Skip to content

Add LSTM/GRU work-area size helper APIs#77

Open
k8ika0s wants to merge 1 commit into
IBM:mainfrom
k8ika0s:qe/issue-08-work-area-size-apis
Open

Add LSTM/GRU work-area size helper APIs#77
k8ika0s wants to merge 1 commit into
IBM:mainfrom
k8ika0s:qe/issue-08-work-area-size-apis

Conversation

@k8ika0s
Copy link
Copy Markdown

@k8ika0s k8ika0s commented Apr 15, 2026

Fixes #47

Summary

  • Add LSTM/GRU work-area size helper APIs.

Why

Changes

  • Implemented on branch qe/issue-08-work-area-size-apis.
  • Includes code and tests scoped to this issue.

Validation

  • s390x integrated battery pass recorded in artifacts/final-validation-20260212T173158Z.
  • Targeted regressions for this scope were validated during branch prep.

Notes

  • DCO signoffs are present on branch commits.

What: Provide helper APIs to compute required work_area bytes for LSTM/GRU.

Why: Deterministic sizing enables caller preallocation and avoids runtime malloc/free overhead in hot inference loops.

Expected impact: Lower allocation overhead and more stable latency; additive opt-in APIs.

Tests: add coverage for sizing helpers and invalid inputs (tests/testDriver_work_area_size_apis.c).
Signed-off-by: Kaitlyn Davis <k8ika0s@gmail.com>
Signed-off-by: Kaitlyn Davis <kaitlyn.davis@ibm.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[Feature Request] Add LSTM/GRU Work-Area Size Helper APIs (Caller Preallocation)

1 participant