Dpcpp port dense #710

yhmtsai · 2021-02-24T08:43:58Z

This PR is WIP. It is runnable on Gen9 GPU in current state.

This is mainly for dense porting and related kernels.

TODO:

dpcpp fix the issue on cpu side in next release.
device config selection?
merge DPCPP cooperative group #757
merge Add common interface for simple kernels #733 (can be separate

codecov · 2021-05-25T22:56:04Z

Codecov Report

Merging #710 (d175116) into develop (c101790) will decrease coverage by 0.01%.
The diff coverage is n/a.

@@             Coverage Diff             @@
##           develop     #710      +/-   ##
===========================================
- Coverage    94.25%   94.24%   -0.02%     
===========================================
  Files          408      408              
  Lines        32756    32757       +1     
===========================================
- Hits         30875    30872       -3     
- Misses        1881     1885       +4

Impacted Files	Coverage Δ
omp/reorder/rcm_kernels.cpp	`95.06% <0.00%> (-3.08%)`	⬇️
include/ginkgo/core/base/range_accessors.hpp	`78.94% <0.00%> (+5.26%)`	⬆️

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 c101790...d175116. Read the comment docs.

yhmtsai · 2021-05-26T09:08:12Z

format!

sonarqubecloud · 2021-06-02T19:36:08Z

Kudos, SonarCloud Quality Gate passed!

0 Bugs
0 Vulnerabilities
0 Security Hotspots
2 Code Smells

0.0% Coverage
3.8% Duplication

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

- public per_kernel link option when static - add find(MKL) when ginkgo is static

thoasm

For the most part, it looks good to me, but I really dislike the C-style casts and want them changed before approving.
Other than that, I mostly have typos and other small nits.

dpcpp/CMakeLists.txt

dpcpp/base/helper.dp.cpp

dpcpp/base/helper.hpp

dpcpp/matrix/dense_kernels.dp.cpp

include/ginkgo/core/synthesizer/containers.hpp

Co-authored-by: Thomas Grützmacher <thomas.gruetzmacher@kit.edu>

yhmtsai · 2021-07-20T15:43:56Z

@ginkgo-project/reviewers I update the code by reviews. please check and review it.
I do not change the name of thread_ids.dp.hpp due to future porting concern.
I also add some // TODO: porting ... to remind that we should revisit them when we are close to fully porting dpcpp or after more porting involving.

upsj

LGTM in general, nice job keeping the interface so consistent w.r.t CUDA.
The auto changes in common decrease readibility quite a lot, so I would prefer to use explicit types. In general, integer literals are typed, so 0 is int, even with auto, which is why we could keep this as it is, but I understand the objection that it obscures the type of the parameters (overflows are not really a risk in almost all of these cases, since we are only dealing with local data or small sizes as in the tests).
Most of the dense_kernels tests are already present in test/matrix/dense_kernels.cpp, so you can probably remove them or move them there if they are not yet present.

upsj · 2021-07-19T10:45:32Z

cmake/create_test.cmake

+    # Note: MKL_ENV is empty on linux. Maybe need to apply MKL_ENV to all test.
+    if (MKL_ENV)
+       set_tests_properties(${test_target_name} PROPERTIES ENVIRONMENT "${MKL_ENV}")
+    endif()


Why does running tests need MKL environment variables? If this is really necessary, we should be fixing it in the environment, not CMake.

this is from the MKL example https://software.intel.com/content/www/us/en/develop/documentation/onemkl-windows-developer-guide/top/getting-started/cmake-config-for-onemkl.html

Setting environment variables is not the responsibility of the build system. I would prefer if we remove this altogether, as we can't guarantee users set it the same way for their application.

the input env is defined by MKL cmake and not all parameters from environment, which means no difference even if user gives same env with different input.
I am not sure whether it is only for test or for all application.
I agree with you if it only contains LD_LIBRARY_PATH
but it also contains MKL_INTERFACE_LAYER, MKL_THREADING_LAYER, MKL_BLACS_MPI depends MKL input or platform

I am not sure whether it is only for test or for all application.

It is not possible to set the environment variables for an application that is built for CMake. Thus, introducing this change would mean that the tests potentially behave differently than application codes.
This is a really bad CMake setup, since it mixes runtime behavior with compile-time options. I should probably bring this up in the next Intel meeting.

I agree with you.

upsj · 2021-07-19T10:48:10Z

dpcpp/CMakeLists.txt

 target_link_libraries(ginkgo_dpcpp PUBLIC ginkgo_device)
+target_link_libraries(ginkgo_dpcpp PRIVATE $<LINK_ONLY:MKL::MKL_DPCPP>)


Why this separation into linker and compiler flags? Does MKL add SYCL-specific flags?

what do you mean separation here?

As far as I understand it, this only links against MKL, but doesn't add any include paths or other compiler flags associated with the MKL_DPCPP target. Why is this necessary, as opposed to plain

Suggested change

target_link_libraries(ginkgo_dpcpp PRIVATE $<LINK_ONLY:MKL::MKL_DPCPP>)

target_link_libraries(ginkgo_dpcpp PRIVATE MKL::MKL_DPCPP)

I try both version. LINK_ONLY is usually used by INTERFACE_LINK_LIBRARIES in set_targert_properties
using target_link_libraries will handle it automatically.
previous one gives $<LINK_ONLY:<$<LINK_ONLY>:MKL::MKL_DPCPP>>
updated one gives $<LINK_ONLY>:MKL::MKL_DPCPP>

common/factorization/par_ilut_select_kernels.hpp.inc

upsj · 2021-07-20T15:44:12Z

dpcpp/components/reduction.dp.hpp

+{
+    auto block_results_val = source;
+    size_type grid_dim = size;
+    auto block_results = Array<ValueType>(exec);


It's a move, not a copy, so if the initialization doesn't happen in-place (which I think it does), it would have negligible overhead.

dpcpp/test/matrix/dense_kernels.cpp

upsj · 2021-07-20T15:54:34Z

dpcpp/matrix/dense_kernels.dp.cpp

+                               size_type *result)
+{
+    const auto num_rows = source->get_size()[0];
+    auto nnz_per_row = Array<size_type>(exec, num_rows);


I am saying: You don't need to worry about this IMO, at worst it just moves a few additional pointers around.

need to revisit these TODO when we are close to fully porting ginkgo 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>

upsj

LGTM! Only the CMake-specific discussions should be resolved from my viewpoint.

dpcpp/test/matrix/dense_kernels.cpp

thoasm

LGTM!

note: gtest_skip can not be in another function call Co-authored-by: Tobias Ribizel <ribizel@kit.edu>

stale

sonarqubecloud · 2021-07-23T03:58:15Z

Kudos, SonarCloud Quality Gate passed!

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
No Duplication information

Ginkgo release 1.4.0 The Ginkgo team is proud to announce the new Ginkgo minor release 1.4.0. This release brings most of the Ginkgo functionality to the Intel DPC++ ecosystem which enables Intel-GPU and CPU execution. The only Ginkgo features which have not been ported yet are some preconditioners. Ginkgo's mixed-precision support is greatly enhanced thanks to: 1. The new Accessor concept, which allows writing kernels featuring on-the-fly memory compression, among other features. The accessor can be used as header-only, see the [accessor BLAS benchmarks repository](https://github.com/ginkgo-project/accessor-BLAS/tree/develop) as a usage example. 2. All LinOps now transparently support mixed-precision execution. By default, this is done through a temporary copy which may have a performance impact but already allows mixed-precision research. Native mixed-precision ELL kernels are implemented which do not see this cost. The accessor is also leveraged in a new CB-GMRES solver which allows for performance improvements by compressing the Krylov basis vectors. Many other features have been added to Ginkgo, such as reordering support, a new IDR solver, Incomplete Cholesky preconditioner, matrix assembly support (only CPU for now), machine topology information, and more! Supported systems and requirements: + For all platforms, cmake 3.13+ + C++14 compliant compiler + Linux and MacOS + gcc: 5.3+, 6.3+, 7.3+, all versions after 8.1+ + clang: 3.9+ + Intel compiler: 2018+ + Apple LLVM: 8.0+ + CUDA module: CUDA 9.0+ + HIP module: ROCm 3.5+ + DPC++ module: Intel OneAPI 2021.3. Set the CXX compiler to `dpcpp`. + Windows + MinGW and Cygwin: gcc 5.3+, 6.3+, 7.3+, all versions after 8.1+ + Microsoft Visual Studio: VS 2019 + CUDA module: CUDA 9.0+, Microsoft Visual Studio + OpenMP module: MinGW or Cygwin. Algorithm and important feature additions: + Add a new DPC++ Executor for SYCL execution and other base utilities [#648](#648), [#661](#661), [#757](#757), [#832](#832) + Port matrix formats, solvers and related kernels to DPC++. For some kernels, also make use of a shared kernel implementation for all executors (except Reference). [#710](#710), [#799](#799), [#779](#779), [#733](#733), [#844](#844), [#843](#843), [#789](#789), [#845](#845), [#849](#849), [#855](#855), [#856](#856) + Add accessors which allow multi-precision kernels, among other things. [#643](#643), [#708](#708) + Add support for mixed precision operations through apply in all LinOps. [#677](#677) + Add incomplete Cholesky factorizations and preconditioners as well as some improvements to ILU. [#672](#672), [#837](#837), [#846](#846) + Add an AMGX implementation and kernels on all devices but DPC++. [#528](#528), [#695](#695), [#860](#860) + Add a new mixed-precision capability solver, Compressed Basis GMRES (CB-GMRES). [#693](#693), [#763](#763) + Add the IDR(s) solver. [#620](#620) + Add a new fixed-size block CSR matrix format (for the Reference executor). [#671](#671), [#730](#730) + Add native mixed-precision support to the ELL format. [#717](#717), [#780](#780) + Add Reverse Cuthill-McKee reordering [#500](#500), [#649](#649) + Add matrix assembly support on CPUs. [#644](#644) + Extends ISAI from triangular to general and spd matrices. [#690](#690) Other additions: + Add the possibility to apply real matrices to complex vectors. [#655](#655), [#658](#658) + Add functions to compute the absolute of a matrix format. [#636](#636) + Add symmetric permutation and improve existing permutations. [#684](#684), [#657](#657), [#663](#663) + Add a MachineTopology class with HWLOC support [#554](#554), [#697](#697) + Add an implicit residual norm criterion. [#702](#702), [#818](#818), [#850](#850) + Row-major accessor is generalized to more than 2 dimensions and a new "block column-major" accessor has been added. [#707](#707) + Add an heat equation example. [#698](#698), [#706](#706) + Add ccache support in CMake and CI. [#725](#725), [#739](#739) + Allow tuning and benchmarking variables non intrusively. [#692](#692) + Add triangular solver benchmark [#664](#664) + Add benchmarks for BLAS operations [#772](#772), [#829](#829) + Add support for different precisions and consistent index types in benchmarks. [#675](#675), [#828](#828) + Add a Github bot system to facilitate development and PR management. [#667](#667), [#674](#674), [#689](#689), [#853](#853) + Add Intel (DPC++) CI support and enable CI on HPC systems. [#736](#736), [#751](#751), [#781](#781) + Add ssh debugging for Github Actions CI. [#749](#749) + Add pipeline segmentation for better CI speed. [#737](#737) Changes: + Add a Scalar Jacobi specialization and kernels. [#808](#808), [#834](#834), [#854](#854) + Add implicit residual log for solvers and benchmarks. [#714](#714) + Change handling of the conjugate in the dense dot product. [#755](#755) + Improved Dense stride handling. [#774](#774) + Multiple improvements to the OpenMP kernels performance, including COO, an exclusive prefix sum, and more. [#703](#703), [#765](#765), [#740](#740) + Allow specialization of submatrix and other dense creation functions in solvers. [#718](#718) + Improved Identity constructor and treatment of rectangular matrices. [#646](#646) + Allow CUDA/HIP executors to select allocation mode. [#758](#758) + Check if executors share the same memory. [#670](#670) + Improve test install and smoke testing support. [#721](#721) + Update the JOSS paper citation and add publications in the documentation. [#629](#629), [#724](#724) + Improve the version output. [#806](#806) + Add some utilities for dim and span. [#821](#821) + Improved solver and preconditioner benchmarks. [#660](#660) + Improve benchmark timing and output. [#669](#669), [#791](#791), [#801](#801), [#812](#812) Fixes: + Sorting fix for the Jacobi preconditioner. [#659](#659) + Also log the first residual norm in CGS [#735](#735) + Fix BiCG and HIP CSR to work with complex matrices. [#651](#651) + Fix Coo SpMV on strided vectors. [#807](#807) + Fix segfault of extract_diagonal, add short-and-fat test. [#769](#769) + Fix device_reset issue by moving counter/mutex to device. [#810](#810) + Fix `EnableLogging` superclass. [#841](#841) + Support ROCm 4.1.x and breaking HIP_PLATFORM changes. [#726](#726) + Decreased test size for a few device tests. [#742](#742) + Fix multiple issues with our CMake HIP and RPATH setup. [#712](#712), [#745](#745), [#709](#709) + Cleanup our CMake installation step. [#713](#713) + Various simplification and fixes to the Windows CMake setup. [#720](#720), [#785](#785) + Simplify third-party integration. [#786](#786) + Improve Ginkgo device arch flags management. [#696](#696) + Other fixes and improvements to the CMake setup. [#685](#685), [#792](#792), [#705](#705), [#836](#836) + Clarification of dense norm documentation [#784](#784) + Various development tools fixes and improvements [#738](#738), [#830](#830), [#840](#840) + Make multiple operators/constructors explicit. [#650](#650), [#761](#761) + Fix some issues, memory leaks and warnings found by MSVC. [#666](#666), [#731](#731) + Improved solver memory estimates and consistent iteration counts [#691](#691) + Various logger improvements and fixes [#728](#728), [#743](#743), [#754](#754) + Fix for ForwardIterator requirements in iterator_factory. [#665](#665) + Various benchmark fixes. [#647](#647), [#673](#673), [#722](#722) + Various CI fixes and improvements. [#642](#642), [#641](#641), [#795](#795), [#783](#783), [#793](#793), [#852](#852) Related PR: #857

Release 1.4.0 to master The Ginkgo team is proud to announce the new Ginkgo minor release 1.4.0. This release brings most of the Ginkgo functionality to the Intel DPC++ ecosystem which enables Intel-GPU and CPU execution. The only Ginkgo features which have not been ported yet are some preconditioners. Ginkgo's mixed-precision support is greatly enhanced thanks to: 1. The new Accessor concept, which allows writing kernels featuring on-the-fly memory compression, among other features. The accessor can be used as header-only, see the [accessor BLAS benchmarks repository](https://github.com/ginkgo-project/accessor-BLAS/tree/develop) as a usage example. 2. All LinOps now transparently support mixed-precision execution. By default, this is done through a temporary copy which may have a performance impact but already allows mixed-precision research. Native mixed-precision ELL kernels are implemented which do not see this cost. The accessor is also leveraged in a new CB-GMRES solver which allows for performance improvements by compressing the Krylov basis vectors. Many other features have been added to Ginkgo, such as reordering support, a new IDR solver, Incomplete Cholesky preconditioner, matrix assembly support (only CPU for now), machine topology information, and more! Supported systems and requirements: + For all platforms, cmake 3.13+ + C++14 compliant compiler + Linux and MacOS + gcc: 5.3+, 6.3+, 7.3+, all versions after 8.1+ + clang: 3.9+ + Intel compiler: 2018+ + Apple LLVM: 8.0+ + CUDA module: CUDA 9.0+ + HIP module: ROCm 3.5+ + DPC++ module: Intel OneAPI 2021.3. Set the CXX compiler to `dpcpp`. + Windows + MinGW and Cygwin: gcc 5.3+, 6.3+, 7.3+, all versions after 8.1+ + Microsoft Visual Studio: VS 2019 + CUDA module: CUDA 9.0+, Microsoft Visual Studio + OpenMP module: MinGW or Cygwin. Algorithm and important feature additions: + Add a new DPC++ Executor for SYCL execution and other base utilities [#648](#648), [#661](#661), [#757](#757), [#832](#832) + Port matrix formats, solvers and related kernels to DPC++. For some kernels, also make use of a shared kernel implementation for all executors (except Reference). [#710](#710), [#799](#799), [#779](#779), [#733](#733), [#844](#844), [#843](#843), [#789](#789), [#845](#845), [#849](#849), [#855](#855), [#856](#856) + Add accessors which allow multi-precision kernels, among other things. [#643](#643), [#708](#708) + Add support for mixed precision operations through apply in all LinOps. [#677](#677) + Add incomplete Cholesky factorizations and preconditioners as well as some improvements to ILU. [#672](#672), [#837](#837), [#846](#846) + Add an AMGX implementation and kernels on all devices but DPC++. [#528](#528), [#695](#695), [#860](#860) + Add a new mixed-precision capability solver, Compressed Basis GMRES (CB-GMRES). [#693](#693), [#763](#763) + Add the IDR(s) solver. [#620](#620) + Add a new fixed-size block CSR matrix format (for the Reference executor). [#671](#671), [#730](#730) + Add native mixed-precision support to the ELL format. [#717](#717), [#780](#780) + Add Reverse Cuthill-McKee reordering [#500](#500), [#649](#649) + Add matrix assembly support on CPUs. [#644](#644) + Extends ISAI from triangular to general and spd matrices. [#690](#690) Other additions: + Add the possibility to apply real matrices to complex vectors. [#655](#655), [#658](#658) + Add functions to compute the absolute of a matrix format. [#636](#636) + Add symmetric permutation and improve existing permutations. [#684](#684), [#657](#657), [#663](#663) + Add a MachineTopology class with HWLOC support [#554](#554), [#697](#697) + Add an implicit residual norm criterion. [#702](#702), [#818](#818), [#850](#850) + Row-major accessor is generalized to more than 2 dimensions and a new "block column-major" accessor has been added. [#707](#707) + Add an heat equation example. [#698](#698), [#706](#706) + Add ccache support in CMake and CI. [#725](#725), [#739](#739) + Allow tuning and benchmarking variables non intrusively. [#692](#692) + Add triangular solver benchmark [#664](#664) + Add benchmarks for BLAS operations [#772](#772), [#829](#829) + Add support for different precisions and consistent index types in benchmarks. [#675](#675), [#828](#828) + Add a Github bot system to facilitate development and PR management. [#667](#667), [#674](#674), [#689](#689), [#853](#853) + Add Intel (DPC++) CI support and enable CI on HPC systems. [#736](#736), [#751](#751), [#781](#781) + Add ssh debugging for Github Actions CI. [#749](#749) + Add pipeline segmentation for better CI speed. [#737](#737) Changes: + Add a Scalar Jacobi specialization and kernels. [#808](#808), [#834](#834), [#854](#854) + Add implicit residual log for solvers and benchmarks. [#714](#714) + Change handling of the conjugate in the dense dot product. [#755](#755) + Improved Dense stride handling. [#774](#774) + Multiple improvements to the OpenMP kernels performance, including COO, an exclusive prefix sum, and more. [#703](#703), [#765](#765), [#740](#740) + Allow specialization of submatrix and other dense creation functions in solvers. [#718](#718) + Improved Identity constructor and treatment of rectangular matrices. [#646](#646) + Allow CUDA/HIP executors to select allocation mode. [#758](#758) + Check if executors share the same memory. [#670](#670) + Improve test install and smoke testing support. [#721](#721) + Update the JOSS paper citation and add publications in the documentation. [#629](#629), [#724](#724) + Improve the version output. [#806](#806) + Add some utilities for dim and span. [#821](#821) + Improved solver and preconditioner benchmarks. [#660](#660) + Improve benchmark timing and output. [#669](#669), [#791](#791), [#801](#801), [#812](#812) Fixes: + Sorting fix for the Jacobi preconditioner. [#659](#659) + Also log the first residual norm in CGS [#735](#735) + Fix BiCG and HIP CSR to work with complex matrices. [#651](#651) + Fix Coo SpMV on strided vectors. [#807](#807) + Fix segfault of extract_diagonal, add short-and-fat test. [#769](#769) + Fix device_reset issue by moving counter/mutex to device. [#810](#810) + Fix `EnableLogging` superclass. [#841](#841) + Support ROCm 4.1.x and breaking HIP_PLATFORM changes. [#726](#726) + Decreased test size for a few device tests. [#742](#742) + Fix multiple issues with our CMake HIP and RPATH setup. [#712](#712), [#745](#745), [#709](#709) + Cleanup our CMake installation step. [#713](#713) + Various simplification and fixes to the Windows CMake setup. [#720](#720), [#785](#785) + Simplify third-party integration. [#786](#786) + Improve Ginkgo device arch flags management. [#696](#696) + Other fixes and improvements to the CMake setup. [#685](#685), [#792](#792), [#705](#705), [#836](#836) + Clarification of dense norm documentation [#784](#784) + Various development tools fixes and improvements [#738](#738), [#830](#830), [#840](#840) + Make multiple operators/constructors explicit. [#650](#650), [#761](#761) + Fix some issues, memory leaks and warnings found by MSVC. [#666](#666), [#731](#731) + Improved solver memory estimates and consistent iteration counts [#691](#691) + Various logger improvements and fixes [#728](#728), [#743](#743), [#754](#754) + Fix for ForwardIterator requirements in iterator_factory. [#665](#665) + Various benchmark fixes. [#647](#647), [#673](#673), [#722](#722) + Various CI fixes and improvements. [#642](#642), [#641](#641), [#795](#795), [#783](#783), [#793](#793), [#852](#852) Related PR: #866

yhmtsai added the 1:ST:WIP This PR is a work in progress. Not ready for review. label Feb 24, 2021

yhmtsai self-assigned this Feb 24, 2021

upsj added this to the Ginkgo 1.4.0 milestone May 5, 2021

yhmtsai force-pushed the dpcpp_port_dense branch 3 times, most recently from b77f1a3 to 8608a7d Compare May 25, 2021 21:07

yhmtsai force-pushed the dpcpp_port_dense branch 5 times, most recently from baef102 to 4fccdbd Compare June 1, 2021 22:04

yhmtsai added 1:ST:ready-for-review This PR is ready for review and removed 1:ST:WIP This PR is a work in progress. Not ready for review. labels Jun 7, 2021

yhmtsai requested review from tcojean, pratikvn, Slaedr, fritzgoebel, hartwiganzt, MarcelKoch, thoasm and upsj June 7, 2021 13:57

yhmtsai and others added 2 commits July 16, 2021 17:59

add cp, update doc, mv mkl bind

4859340

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

MKL cmake, delete unused, simplify func, add job

bb4d648

yhmtsai force-pushed the dpcpp_port_dense branch from 5ee5998 to 318603a Compare July 16, 2021 15:59

yhmtsai added 2 commits July 18, 2021 23:24

MKL static cmake

0b562fd

- public per_kernel link option when static - add find(MKL) when ginkgo is static

use mkl in ncols=1, add cuda descp in dpcpp thread

7ac6d09

yhmtsai force-pushed the dpcpp_port_dense branch from 318603a to 7ac6d09 Compare July 18, 2021 21:33

thoasm requested changes Jul 20, 2021

View reviewed changes

yhmtsai force-pushed the dpcpp_port_dense branch 2 times, most recently from a131660 to 08581f3 Compare July 20, 2021 14:33

improve document, fix auto usage in for, shared_memory usage

6169c2f

Co-authored-by: Thomas Grützmacher <thomas.gruetzmacher@kit.edu>

yhmtsai force-pushed the dpcpp_port_dense branch from 08581f3 to 097f9ce Compare July 20, 2021 15:13

upsj requested changes Jul 20, 2021

View reviewed changes

add some note to indicate the porting TODO

a6b4ccc

need to revisit these TODO when we are close to fully porting ginkgo 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>

yhmtsai force-pushed the dpcpp_port_dense branch from 097f9ce to a6b4ccc Compare July 20, 2021 16:09

upsj approved these changes Jul 20, 2021

View reviewed changes

dpcpp/test/matrix/dense_kernels.cpp Outdated Show resolved Hide resolved

yhmtsai force-pushed the dpcpp_port_dense branch from 9d05c56 to 7802f47 Compare July 20, 2021 21:31

yhmtsai requested a review from tcojean July 21, 2021 08:01

thoasm approved these changes Jul 21, 2021

View reviewed changes

explict type in for, use func not macro to skip, remove dup test

4464c4d

note: gtest_skip can not be in another function call Co-authored-by: Tobias Ribizel <ribizel@kit.edu>

yhmtsai force-pushed the dpcpp_port_dense branch from 7802f47 to 4464c4d Compare July 21, 2021 10:06

yhmtsai 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 Jul 21, 2021

debug/static dpcpp -> debug/shared due to memory

d175116

yhmtsai merged commit 47ba37d into develop Jul 21, 2021

yhmtsai deleted the dpcpp_port_dense branch July 21, 2021 19:31

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Dpcpp port dense #710

Dpcpp port dense #710

yhmtsai commented Feb 24, 2021 •

edited

Loading

codecov bot commented May 25, 2021 •

edited

Loading

yhmtsai commented May 26, 2021

sonarqubecloud bot commented Jun 2, 2021

thoasm left a comment

yhmtsai commented Jul 20, 2021 •

edited

Loading

upsj left a comment

upsj Jul 19, 2021

yhmtsai Jul 20, 2021

upsj Jul 20, 2021

yhmtsai Jul 20, 2021

upsj Jul 20, 2021

yhmtsai Jul 20, 2021

upsj Jul 19, 2021

yhmtsai Jul 20, 2021

upsj Jul 20, 2021

yhmtsai Jul 21, 2021

upsj Jul 20, 2021

upsj Jul 20, 2021

upsj left a comment

thoasm left a comment

sonarqubecloud bot commented Jul 23, 2021

		target_link_libraries(ginkgo_dpcpp PUBLIC ginkgo_device)
		target_link_libraries(ginkgo_dpcpp PRIVATE $<LINK_ONLY:MKL::MKL_DPCPP>)

Dpcpp port dense #710

Dpcpp port dense #710

Conversation

yhmtsai commented Feb 24, 2021 • edited Loading

codecov bot commented May 25, 2021 • edited Loading

Codecov Report

yhmtsai commented May 26, 2021

sonarqubecloud bot commented Jun 2, 2021

thoasm left a comment

Choose a reason for hiding this comment

yhmtsai commented Jul 20, 2021 • edited Loading

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

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

thoasm left a comment

Choose a reason for hiding this comment

sonarqubecloud bot commented Jul 23, 2021

yhmtsai commented Feb 24, 2021 •

edited

Loading

codecov bot commented May 25, 2021 •

edited

Loading

yhmtsai commented Jul 20, 2021 •

edited

Loading