Cherry-pick openxla/xla#44389: [ROCm] Enable float and buffer checker#989
Closed
magaonka-amd wants to merge 1 commit into
Closed
Cherry-pick openxla/xla#44389: [ROCm] Enable float and buffer checker#989magaonka-amd wants to merge 1 commit into
magaonka-amd wants to merge 1 commit into
Conversation
(cherry picked from commit 9b7baa6)
This was referenced Jun 24, 2026
Author
|
Superseded by #993, which combines all four ROCm 0.10.2 cherry-pick PRs into a single PR against rocm-jaxlib-v0.10.2. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Motivation
Requested in #985 review: bring openxla#44389 onto
rocm-jaxlib-v0.10.2. The upstream PR is approved but not yet copybara-merged, so this is cherry-picked from the PR head commit9b7baa6dd5(ROCm:draganm/float_checker) rather than a main commit.Commit (cherry-picked with
-x)9b7baa6dd5— [ROCm] Enable float and buffer checkerFaithfulness note
Cherry-picked onto the pinned base
5a9e73cb. Result matches the upstream PR exactly (17 files, +963/-613). The only manual resolution wasxla/stream_executor/rocm/BUILD: the 3-way merge dragged in unrelated newer-main targets (rocm_memory_reservation,rocm_vmm_allocator,rocm_device_address_vmm_allocator) that are not part of this PR and whose sources don't exist on this base. Those were discarded; only PR openxla#44389's two hunks were applied (the two:buffer_debug_*_kernel_rocmdeps + the two newrocm_librarytargets). All referenced targets (gpu_kernel_registry,gpu:buffer_debug_*_kernel,runtime:buffer_debug_log_structs) resolve; no duplicate targets.Files changed (highlights)
xla/stream_executor/gpu/buffer_debug_{float_check,xor_checksum}_kernel_lib.cu.hxla/stream_executor/rocm/buffer_debug_{float_check,xor_checksum}_kernel_rocm.cu.ccgpu/libs; ROCm BUILD +rocm_executor.ccwired up.Test Plan
rocm-jaxlib-v0.10.2; release-validation CI.