3.3.0.rc1

What's Changed

• YJIT: shrink Context down to 15 bytes by maximecb · Pull Request #8911
• Fix IRB tests by hsbt · Pull Request #8925
• [DOC] More on JSON extensions by BurdetteLamar · Pull Request #8898
• YJIT: make --yjit-max-versions=N option undocumented by maximecb · Pull Request #8962
• YJIT: Print a disasm path to stderr by k0kubun · Pull Request #8967
• Bug #11183: Cumulative error on Complex::I ** 100000000000000000000000000000000
• Bundle rbs-3.3.1 by soutaro · Pull Request #8921
• Bug #20004: Release CI is broken with Ruby master
• Bundle rbs-3.3.2 by soutaro · Pull Request #8974
• YJIT: Skip dump-disasm if it fails to create a file by k0kubun · Pull Request #8968
• [wasm] Fix Asyncify loop exit condition for normal return by kateinoigakukun · Pull Request #9007
• YJIT: add an extra btest for shape too complex by maximecb · Pull Request #9013
• YJIT: Avoid a register spill on arm64 by k0kubun · Pull Request #9014
• windows-arm64 support by pbo-linaro · Pull Request #8995
• YJIT: Fix jmp_ptr_bytes on x86_64 by k0kubun · Pull Request #9016
• YJIT: Apply patches ignoring page_end_reserve by k0kubun · Pull Request #9015
• YJIT: record num_send_cfunc stat by maximecb · Pull Request #9022
• Bug #19928: IPV6 nameserver issues: "Protocol not supported - socket(2) - udp"
• Bug #19994: Seemingly require causes segfault??
• Bug #13513: Resolv::DNS::Message.decode hangs after detecting truncation in UDP messages
• Feature #18151: Incorrect Resolv result when DNS server is unreachable
• Bug #20023: rb_require("noexist") causes segfault
• YJIT: add top C function call counts to --yjit-stats by maximecb · Pull Request #9047
• Feature #18551: Make Range#reverse_each to raise an exception if endless
• YJIT: Assert no patch overlap on pos_marker by k0kubun · Pull Request #9048
• Bug #20019: Ruby 3.3 GVL instrumentation events are fired out of order
• YJIT: reduce default exec-mem-size to 64MiB by maximecb · Pull Request #9054
• YJIT: fix bug in top cfunc logging in --yjit-stats by maximecb · Pull Request #9056
• Bug #20025: Parsing identifiers/constants is case-folding dependent
• Bug #17037: rounding of Rational#to_f
• YJIT: edit yjit.md and bring it up to date by maximecb · Pull Request #9068
• Feature #19630: [RFC] Deprecate Kernel#open("|command-here") due to frequent security issues
• Bug #19877: Non intuitive behavior of syntax only applied to literal value
• YJIT: Bump ec->cfp after setting cfp->jit_return by k0kubun · Pull Request #9072
• Bug #19417: Regexp \p{Word} and [[:word:]] do not match Unicode Other_Number character
• YJIT: optimized codegen for rb_ary_length() by maximecb · Pull Request #9085
• YJIT: Cancel on-stack jit_return on invalidation by k0kubun · Pull Request #9086
• Bug #20030: Ripper.tokenize('"\\C-あ"') separates encoding valid string to encoding invalid string.
• Bug #19838: Ripper nested heredocs
• Bug #20033: /#{''}(?<lvar>)/ =~ s looks like a dynamic regexp match but assigns to local variable
• [DOC] More on What's Here by BurdetteLamar · Pull Request #9099
• [Fix] Support when nil is assigned to variable name by jinroq · Pull Request #9105
• YJIT: Mark and update stubs in invalidated blocks by XrXr · Pull Request #9104
• Don't warn generally, only warn on specific code paths e.g. map. by ioquatix · Pull Request #9131
• Add news for Fiber#kill. by ioquatix · Pull Request #9134
• Fix incorrect "nested_fake_name" documentation. by ioquatix · Pull Request #9135
• YJIT: Avoid register allocation conflict with a higher stack_idx by k0kubun · Pull Request #9143
• Fix SEGV caused by GC::Profiler.raw_data by soutaro · Pull Request #9122
• Bug #20039: Matching US-ASCII string to copied UTF-8 Regexp causes invalid multibyte character error
• Warn it by k0kubun · Pull Request #9152
• Set AI_ADDRCONFIG when making getaddrinfo(3) calls for outgoing conns by KJTsanaktsidis · Pull Request #7295
• Bug #20022: GC.verify_compaction_references does not actually move all objects
• Bug #19779: eval "return" at top level raises LocalJumpError
• Bug #18886: Struct aref and aset don't trigger any tracepoints.
• YJIT: Fix on-stack ISEQ comparison for auto_compact by k0kubun · Pull Request #9164
• Bug #20021: TestGCCompact#test_moving_hashes_down_size_pools is flaky
• [DOC] RDoc for Complex by BurdetteLamar · Pull Request #9181

Note: This list is automatically generated by tool/gen-github-release.rb. Because of this, some commits may be missing.

Full Changelog

v3_3_0_preview3...v3_3_0_rc1