Editorial: Factor out AO RunSuspendedContext() by jmdyck · Pull Request #2664 · tc39/ecma262

jmdyck · 2022-02-13T20:18:38Z

This doesn't address the semantics of Suspend/Resume, just factors out a common chunk of code.

This PR is split into 6 commits where it's easier to confirm the correctness. They're probably fine to be squashed before landing.

michaelficarra · 2022-11-21T17:09:50Z

-            1. Push _asyncContext_ onto the execution context stack; _asyncContext_ is now the running execution context.
-            1. <emu-meta effects="user-code">Resume the suspended evaluation of _asyncContext_</emu-meta> using NormalCompletion(_value_) as the result of the operation that suspended it.
-            1. Assert: When we reach this step, _asyncContext_ has already been removed from the execution context stack and _prevContext_ is the currently running execution context.
+            1. Perform Completion(RunSuspendedContext(_asyncContext_, NormalCompletion(_value_))).


@jmdyck @bakkot Does ecmarkup complain if you remove these Completion(...)s? If so, we should allow them to be omitted when used with Perform since they're not actually entering the algorithm.

Yes it does, e.g.: error: RunSuspendedContext returns a Completion Record, but is not consumed as if it does (typecheck) at spec.html:46018:24

jmdyck · 2023-01-14T14:55:55Z

This PR's RunSuspendedContext (on the 'next' side) is the complementary operation to RunCallerContext (on the 'yield' side), which I proposed in PR #2665, but which got separated out of that PR by editor request. So I'm expecting that this PR will be rejected. If that's the case, hearing it sooner would mean less maintenance work for me.

nicolo-ribaudo · 2026-04-17T08:47:15Z

I know that a few years ago there was some opposition to this, but I think it's a good change. It's a somewhat complex set of steps, and a reader should be able to easily tell that in all those places the spec is doing exactly the same thing.

gibson042

I'm also in favor of this change, and the sequence of commits makes its editorial nature clear.

jmdyck · 2026-04-23T18:20:20Z

As requested at today's editor call: RunCallerContext, the complementary AO extracted from PR #2665, is at https://github.com/jmdyck/ecma262/tree/RunCallerContext. It might make sense to add those commits to this PR (rather than making another PR), since the two are so closely related.

To summarize:
RunSuspendedContext:

invoked from the next side,
pushes a context onto the execution context stack, and
"resumes" that context.

RunCallerContext:

invoked from the yield side,
pops a context from the execution context stack, and
"resumes" the new top of the execution context stack.

I'm not entirely happy with the names of the proposed AOs. They're mostly based on existing terminology, which has issues. In particular, RunSuspendedContext is called that because its cases say "resume the suspended evaluation of context ...", which the cases for RunCallerContext don't. But this is not great, because, in the next/yield distinction, either side is 'suspended' while the other side is running, so "SuspendedContext" could describe both sides. RunGenerativeContext might be better. Or RunCalleeContext, though the caller/callee distinction is maybe misleading (plus it's not great that they differ by only a single letter).

(Note that PR #2962 changes some of the terminology, including scrapping the idea of "suspending" and "resuming" execution contexts.)

michaelficarra · 2026-04-23T19:00:00Z

-            1. Push _asyncContext_ onto the execution context stack; _asyncContext_ is now the running execution context.
-            1. <emu-meta effects="user-code">Resume the suspended evaluation of _asyncContext_</emu-meta> using NormalCompletion(_v_) as the result of the operation that suspended it.
-            1. Assert: When we reach this step, _asyncContext_ has already been removed from the execution context stack and _prevContext_ is the currently running execution context.
+            1. Perform Completion(RunSuspendedContext(_asyncContext_, NormalCompletion(_v_))).


I think we need to give an explicit NOTE that this completion record is intentionally ignored. Same below.

Done. It's also ignored in AsyncGeneratorResume; would you like the same Note there?

In AsyncGeneratorResume, it's not ignored, it's asserted to be normal. So I think we only need the note here.

michaelficarra · 2026-04-23T19:01:04Z

@jmdyck I think both RunSuspendedContext and RunCallerContext are an improvement, and I'm looking forward to getting #2962 rebased on top of them.

edit: Though I also think we need to find new names for these operations.

github-actions · 2026-05-04T20:40:57Z

The rendered spec preview for this PR is available as a single page at https://tc39.es/ecma262/pr/2664 and as multiple pages at https://tc39.es/ecma262/pr/2664/multipage .

jmdyck · 2026-05-04T20:45:15Z

@jmdyck I think both RunSuspendedContext and RunCallerContext are an improvement, and I'm looking forward to getting #2962 rebased on top of them.

So would you like RunCallerContext in this PR or a separate one?

michaelficarra · 2026-05-04T21:04:43Z

@jmdyck Either is fine, but separate is probably better to make the review process simpler.

jmdyck · 2026-05-05T01:55:45Z

@jmdyck Either is fine, but separate is probably better to make the review process simpler.

Done: PR #3846.

linusg

Mild preference for s/Run/Resume/ here and in #3846 but overall this LGTM.

michaelficarra · 2026-05-05T16:33:50Z

@linusg But it's called the "running" execution context 😉. More seriously, #2962 will probably want to name it something about transferring control, so these are probably short-lived names.

jmdyck added the editorial change label Feb 13, 2022

jmdyck force-pushed the RunSuspendedContext branch from e3e0a00 to 2586002 Compare February 28, 2022 04:02

jmdyck force-pushed the RunSuspendedContext branch from 2586002 to 1bb271e Compare March 10, 2022 02:45

jmdyck force-pushed the RunSuspendedContext branch from 1bb271e to 4e953c2 Compare August 16, 2022 01:24

jmdyck mentioned this pull request Nov 9, 2022

Editorial: Eliminate weird Return notes #2665

Merged

jmdyck force-pushed the RunSuspendedContext branch from 4e953c2 to fe05833 Compare November 19, 2022 04:39

michaelficarra reviewed Nov 21, 2022

View reviewed changes

michaelficarra added the editor call to be discussed in the next editor call label Nov 21, 2022

michaelficarra removed the editor call to be discussed in the next editor call label Dec 7, 2022

jmdyck force-pushed the RunSuspendedContext branch 2 times, most recently from ed3013c to 4cfe666 Compare January 14, 2023 14:29

jmdyck force-pushed the RunSuspendedContext branch from 4cfe666 to a05d197 Compare March 1, 2023 04:18

jmdyck force-pushed the RunSuspendedContext branch from a05d197 to c6bc612 Compare May 23, 2024 14:13

jmdyck force-pushed the RunSuspendedContext branch from c6bc612 to 27a2df5 Compare April 7, 2025 18:57

jmdyck force-pushed the RunSuspendedContext branch from 27a2df5 to 98d3b24 Compare July 16, 2025 21:17

jmdyck force-pushed the RunSuspendedContext branch from 98d3b24 to 308349d Compare April 16, 2026 03:29

nicolo-ribaudo approved these changes Apr 17, 2026

View reviewed changes

nicolo-ribaudo added the editor call to be discussed in the next editor call label Apr 23, 2026

michaelficarra removed the editor call to be discussed in the next editor call label Apr 23, 2026

gibson042 approved these changes Apr 23, 2026

View reviewed changes

michaelficarra reviewed Apr 23, 2026

View reviewed changes

jmdyck force-pushed the RunSuspendedContext branch from 308349d to 3e5f088 Compare May 4, 2026 20:35

michaelficarra approved these changes May 4, 2026

View reviewed changes

linusg requested changes May 4, 2026

View reviewed changes

Comment thread spec.html Outdated

jmdyck mentioned this pull request May 5, 2026

Editorial: Factor out AO RunCallerContext() #3846

Open

linusg approved these changes May 5, 2026

View reviewed changes

michaelficarra added the ready to merge Editors believe this PR needs no further reviews, and is ready to land. label May 5, 2026

jmdyck force-pushed the RunSuspendedContext branch from c9d6aab to f36acf9 Compare May 5, 2026 23:47

ljharb pushed a commit to jmdyck/ecma262 that referenced this pull request May 6, 2026

Editorial: Factor out RunSuspendedContext abstract operation (tc39#2664)

875e880

ljharb force-pushed the RunSuspendedContext branch from f36acf9 to 875e880 Compare May 6, 2026 00:23

Editorial: Factor out RunSuspendedContext abstract operation (tc39#2664)

e07ec27

ljharb force-pushed the RunSuspendedContext branch from 875e880 to e07ec27 Compare May 6, 2026 00:24

ljharb merged commit e07ec27 into tc39:main May 6, 2026
10 checks passed

jmdyck deleted the RunSuspendedContext branch May 6, 2026 03:02

jmdyck mentioned this pull request May 8, 2026

Editorial: Various changes around execution contexts #2962

Open

Conversation

jmdyck commented Feb 13, 2022

Uh oh!

michaelficarra Nov 21, 2022

Choose a reason for hiding this comment

Uh oh!

jmdyck Jan 14, 2023

Choose a reason for hiding this comment

Uh oh!

jmdyck commented Jan 14, 2023

Uh oh!

nicolo-ribaudo commented Apr 17, 2026

Uh oh!

gibson042 left a comment

Choose a reason for hiding this comment

Uh oh!

jmdyck commented Apr 23, 2026

Uh oh!

michaelficarra Apr 23, 2026

Choose a reason for hiding this comment

Uh oh!

jmdyck May 4, 2026

Choose a reason for hiding this comment

Uh oh!

michaelficarra May 4, 2026

Choose a reason for hiding this comment

Uh oh!

michaelficarra commented Apr 23, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented May 4, 2026

Uh oh!

jmdyck commented May 4, 2026

Uh oh!

michaelficarra commented May 4, 2026

Uh oh!

Uh oh!

jmdyck commented May 5, 2026

Uh oh!

linusg left a comment

Choose a reason for hiding this comment

Uh oh!

michaelficarra commented May 5, 2026

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

6 participants

michaelficarra commented Apr 23, 2026 •

edited

Loading