Age | Commit message (Collapse) | Author |
|
* ingela/ssl/transport-transparance/ERL-861/OTP-15679:
ssl: Fix transport transparancy
|
|
* ingela/ssl/recv-timeout-bug/ERL-884/ERL-883/OTP-14701:
ssl: Cancel recv timer in all places
|
|
During socket stop the close environment somehow has
been NULLed. Check this before using when sending the
close message.
|
|
* john/compiler/fix-eq-type-infererence-in-validator/ERL-886:
beam_validator: Infer types on both sides of '=:='
|
|
Tracing would be left hanging if the test timed out, causing later
tests to fail.
|
|
be ignored
|
|
|
|
|
|
maint-21
* ingela/ssl/recv-timeout-bug/ERL-884/ERL-883/OTP-14701:
ssl: Cancel recv timer in all places
|
|
* ingela/ssl/transport-transparance/ERL-861/OTP-15679:
ssl: Fix transport transparancy
|
|
* rickard/erl_call/ERL-881/OTP-15676:
Add smoke test for erl_call
Add new api functions to ei_fake_prog
Fix timeout value when waiting for emulator start
Fix initialization of erl_call
|
|
|
|
|
|
|
|
Store cipher suite information in session record.
Test ssl:connection_information/1 in a TLS 1.3 connection.
Change-Id: I7193e6dd2544540e446b5777b5768806cecf2bd3
|
|
We use clock_gettime(CLOCK_REALTIME, ...) to create a simple
timestamp. This is used both for debugging and when writing
error/warning messages (from the nif-code).
The workaround for the absence of this function (on some platforms,
old versions of darwin for example) is to simply skip the timestamp.
|
|
|
|
These functions were added in erl_interface-3.11
|
|
|
|
Allow list of chunks to be given to strip*()
OTP-15680
|
|
* ingela/ssl/public_key/verify_fun-doc-error/ERL-887:
public_key: Correct documentation of verify_fun
|
|
|
|
The script would hang if no local hex had been installed previously.
|
|
Previously erl_call relied on the implicit initialization made
of the ei-lib if no explicit initialization had been done. This
implicit initialization was utterly broken and was removed in
erl_interface-3.11 (OTP 21.3) since it has been documented for
a very long time that an explicit initialization is required.
|
|
Modules like 'merl' and 'erl_prettypr' could be loaded during the
actual test, causing them to be traced too.
|
|
* peterdmv/ssl/client-auth/OTP-15591:
ssl: Improve ssl logging
ssl: Test handling of signature algorithms
ssl: Handle unencrypted Alert (Illegal Parameter)
ssl: Improve verification of received Certificate
ssl: Fix Alert logging
ssl: Fix get_handshake_context/2
ssl: Test HelloRetryRequest with client auth
ssl: Verify signature algorithm in CV
ssl: Verify CertificateVerify
ssl: Test client authentication with certificate
ssl: Validate client certificates (TLS 1.3)
ssl: Test client authentication (empty cert)
ssl: Implement state 'wait_cert'
ssl: Add ssl logger support for CertificateRequest
ssl: Fix ssl alerts
Change-Id: Id4ba14d373f116038a7cb3ff9fc33faed23031c8
|
|
|
|
|
|
|
|
|
|
|
|
Fixed an issue (I think) that during socket close
could cause a core.
|
|
* hans/crypto/fix_chacha20_bug/OTP-15678:
crypto: Disable chacha20 if OpenSSL 1.1.0-1.1.0c
|
|
* hans/crypto/fix_ret_bug_poly1305/OTP-15677:
crypto: Fix bug in error return for crypto:poly1305/2
|
|
|
|
The parsing of the OTP doc seldom fails but in the rare cases it does
not find a function signature hide the error by default. If the user
wish to be notified they may customize `erldoc-no-signature-function'.
Sorry for breaking the "Let it crash!" rule ;)
|
|
Before this commit some function signatures were missed when parsing
the html doc of OTP 21.3. After this commit more are collected but
still two signatures are missed (erlang:statistics/1 and
filename:basedir/3).
|
|
Since Emacs 26 make-symbolic-link will by default not expand a leading
~ in the target. Let's do it ourselves then i guess.
|
|
Would cause connection to crash although all data was delivered to ssl:recv
|
|
* sverker/ets-select-fixation-owner-change-bug/OTP-15672:
erts: Fix bug for yielding ets:replace
|
|
Found by valgrind:
Conditional jump or move depends on uninitialised value(s)
Suspected: ets_select_replace_1:3034 [erl_db.c]
Bug introduced by already merged parent commit
0d550c80d4f19cc432e7de056169695d436c02a0.
|
|
due to a bug. The cipher first appeared in 1.1.0
|
|
Returned the atom notsup instead of the exception notsup.
|
|
Fixed a memory leak in the function for address infos encoding.
|
|
Wait for resource-holding processes to garbage collect before
exiting to know destructors have been called.
|
|
* sverker/ets-select-fixation-owner-change-bug/OTP-15672:
erts: Fix ets:select table fixation leak at owner change
erts: Refactor common things into traverse_context_t
stdlib: Clarify docs for ets:info(_, safe_fixed)
|
|
A debug printout in the socket_stop fucntion used the wrong
requestor record for the pid, which may leasd to a confusing
debug printout.
|
|
The activate-next (reader|writer|acceptor) always
*incorrectly* used the read select function.
This has now been changed so that the functions
are created via a macro with an argument specifying
which select function to use.
|
|
retranspissions -> retransmissions
|
|
|