Skip to content

Forward-merge main into pandas3 - #22533

Merged
AyodeAwe merged 7 commits into
pandas3from
main
May 16, 2026
Merged

Forward-merge main into pandas3#22533
AyodeAwe merged 7 commits into
pandas3from
main

Conversation

@AyodeAwe

Copy link
Copy Markdown
Contributor

Forward-merge triggered by automated cron job to keep pandas3 up-to-date with main.

If this PR has conflicts, it will remain open for manual resolution.

See forward-merger docs for more info.

Matt711 and others added 7 commits May 15, 2026 19:41
Need to make sure `insert_finished` is called for collective operations.

Authors:
  - Matthew Murray (https://github.com/Matt711)

Approvers:
  - Matthew Roeschke (https://github.com/mroeschke)
  - Mads R. B. Kristensen (https://github.com/madsbk)
  - Lawrence Mitchell (https://github.com/wence-)

URL: #22516
Forward-merge release/26.06 into main
Follow up #22319 

Use `cudf::tables_equal` API in libcudf examples instead of anti join to check if two tables are equal

Authors:
  - Muhammad Haseeb (https://github.com/mhaseeb123)
  - Lawrence Mitchell (https://github.com/wence-)
  - Vukasin Milovanovic (https://github.com/vuule)

Approvers:
  - Shruti Shivakumar (https://github.com/shrshi)
  - David Wendt (https://github.com/davidwendt)

URL: #22428
This PR checks for potential malformed headers in Parquet delta decoder and propagates it in decoders.

Authors:
  - Muhammad Haseeb (https://github.com/mhaseeb123)

Approvers:
  - Paul Mattione (https://github.com/pmattione-nvidia)
  - Bradley Dice (https://github.com/bdice)
  - Nghia Truong (https://github.com/ttnghia)

URL: #22275
Forward-merge release/26.06 into main
…22279)

Contributes to #13995

This PR enables the Parquet writer to assign lower dictionary indices to elements appearing earlier in each column chunk thereby writing variable (reduced) number of dictionary key bits for earlier pages.

New algorithm:
1. `map_insert_fn` now inserts this pair in the static map: `{row_idx, frag_idx}` instead of `{row_idx, row_idx}`. `frag_idx` is just `blockIdx.x` and indicates which page fragment actually inserted this entry in the static map (CAS race dependent but earlier thread blocks generally win this race)
2. `map_insert_fn` also writes the `PageFragment::num_dict_vals` field for each page fragment to keep a record of number of unique values inserted by it.
3. Then in `collect_map_entries_kernel` (where we assign keys to dict values), we optionally assign spatially-local keys to dictionary values inserted by the same page fragment. See algorithm details [here](#22279 (comment))
4. Finally, we launch `compute_page_dict_bits_kernel` to compute the max bit width required to encode dict keys for each parquet page. See algorithm [here](#22279 (comment))

See follow up PR #22323 that deterministically writes the index of the first page fragment that sees each dictionary key

Authors:
  - Muhammad Haseeb (https://github.com/mhaseeb123)

Approvers:
  - Vukasin Milovanovic (https://github.com/vuule)
  - Paul Mattione (https://github.com/pmattione-nvidia)
  - Nghia Truong (https://github.com/ttnghia)

URL: #22279
@AyodeAwe
AyodeAwe requested review from a team as code owners May 16, 2026 06:00
@AyodeAwe
AyodeAwe requested review from brandon-b-miller, mhaseeb123 and vyasr and removed request for a team May 16, 2026 06:01
@AyodeAwe
AyodeAwe merged commit cb7520c into pandas3 May 16, 2026
46 of 47 checks passed
@AyodeAwe

Copy link
Copy Markdown
Contributor Author

SUCCESS - forward-merge complete (automated).

@github-actions github-actions Bot added libcudf Affects libcudf (C++/CUDA) code. Python Affects Python cuDF API. CMake CMake build issue cudf-polars Issues specific to cudf-polars pylibcudf Issues specific to the pylibcudf package labels May 16, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CMake CMake build issue cudf-polars Issues specific to cudf-polars libcudf Affects libcudf (C++/CUDA) code. pylibcudf Issues specific to the pylibcudf package Python Affects Python cuDF API.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants