-
Notifications
You must be signed in to change notification settings - Fork 86
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.
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
Fix the automatical strategy when shared. #559
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM!
@@ -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 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
unrelated to your changes: Is there a reason why this function (and move_to(...)) is implemented in the header instead of csr.cpp?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
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?
4af73dc
to
048395f
Compare
Codecov Report
@@ 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
Continue to review full report at Codecov.
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
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.
@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 |
Originally, I think the case about amd_csr->copy_from(nvidia_csr) directly. |
@yhmtsai Indeed, so far we do not even support 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. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
@tcojean I am fine with these changes. I do not have any question in this PR. |
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]>
Kudos, SonarCloud Quality Gate passed!
|
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
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:
copy
function to the strategies in order to createa new shared pointer polymorphically.
matrix is instantiated or copied. Thankfully, the strategies are
currently light objects so this should create few overhead.
correctly with the automatical strategy.
Fixes #426