-
Notifications
You must be signed in to change notification settings - Fork 308
Backport changes for XGBoost compatibility #6727
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Backport changes for XGBoost compatibility #6727
Conversation
|
I got these patches from @trxcllnt. @trxcllnt Can you answer @bernhardmgruber's question? I believe the answer is that the commits on |
🟩 CI finished in 2h 02m: Pass: 100%/157 | Total: 3d 19h | Avg: 35m 04s | Max: 1h 49m | Hits: 77%/173046
|
| Project | |
|---|---|
| CCCL Infrastructure | |
| CCCL Packaging | |
| libcu++ | |
| +/- | CUB |
| Thrust | |
| CUDA Experimental | |
| stdpar | |
| python | |
| CCCL C Parallel Library | |
| Catch2Helper |
Modifications in project or dependencies?
| Project | |
|---|---|
| CCCL Infrastructure | |
| +/- | CCCL Packaging |
| libcu++ | |
| +/- | CUB |
| +/- | Thrust |
| +/- | CUDA Experimental |
| +/- | stdpar |
| +/- | python |
| +/- | CCCL C Parallel Library |
| +/- | Catch2Helper |
🏃 Runner counts (total jobs: 157)
| # | Runner |
|---|---|
| 93 | linux-amd64-cpu16 |
| 17 | windows-amd64-cpu16 |
| 14 | linux-amd64-gpu-rtxa6000-latest-1 |
| 13 | linux-amd64-gpu-h100-latest-1 |
| 10 | linux-arm64-cpu16 |
| 7 | linux-amd64-gpu-rtx2080-latest-1 |
| 3 | linux-amd64-gpu-rtx4090-latest-1 |
…TrivialRuns` (NVIDIA#5252) * streaming non trivial runs * change global offset computation * fixes style * integrate latest bench and test changes * addresses review comments * replaces getters with member var
…A#5440) * adds vsmem to reduce_by_key * adds tests for vsmem * fixes rle, which does not support vsmem yet * addresses review comments
…ypes in `DeviceRunLengthEncode::Encode` (NVIDIA#6451) * adds tests for non default constructible iterators * fixes non default constructible iterators in rle * Simplify generation of `streaming_context` for run_length_encode * Reinstate regression test --------- Co-authored-by: Michael Schellenberger Costa <miscco@nvidia.com>
|
Answering the above question: the only difference is that the patches I got earlier did not include a few changes in |
bd34a2b to
1776c74
Compare
|
This is ready for review. I validated downstream that this fixes xgboost builds. I am still verifying that all of RAPIDS 25.12 builds and that cuDF passes tests with these changes, but I expect that to be done shortly. |
🟩 CI finished in 1h 30m: Pass: 100%/157 | Total: 1d 05h | Avg: 11m 22s | Max: 42m 16s | Hits: 99%/173046
|
| Project | |
|---|---|
| CCCL Infrastructure | |
| CCCL Packaging | |
| libcu++ | |
| +/- | CUB |
| Thrust | |
| CUDA Experimental | |
| stdpar | |
| python | |
| CCCL C Parallel Library | |
| Catch2Helper |
Modifications in project or dependencies?
| Project | |
|---|---|
| CCCL Infrastructure | |
| +/- | CCCL Packaging |
| libcu++ | |
| +/- | CUB |
| +/- | Thrust |
| +/- | CUDA Experimental |
| +/- | stdpar |
| +/- | python |
| +/- | CCCL C Parallel Library |
| +/- | Catch2Helper |
🏃 Runner counts (total jobs: 157)
| # | Runner |
|---|---|
| 93 | linux-amd64-cpu16 |
| 17 | windows-amd64-cpu16 |
| 14 | linux-amd64-gpu-rtxa6000-latest-1 |
| 13 | linux-amd64-gpu-h100-latest-1 |
| 10 | linux-arm64-cpu16 |
| 7 | linux-amd64-gpu-rtx2080-latest-1 |
| 3 | linux-amd64-gpu-rtx4090-latest-1 |
|
Thank you for changing the PR to use proper cherry-picks. May I ask why the last commit, f72b1c7, is necessary? This seems to not come from a commit on |
|
@bernhardmgruber The cherry-picked commits are missing a number of other changes on the |
Admin merging per approval of `build-eng` Updates CCCL to get backports needed for XGBoost compatibility with CCCL 3.1.x. Depends on NVIDIA/cccl#6727
Summary
Backporting changes from PRs #5252, #5440, and #6451. These are needed to compile xgboost with CCCL 3.1.x.
Test plan