Ruby Issue Tracking System

07/25/2026

07:44 AM Ruby Feature #20215: Introduce `IO#readable?`

This issue has come up again in [socketry/async-http#223](https://github.com/socketry/async-http/issues/223).
It is particularly difficult to handle for TLS. The underlying file descriptor of an `SSLSocket` can be readable because it co... ioquatix (Samuel Williams)

07/20/2026

08:02 AM Ruby Feature #22138: Add `RB_NOGVL_PENDING_INTERRUPT_FAIL` flag for `rb_nogvl`.

@ko1 I'd like to backport this, because without it, it is impossible to use pending interrupts with blocking operations / `rb_nogvl`.
Is that acceptable? ioquatix (Samuel Williams)

07/17/2026

03:40 AM Ruby Bug #22196: Heap-use-after-free in `fiber_switch` with transfer-terminated async tasks on 3.4.10

Proposed fix: https://github.com/ruby/ruby/pull/17957 ioquatix (Samuel Williams)

07/16/2026

08:49 AM Ruby Bug #22196 (Assigned): Heap-use-after-free in `fiber_switch` with transfer-terminated async tasks on 3.4.10

Thanks for your report, I'll investigate it. ioquatix (Samuel Williams)

07/12/2026

11:22 PM Ruby Bug #15980: Coverage shows while/until after raise if/unless as uncovered line

I asked GPT 5.5 to fix the issue: https://github.com/ruby/ruby/pull/17825 ioquatix (Samuel Williams)

07:02 AM Ruby Bug #15980: Coverage shows while/until after raise if/unless as uncovered line

Is there any chance we can revisit this issue and propose a better fix than the current "coverage is enabled so avoid doing the optimisation" hack? ioquatix (Samuel Williams)

07/10/2026

08:49 AM Ruby Feature #22138: Add `RB_NOGVL_PENDING_INTERRUPT_FAIL` flag for `rb_nogvl`.

It was suggested to use `RB_NOGVL_MASKED_INTR_FAIL`. However this change is not about masked interrupts but the pending interrupt queue.
The flag was thus renamed from `RB_NOGVL_PENDING_INTERRUPT_FAIL` to `RB_NOGVL_PENDING_INTR_FAIL`. ioquatix (Samuel Williams)

07/02/2026

06:19 AM Ruby Feature #21957: Introduce `Enumerable#close` to free internal resources.

@mame If we decide to use `Fiber#kill` for terminating the fiber (which seems reasonable), it seems bad that it may not return to the point that `rb_fiber_kill` was invoked. WDYT?
It would be the same issue as discussed in https://bugs.... ioquatix (Samuel Williams)

06:18 AM Ruby Feature #21957: Introduce `Enumerable#close` to free internal resources.

Possible implementation: https://github.com/ruby/ruby/pull/17624 ioquatix (Samuel Williams)

06:15 AM Ruby Misc #22107: DevMeeting-2026-07-09

- [Feature #22138] Add `RB_NOGVL_PENDING_INTERRUPT_FAIL` flag for `rb_nogvl`.
- Additive change to protect against race condition. Is it acceptable?
- [Bug #22133] Ruby's default SIGINT handling ignores `Thread.handle_interrupt` maskin... ioquatix (Samuel Williams)