Kelsey Gilbert

WGSL meeting minutes 2022-07-26 * DN: Part of larger “what is infinity”? Looked at ieee794 and …. No comments. Call for review, what should be done here? Seems silly to...

I think it's useful to preserve out-of-memory whenever we can tell that that's a reasonable possibility.

GPU Web 2022-09-07/08 APAC-timed * KN: **haven't had time to get back to this one**. Can discuss a little to get people up to speed. * KG: we can table...

The core difficulty I've generally identified for bindless is validation (particularly in the shader) without throwing away the flexibility and/or perf advantage. I think we'd need to see some compelling...

WGSL meeting minutes 2021-06-08 * MM: Motivation: It would be bad if one browser had different choices of what to reject. * DN: Yes, we just need to pick one...

At some point, didn't we have a proposal for mat.x? E.g. `mat3x3(mat4x4.x.xyz, mat4x4.y.xyz, mat4x4.z.xyz)`

WGSL meeting minutes 2021-09-07 * DN: AB strongly wants us to have consistent behavior for OOB access here, so more discussion is needed here. * DM: The method for OOB...

WGSL meeting minutes 2021-09-14 * AB: The rules are consistent now. There is tension if you want to change them: if you want constexpr to come in now, and if...

I think there might be consensus here that making this a hard error might not be required for mvp/v1, but I want to check with the group during the meeting.

WGSL meeting minutes 2021-09-21 * JG: Suggest not making this a hard error, but can make it hard error later * MM: Not my recollection. My recollection is what Robin...