Simplify third-party integration #786

upsj · 2021-06-06T13:27:56Z

inspired by the changes in #785 and some thoughts on packaging and dependency management:
Right now, we are building our third-party dependencies during configure time, manually pass all CMake parameters and construct targets out of the generated libraries manually.
Since all dependencies except for hwloc are only used inside Ginkgo, I think it might make more sense to build them as subdirectories, since then all CMake-specific flags are forwarded directly and consistently.

This changes the behavior of googletest slightly, making it use shared libraries by default, but otherwise everything should stay the same.

Another thing that we might want to address here is that currently, compiling Ginkgo (even without tests, benchmarks) on CUDA systems without internet connectivity fails, since we try to pull in CAS. Since we only ever need a single file from that repository, why don't we include it directly in our source instead, and set up a sync action that either syncs CAS from our repo or creates a PR if CAS was changed?

Finally, Ginkgo 1.4.0 is the first release with an actual library dependency exceeding vendor libraries (hwloc), so we should check whether the current design is future proof (also thinking of library separation and #715), since we don't seem to be using any hwloc functionality yet, right now. It looks to me like the functionality is most useful for GenericExecutor #704 and MPI support #760 ?

Add a CI job checking compatibility with pre-installed dependencies
Consider vendoring CAS
~~Consider improving separation of hwloc component~~ I think I misunderstood some details there, the separation looks fine

codecov · 2021-06-06T15:26:50Z

Codecov Report

Merging #786 (412e7b3) into develop (9359008) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff            @@
##           develop     #786   +/-   ##
========================================
  Coverage    94.26%   94.26%           
========================================
  Files          400      400           
  Lines        31578    31578           
========================================
  Hits         29768    29768           
  Misses        1810     1810

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 9359008...412e7b3. Read the comment docs.

yhmtsai · 2021-06-07T07:19:34Z

When looking into ExternalProject, I think we do not use ExternalProject fully in ginkgo.
Can FetchContent set the config type of MSVC?
MSVC in linking will require they use same crt in debug mode (with d) or others (without d). or can it mix them together?

upsj · 2021-06-07T07:28:10Z

Since everything is built as part of the same CMake workflow, everything uses the same config. The only issue is GTest, which overwrites the existing CRT flags. Just in general, their CMake setup is a bit of a mess.

third_party/gtest/CMakeLists.txt

yhmtsai · 2021-06-07T13:39:51Z

the google test does not use the value from ginkgo due to CMP0077?
From this https://cmake.org/pipermail/cmake/2019-March/069206.html and also CMP0077 is shown in the log.
make the variable be in the CACHE or to set CMP0077 outside to affect googletest CMakeList?

upsj · 2021-06-08T10:34:01Z

I added CAS directly to the repository and a sync step with the CAS repository. You can see it working in my forks https://github.com/upsj/ginkgo-bot-playground and https://github.com/upsj/CudaArchitectureSelector

yhmtsai · 2021-06-08T10:56:50Z

Could it be also done by github submodule?

upsj · 2021-06-08T11:02:49Z

git submodules have the same issue as FetchContent: You need an internet connection, plus any change in the submodule requires a change in the main repo as well. This way, we make it easier to change CAS and have a self-contained library for builds that don't enable tests and benchmarks.

yhmtsai · 2021-06-08T11:31:01Z

the update from CAS to ginkgo may be true.
but I think the internet somehow depends the download ginkgo in the beginning.
you can use git clone --recurse-submodules to download them together.

upsj · 2021-06-08T11:39:17Z

That doesn't work if you download Ginkgo as a tarball (as many package managers do). I have gotten the impression that git submodules are frowned upon for dependency management (and found them overly complicated myself), and since we control this dependency completely, I think it might be easier to vendor it directly.

yhmtsai · 2021-06-08T11:50:34Z

I see. one nit for the update, could you also add the commit sha of CAS into playground? like you have the commit sha of playground inside CAS

upsj · 2021-06-08T13:55:46Z

After some discussion with @yhmtsai I decided to keep the CAS sync out of the library for now.

tcojean

LGTM in general. I have a few comments and questions.

Nice catch that we can now use FetchContent due to all the CMake version bumps, that is much less awkward to use than ExternalProject.

CMakeLists.txt

tcojean · 2021-06-08T17:00:23Z

core/test/accessor/CMakeLists.txt

-    add_test(NAME "${REL_BINARY_DIR}/${test_name}" COMMAND "${TEST_TARGET_NAME}")
+    add_test(NAME "${REL_BINARY_DIR}/${test_name}"
+        COMMAND "${TEST_TARGET_NAME}"
+        WORKING_DIRECTORY "$<TARGET_FILE_DIR:ginkgo>")


Why set the working directory here, that would make it be build/core if I understand this directive?

There is a longer story behind this which I should mention: Contrary to the CMake default, GTest builds static libraries by default, if BUILD_SHARED_LIBS is unspecified. That means that previously, all tests were linked statically in all settings and we didn't need to worry about the location. With ExternalProject, we didn't forward our BUILD_SHARED_LIBS settings, but with add_subdirectory we do, which means we are now linking dynamically to GTest.
To make tests work on Windows without having to copy around DLLs all the time, we placed all libraries and DLLs into lib/, which is also where libginkgo.dll and more importantly libgtest.dll will be stored. So this change is necessary for all tests to find libgtest.dll

Thanks for the detailed explanation. If that's the case, maybe it makes sense to explicitly add an if (WIN32) to highlight that fact, and keep the defaults otherwise?

The working directory does not matter except for dynamic linking, so I would prefer to keep this, as the explicit conditional would make the code more complex.

tcojean · 2021-06-08T17:02:55Z

third_party/gflags/CMakeLists.txt

-    set(INTEL19_STD_FIX "")
-    if (CMAKE_CXX_COMPILER_ID STREQUAL "Intel" AND CMAKE_CXX_COMPILER_VERSION MATCHES "19")
-        set(INTEL19_STD_FIX "-DCMAKE_CXX_FLAGS=-std=c++14")
+if(NOT gflags_FOUND)


So we reverse the current behavior in that case, which we had previously defined to be use ours first except if explicitly requested. I think that eases package management, but might surprise some users. You should document this in the relevant .md file(s).

Maybe to make this clearer, we could put all these if(NOT xxx_FOUND) in third_party/CMakeLists.txt as well?

good idea. In which cases did we actually use the existing libraries then?

Most of the time that's already in third_party/CMakeLists.txt. Adding AND NOT xxx_FOUND to the existing if conditions should just work.

I think ginkgo_add_subdirectory did the same thing (only include if NOT xxx_FOUND), so this should not change the behavior.

third_party/git-cmake-format/CMakeLists.txt

third_party/gtest/CMakeLists.txt

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

yhmtsai

LGTM.
the gflags/gtest are in build/_dep/<>-src and build/_dep/<>-build not the directory inside the third_party.
I think it is okay?

third_party/rapidjson/CMakeLists.txt

yhmtsai · 2021-06-08T19:26:21Z

could FetchContent handle non-cmake configure project?
change the CONFIGURE_COMMAND to "${TPL_HWLOC_PATH}/src/configure --disable-nvml --disable-cuda --disable-rsmi"

upsj · 2021-06-08T20:01:25Z

Sure, FetchContent could also be used for other tasks. I didn't want to modify the hwloc code though, since that is more complex.
And yes, those are the default paths. I think the only thing remaining until this can be merged is that it seems like the enroot runner is currently not downloading the new images, since CMake doesn't find GTest etc.

* remove MSVC support completely * only find packages if they are needed

sonarqubecloud · 2021-06-10T13:41:15Z

Kudos, SonarCloud Quality Gate passed!

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

No Coverage information
No Duplication information

@tcojean

So I think my changes here are now complete, I simplified the HWLOC integration by

Removing all MSVC-specific code, since we disable it for Windows/Mac anyways.
Merging the GINKGO_USE_EXTERNAL_HWLOC variable with HWLOC_FOUND, because they will both always have the same value
Removing GINKGO_USE_EXTERNAL_* and TPL_* variables from the installed Ginkgo config, since the test and benchmark environment is not installed, and users should not care whether the hwloc we are linking to internally was existing or built by Ginkgo.
Removing unused CMake code
Checking for packages with find_package only if they are needed
@tcojean or @yhmtsai could you give the latest version another read?

CMakeLists.txt

tcojean

LGTM I think. I will need to check what spack thinks of these changes.

tcojean · 2021-06-10T14:59:27Z

cmake/GinkgoConfig.cmake.in

-set(GINKGO_USE_EXTERNAL_GTEST "@GINKGO_USE_EXTERNAL_GTEST@")
-set(GINKGO_USE_EXTERNAL_GFLAGS "@GINKGO_USE_EXTERNAL_GFLAGS@")
-set(GINKGO_USE_EXTERNAL_RAPIDJSON "@GINKGO_USE_EXTERNAL_RAPIDJSON@")
-set(GINKGO_USE_EXTERNAL_HWLOC "@GINKGO_USE_EXTERNAL_HWLOC@")


Having this also helped debugging in case something failed with an external package, knowing it's external would help pinpoint the problem. With the current setup, I'm not sure if we would have a (simple) way to keep something like that. Maybe storing the actual version?

CMakeLists.txt

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

upsj added this to the Ginkgo 1.4.0 milestone Jun 6, 2021

upsj self-assigned this Jun 6, 2021

ginkgo-bot added mod:cuda This is related to the CUDA module. reg:build This is related to the build system. reg:example This is related to the examples. labels Jun 6, 2021

upsj added the 1:ST:run-full-test label Jun 6, 2021

upsj force-pushed the simplify_third_party branch from 7129237 to 12f6ccc Compare June 7, 2021 06:42

yhmtsai reviewed Jun 7, 2021

View reviewed changes

third_party/gtest/CMakeLists.txt Outdated Show resolved Hide resolved

upsj added 4 commits June 7, 2021 17:22

silence CMake CUDA_ARCHITECTURES warnings

91cb452

use FetchContent instead of ExternalProject

dca41a3

set GTest location to lib/

55f2b20

set accessor test working directory

ad9b550

upsj force-pushed the simplify_third_party branch from 87391d6 to b0cb052 Compare June 7, 2021 15:22

fix third-party parameter passing

d559ecb

upsj force-pushed the simplify_third_party branch from b0cb052 to d559ecb Compare June 7, 2021 20:48

upsj added 2 commits June 8, 2021 15:54

vendor CAS module

a15b17b

fix find_package parameter order

e4ac35d

upsj force-pushed the simplify_third_party branch from 6ee144a to e4ac35d Compare June 8, 2021 13:55

tcojean approved these changes Jun 8, 2021

View reviewed changes

review updates

abf5854

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

yhmtsai approved these changes Jun 8, 2021

View reviewed changes

third_party/rapidjson/CMakeLists.txt Show resolved Hide resolved

upsj added 1:ST:ready-to-merge This PR is ready to merge. and removed 1:ST:run-full-test labels Jun 9, 2021

simplify hwloc integration

412e7b3

* remove MSVC support completely * only find packages if they are needed

upsj force-pushed the simplify_third_party branch from 84ccc43 to 412e7b3 Compare June 10, 2021 11:15

yhmtsai reviewed Jun 10, 2021

View reviewed changes

CMakeLists.txt Show resolved Hide resolved

tcojean approved these changes Jun 10, 2021

View reviewed changes

upsj merged commit fc06232 into develop Jun 11, 2021

upsj deleted the simplify_third_party branch June 11, 2021 08:06

Simplify third-party integration #786

Simplify third-party integration #786

Uh oh!

Conversation

Uh oh!

Uh oh!

Uh oh!

Codecov Report

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!