Multigrid improvement and fix #982

yhmtsai · 2022-03-04T10:52:07Z

This pr is mainly for the status and the collection of recent multigrid improvement or fix.
the updates are splitted into #981 #980. Some changes related multigrid itself will be put here.
Please do not focus on this pr until #981 #980 are reviewed and merged

uses the MultigridState to cache the internal vector
changes the mid_smoother default behavior to standalone with nullptr. i.e. no smoother in the middle of F cycle and W cycle
deletes the KCycle from the public interface because I can not make it work well with the current mixed type design. I need more time to investigate it.
uses the same interface from ApplyWithInitialGuess not zero_guess now. i.e. factory.with_default_initial_guess(initial_guess_mode::zero) for preconditioned.
change the three bool (x_is_zero, is_first, is_end) to an enum.

TODO:

core/solver/multigrid.cpp

examples/mixed-multigrid-solver/mixed-multigrid-solver.cpp

tcojean

Some comments

cuda/CMakeLists.txt

examples/mixed-multigrid-solver/mixed-multigrid-solver.cpp

include/ginkgo/core/solver/solver_base.hpp

core/solver/multigrid.cpp

include/ginkgo/core/solver/multigrid.hpp

thoasm

Initial review. So far, only nits and nothing serious found.

core/solver/multigrid.cpp

thoasm · 2022-11-07T04:36:49Z

core/solver/multigrid.cpp

+     * @param b  the right hand side
+     * @param x  the input vectors
+     */


nit: missing documentation for x_is_zero, is_first and is_end

change it to the enum cycle info

include/ginkgo/core/solver/multigrid.hpp

core/solver/multigrid.cpp

examples/mixed-multigrid-solver/mixed-multigrid-solver.cpp

examples/multigrid-preconditioned-solver/multigrid-preconditioned-solver.cpp

tcojean

LGTM

reference/test/solver/multigrid_kernels.cpp

core/solver/multigrid.cpp

pratikvn

Some minor comments. LGTM!

pratikvn · 2022-11-07T14:21:48Z

examples/multigrid-preconditioned-solver/multigrid-preconditioned-solver.cpp

    auto smoother_gen = gko::share(
        ir::build()
            .with_solver(inner_solver_gen)
-            .with_relaxation_factor(0.9)
+            .with_relaxation_factor(static_cast<ValueType>(0.9))


These static casts are a bit annoying. Can we remove them here by moving them to where the parameter is being initialized ?

Related to #1162 ;)

core/solver/multigrid.cpp

pratikvn · 2022-11-07T14:44:14Z

core/solver/multigrid.cpp

+    // origin or next or first
+    bool use_pre = (info & cycle_info::first_of_cycle) ||
+                   mid_case == multigrid::mid_smooth_type::both ||
+                   mid_case == multigrid::mid_smooth_type::pre_smoother;


Do we check for the case if the user wants a pre-smoother (has used the mid_smooth_type::pre_smoother option) but has not provided a pre_smoother list ?

it will behave like pre_smoother, so it will not process smoother

upsj

Sorry I didn't get to this sooner, a few comments before this gets merged, I'll try to finish the rest tomorrow

upsj · 2022-11-07T16:57:36Z

include/ginkgo/core/solver/multigrid.hpp

    {
-        return !parameters_.zero_guess;
-    }
-
-    /**
-     * Gets the stopping criterion factory of the solver.
-     *
-     * @return the stopping criterion factory
-     */
-    std::shared_ptr<const stop::CriterionFactory> get_stop_criterion_factory()
-        const
-    {
-        return stop_criterion_factory_;
-    }
-
-    /**
-     * Sets the stopping criterion of the solver.
-     *
-     * @param other  the new stopping criterion factory
-     */
-    void set_stop_criterion_factory(
-        std::shared_ptr<const stop::CriterionFactory> other)
-    {
-        stop_criterion_factory_ = std::move(other);
-    }
-
-    /**
-     * Gets the system operator of the linear system.
-     *
-     * @return the system operator
-     */
-    std::shared_ptr<const LinOp> get_system_matrix() const
-    {
-        return system_matrix_;
+        return this->get_default_initial_guess() ==
+               initial_guess_mode::provided;


Nit: This can go into the source file

I moved the longer implementation to the source file

include/ginkgo/core/solver/solver_base.hpp

upsj · 2022-11-07T17:01:03Z

reference/test/solver/multigrid_kernels.cpp

-}
-
-
-TYPED_TEST(Multigrid, SolvesStencilSystem2ByFCycle)


F cycles are still present, is this still obsolete?

Deleting System2 is mostly due to Mixed type. It will introduce single precision error when mixed_precision is off. It seems to make sense, but I want to check it more.

upsj · 2022-11-07T17:03:07Z

examples/multigrid-preconditioned-solver/multigrid-preconditioned-solver.cpp

    auto smoother_gen = gko::share(
        ir::build()
            .with_solver(inner_solver_gen)
-            .with_relaxation_factor(0.9)
+            .with_relaxation_factor(static_cast<ValueType>(0.9))


Related to #1162 ;)

core/solver/multigrid.cpp

upsj · 2022-11-07T21:19:37Z

core/solver/multigrid.cpp

@@ -52,6 +52,7 @@ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 #include "core/components/fill_array_kernels.hpp"
 #include "core/solver/ir_kernels.hpp"
 #include "core/solver/multigrid_kernels.hpp"
+#include "core/solver/solver_base.hpp"


 namespace gko {


kcycle kernels below can be removed everywhere

I think the kernel can be used when we add the kcycle back. It still contains the corresponding tests, so I will prefer keeping them for find them easily

upsj · 2022-11-07T21:22:43Z

core/solver/multigrid.cpp

+    // current level's nrows x nrhs
+    std::vector<std::shared_ptr<LinOp>> r_list;
+    // next level's nrows x nrhs
+    std::vector<std::shared_ptr<LinOp>> g_list;
+    std::vector<std::shared_ptr<LinOp>> e_list;
+    // constant 1 x 1


It seems like the comments no longer match the members? Was this left-over from the recursive storage attempt?

the comments is described in the current level fashion, so it also fits in the recursive thought.
level 0 -> r_list[0] has current level's #rows and g_list[0], e_list[0] has next level's #rows

…residual criterion

precision changes happen in restrict/prolong

note: when disable mixed precision, some of them can not reach double criterion

- delete unsafe opt and unused var - move MultigridState into experimental - change the mid smoother default

Co-authored-by: Pratik Nayak <pratikvn@protonmail.com> Co-authored-by: Terry Cojean <terry.cojean@kit.edu>

Co-authored-by: Terry Cojean <terry.cojean@kit.edu>

- use cycle_mode - clean the example and use determinstic - move the MultigridState into multigrid::detail namespace - update the documentation - fix the doxygen Co-authored-by: Pratik Nayak <pratikvn@protonmail.com> Co-authored-by: Terry Cojean <terry.cojean@kit.edu> Co-authored-by: Thomas Grützmacher <thomas.gruetzmacher@kit.edu> Co-authored-by: Tobias Ribizel <ribizel@kit.edu>

Co-authored-by: Tobias Ribizel <ribizel@kit.edu>

ginkgo-bot · 2022-11-08T12:40:51Z

Note: This PR changes the Ginkgo ABI:

Functions changes summary: 187 Removed, 1 Changed (36 filtered out), 295 Added functions
Variables changes summary: 0 Removed, 0 Changed, 0 Added variable

For details check the full ABI diff under Artifacts here

sonarcloud · 2022-11-09T04:38:54Z

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))

yhmtsai added the 1:ST:WIP This PR is a work in progress. Not ready for review. label Mar 4, 2022

yhmtsai self-assigned this Mar 4, 2022

tcojean force-pushed the improve_multigrid branch from aae62f5 to 3bf1057 Compare April 21, 2022 06:57

tcojean added this to the Ginkgo 1.5.0 milestone Oct 20, 2022

yhmtsai added the 1:ST:high-importance This issue/PR is of high importance and must be addressed as soon as possible. label Oct 28, 2022

thoasm requested review from thoasm and pratikvn October 28, 2022 12:35

yhmtsai force-pushed the improve_multigrid branch from 3bf1057 to 63c9f70 Compare November 4, 2022 07:30

pratikvn reviewed Nov 4, 2022

View reviewed changes

core/solver/multigrid.cpp Outdated Show resolved Hide resolved

examples/mixed-multigrid-solver/mixed-multigrid-solver.cpp Outdated Show resolved Hide resolved

tcojean reviewed Nov 4, 2022

View reviewed changes

yhmtsai force-pushed the improve_multigrid branch 4 times, most recently from 56764af to b609166 Compare November 6, 2022 18:19

thoasm reviewed Nov 7, 2022

View reviewed changes

yhmtsai force-pushed the improve_multigrid branch from a84951e to 00a4e75 Compare November 7, 2022 09:44

tcojean approved these changes Nov 7, 2022

View reviewed changes

reference/test/solver/multigrid_kernels.cpp Show resolved Hide resolved

core/solver/multigrid.cpp Outdated Show resolved Hide resolved

pratikvn approved these changes Nov 7, 2022

View reviewed changes

yhmtsai added 1:ST:ready-to-merge This PR is ready to merge. and removed 1:ST:WIP This PR is a work in progress. Not ready for review. labels Nov 7, 2022

yhmtsai force-pushed the improve_multigrid branch from 716de16 to ed2ea48 Compare November 7, 2022 16:52

upsj reviewed Nov 7, 2022

View reviewed changes

upsj approved these changes Nov 7, 2022

View reviewed changes

yhmtsai and others added 18 commits November 8, 2022 13:10

add mixed

deabe71

avoid residual computation in multigrid NOTE: check how to deal with …

277db87

…residual criterion

add nvtx to measure performance

9c2cd0f

Comment nvtx, use zero_guess parameter

560c71c

add static into multigrid

0a17db5

Reuse the workspace correctly

e17782b

use bad way to add b,x workspace

c2295df

generate g, e with next level precision

3257726

precision changes happen in restrict/prolong

fixed compiler mixed type issue

92091a4

update example to allow A and b input

b41cf8d

fixes type/executor error and update example

3a77fd9

WIP: update the test

c694706

note: when disable mixed precision, some of them can not reach double criterion

fix compile error

d209e74

disable nvtx, mixed test, kcycle

9a7a5e1

add EnableApplyWithIntialGuess, solver/iterative base

31c82c0

clear codes and add short description

e16f183

- delete unsafe opt and unused var - move MultigridState into experimental - change the mid smoother default

clear up function, update example

af33452

Co-authored-by: Pratik Nayak <pratikvn@protonmail.com> Co-authored-by: Terry Cojean <terry.cojean@kit.edu>

use shared_ptr to forward declaration of MultigridState

6b908e2

Co-authored-by: Terry Cojean <terry.cojean@kit.edu>

yhmtsai force-pushed the improve_multigrid branch 2 times, most recently from d3fbfd3 to a2f84eb Compare November 8, 2022 05:14

yhmtsai and others added 2 commits November 8, 2022 20:27

use decltype(*mg_level) to avoid pointee usage

cf02395

Co-authored-by: Tobias Ribizel <ribizel@kit.edu>

yhmtsai force-pushed the improve_multigrid branch from a2f84eb to cf02395 Compare November 8, 2022 12:28

tcojean merged commit 03b6a5d into develop Nov 8, 2022

tcojean deleted the improve_multigrid branch November 8, 2022 17:34

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Multigrid improvement and fix #982

Multigrid improvement and fix #982

yhmtsai commented Mar 4, 2022 •

edited

Loading

tcojean left a comment

thoasm left a comment

thoasm Nov 7, 2022

yhmtsai Nov 7, 2022

tcojean left a comment

pratikvn left a comment

pratikvn Nov 7, 2022

upsj Nov 7, 2022

pratikvn Nov 7, 2022

yhmtsai Nov 7, 2022

upsj left a comment

upsj Nov 7, 2022

yhmtsai Nov 8, 2022

upsj Nov 7, 2022

yhmtsai Nov 7, 2022

upsj Nov 7, 2022

upsj Nov 7, 2022

yhmtsai Nov 8, 2022

upsj Nov 7, 2022

yhmtsai Nov 8, 2022

ginkgo-bot commented Nov 8, 2022

sonarcloud bot commented Nov 9, 2022

		}


		TYPED_TEST(Multigrid, SolvesStencilSystem2ByFCycle)

Multigrid improvement and fix #982

Multigrid improvement and fix #982

Conversation

yhmtsai commented Mar 4, 2022 • edited Loading

tcojean left a comment

Choose a reason for hiding this comment

thoasm left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

tcojean left a comment

Choose a reason for hiding this comment

pratikvn left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

upsj left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

ginkgo-bot commented Nov 8, 2022

sonarcloud bot commented Nov 9, 2022

yhmtsai commented Mar 4, 2022 •

edited

Loading