Repository navigation
Reduce packed-world memory and release bootstrap resources - #19
Merged
Merged
Conversation
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Packed-world streaming reserved a 128 KiB source buffer and full-width ID tables regardless of the pack. Size staging once from the supported records and configured coalescing ceiling, compact private ID indexes, and defer ARAM metadata until its first successful allocation. The 649-record validation world saves 57,088 native staging bytes and 9,362 bytes in selected managed array payloads while its cold cache is unused (4,162 array-payload bytes once active). Managed figures exclude object field changes, headers and allocator rounding.
Scope the bootstrap file to construction and release partially created textures, meshes and asset-stream ownership on failure. Preserve the original WorldStreamer constructor and expose its actual staging size for teardown accounting. The audit documents API tradeoffs, SDK opportunities, reproducible payload reports and before/after CLR probes; managed heap reservations remain unchanged.
Validation: full host suite (56 Python tests, Debug/Release Core, ownership/cache checks and both native/fallback ordered-rendering paths) passed. Managed Release builds passed against SDK 318 and the SDK memory branch in capz/DolphinDotNet#171. A PPC native build with that SDK passed all eleven Dolphin streaming scenarios, including faults, pressure, ARAM, cancellation and teardown. Occlusion matched all 199,680 compared pixels exactly, and 140.8 seconds of DSP output had no silent windows. Source and published-package pins remain matched on SDK 318 until a new SDK release is available.