We need to use the same options for both C compiler and assembler
when `-mbranch-protection` is guessed by configure. Otherwise,
`coroutine/arm64/Context.{h,S}` will use incompatible PAC strategies.
* Allows macOS users to use M:N threads (and technically FreeBSD, though it has not been verified on FreeBSD)
* Include sys/event.h header check for macros, and include sys/event.h when present
* Rename epoll_fd to more generic kq_fd (Kernel event Queue) for use by both epoll and kqueue
* MAP_STACK is not available on macOS so conditionall apply it to mmap flags
* Set fd to close on exec
* Log debug messages specific to kqueue and epoll on creation
* close_invalidate raises an error for the kqueue fd on child process fork. It's unclear rn if that's a bug, or if it's kqueue specific behavior
Use kq with rb_thread_wait_for_single_fd
* Only platforms with `USE_POLL` (linux) had changes applied to take advantage of kernel event queues. It needed to be applied to the `select` so that kqueue could be properly applied
* Clean up kqueue specific code and make sure only flags that were actually set are removed (or an error is raised)
* Also handle kevent specific errnos, since most don't apply from epoll to kqueue
* Use the more platform standard close-on-exec approach of `fcntl` and `FD_CLOEXEC`. The io-event gem uses `ioctl`, but fcntl seems to be the recommended choice. It is also what Go, Bun, and Libuv use
* We're making changes in this file anyways - may as well fix a couple spelling mistakes while here
Make sure FD_CLOEXEC carries over in dup
* Otherwise the kqueue descriptor should have FD_CLOEXEC, but doesn't and fails in assert_close_on_exec
Previously, the embedded semicolon in BASERUBY if BASERUBY is
not available breaks tarball builds without BASERUBY when using
OpenBSD make, due to the inability to escape MFLAGS correctly.
This moves the same BASERUBY code into a separate file, avoiding
the MFLAGS quoting issue.
BASERUBY must be passed to build-ext because it is required
by ripper since the introduction of lrama.
Fixes [Bug #19683]
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org>
* [win32] fix compilation for windows-arm64
Credits to MSYS2 Ruby package using this patch.
* [win32] nm use full options
Fix compilation error when using MSYS2 environment.
Credits to MSYS2 Ruby package using this patch.
* [win32] detect llvm-windres (used for windows-arm64)
When adding preprocessor option for llvm-windres (using clang as
parameter), it fails. Thus, do not add this.
It's needed to be able to compile windows-arm64 version, because MSYS2
toolchain is LLVM based (instead of GCC/binutils).
* [win32] pioinfo detection for windows-arm64
This fixes "unexpected ucrtbase.dll" for native windows-arm64 ruby
binary. It does not solve issue with x64 version emulated on this
platform.
Value of pioinfo pointer can be found in ucrtbase.dll at latest adrp/add
sequence before return of _isatty function. This works for both release
and debug ucrt.
Due to the nature of aarch64 ISA (vs x86 or x64), it's needed to
disassemble instructions to retrieve offset value, which is a bit more
complicated than matching specific string patterns.
Details about adrp/add usage can be found in this blog post:
https://devblogs.microsoft.com/oldnewthing/20220809-00/?p=106955
For instruction decoding, the Arm documentation was used as a reference.
During the build, Ruby has special logic to serialize its own builtin
module to disk using the binary iseq format during the build (I assume
for speed so it doesn't have to parse builtin every time it starts
up).
However, since iseq format is architecture-specific, when building on
x86_64 for universal x86_64 + arm64, the serialized builtin module is
written with the x86_64 architecture of the build machine, which fails
this check whenever ruby imports the builtin module on arm64:
1fdaa06660/compile.c (L13243)
Thankfully, there's logic to disable this feature for cross-compiled builds:
1fdaa06660/builtin.c (L6)
This disables the iseq logic for universal builds as well.
Fixes [Bug #18286]
I check the s390s-linux/Ubuntu system and I found that MN threads
doesn't work with:
* function inlining (-O0 doesn't repro)
* Thread local specifier (`_Thread_local`.
With `pthread_get_specific works)
(it is not an issue of `__tls_get_addr()` written in thread.c)
* swap context with ucontext (coroutine/ucontext)
I couldn't find out what is the root cause of this issue but
disable MN threads to make CI healthy.
On some platforms, such as FreeBSD and Oracle Linux, symbols defined
in the crt0 setup routine are exported from shared libraries. So
ignore the symbols that would be exported even in an empty shared
library.
This patch introduce M:N thread scheduler for Ractor system.
In general, M:N thread scheduler employs N native threads (OS threads)
to manage M user-level threads (Ruby threads in this case).
On the Ruby interpreter, 1 native thread is provided for 1 Ractor
and all Ruby threads are managed by the native thread.
From Ruby 1.9, the interpreter uses 1:1 thread scheduler which means
1 Ruby thread has 1 native thread. M:N scheduler change this strategy.
Because of compatibility issue (and stableness issue of the implementation)
main Ractor doesn't use M:N scheduler on default. On the other words,
threads on the main Ractor will be managed with 1:1 thread scheduler.
There are additional settings by environment variables:
`RUBY_MN_THREADS=1` enables M:N thread scheduler on the main ractor.
Note that non-main ractors use the M:N scheduler without this
configuration. With this configuration, single ractor applications
run threads on M:1 thread scheduler (green threads, user-level threads).
`RUBY_MAX_CPU=n` specifies maximum number of native threads for
M:N scheduler (default: 8).
This patch will be reverted soon if non-easy issues are found.
[Bug #19842]
To get rid of mysterious errors such as:
```
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/nm: error: libruby.3.3-static.a(/): The end of the file was unexpectedly encountered
```
and
```
ld: warning: ignoring file ../../libruby.3.3-static.a, building for macOS-x86_64 but attempting to link with file built for macOS-x86_64
Undefined symbols for architecture x86_64:
"_rb_rational_num", referenced from:
```
Checking by `__STDC_VERSION__` is unreliable because old gcc 4.8
supports `-std=gnu11` option but does not implement `_Thread_local`.
Check the implementation directly instead.
This is useful for passing directory file descriptors over UNIX
sockets or to child processes to avoid TOCTOU vulnerabilities.
The implementation follows the Dir.chdir code.
This will raise NotImplementedError on platforms not supporting
both fchdir and dirfd.
Implements [Feature #19347]
[Misc #16671]
I'd like to bump it to 2.7 to use pattern matching in
tool/mk_builtin_loader.rb.
However, I experienced a few blockers. 2.5 seems like the closest
version that is easy enough to use on CIs, so let me bump the version to
it as an intermediate step for it. I want to use &. and <<~ in 2.3 too.
Known blockers:
* AppVeyor Visual Studio 2015 doesn't have Ruby 2.7. You'd need to bump
the version to Visual Studio 2019.
* GitHub Actions windows-2019 doesn't have Ruby 2.7 either. You
can use ruby/setup-ruby, but configure doesn't seem to work with it.
* For ruby/ruby-ci-imaage, bionic doesn't have Ruby 2.7. I tried using
ruby-build to build Ruby 2.7 from package, but the build on its CI
seems to somehow loop forever when I do that. So I gave it up for now.
We might want to wait until bionic becomes EOL.
Note:
* AppVeyor Visual Studio 2015 has Ruby <= 2.6.3
https://www.appveyor.com/docs/windows-images-software/#ruby
* GitHub Actions windows-2019 uses Ruby 2.5.9
https://github.com/actions/runner-images/blob/main/images/win/Windows2019-Readme.md
Nobuyoshi Nakada
Takashi Kokubun
Yuta Saito
JP Camara
Jeremy Evans
Pierrick Bouvier
Hiroshi SHIBATA
Ben Hamilton
Koichi Sasada
Jean Boussier
Jemma Issroff