Add `device_matrix_data` and device-side `matrix::read` #886

upsj · 2021-09-13T14:32:10Z

This PR adds a new device_matrix_data that works equivalently to matrix_data except that it uses a device-side Array. I provide a default implementation using the old host-side initialization.

There are still some things to do, but I wanted to ask for comments already.

merge Simple kernel reduction #833 to implement Ell compute_max_nnz
add Ell::read
add Hybrid::read
add Sellp::read
~~modify Ell/Sellp sentinel values~~ Format conversion, SpMV and write improvements #905
~~modify conversions for numerical zeros~~ This will be taken care of with Format conversion, SpMV and write improvements #905
~~move read tests from core to reference~~ (this will no longer be necessary with Always enable GKO_BUILD_REFERENCE for tests #889)
~~add tests checking zero behavior~~ This will be taken care of with Add generic matrix format test #904
add tests for remove_zeros and device_matrix_data functionality

Fixes #729

yhmtsai

LGTM. only some nit/comments

common/cuda_hip/matrix/fbcsr_kernels.hpp.inc

yhmtsai · 2021-11-18T07:24:31Z

common/cuda_hip/matrix/hybrid_kernels.hpp.inc

+
+
+template <typename ValueType, typename IndexType>
+struct hybrid_tuple_unpack_functor {


Because it always extract the second element of the tuple, maybe use extract_device_entry for name or make the 1 as template?

This is only necessary because a Lambda wouldn't be copy-assignable, which one of the algorithms requires. So this is only a local workaround without any generic usage.

yhmtsai · 2021-11-18T07:52:53Z

core/matrix/hybrid.cpp

- }
+ auto exec = this->get_executor();
+ auto local_data = make_temporary_clone(exec, &data.nonzeros);
+ Array<int64> row_ptrs{exec, data.size[0] + 1};


note. if the row index is larger than IndexType and the entry is stored to coo, coo still use IndextType to store it.

That is true, though I would separate between row pointers (indexing nonzeros, why I'm using int64 here) and row indices (indexing rows) here. The former are more likely to overflow than the latter.

omp/matrix/fbcsr_kernels.cpp

omp/matrix/hybrid_kernels.cpp

reference/matrix/fbcsr_kernels.cpp

omp/matrix/fbcsr_kernels.cpp

This should never have been possible in the first place

* rename from_matrix_data to fill_in_matrix_data * explicitly specify fake_complex alignment * improve ELL and SELL-P fill_in_matrix_data implementation * fix Fbcsr fill_in_matrix_data implementation * simplify kernel declaration Co-authored-by: Thomas Grützmacher <[email protected]> Co-authored-by: Yuhsiang Tsai <[email protected]>

* use ref split_matrix_data for OpenMP * fix fbcsr from_matrix_data * fix gdb pretty printers for const Array refs Co-authored-by: Yuhsiang Tsai <[email protected]>

ginkgo-bot · 2021-11-20T08:42:04Z

Note: This PR changes the Ginkgo ABI:

Functions changes summary: 2720 Removed, 1843 Changed (11694 filtered out), 4805 Added functions
Variables changes summary: 0 Removed, 0 Changed, 0 Added variable

For details check the full ABI diff under Artifacts here

ginkgo-bot · 2021-11-20T08:53:44Z

Note: This PR changes the Ginkgo ABI:

Functions changes summary: 2720 Removed, 1843 Changed (11694 filtered out), 4805 Added functions
Variables changes summary: 0 Removed, 0 Changed, 0 Added variable

For details check the full ABI diff under Artifacts here

sonarcloud · 2021-11-20T20:29:36Z

Kudos, SonarCloud Quality Gate passed!

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
No Duplication information

Advertise release 1.5.0 and last changes + Add changelog, + Update third party libraries + A small fix to a CMake file See PR: #1195 The Ginkgo team is proud to announce the new Ginkgo minor release 1.5.0. This release brings many important new features such as: - MPI-based multi-node support for all matrix formats and most solvers; - full DPC++/SYCL support, - functionality and interface for GPU-resident sparse direct solvers, - an interface for wrapping solvers with scaling and reordering applied, - a new algebraic Multigrid solver/preconditioner, - improved mixed-precision support, - support for device matrix assembly, and much more. If you face an issue, please first check our [known issues page](https://github.com/ginkgo-project/ginkgo/wiki/Known-Issues) and the [open issues list](https://github.com/ginkgo-project/ginkgo/issues) and if you do not find a solution, feel free to [open a new issue](https://github.com/ginkgo-project/ginkgo/issues/new/choose) or ask a question using the [github discussions](https://github.com/ginkgo-project/ginkgo/discussions). Supported systems and requirements: + For all platforms, CMake 3.13+ + C++14 compliant compiler + Linux and macOS + GCC: 5.5+ + clang: 3.9+ + Intel compiler: 2018+ + Apple LLVM: 8.0+ + NVHPC: 22.7+ + Cray Compiler: 14.0.1+ + CUDA module: CUDA 9.2+ or NVHPC 22.7+ + HIP module: ROCm 4.0+ + DPC++ module: Intel OneAPI 2021.3 with oneMKL and oneDPL. Set the CXX compiler to `dpcpp`. + Windows + MinGW and Cygwin: GCC 5.5+ + Microsoft Visual Studio: VS 2019 + CUDA module: CUDA 9.2+, Microsoft Visual Studio + OpenMP module: MinGW or Cygwin. Algorithm and important feature additions: + Add MPI-based multi-node for all matrix formats and solvers (except GMRES and IDR). ([#676](#676), [#908](#908), [#909](#909), [#932](#932), [#951](#951), [#961](#961), [#971](#971), [#976](#976), [#985](#985), [#1007](#1007), [#1030](#1030), [#1054](#1054), [#1100](#1100), [#1148](#1148)) + Porting the remaining algorithms (preconditioners like ISAI, Jacobi, Multigrid, ParILU(T) and ParIC(T)) to DPC++/SYCL, update to SYCL 2020, and improve support and performance ([#896](#896), [#924](#924), [#928](#928), [#929](#929), [#933](#933), [#943](#943), [#960](#960), [#1057](#1057), [#1110](#1110), [#1142](#1142)) + Add a Sparse Direct interface supporting GPU-resident numerical LU factorization, symbolic Cholesky factorization, improved triangular solvers, and more ([#957](#957), [#1058](#1058), [#1072](#1072), [#1082](#1082)) + Add a ScaleReordered interface that can wrap solvers and automatically apply reorderings and scalings ([#1059](#1059)) + Add a Multigrid solver and improve the aggregation based PGM coarsening scheme ([#542](#542), [#913](#913), [#980](#980), [#982](#982), [#986](#986)) + Add infrastructure for unified, lambda-based, backend agnostic, kernels and utilize it for some simple kernels ([#833](#833), [#910](#910), [#926](#926)) + Merge different CUDA, HIP, DPC++ and OpenMP tests under a common interface ([#904](#904), [#973](#973), [#1044](#1044), [#1117](#1117)) + Add a device_matrix_data type for device-side matrix assembly ([#886](#886), [#963](#963), [#965](#965)) + Add support for mixed real/complex BLAS operations ([#864](#864)) + Add a FFT LinOp for all but DPC++/SYCL ([#701](#701)) + Add FBCSR support for NVIDIA and AMD GPUs and CPUs with OpenMP ([#775](#775)) + Add CSR scaling ([#848](#848)) + Add array::const_view and equivalent to create constant matrices from non-const data ([#890](#890)) + Add a RowGatherer LinOp supporting mixed precision to gather dense matrix rows ([#901](#901)) + Add mixed precision SparsityCsr SpMV support ([#970](#970)) + Allow creating CSR submatrix including from (possibly discontinuous) index sets ([#885](#885), [#964](#964)) + Add a scaled identity addition (M <- aI + bM) feature interface and impls for Csr and Dense ([#942](#942)) Deprecations and important changes: + Deprecate AmgxPgm in favor of the new Pgm name. ([#1149](#1149)). + Deprecate specialized residual norm classes in favor of a common `ResidualNorm` class ([#1101](#1101)) + Deprecate CamelCase non-polymorphic types in favor of snake_case versions (like array, machine_topology, uninitialized_array, index_set) ([#1031](#1031), [#1052](#1052)) + Bug fix: restrict gko::share to rvalue references (*possible interface break*) ([#1020](#1020)) + Bug fix: when using cuSPARSE's triangular solvers, specifying the factory parameter `num_rhs` is now required when solving for more than one right-hand side, otherwise an exception is thrown ([#1184](#1184)). + Drop official support for old CUDA < 9.2 ([#887](#887)) Improved performance additions: + Reuse tmp storage in reductions in solvers and add a mutable workspace to all solvers ([#1013](#1013), [#1028](#1028)) + Add HIP unsafe atomic option for AMD ([#1091](#1091)) + Prefer vendor implementations for Dense dot, conj_dot and norm2 when available ([#967](#967)). + Tuned OpenMP SellP, COO, and ELL SpMV kernels for a small number of RHS ([#809](#809)) Fixes: + Fix various compilation warnings ([#1076](#1076), [#1183](#1183), [#1189](#1189)) + Fix issues with hwloc-related tests ([#1074](#1074)) + Fix include headers for GCC 12 ([#1071](#1071)) + Fix for simple-solver-logging example ([#1066](#1066)) + Fix for potential memory leak in Logger ([#1056](#1056)) + Fix logging of mixin classes ([#1037](#1037)) + Improve value semantics for LinOp types, like moved-from state in cross-executor copy/clones ([#753](#753)) + Fix some matrix SpMV and conversion corner cases ([#905](#905), [#978](#978)) + Fix uninitialized data ([#958](#958)) + Fix CUDA version requirement for cusparseSpSM ([#953](#953)) + Fix several issues within bash-script ([#1016](#1016)) + Fixes for `NVHPC` compiler support ([#1194](#1194)) Other additions: + Simplify and properly name GMRES kernels ([#861](#861)) + Improve pkg-config support for non-CMake libraries ([#923](#923), [#1109](#1109)) + Improve gdb pretty printer ([#987](#987), [#1114](#1114)) + Add a logger highlighting inefficient allocation and copy patterns ([#1035](#1035)) + Improved and optimized test random matrix generation ([#954](#954), [#1032](#1032)) + Better CSR strategy defaults ([#969](#969)) + Add `move_from` to `PolymorphicObject` ([#997](#997)) + Remove unnecessary device_guard usage ([#956](#956)) + Improvements to the generic accessor for mixed-precision ([#727](#727)) + Add a naive lower triangular solver implementation for CUDA ([#764](#764)) + Add support for int64 indices from CUDA 11 onward with SpMV and SpGEMM ([#897](#897)) + Add a L1 norm implementation ([#900](#900)) + Add reduce_add for arrays ([#831](#831)) + Add utility to simplify Dense View creation from an existing Dense vector ([#1136](#1136)). + Add a custom transpose implementation for Fbcsr and Csr transpose for unsupported vendor types ([#1123](#1123)) + Make IDR random initilization deterministic ([#1116](#1116)) + Move the algorithm choice for triangular solvers from Csr::strategy_type to a factory parameter ([#1088](#1088)) + Update CUDA archCoresPerSM ([#1175](#1116)) + Add kernels for Csr sparsity pattern lookup ([#994](#994)) + Differentiate between structural and numerical zeros in Ell/Sellp ([#1027](#1027)) + Add a binary IO format for matrix data ([#984](#984)) + Add a tuple zip_iterator implementation ([#966](#966)) + Simplify kernel stubs and declarations ([#888](#888)) + Simplify GKO_REGISTER_OPERATION with lambdas ([#859](#859)) + Simplify copy to device in tests and examples ([#863](#863)) + More verbose output to array assertions ([#858](#858)) + Allow parallel compilation for Jacobi kernels ([#871](#871)) + Change clang-format pointer alignment to left ([#872](#872)) + Various improvements and fixes to the benchmarking framework ([#750](#750), [#759](#759), [#870](#870), [#911](#911), [#1033](#1033), [#1137](#1137)) + Various documentation improvements ([#892](#892), [#921](#921), [#950](#950), [#977](#977), [#1021](#1021), [#1068](#1068), [#1069](#1069), [#1080](#1080), [#1081](#1081), [#1108](#1108), [#1153](#1153), [#1154](#1154)) + Various CI improvements ([#868](#868), [#874](#874), [#884](#884), [#889](#889), [#899](#899), [#903](#903), [#922](#922), [#925](#925), [#930](#930), [#936](#936), [#937](#937), [#958](#958), [#882](#882), [#1011](#1011), [#1015](#1015), [#989](#989), [#1039](#1039), [#1042](#1042), [#1067](#1067), [#1073](#1073), [#1075](#1075), [#1083](#1083), [#1084](#1084), [#1085](#1085), [#1139](#1139), [#1178](#1178), [#1187](#1187))

upsj requested a review from a team September 13, 2021 14:32

upsj added this to the Ginkgo 1.5.0 milestone Sep 13, 2021

upsj added this to In Progress in Ginkgo development Sep 13, 2021

upsj self-assigned this Sep 13, 2021

upsj added the 1:ST:ready-for-review This PR is ready for review label Sep 13, 2021

upsj changed the title ~~device_matrix_data and device-side matrix::read~~ Add device_matrix_data and device-side matrix::read Sep 13, 2021

upsj mentioned this pull request Sep 29, 2021

Always enable GKO_BUILD_REFERENCE for tests #889

Merged

upsj force-pushed the device_matrix_data branch from 98a9d98 to f93a77c Compare October 9, 2021 08:59

upsj force-pushed the device_matrix_data branch 5 times, most recently from 167c2be to 3c39b8c Compare November 8, 2021 16:52

upsj added 1:ST:ready-for-review This PR is ready for review and removed 1:ST:need-feedback The PR is somewhat ready but feedback on a blocking topic is required before a proper review. labels Nov 8, 2021

upsj moved this from In Progress to Awaiting Review in Ginkgo development Nov 8, 2021

upsj force-pushed the device_matrix_data branch 4 times, most recently from 03bf993 to f1a9528 Compare November 11, 2021 19:48

yhmtsai approved these changes Nov 18, 2021

View reviewed changes

Ginkgo development automation moved this from Awaiting Review to Awaiting Merge Nov 18, 2021

upsj added 1:ST:ready-to-merge This PR is ready to merge. and removed 1:ST:ready-for-review This PR is ready for review labels Nov 18, 2021

upsj force-pushed the device_matrix_data branch 4 times, most recently from b89d2c7 to 633e69f Compare November 20, 2021 08:13

upsj and others added 14 commits November 20, 2021 09:22

move components to _kernels suffix and common

5a1762e

make ISAI storage sparse

2479535

fix triangular matrix generator

f1a817b

remove Fbcsr::set_block_size

731078a

This should never have been possible in the first place

fix submatrix extraction warning

52b2f01

add device_matrix_data

2b26120

remove zeros from matrix read tests

babf299

add matrix read kernels

a6baf1d

add documentation for device_matrix_data

9f4b20c

add random matrix data generation functions

6678a9f

review updates

c944b74

* use ref split_matrix_data for OpenMP * fix fbcsr from_matrix_data * fix gdb pretty printers for const Array refs Co-authored-by: Yuhsiang Tsai <[email protected]>

test device_matrix_data_kernels for reference

56de3ab

add device_matrix_data to test_install

3c570da

upsj force-pushed the device_matrix_data branch from 633e69f to 3c570da Compare November 20, 2021 08:23

upsj merged commit 2a34e0e into develop Nov 20, 2021

upsj deleted the device_matrix_data branch November 20, 2021 11:16

upsj mentioned this pull request Jul 28, 2023

Is custom initial parameters of function "read" needed? #96

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add `device_matrix_data` and device-side `matrix::read` #886

Add `device_matrix_data` and device-side `matrix::read` #886

upsj commented Sep 13, 2021 •

edited

Loading

yhmtsai left a comment

yhmtsai Nov 18, 2021

upsj Nov 19, 2021

yhmtsai Nov 18, 2021

upsj Nov 19, 2021

ginkgo-bot commented Nov 20, 2021

ginkgo-bot commented Nov 20, 2021

sonarcloud bot commented Nov 20, 2021



		template <typename ValueType, typename IndexType>
		struct hybrid_tuple_unpack_functor {

Add device_matrix_data and device-side matrix::read #886

Add device_matrix_data and device-side matrix::read #886

Conversation

upsj commented Sep 13, 2021 • edited Loading

yhmtsai left a comment

Choose a reason for hiding this comment

yhmtsai Nov 18, 2021

Choose a reason for hiding this comment

upsj Nov 19, 2021

Choose a reason for hiding this comment

yhmtsai Nov 18, 2021

Choose a reason for hiding this comment

upsj Nov 19, 2021

Choose a reason for hiding this comment

ginkgo-bot commented Nov 20, 2021

ginkgo-bot commented Nov 20, 2021

sonarcloud bot commented Nov 20, 2021

Add `device_matrix_data` and device-side `matrix::read` #886

Add `device_matrix_data` and device-side `matrix::read` #886

upsj commented Sep 13, 2021 •

edited

Loading