| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
| |
Changes: https://github.com/ruby/ruby/releases
https://github.com/ruby/ruby/blob/master/NEWS.md
https://www.ruby-lang.org/en/news/2026/09/15/ruby-4-0-7-released/
|
| |
|
|
|
|
| |
Changes: https://github.com/ruby/ruby/releases
https://github.com/ruby/ruby/blob/master/NEWS.md
https://www.ruby-lang.org/en/news/2026/07/14/ruby-4-0-6-released/
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Two ports, devel/rubygem-glib2 and graphics/rubygem-cairo, have been
broken on armv7 for a long time. Recently, pkubaj fixed a problem with
Ruby's coroutine code on ppc64* (see 3bb0a9335) affecting the very same
ports. This naturally raises the question if the armv7 problems are
related.
Turns out that due to an oversight, we use the generic ucontext.h-based
coroutine code on armv7. Switching to the assembly-based code fixes
the ports! Further digging suggests that there may be a problem in the
freebsd32 support for these calls on aarch64 causing the observed
issues. We will investigate those separately.
Tested by: fuz
Approved by: portmgr (build fix blanket)
MFH: 2026Q2
|
| |
|
|
|
|
|
|
|
|
| |
- Remove USE_LOCALE workaround [1]
Changes: https://github.com/ruby/ruby/releases
https://github.com/ruby/ruby/blob/master/NEWS.md
https://www.ruby-lang.org/en/news/2026/05/20/ruby-4-0-5-released/
Security: CVE-2026-46727
Reference: https://bugs.ruby-lang.org/issues/22065 [1]
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The hand-written `coroutine_transfer` in `coroutine/ppc64le/Context.S`
lacks the ELFv2 ABI global entry prologue. When the function is reached
via the PLT (lazy resolution path), r2 (TOC pointer) is left pointing
at whatever the resolver had loaded — typically the dynamic linker's
own TOC — instead of libruby's TOC.
The wrong r2 propagates through the fiber switch into `fiber_entry`,
which jumps into `fiber_restore_thread` → `rb_current_ec_set`. The
first PLT thunk inside the new fiber (`__plt___tls_get_addr` for the
`ruby_current_ec` thread-local) computes its GOT entry as
`r2 + offset` and segfaults dereferencing the bogus address.
Fixes build of textproc/rubygem-nokogiri and devel/rubygem-glib2.
|
| |
|
|
|
|
| |
Changes: https://github.com/ruby/ruby/releases
https://github.com/ruby/ruby/blob/master/NEWS.md
https://www.ruby-lang.org/en/news/2026/05/11/ruby-4-0-4-released/
|
| |
|
|
|
|
| |
Changes: https://github.com/ruby/ruby/releases
https://github.com/ruby/ruby/blob/master/NEWS.md
https://www.ruby-lang.org/en/news/2026/04/21/ruby-4-0-3-released/
|
| |
|
|
|
|
| |
Changes: https://github.com/ruby/ruby/releases
https://github.com/ruby/ruby/blob/master/NEWS.md
https://www.ruby-lang.org/en/news/2026/03/16/ruby-4-0-2-released/
|
| |
|
|
|
|
| |
Changes: https://github.com/ruby/ruby/releases
https://github.com/ruby/ruby/blob/master/NEWS.md
https://www.ruby-lang.org/en/news/2026/01/13/ruby-4-0-1-released/
|
| |
|
|
|
| |
RUBY_PORTREVISION is only used for lang/ruby{32,33,34,40}. These ports can set
PORTREVISION directly rather than defined in the shared Mk/Uses/ruby.mk.
|
| |
|
|
|
|
| |
Changes: https://github.com/ruby/ruby/releases
https://github.com/ruby/ruby/blob/master/NEWS.md
https://www.ruby-lang.org/en/news/2025/12/25/ruby-4-0-0-released/
|
| |
|
|
|
| |
- Use LIB_DEPENDS for libffi
- Cosmetic change
|
| |
|
|
| |
ReleaseNotes: https://www.ruby-lang.org/en/news/2025/11/17/ruby-4-0-0-preview2-released/
|
|
|
Upstrem changed version of master branch in their repository from 3.5
to 4.0.
|