Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
6,867 Commits 32 Branches 59 Tags
2d7ffbdf3dfeac6490f3068e7ac207e1fc3aa818
Commit Graph
89 Commits
Author SHA1 Message Date
Hui Liu 2d7ffbdf3d merge latest changes 2023-12-13 21:06:28 -08:00
Jiaming Yuan 0715ab3c10 Use dlopen to load NCCL. (#9796) 
This PR adds optional support for loading nccl with `dlopen` as an alternative of compile time linking. This is to address the size bloat issue with the PyPI binary release.
- Add CMake option to load `nccl` at runtime.
- Add an NCCL stub.

After this, `nccl` will be fetched from PyPI when using pip to install XGBoost, either by a user or by `pyproject.toml`. Others who want to link the nccl at compile time can continue to do so without any change.

At the moment, this is Linux only since we only support MNMG on Linux.
 2023-11-22 19:27:31 +08:00
Hui Liu d7f1235b7d Merge branch 'master' into sync-condition-2023Oct11 2023-10-30 13:19:33 -07:00
James Lamb 0c621094b3 [CI] enforce cmakelint checks (#9728) 2023-10-28 05:38:04 +08:00
Hui Liu f9f39b092b add HIP LIB PATH 2023-10-23 16:52:33 -07:00
Hui Liu 6ba66463b6 fix uuid and Clear/SetValid 2023-10-23 16:32:26 -07:00
Hui Liu 55994b1ac7 enable ROCm on latest XGBoost 2023-10-23 11:15:04 -07:00
Hui Liu 15421e40d9 enable ROCm on latest XGBoost 2023-10-23 11:07:08 -07:00
Chuck Atkins 83cdf14b2c CMake LTO and CUDA arch (#9677) 2023-10-20 13:01:37 +08:00
James Lamb eb562d3829 [CI] address cmakelint warnings about whitespace (#9674) 2023-10-14 12:46:07 +08:00
Your Name ea19555474 temp merge, disable 1 line, SetValid 2023-10-12 16:16:44 -07:00
James Lamb 2e42f33fc1 [CI] standardize else() and enfunction() calls in CMake scripts (#9653) 2023-10-12 11:14:19 +08:00
James Lamb db8d117f7e [CI] standardize endif() calls in CMake scripts (#9637) 2023-10-08 11:45:20 +08:00
Jiaming Yuan 90ef250ea1 [rabit] Drop support for MPI backend. (#9525) 
- Add checks in cmake.
- Remove mpi related code.
 2023-08-28 21:01:22 +08:00
Jiaming Yuan f380c10a93 Use hint for find nccl. (#9490) 2023-08-16 16:08:41 +08:00
Philip Hyunsu Cho 5bd163aa25 Explicitly specify libcudart_static in CMake config (#9436) 2023-08-05 14:15:44 -07:00
Rong Ou 7579905e18 Retry switching to per-thread default stream (#9416) 2023-07-26 07:09:12 +08:00
Jiaming Yuan 3a9996173e Revert "Switch to per-thread default stream (#9396)" (#9413) 
This reverts commit f7f673b00c.
 2023-07-24 12:03:28 -07:00
Rong Ou f7f673b00c Switch to per-thread default stream (#9396) 2023-07-20 08:21:00 +08:00
Jiaming Yuan 7a0ccfbb49 Add compute 90. (#9397) 2023-07-19 13:42:38 +08:00
Your Name 42867a4805 sync Jun 1 2023-06-01 15:55:06 -07:00
Jiaming Yuan 097f11b6e0 Support CUDA f16 without transformation. (#9207) 
- Support f16 from cupy.
- Include CUDA header explicitly.
- Cleanup cmake nvtx support.
 2023-05-30 20:54:31 +08:00
amdsc21 e961016e71 rm HIPCUB 2023-03-10 22:21:37 +01:00
amdsc21 f0b8c02f15 merge latest changes 2023-03-10 22:10:20 +01:00
Jiaming Yuan c5c8f643f2 Remove the cub submodule. (#8888) 
XGBoost now uses CTK-11.8 for binary packages, there's no need to maintain a cub
submodule anymore.
 2023-03-09 19:43:02 -08:00
amdsc21 f0febfbcac finish gpu_predictor.cu 2023-03-10 01:29:54 +01:00
amdsc21 53b5cd73f2 add hip flags 2023-03-08 03:42:51 +01:00
amdsc21 52b05d934e add hip 2023-03-08 03:32:19 +01:00
Jiaming Yuan f7ce0ec0df Upgrade gcc toolchain to 9.x. (#8878) 
* Use new tool chain.

* Use gcc-9.

* Use cmake from system.

* DOn't link leak.
 2023-03-07 08:25:23 -08:00
Jiaming Yuan 173096a6a7 Discover libasan.so.6. (#8864) 2023-03-06 18:56:54 +08:00
Jiaming Yuan 4d665b3fb0 Restore clang tidy test. (#8861) 2023-03-03 13:47:04 -08:00
Philip Hyunsu Cho 6d8afb2218 [CI] Require C++17 + CMake 3.18; Use CUDA 11.8 in CI (#8853) 
* Update to C++17

* Turn off unity build

* Update CMake to 3.18

* Use MSVC 2022 + CUDA 11.8

* Re-create stack for worker images

* Allocate more disk space for Windows

* Tempiorarily disable clang-tidy

* RAPIDS now requires Python 3.10+

* Unpin cuda-python

* Use latest NCCL

* Use Ubuntu 20.04 in RMM image

* Mark failing mgpu test as xfail
 2023-03-01 09:22:24 -08:00
Jiaming Yuan 284dcf8d22 Add script for change version. (#8443) 
- Replace jvm regex replacement script with mvn command.
- Replace cmake script for python version with python script.
- Automate rest of the manual steps.

The script can handle dev branch, rc release, and formal release version.
 2022-11-24 00:06:39 +08:00
Jiaming Yuan bc818316f2 Prepare for improving Windows networking compatibility. (#8234) 
* Prepare for improving Windows networking compatibility.

* Include dmlc filesystem indirectly as dmlc/filesystem.h includes windows.h, which
  conflicts with winsock2.h
* Define `NOMINMAX` conditionally.
* Link the winsock library when mysys32 is used.
* Add config file for read the doc.
 2022-09-10 15:16:49 +08:00
Rory Mitchell f421c26d35 Tune cuda architectures (#8152) 2022-08-11 13:36:47 -07:00
Jiaming Yuan 142a208a90 Fix compiler warnings. (#8022) 
- Remove/fix unused parameters
- Remove deprecated code in rabit.
- Update dmlc-core.
 2022-06-22 21:29:10 +08:00
Jiaming Yuan 1a33b50a0d Fix compiler warnings. (#7974) 
- Remove unused parameters. There are still many warnings that are not yet
addressed. Currently, the warnings in dmlc-core dominate the error log.
- Remove `distributed` parameter from metric.
- Fixes some warnings about signed comparison.
 2022-06-06 22:56:25 +08:00
Jiaming Yuanandjakirkham d48123d23b Fix rmm build (#7973) 
- Optionally switch to c++17
- Use rmm CMake target.
- Workaround compiler errors.
- Fix GPUMetric inheritance.
- Run death tests even if it's built with RMM support.

Co-authored-by: jakirkham <jakirkham@gmail.com>
 2022-06-06 20:18:32 +08:00
Jiaming YuanandPhilip Hyunsu Cho fd78af404b Drop support for deprecated CUDA architectures. (#7774) 
Co-authored-by: Philip Hyunsu Cho <chohyu01@cs.washington.edu>
 2022-03-31 21:42:23 +08:00
AJ Schmidtandptaylor 511805c981 Compress fatbins (#7601) 
* compress CUDA device code

Co-authored-by: ptaylor <paul.e.taylor@me.com>
 2022-01-25 18:30:59 +08:00
Philip Hyunsu Cho 2a0368b7ca Add CMake option to use /MD runtime (#7277) 2021-09-30 13:13:57 +08:00
Jiaming Yuan c311a8c1d8 Enable compiling with system cub. (#7232) 
- Tested with all CUDA 11.x.
- Workaround cub scan by using discard iterator in AUC.
- Limit the size of Argsort when compiled with CUDA cub.
 2021-09-17 14:28:18 +08:00
Jiaming Yuan cdfaa705f3 Fix building on CUDA 11.0. (#7187) 2021-08-25 02:57:53 -07:00
Jiaming Yuan 9c64618cb6 [breaking] Remove CUDA sm_35, add sm_86 (#7182) 2021-08-25 16:04:23 +08:00
Jiaming Yuan 345796825f Optional find dependency in installed cmake config. (#7099) 
* Find dependency only when xgboost is built as static library.
* Resolve msvc warning.
* Add test for linking shared library.
 2021-07-11 17:20:55 +08:00
Jiaming Yuan ab6fd304c4 [Python] Change development release postfix to dev (#6988) 2021-05-27 16:06:51 +08:00
James LambandHyunsu Cho 894e9bc5d4 [R-package] remove dependency on {magrittr} (#6928) 
Co-authored-by: Hyunsu Cho <chohyu01@cs.washington.edu>
 2021-05-13 04:34:59 +08:00
James Lamb 20f34d9776 [R-package] Update dependencies from CMake-based installation (#6906) 
* remove stringi
* add Matrix and jsonlite
 2021-04-29 01:32:01 +08:00
Philip Hyunsu Cho 0c85b90671 [R] Fix R package installation via CMake (#6423) 2020-11-22 05:49:09 -08:00
Rory Mitchell 29745c6df2 Fix inclusive scan for large sizes (#6234) 2020-11-03 17:01:43 +13:00