Renderer source fixtures fail after stream and descriptor admission merges #1813
Labels
No labels
accessibility
autorelease: pending
autorelease: tagged
bug
dependencies
documentation
duplicate
enhancement
github_actions
good first issue
help wanted
invalid
question
testing wanted
wontfix
No milestone
No assignees
1 participant
Due date
No due date set.
Dependencies
No dependencies set
Reference
MagicBeanDev/Hookshot#1813
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Two game-free source fixtures no longer compile after the renderer admission changes in #1658 and #1754 merged. Reproduced on untouched main
52ac35cawith Windows LLVM:tests/test_draw_stream_visibility.py:use of undeclared identifier 'stream_insert'. The real stream allocation and recording reset now depend on that production helper, but the fixture does not extract it.tests/test_sampler_admission.py:use of undeclared identifier 'vkDestroyDescriptorPool'. The complete descriptor binder now references owning-pool rollback, and the fixture lacks that Vulkan environment entry point.The exact main full suite reports268 passed/7 failed/20 skipped; a separate candidate adds one passing check and has the same seven failures. This issue covers only these two renderer fixture integration failures. #1793 owns PREPARE and saved-key aliases; a long-path fixture finding is also reported there.
Proposed fix: restore the actual helper dependency and the modeled Vulkan environment without changing production behavior, suppressing warnings or weakening existing assertions. Validate all16 stream scenarios and108 sampler cases on Windows and Linux, plus sanitizers and exact-head CI. No game/runtime improvement is claimed for a test-only repair.
Claim: issai / Arkex-ai in #dev-work, before implementation. Priority: CI-first under the workflow; S3 x reach1.0 x confidence1.0 =1.0. I can test Windows/RTX5070Ti and Linux CPU fixtures. No native failure injection is involved.