Fix the automatical strategy when shared. #559

tcojean · 2020-06-05T18:55:02Z

The automatical strategy has an issue when shared, it behaves as the
wrong strategy if the matrices using it have different properties.
This fixes this issue only by extending the current interface, and not
changing any existing function. All other (more elegant) approaches
I thought of would break the current interface. In effect:

Add a new virtual copy function to the strategies in order to create
a new shared pointer polymorphically.
The approach used is to always copy the strategy, whenever a CSR
matrix is instantiated or copied. Thankfully, the strategies are
currently light objects so this should create few overhead.
Add tests which ensure that matrices with difference properties work
correctly with the automatical strategy.

Fixes #426

upsj

LGTM!

upsj · 2020-06-05T19:05:05Z

include/ginkgo/core/matrix/csr.hpp

@@ -549,7 +598,13 @@ class Csr : public EnableLinOp<Csr<ValueType, IndexType>>,
 void convert_to(Csr<ValueType, IndexType> *result) const override
 {
 bool same_executor = this->get_executor() == result->get_executor();
- EnableLinOp<Csr>::convert_to(result);
+ // NOTE: as soon as strategies are improved, this can be reverted


unrelated to your changes: Is there a reason why this function (and move_to(...)) is implemented in the header instead of csr.cpp?

I'm unsure, usually we do, for example for Coo:

using EnableLinOp<Coo>::convert_to; using EnableLinOp<Coo>::move_to;

So they are defined in the header in that case, maybe that was done intentionally to keep that situation when we fixed some strategy issues?

include/ginkgo/core/matrix/csr.hpp

hip/test/matrix/csr_kernels.hip.cpp

codecov · 2020-06-06T15:45:41Z

Codecov Report

Merging #559 into develop will decrease coverage by 0.20%.
The diff coverage is 13.23%.

@@             Coverage Diff             @@
##           develop     #559      +/-   ##
===========================================
- Coverage    84.36%   84.15%   -0.21%     
===========================================
  Files          296      296              
  Lines        19799    19835      +36     
===========================================
- Hits         16704    16693      -11     
- Misses        3095     3142      +47

Impacted Files	Coverage Δ
reference/test/factorization/par_ict_kernels.cpp	`92.55% <ø> (ø)`
reference/test/factorization/par_ilut_kernels.cpp	`96.17% <ø> (ø)`
reference/test/matrix/coo_kernels.cpp	`100.00% <ø> (ø)`
reference/test/matrix/dense_kernels.cpp	`100.00% <ø> (ø)`
reference/test/matrix/ell_kernels.cpp	`100.00% <ø> (ø)`
reference/test/matrix/hybrid_kernels.cpp	`100.00% <ø> (ø)`
reference/test/matrix/sellp_kernels.cpp	`100.00% <ø> (ø)`
include/ginkgo/core/matrix/csr.hpp	`13.46% <6.34%> (-4.68%)`	⬇️
core/matrix/csr.cpp	`84.29% <100.00%> (+6.82%)`	⬆️
core/test/matrix/csr_builder.cpp	`100.00% <100.00%> (ø)`
... and 4 more

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 1af71c3...b233808. Read the comment docs.

pratikvn

LGTM!

yhmtsai

Sorry for revert the label. I would like to check some consideration.
the automatical strategy copy can not change the parameter depends on executor?
I am not sure whether we need to consider a AMD GPU and a NVIDIA GPU in same system.
copy in the convert_to uses same nwarps, warp_size on AMD GPU and NVIDIA GPU.

core/test/matrix/csr_builder.cpp

tcojean · 2020-06-09T12:35:00Z

@yhmtsai I do not understand your issue properly. I think with the current implementation there is no way to change the executor dynamically. If the user wants to use automatical for both AMD and NVIDIA hardware, then he will need to anyway declare two different make_shared<Csr::automatical>(cudaExec) and make_shared<Csr::automatical>(hipExec) to have the proper values. In this case, there should be no issue with copies happening with the wrong values?

yhmtsai · 2020-06-09T12:54:20Z

Originally, I think the case about amd_csr->copy_from(nvidia_csr) directly.
we do not support copy/move between amd/nvidia, right?
in host matrix -> amd matrix, the host matrix has warp_size = 32 by default. It will pass this to amd_csr, which leads inconsistent warp_size. strategy_rebuild_helper to reset the strategy by executor to make sure that the strategy uses the parameters from executor not others.

tcojean · 2020-06-09T13:26:50Z

@yhmtsai Indeed, so far we do not even support raw_copy_to between AMD hardware and CUDA hardware. I think there would some work required to make it work.

Now I see the issue, I will try to see what I can do for that nonetheless. Or at least put a warning/fixme tag.

fritzgoebel

LGTM

yhmtsai

LGTM

include/ginkgo/core/matrix/csr.hpp

tcojean · 2020-06-09T21:42:47Z

@yhmtsai I should have fixed that issue now. In the end, I put @upsj code from the convert_to<Csr<next_precision...>> into a helper, as that is exactly what needs to be called to fix this issue I believe.

Please tell me if I'm missing anything.

yhmtsai · 2020-06-10T11:51:05Z

@tcojean I am fine with these changes. I do not have any question in this PR.

include/ginkgo/core/matrix/csr.hpp

The automatical strategy has an issue when shared, it behaves as the wrong strategy if the matrices using it have different properties. This fixes this issue only by extending the current interface, and not changing any existing function. In effect: + Add a new virtual `copy` function to the strategies in order to create a new shared pointer polymorphically. + The approach used is to always copy the strategy, whenever a CSR matrix is instantiated or copied. Thankfully, the strategies are currently light objects so this should create few overhead. + Add tests which ensure that matrices with difference properties work correctly with the automatical strategy. Fixes #426

Co-authored-by: upsj <[email protected]>

+ Use usual formatting for the `csr_builder` subclass. + Put the strategy management code from `convert_to(Csr<next_precision<...>>)` into a `convert_strategy_helper` to properly rebuild the correct strategy for another executor. Co-authored-by: Yuhsiang M. Tsai <[email protected]>

If the destination executor is not hip or cuda, use the current executor's information if available, otherwise fallback to CUDA.

Co-authored-by: Pratik Nayak <[email protected]>

sonarcloud · 2020-06-12T15:26:04Z

Kudos, SonarCloud Quality Gate passed!

0 Bugs
0 Vulnerabilities (and 0 Security Hotspots to review)
12 Code Smells

6.2% Coverage
0.0% Duplication

The version of Java (1.8.0_121) you have used to run this analysis is deprecated and we will stop accepting it from October 2020. Please update to at least Java 11.
Read more here

The Ginkgo team is proud to announce the new minor release of Ginkgo version 1.2.0. This release brings full HIP support to Ginkgo, new preconditioners (ParILUT, ISAI), conversion between double and float for all LinOps, and many more features and fixes. Supported systems and requirements: + For all platforms, cmake 3.9+ + Linux and MacOS + gcc: 5.3+, 6.3+, 7.3+, all versions after 8.1+ + clang: 3.9+ + Intel compiler: 2017+ + Apple LLVM: 8.0+ + CUDA module: CUDA 9.0+ + HIP module: ROCm 2.8+ + Windows + MinGW and CygWin: gcc 5.3+, 6.3+, 7.3+, all versions after 8.1+ + Microsoft Visual Studio: VS 2017 15.7+ + CUDA module: CUDA 9.0+, Microsoft Visual Studio + OpenMP module: MinGW or CygWin. The current known issues can be found in the [known issues page](https://github.com/ginkgo-project/ginkgo/wiki/Known-Issues). # Additions Here are the main additions to the Ginkgo library. Other thematic additions are listed below. + Add full HIP support to Ginkgo [#344](#344), [#357](#357), [#384](#384), [#373](#373), [#391](#391), [#396](#396), [#395](#395), [#393](#393), [#404](#404), [#439](#439), [#443](#443), [#567](#567) + Add a new ISAI preconditioner [#489](#489), [#502](#502), [#512](#512), [#508](#508), [#520](#520) + Add support for ParILUT and ParICT factorization with ILU preconditioners [#400](#400) + Add a new BiCG solver [#438](#438) + Add a new permutation matrix format [#352](#352), [#469](#469) + Add CSR SpGEMM support [#386](#386), [#398](#398), [#418](#418), [#457](#457) + Add CSR SpGEAM support [#556](#556) + Make all solvers and preconditioners transposable [#535](#535) + Add CsrBuilder and CooBuilder for intrusive access to matrix arrays [#437](#437) + Add a standard-compliant allocator based on the Executors [#504](#504) + Support conversions for all LinOp between double and float [#521](#521) + Add a new boolean to the CUDA and HIP executors to control DeviceReset (default off) [#557](#557) + Add a relaxation factor to IR to represent Richardson Relaxation [#574](#574) + Add two new stopping criteria, for relative (to `norm(b)`) and absolute residual norm [#577](#577) ### Example additions + Templatize all examples to simplify changing the precision [#513](#513) + Add a new adaptive precision block-Jacobi example [#507](#507) + Add a new IR example [#522](#522) + Add a new Mixed Precision Iterative Refinement example [#525](#525) + Add a new example on iterative trisolves in ILU preconditioning [#526](#526), [#536](#536), [#550](#550) ### Compilation and library changes + Auto-detect compilation settings based on environment [#435](#435), [#537](#537) + Add SONAME to shared libraries [#524](#524) + Add clang-cuda support [#543](#543) ### Other additions + Add sorting, searching and merging kernels for GPUs [#403](#403), [#428](#428), [#417](#417), [#455](#455) + Add `gko::as` support for smart pointers [#493](#493) + Add setters and getters for criterion factories [#527](#527) + Add a new method to check whether a solver uses `x` as an initial guess [#531](#531) + Add contribution guidelines [#549](#549) # Fixes ### Algorithms + Improve the classical CSR strategy's performance [#401](#401) + Improve the CSR automatical strategy [#407](#407), [#559](#559) + Memory, speed improvements to the ELL kernel [#411](#411) + Multiple improvements and fixes to ParILU [#419](#419), [#427](#427), [#429](#429), [#456](#456), [#544](#544) + Fix multiple issues with GMRES [#481](#481), [#523](#523), [#575](#575) + Optimize OpenMP matrix conversions [#505](#505) + Ensure the linearity of the ILU preconditioner [#506](#506) + Fix IR's use of the advanced apply [#522](#522) + Fix empty matrices conversions and add tests [#560](#560) ### Other core functionalities + Fix complex number support in our math header [#410](#410) + Fix CUDA compatibility of the main ginkgo header [#450](#450) + Fix isfinite issues [#465](#465) + Fix the Array::view memory leak and the array/view copy/move [#485](#485) + Fix typos preventing use of some interface functions [#496](#496) + Fix the `gko::dim` to abide to the C++ standard [#498](#498) + Simplify the executor copy interface [#516](#516) + Optimize intermediate storage for Composition [#540](#540) + Provide an initial guess for relevant Compositions [#561](#561) + Better management of nullptr as criterion [#562](#562) + Fix the norm calculations for complex support [#564](#564) ### CUDA and HIP specific + Use the return value of the atomic operations in our wrappers [#405](#405) + Improve the portability of warp lane masks [#422](#422) + Extract thread ID computation into a separate function [#464](#464) + Reorder kernel parameters for consistency [#474](#474) + Fix the use of `pragma unroll` in HIP [#492](#492) ### Other + Fix the Ginkgo CMake installation files [#414](#414), [#553](#553) + Fix the Windows compilation [#415](#415) + Always use demangled types in error messages [#434](#434), [#486](#486) + Add CUDA header dependency to appropriate tests [#452](#452) + Fix several sonarqube or compilation warnings [#453](#453), [#463](#463), [#532](#532), [#569](#569) + Add shuffle tests [#460](#460) + Fix MSVC C2398 error [#490](#490) + Fix missing interface tests in test install [#558](#558) # Tools and ecosystem ### Benchmarks + Add better norm support in the benchmarks [#377](#377) + Add CUDA 10.1 generic SpMV support in benchmarks [#468](#468), [#473](#473) + Add sparse library ILU in benchmarks [#487](#487) + Add overhead benchmarking capacities [#501](#501) + Allow benchmarking from a matrix list file [#503](#503) + Fix benchmarking issue with JSON and non-finite numbers [#514](#514) + Fix benchmark logger crashers with OpenMP [#565](#565) ### CI related + Improvements to the CI setup with HIP compilation [#421](#421), [#466](#466) + Add MacOSX CI support [#470](#470), [#488](#488) + Add Windows CI support [#471](#471), [#488](#488), [#510](#510), [#566](#566) + Use sanitizers instead of valgrind [#476](#476) + Add automatic container generation and update facilities [#499](#499) + Fix the CI parallelism settings [#517](#517), [#538](#538), [#539](#539) + Make the codecov patch check informational [#519](#519) + Add support for LLVM sanitizers with improved thread sanitizer support [#578](#578) ### Test suite + Add an assertion for sparsity pattern equality [#416](#416) + Add core and reference multiprecision tests support [#448](#448) + Speed up GPU tests by avoiding device reset [#467](#467) + Change test matrix location string [#494](#494) ### Other + Add Ginkgo badges from our tools [#413](#413) + Update the `create_new_algorithm.sh` script [#420](#420) + Bump copyright and improve license management [#436](#436), [#433](#433) + Set clang-format minimum requirement [#441](#441), [#484](#484) + Update git-cmake-format [#446](#446), [#484](#484) + Disable the development tools by default [#442](#442) + Add a script for automatic header formatting [#447](#447) + Add GDB pretty printer for `gko::Array` [#509](#509) + Improve compilation speed [#533](#533) + Add editorconfig support [#546](#546) + Add a compile-time check for header self-sufficiency [#552](#552) # Related PR: #583

tcojean added is:bug Something looks wrong. mod:core This is related to the core module. type:matrix-format This is related to the Matrix formats 1:ST:ready-for-review This PR is ready for review labels Jun 5, 2020

tcojean added this to the Ginkgo 1.2.0 milestone Jun 5, 2020

tcojean requested review from upsj, pratikvn, thoasm, yhmtsai and fritzgoebel June 5, 2020 18:55

tcojean self-assigned this Jun 5, 2020

tcojean force-pushed the fix_automatical branch from 400770b to f768210 Compare June 5, 2020 18:59

upsj approved these changes Jun 5, 2020

View reviewed changes

upsj reviewed Jun 5, 2020

View reviewed changes

include/ginkgo/core/matrix/csr.hpp Outdated Show resolved Hide resolved

tcojean force-pushed the fix_automatical branch 3 times, most recently from 4af73dc to 048395f Compare June 6, 2020 08:45

upsj reviewed Jun 6, 2020

View reviewed changes

hip/test/matrix/csr_kernels.hip.cpp Outdated Show resolved Hide resolved

tcojean force-pushed the fix_automatical branch from 048395f to 1d49c5a Compare June 6, 2020 11:32

pratikvn approved these changes Jun 9, 2020

View reviewed changes

tcojean 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 Jun 9, 2020

tcojean force-pushed the fix_automatical branch from 1d49c5a to 9114280 Compare June 9, 2020 11:32

yhmtsai requested changes Jun 9, 2020

View reviewed changes

core/test/matrix/csr_builder.cpp Show resolved Hide resolved

tcojean force-pushed the fix_automatical branch from 94a219a to 50f1a96 Compare June 9, 2020 13:35

fritzgoebel approved these changes Jun 9, 2020

View reviewed changes

tcojean removed the 1:ST:ready-to-merge This PR is ready to merge. label Jun 9, 2020

yhmtsai approved these changes Jun 9, 2020

View reviewed changes

yhmtsai reviewed Jun 9, 2020

View reviewed changes

include/ginkgo/core/matrix/csr.hpp Show resolved Hide resolved

tcojean force-pushed the fix_automatical branch from 50f1a96 to 66b749b Compare June 9, 2020 21:40

tcojean force-pushed the fix_automatical branch from 66b749b to bfbaf5f Compare June 9, 2020 21:44

tcojean added the 1:ST:ready-for-review This PR is ready for review label Jun 9, 2020

yhmtsai reviewed Jun 10, 2020

View reviewed changes

include/ginkgo/core/matrix/csr.hpp Show resolved Hide resolved

tcojean force-pushed the fix_automatical branch from 91982ab to 725d246 Compare June 11, 2020 11:37

tcojean and others added 5 commits June 12, 2020 13:31

Fix convert_to for CSR and drop useless move

778e331

Co-authored-by: upsj <[email protected]>

Preserve the executor information when copying strategies

fdc2f6f

If the destination executor is not hip or cuda, use the current executor's information if available, otherwise fallback to CUDA.

Set contributing guidelines as a markdown for doc.

b233808

Co-authored-by: Pratik Nayak <[email protected]>

tcojean force-pushed the fix_automatical branch from 725d246 to b233808 Compare June 12, 2020 11:31

tcojean 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 Jun 12, 2020

tcojean merged commit 07a0dac into develop Jun 13, 2020

tcojean deleted the fix_automatical branch June 13, 2020 09:53

tcojean mentioned this pull request Jun 23, 2020

Release/1.2.0 #576

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Fix the automatical strategy when shared. #559

Fix the automatical strategy when shared. #559

tcojean commented Jun 5, 2020

upsj left a comment

upsj Jun 5, 2020

tcojean Jun 5, 2020

codecov bot commented Jun 6, 2020 •

edited

Loading

pratikvn left a comment

yhmtsai left a comment

tcojean commented Jun 9, 2020

yhmtsai commented Jun 9, 2020

tcojean commented Jun 9, 2020 •

edited

Loading

fritzgoebel left a comment

yhmtsai left a comment

tcojean commented Jun 9, 2020

yhmtsai commented Jun 10, 2020

sonarcloud bot commented Jun 12, 2020

Fix the automatical strategy when shared. #559

Fix the automatical strategy when shared. #559

Conversation

tcojean commented Jun 5, 2020

upsj left a comment

Choose a reason for hiding this comment

upsj Jun 5, 2020

Choose a reason for hiding this comment

tcojean Jun 5, 2020

Choose a reason for hiding this comment

codecov bot commented Jun 6, 2020 • edited Loading

Codecov Report

pratikvn left a comment

Choose a reason for hiding this comment

yhmtsai left a comment

Choose a reason for hiding this comment

tcojean commented Jun 9, 2020

yhmtsai commented Jun 9, 2020

tcojean commented Jun 9, 2020 • edited Loading

fritzgoebel left a comment

Choose a reason for hiding this comment

yhmtsai left a comment

Choose a reason for hiding this comment

tcojean commented Jun 9, 2020

yhmtsai commented Jun 10, 2020

sonarcloud bot commented Jun 12, 2020

codecov bot commented Jun 6, 2020 •

edited

Loading

tcojean commented Jun 9, 2020 •

edited

Loading