Ecma/TC57/2026/XY Ecma/GA/2026/XY

1 Welcome, opening and meeting logistics

Chris Bieneman, chair of TC57, began the teleconference and welcomed attendees. Attendees were reminded of the committee’s code of conduct, Ecma’s IPR policy, and TC57’s royalty-free patent policy extension. All attendees were informed that the meeting was recorded and transcribed by Zoom for the purpose of note-taking, and directed to review the video recording and legal disclaimer if they were not familiar.

2 Approval of the minutes of past meetings

There were no minutes to approve at this meeting.

3 Review of the agenda

The meeting’s agenda pull request was proposed and approved at the start of the meeting.

Editors’ report

• Roadmap View - https://github.com/orgs/hlsl-tc57/projects/1/
• Draft Status - #75

3.2 Review new proposals

3.2.1 New proposals for stage advancement today

• New Proposal - Strict Initializer Lists - #54
• New Proposal - unroll factor - #68

3.2.2 New proposals to be advanced on 07 July

• New Proposal - Named casts - #72
• New Proposal - groupshared arguments - #73
• New Proposal - const non-static member functions - #74
• New Proposal - non-member operator overloading - #115

3.3 Discussions

• Normative language changes
  • Basic.Scope - #63
  • Resources.TypedBuffer - #65
    • How do we want to write the “standard library clauses”?
      • Should we use C++ syntax for inexpressible things? Consider #65
      • Is it a goal to make HLSL able to express its own standard library? - #116
      • Should our size type be signed or unsigned? #67
• Overload Resolution - #62
  • Conversions in initializer lists
  • Question about conversion operators - https://hlsl.godbolt.org/z/d4zWo59vf - what should we allow?

4 Editors’ report

4.1 Roadmap review

There are several new proposals and normative changes coming up for the next meeting. Delegates can see what’s coming up clearly visualised on GitHub.

4.2 Draft tracking #75

The chair has created a meta-issue to track incomplete clauses and subclauses as carefully as possible, including any scenario where there is text that the committee knows is not fully developed, including and especially placeholder content.

There was also a conversation about how—especially in this push for a first edition of the specification—the committee acknowledges final text as final, or makes modifications on normative text that’s already been approved but is being influenced by new information.

The chair confirmed that any normative language that is merged into the draft must be ready to be in final standard, i.e. no scenarios where the author or editor suggests they will fix concerns later and merge regardless of opposition.

5 New proposals

5.1 New Proposal - Strict Initializer Lists - #54

The chair presented the Strict Initializer Lists proposal for Under Consideration. Gregory Roth (Nvidia) had left comments on the proposal and had some unresolved comments about the content. Chris Bieneman (Microsoft, Chair) believed they were concerns to be resolved during the Under Consideration stage. (Comments and responses are available at hlsl-tc57#54)

The intention for this proposal is to ultimately be added to the 202y version (or later) of the specification, but is being submitted for consideration now due to the amount of interest that has been displayed for it outside the committee and the number of future language features that would be dependent on it.

Greg will take responsibility for how to communicate the meta of the above information on the proposal template, and then the proposal will be merged as Under Consideration.

5.2 New Proposal - unroll factor - #68

The chair presented the loop unroll factor proposal for Under Consideration. The author (Farzon Lotfi, Microsoft) resolved all comments ahead of the meeting, and the merge was approved.

6 Discussions

6.1 Basic.Scope - #63

This pull request was approved by several delegates ahead of the meeting, and will be merged.

6.2 Should our size type be signed or unsigned? #67

Chris did some digging and concluded that DXC is inconsistent with its use of IntTy and UnsignedIntTy. He is of the opinion that HLSL should above all else be consistent, and presumably unsigned will be the most compatible choice.

The scenarios where this was a breaking change were limited, though Hai Nguyen (Nvidia) was concerned about introducing warnings which could break build systems.

The committee reached consensus on specifying on unsigned integers.

6.3 Is it a goal to make HLSL able to express its own standard library? - #116

The question was posed whether expressing the stdlib in HLSL syntax should be a long-term goal. It would require breaking changes and therefore is not necessarily a 202x concern, but if accepted as a design principle could ultimately provide more consistent behaviour between built-in types and user-defined types.

The committee reached consensus to add this to the design guidelines to be referenced as the language evolves.

6.4 Resources.TypedBuffer - #65

6.4.1 Should we use C++ syntax for inexpressible things? Consider #65

Greg left comments on this PR to the effect of expressing a belief that information is better expressed in a plain language. He suggested utilizing C++ syntax in an annex to cover the problem this PR is attempting to solve. Chris emphasises that the specification is not written for an audience of users, but rather implementers, and that the C++ syntax provides clarity. Alexander Johnston remarked that both could be useful, and if syntax that is helpful for users is to be added in the future, it’s easier to start it now. Tex Riddell suggested making codeblocks non-normative when providing them within the specification may provide a compromise.

The committee did not reach consensus to merge this PR, and will continue to discuss between meetings.

6.5 Overload Resolution - #62

The chair requests that delegates review this PR along with the new proposals listed in the agenda before the next meeting.

7 Any other business

Delegates are requested to provided explanatory text in PRs. Delegates are also requested to actively engage with PR reviews so comments may be resolved before the next meeting.

8 Next meetings

The next meetings will take place as Zoom video calls fortnightly. The next scheduled is on 07 July.

8.1 Past and Future Agendas

Agendas are viewable in GitHub and remain available in perpetuity.

9 Conclusion

The Chair wrapped up the meeting and the call concluded.