Repository navigation
Conversation
Preserve exact geometry and fixed dynamic shared-memory requirements through artifact loading so preparation can select and validate the compiled allocation. Keep artifact writers and readers on one in-tree implementation while preserving legacy bundles without launch facts. Signed-off-by: Trent Nelson <trent@trent.me>
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Prepared launches currently know only source-declared shared-memory sizes. This adds compiler-produced facts for each concrete kernel, so
dynamic_shared = inferredandLaunchConfig2D::auto_shared(...)select the allocation from the loaded artifact and reject incompatible block shapes or insufficient explicit sizes.Fixed
DynamicSharedArray::get_fixedviews describe one aliasing base region; their extents and alignments combine by maximum. Artifact v3 carries those facts through PTX/cubin loading, including distinct generic PTX entries. Bundles without facts keep their v1/v2 encoding. Unsafe device code remains responsible for initialization, bounds, aliasing, synchronization, and declaring any externally accessed dynamic region.Validation: artifact compatibility and malformed-record checks; host runtime, macro, backend, Cargo tooling and PTX tooling tests; clippy/rustdoc; native helper layouts, exact 2D geometry, automatic/explicit allocations, known-zero scratch and owned async completion; memcheck for PTX and materialized cases. Generic cubin loading remains unsupported.
This source draft keeps writers and readers on the same in-tree
oxide-artifactsimplementation. The 246 existing lockfile edits change only that source identity; one new fixture lock is separate. Package version/release coordination is still required before publication.