summaryrefslogtreecommitdiff
path: root/src/lib/libssl/ssl_transcript.c (unfollow)
Commit message (Collapse)AuthorFilesLines
2020-01-30Revise for TLSv1.3 key share changes.jsing1-8/+27
2020-01-30Provide struct/functions for handling TLSv1.3 key shares.jsing8-130/+299
Pull out the key share handling code and provide a clean/self contained interface. This will make it easier to support groups other than X25519. ok beck@ inoguchi@ tb@
2020-01-30Factor out/rewrite the ECDHE EC point key exchange code.jsing5-173/+188
This reduces replication between the existing TLS client/server and allows the code to soon be reused for TLSv1.3. With feedback from inoguchi@ and tb@ ok inoguchi@ tb@
2020-01-30Enable t_ptrace with an errno change compared to NetBSD.mpi4-19/+31
Note that the last test triggers a kernel bug related to waitpid(9) and ptraced processes. This is now visible thanks to recent make(1) changes. guenther@ suggests to look at the logic behind `p_orphan' in FreeBSD to fix this bug.
2020-01-29Remove dead prototypes.jsing1-10/+1
2020-01-29If the TLSv1.3 code has not recorded an error and something already existsjsing3-3/+9
on the error stack, refrain from pushing an 'unknown' error on the stack. This should allow libcrypto errors (including bio) to be visible, where we have nothing better to offer. ok tb@
2020-01-29Remove unused stub implementation of tls13_accept(). The correcttb1-11/+1
and used version is in tls13_server.c. ok inoguchi jsing
2020-01-27Some of the test vectors that were added in the upstream committb1-7/+9
4672ff74d68766e7785c2cac4c597effccef2c5c have a zero byte prepended. Run the secp224k1 ECDH tests and adjust this if needed.
2020-01-26revert previous nc loop refactor from 1.211, breaks bluhm's stuffbeck1-64/+36
will attempt again later, now that there is new regress
2020-01-26Fix SSL_CIPHER_descriptioninoguchi1-2/+2
ok jsing@
2020-01-26Restrict to use up to TLSv1.2 for test_server_client in appstest.shinoguchi1-7/+7
Currently, only s_client has TLSv1.3 and s_server does not.
2020-01-26Avoid 32 bit right shift with unsigned int in crypto/cast/cast_lcl.hinoguchi1-2/+2
ok tb@
2020-01-26tweak previous; ok tbjmc1-3/+3
2020-01-26typotb1-2/+2
2020-01-26Document the change in EVP_chacha20(3).tb1-3/+5
Discussed with jsing
2020-01-26Improve the comment explaining why the previous change matches OpenSSL'stb1-8/+15
behavior. ok jsing
2020-01-26When an SSL method is set, bump the max version back to that of thejsing1-1/+10
incoming method if it is a client. This addresses the case where TLS_method() is used to initialise a SSL_CTX, then a TLS_client_method() is then set, resulting in TLSv1.2 being used instead of TLSv1.3. This is observable in smtpd. ok beck@
2020-01-26When switching back to a legacy client or server, ensure we reset thejsing2-2/+4
handshake function pointer. Fixes an isssue found by jca@ with OpenVPN. ok beck@ tb@
2020-01-26Fix basement bug where record layer would not correctly deal withbeck1-1/+4
traffic retries when not yet encrypting. ok jsing@
2020-01-26Add server side support for requesting client certificates in tls 1.3beck1-4/+173
ok jsing@
2020-01-26Add client certificate support for tls 1.3beck2-15/+149
ok jsing@
2020-01-26Add back the tests that were deleted in previous but not containedtb1-1/+9
in OpenSSL's test suite.
2020-01-26Add sigalgs for server side to enable client certificate processingbeck1-5/+34
in tls 1.3 Will be used in a follow on commit to enable tls1.3 client certificates ok jsing@
2020-01-26server sigalgs test is now bogus, disable for nowbeck1-2/+3
2020-01-26Adjust tests to match the change in EVP_chacha20().tb2-11/+22
Test vectors taken from OpenSSL 1.1.1d (under OpenSSL's old license).
2020-01-26Move pad and verify context into tls13_lib.cbeck4-68/+60
ok jsing@
2020-01-26Adjust EVP_chacha20()'s behavior to match OpenSSL's semantics:tb1-7/+14
The new IV is 128 bit long and is actually the 64 bit counter followed by 64 the bit initialization vector. This is needed by an upcoming change in OpenSSH and is a breaking change for all current callers. There are language bindings for Node.js, Rust and Erlang, but none of our ports use them. Note that EVP_chacha20() was first introduced in LibreSSL on May 1, 2014 while the entirely incompatible version in OpenSSL was committed on Dec 9, 2015. Initial diff from djm and myself, further refinements by djm. Ports grepping by sthen ok jsing
2020-01-25Add an underbar for consistency.tb1-2/+2
2020-01-25Disable cert interop tests for now.jsing1-2/+2
The libressl TLSv1.3 client and server currently lack client certificate authentication support and this test expects all clients can auth with all servers. We can likely turn this back on in the near future.
2020-01-25Actually disable cipher interop tests.jsing1-3/+3
2020-01-25Disable the cipher interop tests.jsing1-3/+4
These make far too many assumptions about cipher suites - TLSv1.3 cipher suites can only be used with TLSv1.3 and there is tests using TLSv1.3 cipher suites with TLSv1.2 will not work. Likewise, expecting TLSv1.2 cipher suites to work with TLSv1.3 is futile. Additionally, eopenssl11 lists TLSv1.3 cipher suites with different names to libressl. Futher work will be necessary before this can be re-enabled.
2020-01-25Accept both TLSv1.2 and TLSv1.3 protocols for netcat.jsing1-4/+3
This can potentially be improved by adding knowledge about which libraries support which versions and handle differences between clients and servers.
2020-01-25Disable session regress for libressl client talking to openssl11 server.jsing1-1/+2
This is now talking over TLSv1.3 and needs session support.
2020-01-25Revert change to certificate request check from r1.45.jsing1-3/+3
This code was correct, it was the entry in the table that was incorrect. ok beck@
2020-01-25Only perform the downgrade check if our max version is less than TLSv1.3.jsing1-15/+17
Issue noticed by kn@ when talking to a TLSv1.3 capable mail server, but with smtpd capping max version to TLSv1.2. ok beck@
2020-01-25Preserve the transcript hash for the client finished message,beck1-2/+3
and correct the message type for certificate request. ok jsing@
2020-01-25Support legacy message callbacks. First step for SSL_set_msg_callback(3)tb3-3/+40
support. Makes openssl s_client -msg work for handshake messages. ok beck jsing
2020-01-25Correct value for SSL_TLSEXT_MSG_HRR.jsing1-2/+2
ok beck@ tb@
2020-01-25Only discard the extension block for client hello and server hellojsing1-2/+3
messages. TLSv1.3 messages that include extensions need a length prefixed field with zero bytes, rather than no data at all. ok beck@ tb@
2020-01-25Only send an RI extension for pre-TLSv1.3 versions.jsing1-2/+2
ok beck@
2020-01-25It is possible to receive a pre-TLSv1.3 alert in response to a TLSv1.3jsing3-4/+24
client hello. Allow pre-TLSv1.3 alerts (including warnings) to be received before the server hello message. Disallow pre-TLSv1.3 alerts as soon as we know that we are using TLSv1.3. Noticed by ajacoutot@ while connecting to www.openprinting.org. ok tb@
2020-01-25Correct backwards test so that we may accept a certificate requstbeck1-3/+3
from the server. ok jsing@
2020-01-25add a couple of XXX for future cleanuptb1-1/+4
2020-01-25Disable the client hello message regress test for now.jsing1-2/+2
The golden values have changed due to TLSv1.3 and will likely change more in the near future. This will be updated and re-enabled when things settle. Discussed with beck@
2020-01-25Ensure that TLSv1.0 and TLSv1.1 are enabled before running SSLv2 clientjsing1-5/+10
hello tests.
2020-01-24Teach openssl s_client a bit about TLSv1.3.tb1-2/+17
ok beck jsing
2020-01-24Complete the initial TLSv1.3 implementation.jsing3-14/+300
ok beck@ tb@
2020-01-24Preserve the TLS transcript at additional points.jsing1-9/+23
This is needed for the TLSv1.3 server and will also be needed for client certificate authentication. Note that we preserve on receive but before recording the new handshake message, whereas we preserve on send after recording the new handshake message. ok tb@
2020-01-24Permit 0 length writes, because openssl s_client is specialbeck1-2/+2
ok jsing@
2020-01-24Store the legacy session identifier from the ClientHello so we can actuallyjsing1-1/+10
echo it. ok beck@ tb@
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-04-28 20:13:07 +0000