summaryrefslogtreecommitdiff
path: root/src/lib/libssl/ssl_init.c (unfollow)
Commit message (Collapse)AuthorFilesLines
2019-02-21Fix a few cases where int was used instead of ssize_t.jsing1-3/+3
2019-02-17Convert openssl(1) pkeyutl to the newer style of option handling.inoguchi1-198/+278
ok jsing@ tb@
2019-02-16Add tests for pkeyutlinoguchi1-6/+46
2019-02-14Correct guards.jsing1-3/+3
2019-02-14Notify the record layer upon handshake completion.jsing1-2/+4
ok tb@
2019-02-14Switch to application traffic keys as appropriate.jsing1-1/+17
Switch the read traffic key to the server application traffic key once the server finished message has been processed. Switch the write traffic key to the client application traffic key after sending the client finished message. ok tb@
2019-02-14Split tls13_record_layer_set_traffic_keys() into two separate functions.jsing3-17/+23
This allows the read traffic key to be set independently of the write traffic key. This will become necessary for KeyUpdate handling, however also allows for switching to application traffic keys at more appropriate stages of the handshake. ok tb@
2019-02-14Provide a TLS 1.3 capable client method.jsing3-5/+50
ok tb@
2019-02-13one more error message that should go to stderrtb1-2/+3
2019-02-13Clean up some pesky trailing whitespace.jsing1-10/+10
2019-02-13Handle finished messages in TLSv1.3 client.jsing2-14/+131
This adds support for processing of the server finished message and generation of the client finished message. ok tb@
2019-02-13New futex(2) based rwlock implementation based on the mutex code.mpi1-1/+12
This implementation reduces contention because threads no longer need to spin calling sched_yield(2) before going to sleep. Tested by many, thanks! ok visa@, pirofti@
2019-02-13Xr the byteorder funcs; from tim kuijstenjmc1-3/+4
2019-02-11Implement handling of Certificate and CertificateVerify messages.jsing2-31/+214
This allows the TLS 1.3 client to process the certificates that the server has sent and verify that the server has possession of the private key. ok tb@
2019-02-11getopt(3) returns int, not char. Fix type of ch variable to preventbluhm2-6/+6
sign error during arm regress.
2019-02-10"non-existant" is one of those words that don't exist, so use "non-existent"phessler1-2/+2
instead From Pamela Mosiejczuk, many thanks! OK phessler@ deraadt@
2019-02-10Preserve the transcript hash at specific stages of the TLSv1.3 handshake.jsing2-2/+16
There are various points where we need the hash of all messages prior to the current message. Support this by having the handshake code preserve the transcript hash prior to recording the current message, which avoids the need to sprinkle this throughout multiple handlers. ok inoguchi@ tb@
2019-02-09Fix weird wrap showing cipher list in interactive modeinoguchi6-13/+23
ok jsing@ tb@
2019-02-09Rename the file that contains the transcript handling code.jsing2-3/+3
ok inoguchi@ tb@
2019-02-09Rename tls1_handshake_hash*() to tls1_transcript_hash*().jsing7-29/+29
While handshake hash is correct (in as far as it is a hash of handshake messages), using tls1_transcript_hash*() aligns them with the naming of the tls1_transcript*() functions. Additionally, the TLSv1.3 specification uses Transcript-Hash and "transcript hash", which this matches. ok inoguchi@ tb@
2019-02-09Provide a reference from the TLSv1.3 context to the handshake state.jsing2-19/+25
This allows ctx->hs to be used throughout the TLSv1.3 code, rather than S3I(ctx->ssl)->hs_tls13. ok inoguchi@ tb@
2019-02-09Summarize the 4 same name functions and move it to apps.cinoguchi7-72/+26
ok tb@ jsing@
2019-02-07Implement processing of EncryptedExtensions in the TLS 1.3 client.jsing2-8/+28
ok bcook@ tb@
2019-02-05Add islower check to show_ciphers in pkey.c and rsa.c.inoguchi2-2/+10
suggested by jsing@ ok tb@
2019-02-05Convert openssl(1) pkey to the newer style of option handling.inoguchi1-96/+171
ok jsing@
2019-02-04Implement parsing and processing of TLSv1.3 ServerHello messages.jsing3-11/+187
ok tb@
2019-02-04Make it easier to build with LIBRESSL_HAS_TLS1_3 enabled.jsing1-1/+4
ok tb@
2019-02-04Document missing command line options for the rsautl command.tb2-9/+16
ok jsing, "looks good!" jmc
2019-02-03zap spaces before tabstb1-5/+5
2019-02-03Convert openssl(1) rsautl to the newer style of option handling.jsing1-121/+196
ok beck@ inoguchi@ tb@
2019-02-03Revert r1.38 as it introduces use of a stack value post function return.jsing1-50/+86
The deduplication is also not quite right - this will be revisited in due course.
2019-02-03Use malloc() and memcpy() the test X25519 x25519_peer_public value.jsing1-3/+6
Otherwise, if tlsext_keyshare_server_build() fails we call free with a pointer to static memory and bad things happen. Reported by bcook@
2019-01-31unwrap a line introduced in previous.tb1-3/+2
2019-01-30Correct handling of TLS sigalgs extension for TLSv1.0/TLSv1.1.jsing1-33/+19
When operating as a TLSv1.0 or TLSv1.1 server, we still have to parse the TLS sigalgs extension if presented by the client (which might be TLSv1.2 capable), rather than treating its presence as an error. While here, remove future version dependence issues by avoiding explicit version equality checks. Issue reported by bluhm@. ok bluhm@ tb@
2019-01-29instead of calling abort() for 3 "must not use anymore" interfaces,deraadt1-7/+4
call OPENSSL_assert() with the specific message, which due to recent changes will end up syslog :) ok jsing
2019-01-29Remove SSLv23 padding mode from rsautl's usage (left behind in a removaltb1-2/+1
from code and manual in 2017). Reported by KEINOS in github issue #101.
2019-01-28spelling;jmc1-3/+3
2019-01-28Deduplicate a bunch of replicated code in the extension handlingbeck1-86/+50
ok tb@
2019-01-28Add tls_extension_seen(), a utility to know if a particular extensionbeck2-9/+15
has been seen in the handshake so far. Use it for keyshare. ok tb@
2019-01-27zap a commented out line. 0RTT will need more thought thantb1-2/+1
just uncommenting this.
2019-01-27refactor and clean up the code generating dot output.tb1-53/+76
2019-01-26Recent discussions about abort() potentially leaving key material inderaadt1-5/+8
core files (which can depend upon various file layouts) have resonated with my hate for this function outside a purely debugging context. I also dislike how the report goes to stderr which may get lost or ignored. Increase the noise (with syslog_r) and use _exit(1) to gaurantee termination. ok jsing
2019-01-25print errors to stderrtb1-19/+21
2019-01-25sort output suffixestb1-2/+2
2019-01-25I am retiring my old email address; replace it with my OpenBSD one.millert27-61/+61
2019-01-24Add code to visualize the state machine. Both the state machine and thetb2-5/+79
output will have to be tweaked, but this may as well happen in-tree. To try it, pkg_add graphviz and run 'make handshake.svg' in this directory. Committing early so Bob's followers can play.
2019-01-24Set pointer variables in tls13 handshake to NULL on clearbeck1-1/+5
ok jsing@ bcook@
2019-01-24set the NEGOTIATED flag in the flags argument rather thantb1-4/+4
squeezing it into the table.
2019-01-24Add server side of versions, keyshare, and client and server of cookiebeck5-28/+742
extensions for tls1.3. versions is currently defanged to ignore its result until tls13 server side wired in full, so that server side code still works today when we only support tls 1.2 ok bcook@ tb@ jsing@
2019-01-24move the extensions_seen into the handshake structbeck3-6/+12
ok jsing@