Changes in matseqdensecupm.hpp

901f93825bbaa60582af604c6700caf57884a2e1 - Merge branch 'release'

Fri, 02 Jan 2026 00:10:15 +0000

Merge branch 'release' List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

55e133c907ce828ecb365bf747914890c489d889 - Merge branch 'jolivet/minor-docs' into 'release'

Mon, 22 Dec 2025 17:52:03 +0000

Merge branch 'jolivet/minor-docs' into 'release'Minor documentation fixesSee merge request petsc/petsc!8908 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

758d473275857be73ec985adb2a285ab97907225 - Minor documentation fixes

Mon, 22 Dec 2025 06:54:19 +0000

Minor documentation fixes List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

cd871708d6ae82bd70cc1a9e2138f9b57839fe75 - Merge branch 'cuda' into 'main'

Tue, 16 Sep 2025 04:23:51 +0000

Merge branch 'cuda' into 'main'Fix CUDA 13 API incompatibilitiesSee merge request petsc/petsc!8632 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

b547d71c1c518418f5bd567d3db9f86a4dbb9f85 - Fix CUDA 13 API incompatibilities

Tue, 05 Aug 2025 21:28:28 +0000

Fix CUDA 13 API incompatibilitiesCo-authored-by: Satish Balay List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

58bddbc0aeb8e2276be3739270a4176cb222ba3a - Merge branch 'tisaac/feature-lmvm-rework' into 'main'

Fri, 11 Apr 2025 04:14:39 +0000

Merge branch 'tisaac/feature-lmvm-rework' into 'main'MatLMVM reworkSee merge request petsc/petsc!6401 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

d016bdde269de9549a736fe23cc3868ea52c341b - Mat: Fix and improve the performance of dense matrix multiplication

Wed, 26 Mar 2025 16:45:16 +0000

Mat: Fix and improve the performance of dense matrix multiplicationMat: Add MATDENSEFROMVECTYPE constructor typeNow in a tests set you can do``` testset: args: -mat_type densefromvectype test: test_cuda requires: cuda args: -vec_type veccuda test: test_hip requires: hip args: -vec_type vechip```(This assumes that you call `MatSetVecType()` before you call`MatSetFromOptions()`)Mat_MPIDense: Cache offsets of MatDenseGetSubMatrix() to avoid communication in more casesMat: Add missing implementations for internal "MatMultColumnRange()" interfaceMat_MPIDense: Fix the zeroing of buffers in multiplication routinesMat_MPIDense: Add optimization of MatMatMult routines when all columns are owned by rank 0The communication for intermediate buffers can be handled with allreduce / bcast operations,but we use the PetscSF matvec context instead of MPI routines directly so that we willuse gpu-aware MPI if possible. List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

3853def2c757d2edd67eca368879a1c0942aa1f7 - CUPM: Fix some mult routines and make some small performance improvements

Wed, 26 Mar 2025 16:19:55 +0000

CUPM: Fix some mult routines and make some small performance improvementsVecSeq_CUPM: Fix ::Dot, ::TDot, ::WAXPYAsync, and ::AXPBYAsync to work with non-device vectorsMat_SeqDenseCUPM: fix ::SetRandom for hipMat_SeqDenseCUPM: implement conjugate on the deviceMat_SeqDenseCUPM: Fix MatMult() (and friends) when the vector is not on the deviceThis implementation assumes that the cost of moving the matrix to thehost is more expensive that moving the vector to the device, so temporarydevice copies of the host vectors are used.VecCUPM: Avoid device synchronization in some cases of ResetArray()The documentation says that the PetscDeviceSynchronization() isonly needed if CopyToDevice_() resulted in a HtoD memcpy,which would only happen if v->offloadmask == PETSC_OFFLOAD_CPU.If we test this condition, we can avoid the synchronization.This improves the peformance of MatDenseRestoreColumnVecWrite() inperformance critical loops.CUPM: use thrust::hip::par_nosync List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

d0e6bf2ad94dcc89b258ce16c7987200a4714786 - Merge branch 'jolivet/clang-format-19' into 'main'

Wed, 25 Sep 2024 06:44:18 +0000

Merge branch 'jolivet/clang-format-19' into 'main'clang-format-19See merge request petsc/petsc!7644 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

57508ece14a6b1339c0bbf016ecd72f673a062b0 - Remove unneeded parentheses

Tue, 17 Sep 2024 23:49:15 +0000

Remove unneeded parentheses List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

bb59972c36a4dbc3a3dca65232097048bda6fd19 - Merge branch 'jed/fix-cupm-visibility' into 'main'

Sun, 05 May 2024 14:28:01 +0000

Merge branch 'jed/fix-cupm-visibility' into 'main'cupm: fix visibility to build without warningsSee merge request petsc/petsc!7526 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

85f25e71687eef93ec659366612fd9d9ee335aac - cupm: fix visibility to build without warnings

Sat, 04 May 2024 14:24:10 +0000

cupm: fix visibility to build without warningsThanks-to: Lawrence Mitchell List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

8c77480b6f59861b6f79a6892fa2bb8bff302065 - Merge remote-tracking branch 'origin/release'

Thu, 04 Apr 2024 21:36:10 +0000

Merge remote-tracking branch 'origin/release' List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

74577c92ea46b005e87f1ee1e3a49c11c6f63a12 - Merge branch 'jose/release/cupm-int-mult' into 'release'

Tue, 02 Apr 2024 12:43:32 +0000

Merge branch 'jose/release/cupm-int-mult' into 'release'CUPM: do not forbid allocations larger than PetscInt sizeSee merge request petsc/petsc!7427 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

161307759e0498636771c3d62e98c118d017b0d3 - CUPM: do not forbid allocations larger than PetscInt size

Tue, 02 Apr 2024 06:59:20 +0000

CUPM: do not forbid allocations larger than PetscInt sizeThis fixes an error triggered in SLEPc when creating a BV with largesizes, i.e., a dense Mat where the product of rows and columns overflowsPetscInt, for instance 134217728 x 17Reported-by: DenverCoder9 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

5fe01c21d727d5e61b58091b225b37a3f96d9fc7 - Merge branch 'hsuh/dense-lmvm-squashed' into 'main'

Fri, 22 Mar 2024 14:25:57 +0000

Merge branch 'hsuh/dense-lmvm-squashed' into 'main'Dense LMVMSee merge request petsc/petsc!7371 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

0be0d8bde7cbed9986facbcd3689d3188b7894db - MatMult-ColumnRanges support added

Thu, 14 Mar 2024 22:44:06 +0000

MatMult-ColumnRanges support added List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

3c4b05c55102fd6faf01d2bfa02f0ea34b12f7db - Merge branch 'blanca/mat-mult-hermitian-transpose' into 'main'

Wed, 10 Jan 2024 16:33:16 +0000

Merge branch 'blanca/mat-mult-hermitian-transpose' into 'main'Support for MatMultHermitianTranspose with MATDENSE, MATNEST and MATSCALAPACKSee merge request petsc/petsc!7171 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

9557186961ced456c90f2422459de4a750b9b85b - Added MatMultHermitianTranspose and MatMultHermitianTransposeAdd for MATDENSECUPM

Fri, 05 Jan 2024 11:58:45 +0000

Added MatMultHermitianTranspose and MatMultHermitianTransposeAdd for MATDENSECUPM List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp

9dd11ecf0918283bb567d8b33a92f53ac4ea7840 - Merge branch 'jacobf/2023-08-17/header-guard-check' into 'main'

Fri, 25 Aug 2023 13:49:03 +0000

Merge branch 'jacobf/2023-08-17/header-guard-check' into 'main'Check header guardsSee merge request petsc/petsc!6822 List of files: /petsc/src/mat/impls/dense/seq/cupm/matseqdensecupm.hpp