Repository navigation
Conversation
build.bat gives each worktree its own _MSPDBSRV_ENDPOINT_ and kills its own mspdbsrv.exe before building. build_quiet.py --keep-going calls cmakew directly and syntax_check.py calls cl.exe directly, so neither inherited that, and a server already wedged before the build started survived build.bat kill and still failed with C1090. common.build_env derives the same endpoint name; common.kill_pdb_server stops only this worktree instance; build_quiet.py retries the build once after a C1090/C1033. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NaSkbxngnm28vFqxp3wNZs (cherry picked from commit ac343b4)
The cherry-picked fix recovers from a wedged mspdbsrv.exe by stopping this worktree''s instance and building again. C1090 has a second cause that recovery cannot reach: the endpoint name itself can wedge, outliving every mspdbsrv.exe that uses it, so retrying into the same name fails forever. Six consecutive builds failed that way here. Killing the server did not help, nor did deleting the PCH and both PDBs; the endpoint was correctly unique, not inherited from the environment, and the toolchain binaries matched a worktree that was building fine at that moment. The same tree then built first try under a different endpoint name, which is what identified the cause. build_quiet.py now takes a second retry on a fresh endpoint, and build_env() accepts an explicit one so a caller can pin it for a session. Co-Authored-By: Claude Opus 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01A4PvfebpUzaMXy93EutBjJ (cherry picked from commit 2826353)
…State # Conflicts: # src/OpenSHC/Rendering/ViewportRenderState/computeTileXOffset.cpp # src/OpenSHC/Rendering/ViewportRenderState/createFloatingTextElement.cpp # src/OpenSHC/Rendering/ViewportRenderState/focusOnCoordinate.cpp # src/OpenSHC/Rendering/ViewportRenderState/resetupViewportThunk.cpp # src/OpenSHC/Rendering/ViewportRenderState/restoreFocusTile.cpp # src/OpenSHC/Rendering/ViewportRenderState/translateXYToTile.cpp # src/OpenSHC/Rendering/ViewportRenderState/xyAreValid.cpp # tools/reimplementation-helper/batch/README.md # tools/reimplementation-helper/batch/build_quiet.py # tools/reimplementation-helper/batch/syntax_check.py
…urrent main 135 commits behind main and uncompilable against its regenerated headers, so this branch had never been measured. All 32 sources now build, link and score, with none excluded. The merge conflicted on seven ViewportRenderState sources that main has since gained independently and on three batch tools; all took main's side. 452 stale generated field references were rewritten over two passes by the project's own refactor_struct_fields.py. That tool builds its solution list eagerly, so a single field name with no offset encoded in it aborts every other fix in the same log; feeding it only the C2039 lines whose names carry an offset lets it resolve the rest, and the leftovers get handled by hand: - ownerFlagFrame -> flagSlot.overlayImage. The field at 0x80 is now a BuildingFlagSlot, which is a 4-byte struct wrapping one int, so the member access is the same four bytes. - shouldRenderRoof -> extraOverlayImage2 (0x60), shouldRenderSomeOverlay -> extraOverlayImage3 (0x64), both confirmed against this branch's own pre-merge Building.hpp. - tileRef2 -> tileRefs[1]: 0x1CC is one int into tileRefs[36] at 0x1C8. - renderMap's MiscDisplayLayer walker was ushort*, but the header declares that layer short[]. It only bit-tests and widens through that pointer, so short* changes no behaviour. - updateBuildingPreviewPosition built an all-zero TextMessageBLLookupStructUnion to pass to setBottomLeftTextDisplayText, whose param_4 is now int; that is 0. First measurement, reimplemented data and all 32 resolvers enabled: BUILD_OK, avg 78.33 over 32, 17 functions below 100. No match work is attempted here - none was possible until it built. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_013Fzi11pcBU1Y7Y4oanrZQL
gynt
force-pushed
the
reimpl/ViewportRenderState
branch
from
October 23, 2026 09:56
182c3fc to
14128b9
Compare
…4.2% Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NVYWDtHk2xzvq9hszJRS9r
…llaps Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01NVYWDtHk2xzvq9hszJRS9r
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.