv1.1 and
TLSv1.2). Change all users of libtls to use TLS_PROTOCOLS_ALL so that they
maintain existing behaviour.
Discussed with tedu@ and reyk@.
|
|
|
|
|
|
|
|
|
| |
as tls_connect(), however allows the name to use for verification to be
explicitly provided, rather than being inferred from the host value.
Requested by reyk@
ok reyk@ tedu@
|
|
|
|
|
|
|
|
|
|
| |
to an actual host and use "servername" when referring to the name of the
TLS server that we expect to be indentified in the server certificate.
Likewise, rename verify_host to verify_name and use the term "name"
throughout the verification code (rather than host or hostname).
Requested by and ok tedu@
|
|
|
|
|
|
|
|
|
| |
IPv4 or IPv6 address before trying to resolve the address with the
AI_ADDRCONFIG flag set. This makes sure that attempts to connect to
numeric IPs or loopback addresses are always possible and not
prevented by AI_ADDRCONFIG.
OK jsing@ tedu@
|
|
|
|
|
|
|
| |
to ignore unsupported address families - eg. don't resolv IPv6 on
IPv4-only hosts.
OK jsing@
|
| |
|
|
|
|
|
|
|
|
| |
The tls_accept_socket() has been previously removed because the API is
not fixed yet; but it is also already used by httpd(8) and spamd(8) so
it is time to add it again and eventually change it later.
OK tedu@
|
|
|
|
|
|
|
|
|
| |
into memory. This can be used for tls_config_set_ca_mem(),
tls_config_set_cert_mem() or tls_config_set_key_mem().
With input from jsing@, tedu@ and henning@
OK tedu@
|
|
|
|
|
|
|
| |
dance handling code. This means that we get slightly useful messages when
a TLS connection or accept fails.
Requested by reyk@
|
|
|
|
|
|
|
|
|
| |
use for DHE. This enables the use of DHE cipher suites.
Rename tls_config_set_ecdhcurve() to tls_config_set_ecdhecurve() since it
is only used to specify the curve for ephemeral ECDH.
Discussed with reyk@
|
|
|
|
|
|
| |
failures return something that is actually useful to the caller.
ok reyk@
|
|
|
|
|
| |
As discussed with beck@ jsing@ and others
OK beck@
|
|
|
|
| |
OK jsing@
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
specifying a file. This enables CA verification in privsep'ed
processes that are running chroot'ed without direct access to the
certificate files.
With feedback, tests, and OK from bluhm@
|
|
|
|
|
|
|
|
| |
using the name of relayd relay or smtpd pki, use a 32 byte arc4random
buffer that should be unique for the context. This fixes an issue in
OpenSMTPD when a long pki name could break the configuration.
OK gilles@ benno@
|