Add MetaHeader parity test addressing D108793458 review comment (#5935)#5935
Closed
lizhe-ji wants to merge 1 commit into
Closed
Add MetaHeader parity test addressing D108793458 review comment (#5935)#5935lizhe-ji wants to merge 1 commit into
lizhe-ji wants to merge 1 commit into
Conversation
meta-codesync
Bot
changed the title
lizhe-ji
pushed a commit
to lizhe-ji/FBGEMM-1
that referenced
this pull request
Jun 19, 2026
…rch#5935) Summary: This diff addresses the review comment in D108793458 inline 1189-1193 about hardcoded magic number 16 in values[j].size() >= 16 and sizeof(int64_t) offset baking in MetaHeader layout assumptions. Adds SSDTableBatchedEmbeddingsTest.MetaHeaderParityWithDRAM to verify SSD hard-coded 16-byte header parsing produces identical output to DRAM FixedBlockPool::get_metaheader_raw for same MetaHeader bytes. Keeps production ssd_table_batched_embeddings.h unchanged per follow-up discussion — hard-coded 16 / 8 remains with comment, test is source of truth for cross-backend parity. Updates test BUCK dep to include dram_kv_embedding_inference for FixedBlockPool header access. No production behavior change. Differential Revision: D109084819
lizhe-ji
force-pushed
the
export-D109084819
branch
from
246c305 to
b3d9466
Compare
lizhe-ji
pushed a commit
to lizhe-ji/FBGEMM-1
that referenced
this pull request
Jun 19, 2026
…rch#5935) Summary: X-link: facebookresearch/FBGEMM#2853 This diff addresses the review comment in D108793458 inline 1189-1193 about hardcoded magic number 16 in values[j].size() >= 16 and sizeof(int64_t) offset baking in MetaHeader layout assumptions. Adds SSDTableBatchedEmbeddingsTest.MetaHeaderParityWithDRAM to verify SSD hard-coded 16-byte header parsing produces identical output to DRAM FixedBlockPool::get_metaheader_raw for same MetaHeader bytes. Keeps production ssd_table_batched_embeddings.h unchanged per follow-up discussion — hard-coded 16 / 8 remains with comment, test is source of truth for cross-backend parity. Updates test BUCK dep to include dram_kv_embedding_inference for FixedBlockPool header access. No production behavior change. Differential Revision: D109084819
lizhe-ji
force-pushed
the
export-D109084819
branch
2 times, most recently
from
5f3cdc6 to
a4ee9d4
Compare
lizhe-ji
pushed a commit
to lizhe-ji/FBGEMM-1
that referenced
this pull request
Jun 19, 2026
…rch#5935) Summary: X-link: facebookresearch/FBGEMM#2853 This diff addresses the review comment in D108793458 inline 1189-1193 about hardcoded magic number 16 in values[j].size() >= 16 and sizeof(int64_t) offset baking in MetaHeader layout assumptions. Adds SSDTableBatchedEmbeddingsTest.MetaHeaderParityWithDRAM to verify SSD hard-coded 16-byte header parsing produces identical output to DRAM FixedBlockPool::get_metaheader_raw for same MetaHeader bytes. Keeps production ssd_table_batched_embeddings.h unchanged per follow-up discussion — hard-coded 16 / 8 remains with comment, test is source of truth for cross-backend parity. Updates test BUCK dep to include dram_kv_embedding_inference for FixedBlockPool header access. No production behavior change. Differential Revision: D109084819
lizhe-ji
pushed a commit
to lizhe-ji/FBGEMM-1
that referenced
this pull request
Jun 19, 2026
…rch#5935) Summary: X-link: facebookresearch/FBGEMM#2853 This diff addresses the review comment in D108793458 inline 1189-1193 about hardcoded magic number 16 in values[j].size() >= 16 and sizeof(int64_t) offset baking in MetaHeader layout assumptions. Adds SSDTableBatchedEmbeddingsTest.MetaHeaderParityWithDRAM to verify SSD hard-coded 16-byte header parsing produces identical output to DRAM FixedBlockPool::get_metaheader_raw for same MetaHeader bytes. Keeps production ssd_table_batched_embeddings.h unchanged per follow-up discussion — hard-coded 16 / 8 remains with comment, test is source of truth for cross-backend parity. Updates test BUCK dep to include dram_kv_embedding_inference for FixedBlockPool header access. No production behavior change. Reviewed By: vinares Differential Revision: D109084819
lizhe-ji
force-pushed
the
export-D109084819
branch
from
a4ee9d4 to
68b13a6
Compare
…rch#5935) Summary: X-link: facebookresearch/FBGEMM#2853 This diff addresses the review comment in D108793458 inline 1189-1193 about hardcoded magic number 16 in values[j].size() >= 16 and sizeof(int64_t) offset baking in MetaHeader layout assumptions. Adds SSDTableBatchedEmbeddingsTest.MetaHeaderParityWithDRAM to verify SSD hard-coded 16-byte header parsing produces identical output to DRAM FixedBlockPool::get_metaheader_raw for same MetaHeader bytes. Keeps production ssd_table_batched_embeddings.h unchanged per follow-up discussion — hard-coded 16 / 8 remains with comment, test is source of truth for cross-backend parity. Updates test BUCK dep to include dram_kv_embedding_inference for FixedBlockPool header access. No production behavior change. Reviewed By: vinares Differential Revision: D109084819
lizhe-ji
force-pushed
the
export-D109084819
branch
from
68b13a6 to
534e555
Compare
Contributor
|
This pull request has been merged in 17dd8e3. |
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.
Summary:
X-link: https://github.com/facebookresearch/FBGEMM/pull/2853
This diff addresses the review comment in D108793458 inline 1189-1193
about hardcoded magic number 16 in values[j].size() >= 16 and
sizeof(int64_t) offset baking in MetaHeader layout assumptions.
Adds SSDTableBatchedEmbeddingsTest.MetaHeaderParityWithDRAM to verify
SSD hard-coded 16-byte header parsing produces identical output to
DRAM FixedBlockPool::get_metaheader_raw for same MetaHeader bytes.
Keeps production ssd_table_batched_embeddings.h unchanged per follow-up
discussion — hard-coded 16 / 8 remains with comment, test is source of
truth for cross-backend parity.
Updates test BUCK dep to include dram_kv_embedding_inference for
FixedBlockPool header access. No production behavior change.
Reviewed By: vinares
Differential Revision: D109084819