Prior to this patch, if I ran:
ruby setup.rb --destdir /foo
Then Bundler files would be written into /foo/foo, because destdir was
being prepended, even though `bundler_spec.bin_dir` already included
destdir.
https://github.com/rubygems/rubygems/commit/9e857ffb52
* New code page allocation logic
* Fix leaked globals
* Fix leaked symbols, yjit asm tests
* Make COUNTED_EXIT take a jit argument, so we can eliminate global ocb
* Remove extra whitespace
* Change block start_pos/end_pos to be pointers instead of uint32_t
* Change branch end_pos and start_pos to end_addr, start_addr
The compaction tests get stuck randomly on s390x for unknown reason.
http://rubyci.s3.amazonaws.com/s390x/ruby-master/log/20211104T030003Z.fail.html.gz
```
[13715/21145] TestGCCompact#test_gc_compact_statstimeout: output interval exceeds 1800.0 seconds.
```
We spent some time to investigate this issue, but we can't figure out
why, and it is unlikely that we'll be able to fix it anytime soon.
This random failure makes the CI unuseful, so tentatively we suppress
this test for a while. A contribution from those who are familiar with
s390x is welcome.
The last_communicated timestamp is for HTTP persistent connection, to
decide whether the current TCP connection may be reused for the
subsequent requests or not. Naturally, the timer must be reset if the
connection is recreated since it is no longer relevant.
https://github.com/ruby/net-http/commit/0a013de42d
This fixes the following test error testing against OpenSSL 3.x:
~~~
2) Failure:
TestGemRequest#test_verify_certificate_extra_message [/builddir/build/BUILD/ruby-3.0.2/test/rubygems/test_gem_request.rb:358]:
<"ERROR: SSL verification error at depth 0: invalid CA certificate (24)\n" +
"ERROR: Certificate is an invalid CA certificate\n"> expected but was
<"ERROR: SSL verification error at depth 0: invalid CA certificate (79)\n" +
"ERROR: Certificate is an invalid CA certificate\n">.
~~~
Where the root cause is this OpenSSL commit:
1e41dadfa7
It seems that OpenSSL upstream considers the constant value just an
implementation detail and therefore this changes the test case to
follow the suite.
https://github.com/rubygems/rubygems/commit/8acf8e95dc
Readline seems to emit an escape sequence `"\e[?1034h` at an
unpredictable timing when `TERM=xterm` or something.
This leads to a noisy random test failure.
http://rubyci.s3.amazonaws.com/amazon2/ruby-master/log/20211102T063005Z.fail.html.gz
```
1) Failure:
TestReadline#test_modify_text_in_pre_input_hook [/home/chkbuild/chkbuild/tmp/build/20211102T063005Z/ruby/test/readline/test_readline.rb:404]:
<"> hello world\n"> expected but was
<"\e[?1034h> hello world\n">.
```
This test is for Ruby's readline extension, not for readline library
itself. So here this change force to use "TERM=vt100" to run the tests.
* YJIT: Support kwargs sends with all defaults
Previously keyword argument methods were only compiled by YJIT when all
keywords were specified in the caller.
This adds support for calling methods with keyword arguments when no
keyword arguments are specified and all are filled with the defaults.
* Remove unused send_iseq_kwargs_none_passed
OpenSSL::SSL::SSLSocket allowed #read and #write to be called before an
SSL/TLS handshake is completed. They passed unencrypted data to the
underlying socket.
This behavior is very odd to have in this library. A verbose mode
warning "SSL session is not started yet" was emitted whenever this
happened. It also didn't behave well with OpenSSL::Buffering. Let's
just get rid of it.
Fixes: https://github.com/ruby/openssl/issues/9https://github.com/ruby/openssl/commit/bf780748b3
Calling `Bundler.definition.specs` will memoize materialized specs.
However, requiring `bundler/setup` will end up materializing the same
set of specs, but not memoize them.
This change makes things consistent.
https://github.com/rubygems/rubygems/commit/e4c2b52824
