2 Nov
2023
2 Nov
'23
12:02 a.m.
It's true that this is effectively unused now, but it's not clear to me that we have an equivalent barrier anywhere. I.e., does this being unused indicate a (potentially hard to debug) bug, instead of simply being some leftover code?
After struggling to understand tessellation again, I believe the answer is yes. I've pushed a new version that moves the check for vocp into the normalization pass and stores it in the vsir shader desc. Alternate approaches work, though, so let me know if you prefer a different way of detecting this case.