Theresa Foley

Thanks for posting this as its own issue. I can probably try to tackle this, but I'm wondering what kind of timeline you have on when you'd need the implementation...

I'm starting to work on this and have made some positive steps toward getting things fixed in my branch. When this stuff lands the names and signatures are likely to...

PR #1894 should fix up a lot of the compiler's behavior around these combined types, and those fixes should be available in the `v0.19.0` and later releases. However, the change...

There's really a few issues here (and we might need to split this so we can track them individually). A Dynamic-Dispatch Lowering Bug? ======================= @csyonghe I think the most important...

The overhead of `spCompile` comes from multiple steps in the Slang compiler stack, which includes generating GLSL and then invoking the Khronos glslang compiler to produce SPIR-V (and in recent...

The second error message you are seeing is because Slang doesn't understand the `[unroll(n)]` attribute; we will need to fix that. The first error message is coming from fxc, because...

Okay, as a workaround for now, you could consider using a Slang-specific feature for "compile-time loops" that can guarantee unrolling. For example, here is some code that uses the `Sample()`...

I was just about to come here and comment that a `ParameterBlock` where `X` contains only resources won't generate a constant buffer, and Falcor probably doesn't account for that. Putting...

This looks like a Slang issue, yes. The Slang front-end is failing to look for (and therefore preserve) system-value semantics on the patch-constant function.

@nbenty It is possible that we could issue a hotfix to the version of Slang that is currently used by Falcor 3.2. I have a good sense of what is...