On Thu Feb 1 10:48:17 2024 +0000, Giovanni Mascellani wrote:
Doesn't that mean that as soon as we'll want to return the data stored there we'll have a problem with the vkd3d-shader interface? Maybe we should design the vkd3d-shader interface so that it already allows returning the data we're skipping here?
The idea is that it'd go through a different vkd3d-shader interface, in this case specifically a (slightly extended) vkd3d_shader_scan_descriptor_info1. We already have vkd3d_shader_scan_signature_info which covers a different part of ID3D12ShaderReflection. But that's also one of the topics that I had proposed for discussion, maybe we do want everything in the same scan struct after all, or maybe we just want all the RDEF information together, or I don't know...