Repository navigation
Handle duplicate allocation labels during replay - #53
Merged
Merged
Conversation
Match input allocations and reference outputs by captured allocation pointer and memory space. Expose allocation descriptors for unambiguous comparisons, including flat views whose extent is inferred from the recorded byte count. Reject ambiguous label-only lookups and cover duplicate labels, reordered references, empty allocations, and comparison size validation.
Signed-off-by: daxmawal <jeanfrancoismanutea@gmail.com>
daxmawal
commented
Sep 12, 2026
Signed-off-by: daxmawal <jeanfrancoismanutea@gmail.com>
Signed-off-by: daxmawal <jeanfrancoismanutea@gmail.com>
daxmawal
marked this pull request as ready for review
September 14, 2026 13:49
daxmawal
marked this pull request as draft
September 24, 2026 08:11
daxmawal
marked this pull request as ready for review
September 24, 2026 09:47
Signed-off-by: daxmawal <jeanfrancoismanutea@gmail.com>
tretre91
self-requested a review
September 24, 2026 16:18
Member
Author
|
@tretre91 I'm wondering which interface we should prioritize for comparing replay results. This PR fixes handling of allocations that share a label and rejects ambiguous label-based comparisons. Another option would be Should I keep the allocation identification by address and memory space in this PR, drop my changes to label-based |
Member
|
I will try to look at it but I don't think I will finish my review before monday, we'll be able to discuss the changes in person by then |
tretre91
requested changes
Sep 30, 2026
daxmawal
commented
Oct 3, 2026
tretre91
reviewed
Oct 5, 2026
Co-authored-by: Trévis Morvany <63788850+tretre91@users.noreply.github.com>
Signed-off-by: daxmawal <jeanfrancoismanutea@gmail.com>
Member
Author
|
Merged, @tretre91. Thanks a lot ! I'll work on the nested lambda, and I'll come back with another PR to fix the other things we discussed in this PR. |
This was referenced Oct 6, 2026
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
2 participants
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.
Kokkos allocations can share the same label. Previously, replay allocations and reference outputs were indexed by label, so one allocation could hide another and make output comparison unreliable. This change preserves each allocation separately and matches inputs with their reference outputs using the captured allocation pointer and memory space. The new
get_allocationsAPI exposes these pairs throughReplayAllocationdescriptors. Newcompare_viewsoverloads accept a descriptor and support explicit dimensions or a flat view whose size is inferred from the recorded byte count.Follow-up PR will add a
compare_views(replayed_view, comparator)overload that uses the replayed view’s data pointer and memory space to locate the corresponding reference.P.S. I encountered this issue in a Dyablo Kokkos kernel where multiple views shared the same label.