Skip to content

Snapshot of Kokkos and Kokkos-Kernels #2

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.

Sign up for GitHub

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

Draft
github-actions wants to merge 2 commits into
base: develop
Choose a base branch
from
Draft

Snapshot of Kokkos and Kokkos-Kernels #2

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
88d13dd
Snapshot of kokkos from commit 77f357717bc649a341ecb2aac0bb8698325d775c
Apr 19, 2026
a5e4da5
Snapshot of kokkos-kernels from commit 50e836eeaf6cfaad85f1decb6103d1…
Apr 19, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
The diff you're trying to view is too large. We only load the first 3000 changed files.
193 changes: 192 additions & 1 deletion packages/kokkos-kernels/CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,12 +1,203 @@
# Change Log

## [5.1.0](https://github.com/kokkos/kokkos-kernels/tree/5.1.0)
[Full Changelog](https://github.com/kokkos/kokkos-kernels/compare/5.0.2...5.1.0)

### New Features

#### BLAS updates
- Lapack geqrf [\#2858](https://github.com/kokkos/kokkos-kernels/pull/2858)

#### Batched updates
- Implement batched serial syr2 [\#2938](https://github.com/kokkos/kokkos-kernels/pull/2938)
- support batched axpy of 1D [\#2840](https://github.com/kokkos/kokkos-kernels/pull/2840)

#### Sparse updates
- Sell spmv [\#2911](https://github.com/kokkos/kokkos-kernels/pull/2911)
- Sparse - SellMatrix: adding SELL sparse matrix format [\#2864](https://github.com/kokkos/kokkos-kernels/pull/2864)

#### Misc updates
- Add shell sort option for matrix and graph sorting [\#2936](https://github.com/kokkos/kokkos-kernels/pull/2936)

### Enhancements

#### BLAS

#### Batched
- BatchedDblBufGemm: ensure vector_len ≤ vector_length_max [\#2967](https://github.com/kokkos/kokkos-kernels/pull/2967)
- BatchedDblBufGemm: make sure vector_len estimate ≥ 1 [\#2959](https://github.com/kokkos/kokkos-kernels/pull/2959)
- Conj trans support for batched Team/TeamVector Copy [\#2935](https://github.com/kokkos/kokkos-kernels/pull/2935)
- Refactor batched copy [\#2910](https://github.com/kokkos/kokkos-kernels/pull/2910)

#### Sparse
- Avoid calling RCB in each kk_extract_diagonal_blocks_crsmatrix_with_rcb_sequential [\#2963](https://github.com/kokkos/kokkos-kernels/pull/2963)
- SortCRS: pick between shell/radix automatically; speed up radix [\#2949](https://github.com/kokkos/kokkos-kernels/pull/2949)
- spiluk: Limit memory usage for iw work buffer based on input matrix size [\#2753](https://github.com/kokkos/kokkos-kernels/pull/2753)

#### Common utilities
- Transpose changes [\#2904](https://github.com/kokkos/kokkos-kernels/pull/2904)

#### TPL support
- Lapack: adding Apple Accelerate support [\#2940](https://github.com/kokkos/kokkos-kernels/pull/2940)

### Build System
- Fix FindTPLCUBLAS.cmake and friends [\#2956](https://github.com/kokkos/kokkos-kernels/pull/2956)
- Enforce capitalization of `KokkosKernels_` CMake options [\#2961](https://github.com/kokkos/kokkos-kernels/pull/2961)
- Build-system: adding support for Apple Accelerate library [\#2707](https://github.com/kokkos/kokkos-kernels/pull/2707)
- Format cmake to lowercase [\#2874](https://github.com/kokkos/kokkos-kernels/pull/2874)

### Documentation and Testing
- Add spiluk example from docs to build [\#2992](https://github.com/kokkos/kokkos-kernels/pull/2992)
- docs: flesh out component configuration options [\#2965](https://github.com/kokkos/kokkos-kernels/pull/2965)
- [Docs] batched Serial syr2 [\#2948](https://github.com/kokkos/kokkos-kernels/pull/2948)
- [Docs] batched Serial/Team/TeamVector copy [\#2937](https://github.com/kokkos/kokkos-kernels/pull/2937)
- docs: adding wiki example for axpy and including it in docs [\#2934](https://github.com/kokkos/kokkos-kernels/pull/2934)
- Update KUG banner submit → register [\#2941](https://github.com/kokkos/kokkos-kernels/pull/2941)
- Docs req update [\#2916](https://github.com/kokkos/kokkos-kernels/pull/2916)
- Enable support for Google Analytics to track traffic on the doc [\#2909](https://github.com/kokkos/kokkos-kernels/pull/2909)
- [Docs] batched Serial/Team/TeamVector axpy [\#2905](https://github.com/kokkos/kokkos-kernels/pull/2905)
- docs: update urllib3 requirement to build the docs [\#2903](https://github.com/kokkos/kokkos-kernels/pull/2903)
- [Docs] Add batched serial axpy example for documentation [\#2901](https://github.com/kokkos/kokkos-kernels/pull/2901)
- [Docs] batched Serial/Team/TeamVector trsm [\#2900](https://github.com/kokkos/kokkos-kernels/pull/2900)
- Blas1::set docs [\#2599](https://github.com/kokkos/kokkos-kernels/pull/2599)
- Adding badge on README [\#2890](https://github.com/kokkos/kokkos-kernels/pull/2890)
- docs: large update of requirements and dropping cppkokkos [\#2852](https://github.com/kokkos/kokkos-kernels/pull/2852)
- [Docs] Add batched serial trsm example for documentation [\#2841](https://github.com/kokkos/kokkos-kernels/pull/2841)

### Benchmarks
- Benchmark - Spmv: fixing a small bug in input parsing [\#2969](https://github.com/kokkos/kokkos-kernels/pull/2969)
- Upgrade sort_crs perftest [\#2947](https://github.com/kokkos/kokkos-kernels/pull/2947)
- PerfTests: removing raja perf tests [\#2902](https://github.com/kokkos/kokkos-kernels/pull/2902)
- Benchmark: cleaning up some compiler warnings [\#2898](https://github.com/kokkos/kokkos-kernels/pull/2898)
- More config options for par_ilut benchmark [\#2859](https://github.com/kokkos/kokkos-kernels/pull/2859)

### Cleanup
- Sparse - SpMV: removing cusparse < 11500 path [\#2888](https://github.com/kokkos/kokkos-kernels/pull/2888)
- Common - ExecUtils: cleaning a preprocessor guard for x86_64 architecture [\#2996](https://github.com/kokkos/kokkos-kernels/pull/2996)

### Deprecations:
- Deprecate old kk_extract_diagonal_blocks_crsmatrix_with_rcb_sequential interface [\#2980](https://github.com/kokkos/kokkos-kernels/pull/2980)
- Deprecate KOKKOSKERNELS_ENABLE_HOST_ONLY [\#2971](https://github.com/kokkos/kokkos-kernels/pull/2971)
- Deprecate `Level2::Blocked::mb` and `Level3::Blocked::mb` [\#2950](https://github.com/kokkos/kokkos-kernels/pull/2950)
- Refactor KK debug guards [\#2917](https://github.com/kokkos/kokkos-kernels/pull/2917)

### Bug Fixes
- Add missing popRegion() [\#2863](https://github.com/kokkos/kokkos-kernels/pull/2863)
- Add some missing SPMV labels [\#2952](https://github.com/kokkos/kokkos-kernels/pull/2952)
- Workaround issues with legacy View in Trilinos [\#2856](https://github.com/kokkos/kokkos-kernels/pull/2856)
- swap: add missing include [\#2984](https://github.com/kokkos/kokkos-kernels/pull/2984)
- Batched GEMM compatibility fixes [\#2829](https://github.com/kokkos/kokkos-kernels/pull/2829)
- Fix batched getrs [\#2981](https://github.com/kokkos/kokkos-kernels/pull/2981)

## [5.0.2](https://github.com/kokkos/kokkos-kernels/tree/5.0.2)
[Full Changelog](https://github.com/kokkos/kokkos-kernels/compare/5.0.1...5.0.2)

### Deprecations:
- Deprecate unused `KOKKOSKERNELS_ENABLE_HOST_ONLY` macro

### Bug and Warning Fixes:
- Cleanup warning and warning system [\#2861](https://github.com/kokkos/kokkos-kernels/pull/2861)
- Batched - Vector: marking defaulted method properly [\#2908](https://github.com/kokkos/kokkos-kernels/pull/2908)
- Fix template argument in sort_and_merge_graph [\#2918](https://github.com/kokkos/kokkos-kernels/pull/2918)

### CI and Workflow Updates:
- CI - gcc/1050: adding new action with c++20 [\#2844](https://github.com/kokkos/kokkos-kernels/pull/2844)
- Update workflows for cxx20 and kokkos 5.0 [\#2747](https://github.com/kokkos/kokkos-kernels/pull/2747), [\#2862](https://github.com/kokkos/kokkos-kernels/pull/2862), [\#2906](https://github.com/kokkos/kokkos-kernels/pull/2906)
- Release workflow: update versioning scheme [\#2857](https://github.com/kokkos/kokkos-kernels/pull/2857)

## [5.0.1](https://github.com/kokkos/kokkos-kernels/tree/5.0.1)
[Full Changelog](https://github.com/kokkos/kokkos-kernels/compare/5.0.0...5.0.1)

### Bug Fixes:
- SPILUK symbolic and numeric update, fix overflow issue [\#2887](https://github.com/kokkos/kokkos-kernels/pull/2887)
- Fix SYCL issues with MKL on Aurora [\#2865](https://github.com/kokkos/kokkos-kernels/pull/2865)

## [5.0.0](https://github.com/kokkos/kokkos-kernels/tree/5.0.0)
[Full Changelog](https://github.com/kokkos/kokkos-kernels/compare/4.7.01...5.0.0)

### Enhancements and updates:
- Improve performance of par_ilut and improve benchmark [\#2846](https://github.com/kokkos/kokkos-kernels/pull/2846)
- Prefer bit manipulation functions from Kokkos Core [\#2786](https://github.com/kokkos/kokkos-kernels/pull/2786)
- Common - ArithTraits: moving from Kokkos to KokkosKernels [\#2771](https://github.com/kokkos/kokkos-kernels/pull/2771)
- Update headers in preparation for the 5.0 release [\#2795](https://github.com/kokkos/kokkos-kernels/pull/2795)
- Add a member function to BsrMatrix to convert to Crs [\#2809](https://github.com/kokkos/kokkos-kernels/pull/2809)

### Bug Fixes:
- Batched - QR: fixes for complex scalars [\#2590](https://github.com/kokkos/kokkos-kernels/pull/2590)
- Kokkos Batched InnerTrsm: Allocate in inner scope to avoid warnings with Sacado types [\#2810](https://github.com/kokkos/kokkos-kernels/pull/2810)
- Define Fortran interface macros for LAPACK [\#2802](https://github.com/kokkos/kokkos-kernels/pull/2802)
- Fixed blas includes [\#2776](https://github.com/kokkos/kokkos-kernels/pull/2776)
- Minor fix for coefficient type in SPMV_Functor [\#2730](https://github.com/kokkos/kokkos-kernels/pull/2730)
- Drop deprecated double4 type for CUDA 13 [\#2718](https://github.com/kokkos/kokkos-kernels/pull/2718)

### Deprecations:
- Require `KOKKOS_ENABLE_DEPRECATED_CODE_5` to be defined for including `<Kokkos_ArithTraits.hpp>` [\#2798](https://github.com/kokkos/kokkos-kernels/pull/2798)

### Cleanup:
- Batched: Remove deprecated macros [\#2836](https://github.com/kokkos/kokkos-kernels/pull/2836)
- Sparse, Graph: remove deprecated declarations [\#2835](https://github.com/kokkos/kokkos-kernels/pull/2835)
- Avoid opening the Kokkos namespace [\#2800](https://github.com/kokkos/kokkos-kernels/pull/2800)
- Avoid using `Kokkos::reduction_identity` [\#2751](https://github.com/kokkos/kokkos-kernels/pull/2751)
- Rename more reserved identifiers [\#2727](https://github.com/kokkos/kokkos-kernels/pull/2727)
- variety of reserved indentifier cleanups [\#2720](https://github.com/kokkos/kokkos-kernels/pull/2720)
- Rename type HostMirror -> host_mirror_type [\#2713](https://github.com/kokkos/kokkos-kernels/pull/2713)
- Documentation: small clean-ups [\#2749](https://github.com/kokkos/kokkos-kernels/pull/2749)

### Documentation and Testing:
- [Docs] Adding the graph coloring publication [\#2834](https://github.com/kokkos/kokkos-kernels/pull/2834)
- [Docs] Add batched serial trsv documentation [\#2830](https://github.com/kokkos/kokkos-kernels/pull/2830)
- [Docs] batched serial laswp [\#2828](https://github.com/kokkos/kokkos-kernels/pull/2828)
- [Docs] batched serial lacgv [\#2825](https://github.com/kokkos/kokkos-kernels/pull/2825)
- [Docs] batched serial iamax [\#2804](https://github.com/kokkos/kokkos-kernels/pull/2804)
- [Docs] Add batched serial trsv example for documentation [\#2803](https://github.com/kokkos/kokkos-kernels/pull/2803)
- [Docs] Add batched serial lacgv example for documentation [\#2801](https://github.com/kokkos/kokkos-kernels/pull/2801)
- [Docs] Add batched serial laswp example for documentation [\#2797](https://github.com/kokkos/kokkos-kernels/pull/2797)
- [Docs] batched serial tbsv [\#2796](https://github.com/kokkos/kokkos-kernels/pull/2796)
- [Docs] Add serial iamax example for documentation [\#2789](https://github.com/kokkos/kokkos-kernels/pull/2789)
- [Docs] Add serial tbsv example for documentation [\#2788](https://github.com/kokkos/kokkos-kernels/pull/2788)
- [Docs] batched serial syr [\#2787](https://github.com/kokkos/kokkos-kernels/pull/2787)
- [Docs] Add serial syr example for documentation [\#2774](https://github.com/kokkos/kokkos-kernels/pull/2774)
- [Docs] batched serial ger [\#2773](https://github.com/kokkos/kokkos-kernels/pull/2773)
- [Docs] Add batched serial gbtrf/gbtrs [\#2767](https://github.com/kokkos/kokkos-kernels/pull/2767)
- [Docs] Add batched serial getrf/getrs [\#2765](https://github.com/kokkos/kokkos-kernels/pull/2765)
- [Docs] Add serial ger example for documentation [\#2756](https://github.com/kokkos/kokkos-kernels/pull/2756)
- [Docs] List available batched dense functions in a table [\#2750](https://github.com/kokkos/kokkos-kernels/pull/2750)
- [Docs] batched serial pbtrf and pbtrs [\#2748](https://github.com/kokkos/kokkos-kernels/pull/2748)
- [Docs] batched serial pttrf and pttrs [\#2678](https://github.com/kokkos/kokkos-kernels/pull/2678)
- CI: Adding benchmark build and run on Aurora [\#2824](https://github.com/kokkos/kokkos-kernels/pull/2824)
- CI: Adding ci yaml file for Polaris [\#2812](https://github.com/kokkos/kokkos-kernels/pull/2812)
- CI: Add V100 CUDA 12.6.2 build with TPLs [\#2738](https://github.com/kokkos/kokkos-kernels/pull/2738)
- CI: adding new build to test previous minor release [\#2737](https://github.com/kokkos/kokkos-kernels/pull/2737)

## [4.7.02](https://github.com/kokkos/kokkos-kernels/tree/4.7.02)
[Full Changelog](https://github.com/kokkos/kokkos-kernels/compare/4.7.01...4.7.02)

### Bug Fixes:
- Fix SYCL issues with MKL on Aurora [\#2865](https://github.com/kokkos/kokkos-kernels/pull/2865)

## [4.7.01](https://github.com/kokkos/kokkos-kernels/tree/4.7.01)
[Full Changelog](https://github.com/kokkos/kokkos-kernels/compare/4.7.00...4.7.01)

### New Features and Enhancements
- First implementation of recursive coordinate bisection (RCB) in graph [\#2708](https://github.com/kokkos/kokkos-kernels/pull/2708)
- Update the bisect break condition in RCB [\#2766](https://github.com/kokkos/kokkos-kernels/pull/2766)
- Add setNumRows, setNumCols to sparse matrix structures [\#2700](https://github.com/kokkos/kokkos-kernels/pull/2700)
- Add optional argument to configure sorting algorithm used in KokkosSparse:sort_crs_matrix [\#2714](https://github.com/kokkos/kokkos-kernels/pull/2714)

### Bug Fixes:
- Batched - SVD: adding iteration limits [\#2706](https://github.com/kokkos/kokkos-kernels/pull/2706)
- Minor fix for coefficient type in SPMV_Functor [\#2730](https://github.com/kokkos/kokkos-kernels/pull/2730)
- cusparse spmv_mv: use native fallback if y not 16B aligned [\#2746](https://github.com/kokkos/kokkos-kernels/pull/2746)
- Fix a CMake error when benchmarks and perf-tests enabled [\#2729](https://github.com/kokkos/kokkos-kernels/pull/2729)
- Clean up old SortCrs workaround, update RCB for host_mirror_type renaming [\#2721](https://github.com/kokkos/kokkos-kernels/pull/2721)
- Drop deprecated double4 type for CUDA 13 [\#2718](https://github.com/kokkos/kokkos-kernels/pull/2718)

## [4.7.00](https://github.com/kokkos/kokkos-kernels/tree/4.7.00)
[Full Changelog](https://github.com/kokkos/kokkos-kernels/compare/4.6.02...4.7.00)

### New Features

#### Batched updates
- Implement batched serial gbtrf [\#2489](https://github.com/kokkos/kokkos-kernels/pull/2489)
- Implement batched serial gbtrs [\#2539](https://github.com/kokkos/kokkos-kernels/pull/2539)

### Enhancements:
Expand Down
Loading