-
Notifications
You must be signed in to change notification settings - Fork 39
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
Feature/kokkos #188
Open
ajpowelsnl
wants to merge
143
commits into
LLNL:develop
Choose a base branch
from
ajpowelsnl:feature/kokkos
base: develop
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Feature/kokkos #188
Changes from all commits
Commits
Show all changes
143 commits
Select commit
Hold shift + click to select a range
792224f
First pass at Kokkos in the build system
DavidPoliakoff 9456d76
First example
DavidPoliakoff 4e4c2dd
Merge branch 'develop' into feature/kokkos
DavidPoliakoff 19afb97
Reworking to split out the Kokkos implementations
DavidPoliakoff d658a4d
Added in the Kokkos implementation file (whoops)
DavidPoliakoff f7631d1
Moved Kokkos files into their own directory
DavidPoliakoff 0f74831
Working state, can now just stamp out kernels
DavidPoliakoff c70a7ca
Fixed up the sequential Kokkos version
DavidPoliakoff b52d974
WIP: CUDA
DavidPoliakoff 37c2897
Pulled latest Kokkos
DavidPoliakoff ae5007a
Functional DAXPY
DavidPoliakoff 7f707f9
Corrected name
DavidPoliakoff 88f98f9
Update C++ standard
DavidPoliakoff 8b12325
Changes for a minimal build
DavidPoliakoff 8b828f5
Kokkos IF_QUAD basic kernels
ajpowelsnl 529746d
Merge branch 'develop' into temp_kokkos_kernels
ajpowelsnl cfda905
infrastructure changes and Kokkos_ATOMIC_PI
ajpowelsnl b227e70
basic-kokkos IF_QUAD and ATOMIC_PI kernels
ajpowelsnl b3ef7f5
Adding piles o' files
DavidPoliakoff 104ac3d
Removed extra files
DavidPoliakoff c9cf2a7
Rename misnamed files
DavidPoliakoff fddf904
Infrastructure fixes
DavidPoliakoff 7a37806
Small changes to enable correct output
DavidPoliakoff 7aa5d8d
first implementation of a "kokkos-mechanics" test in RajaPerfSuite
ajpowelsnl 6858ba3
Added stream-add with Kokkos Views
DavidPoliakoff 3f8acf9
Adding basic kernels for Kokkos performance testing
ajpowelsnl 8bac825
csv_xml.py: parser for RAJAPerf-timing.csv to xml
ajpowelsnl c1193dc
csv_xml.py: raw xml dump of csv results
ajpowelsnl 9d0501c
csv_xml.py: updates to raw xml output
ajpowelsnl 5ed43c0
KokkosSeq: Kokkos-i-fied basic kernels
ajpowelsnl ca36ef0
KokkosSeq: Kokkos-i-fied basic kernels
ajpowelsnl b171109
csv_xml.py: rm "count", rm print statements
ajpowelsnl 3381d88
Add profiling hooks to enable SPOT
DavidPoliakoff 71901b6
Kokkos Sequential Basic Kernel Implementation
ajpowelsnl f252173
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/RAJAPerf into …
ajpowelsnl 27d63c9
KokkosCuda basic kernels
ajpowelsnl eec425e
Unbroke the build system (hahaoops)
DavidPoliakoff c1f72a5
rm "newline" parameter from open statement
ajpowelsnl 9835eff
rm the second "newline" paramter from open statement
ajpowelsnl 8bdcaf7
Update Kokkos
DavidPoliakoff c69ce1a
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/rajaperf into …
DavidPoliakoff e1b705a
Added metadata calls, for testing
DavidPoliakoff 91f7910
Added in metadata callbacks
DavidPoliakoff 9b4cfdc
KokkosCuda kernels; ATTN: NESTED_INIT not working
ajpowelsnl 7753a5d
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/RAJAPerf into …
ajpowelsnl 233072e
Revert to standard Kokkos nomenclature
DavidPoliakoff a62d462
Fixes for NESTED_INIT
DavidPoliakoff 2d35bc9
Remove debug output
DavidPoliakoff 25d3ac4
adding date stamp to xml file
ajpowelsnl cb4bd10
tidy'ing up changes
ajpowelsnl 6cb52f2
csv_xml.py: lower casing variants
ajpowelsnl c34a43d
re-worked NESTED_INIT kernel in Kokkos design
ajpowelsnl 8fca574
Sed surgery
DavidPoliakoff 8dc5c01
Sed surgery 2
DavidPoliakoff 944e58a
Removed all mentions of Kokkos specific backend implementations from …
DavidPoliakoff a3d4061
tpl/kokkos: unlinking sym link & cloing kokkos
ajpowelsnl 4f738c8
DAXPY kernel with KokkosViews for pointer x, pointer y
ajpowelsnl ce4079e
IF_QUAD.cpp: re-write basic kernels using KokkosView
ajpowelsnl fc4bb07
add Kokkos::fence() to kernel test
ajpowelsnl 1ca2a3c
INIT3 with Kokkos Views and custom types
ajpowelsnl 86c1283
Re-write of basic kernels using KokkosViews
ajpowelsnl 5c32d68
Rewrite of REDUCE3_INT using KokkosViews
ajpowelsnl 497741c
Re-write of TRAP_INT with Kokkos::fences();
ajpowelsnl 5ef80af
At least functional version of this
DavidPoliakoff ce5738a
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/RAJAPerf into …
ajpowelsnl 0be15e4
OpenMPTarget build
DavidPoliakoff 5af2f06
ATOMIC_PI notes added to explain changes
ajpowelsnl 0261b3b
Merge branch 'feature/kokkos' of github.com:DavidPoliakoff/rajaperf i…
ajpowelsnl a4f2ef0
changes needed for diverse builds
e306920
Merge branch 'feature/kokkos' of https://github.com/ajpowelsnl/RAJAPe…
210bd2c
Candidate fixes for builds
DavidPoliakoff 28c76e5
commented incomplete code to fix RPS builds
ajpowelsnl e76979e
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/RAJAPerf into …
ajpowelsnl 97f9e5d
DAXPY: comment out functor
ajpowelsnl 4dbf1d0
interface prototype that builds
ajpowelsnl a3196d6
Added infrastructure for make_kernel_base, and support for it in the …
DavidPoliakoff ae6286d
Added infrastructure for make_kernel_base, and support for it in the …
DavidPoliakoff 37659d6
Merge branch 'develop' into merge_test
DavidPoliakoff 6f16423
Small fix to CMakeLists
DavidPoliakoff 05d6ab7
add KOKKOS_FUNCTION to TRAP_INT-Kokkos.cpp
8c35199
updating rocprim in RAJA
9d8292e
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/rajaperf into …
DavidPoliakoff 18253c3
Added infrastructure for make_kernel_base, and support for it in the …
DavidPoliakoff 4f730cd
Update RAJA with version from develop
DavidPoliakoff 7f17ba4
cleaning up RPS vs Kokkos handling of HIP backend
b406206
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/rajaperf into …
af54587
Changes for HIP config and build
009388c
making RPS use new infrastructure (Executor.cpp) to handle Kokkos
ajpowelsnl 7be49db
Fixes for OpenMPTarget build
DavidPoliakoff f238c1c
IF_QUAD-Kokkos.cpp: tidying up formatting
ajpowelsnl c9ab748
stream-kokkos/ADD: changes to infrastructure and kernel
ajpowelsnl a0873de
Fixing run_params handling in RPS
ajpowelsnl 1514d53
Kokkos translations of stream kernels
ajpowelsnl 2ce3e28
EDUCE3_INT-Kokkos.cpp: formatting for clarity
ajpowelsnl ba8678e
watchr_KokkosConfig.json: config for new watchr
ajpowelsnl 80dea71
watchr_KokkosConfig.json: recurse dir set to true
ajpowelsnl 4093557
basic/IF_QUAD.cpp: fix typos and formatting
ajpowelsnl 12af5ca
RPS infrastructure changes:
ajpowelsnl b9a1381
lcals: Making kernel group Kokkos runable
ajpowelsnl 8d193ed
stream-kokkos/ADD-Kokkos.cpp: fix typos
ajpowelsnl bee1341
lcals-kokkos: lcals kernel group kokkos implement
ajpowelsnl d90c1fa
apps-kokkos: initial Kokkos-i-fication of apps kernels
ajpowelsnl 115a116
Merge branch 'develop' into subfeature/kokkos
ajpowelsnl a26044d
Merge remote-tracking branch 'upstream/develop' into subfeature/kokkos
DavidPoliakoff 2a65ef7
CMakeLists.txt: apps-kokkos fix up
ajpowelsnl 4a24d9a
apps header files: add runKokkosVariant
ajpowelsnl 0611696
RAJAPerfSuite: fix up for apps kernels
ajpowelsnl fa7f247
WIP-COUPLE.cpp: temp fix comment body
ajpowelsnl a8a7553
Attempt at merge conflict resolution
DavidPoliakoff c13b61c
CMakeLists.txt: fix amd arch
ajpowelsnl 298cd4c
watchr_KokkosConfig.json: updated script
ajpowelsnl 3b0ef87
Merge branch 'subfeature/kokkos' into merge_test
ajpowelsnl 92f8f4d
deconflicting merge: apps
ajpowelsnl efbfb4f
watchr_KokkosConfig.json: fix RAJAPerf name
ajpowelsnl cf7e93a
watchr_KokkosConfig.json: fix up config
ajpowelsnl da4f686
watchr_KokkosConfig.json:rm timestamp for plotting
ajpowelsnl 59ecd95
watchr_KokkosConfig.json: regex fix
ajpowelsnl 8f7e27b
watchr_KokkosConfig.json: fixup regex
ajpowelsnl 8d4723c
watchr_KokkosConfig.json: fix lambda_cuda
ajpowelsnl 4c4f103
watchr_KokkosConfig.json: format lambda_cuda
ajpowelsnl 170490d
omptarget: fix up build errors
ajpowelsnl 4766ccb
watchr_KokkosConfig.json: fix lambda_cuda
ajpowelsnl 3e331dc
watchr_KokkosConfig.json: fix lambda_cuda
ajpowelsnl 6af5a71
Kokkos translations: apps, algorithm
ajpowelsnl a9dbc63
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/rajaperf into …
ajpowelsnl 1fd7f0f
kokkos-algorithms: Kokkos translation
ajpowelsnl 8e56aa4
VOL3D-Kokkos.cpp: Fix #ifdefs, #endifs
ajpowelsnl ff48c3a
HALOEXCHANGE-Kokkos.cpp: Kokkos translation
ajpowelsnl 621e382
FIR-Kokkos.cpp: Wrapping coeff array in View
ajpowelsnl f19da0a
Merge remote-tracking branch 'upstream/develop' into feature/kokkos
ajpowelsnl 29c3460
Merge branch 'feature/kokkos' of github.com:ajpowelsnl/rajaperf into …
ajpowelsnl 038c34e
commenting features not integrated into Kokkos testing
ajpowelsnl f9d829d
Addressing David B.'s first batch of PR Comments
ajpowelsnl ec0421b
Fix some compiler warnings
ajpowelsnl 520357d
Merge 1: RPS + RPS AS INFRASTRUCTURE
ajpowelsnl c064d50
Merge 2: RPS + INFRASTRUCTURE
ajpowelsnl 29dfbfd
Fixing test reps and problem size numbers
ajpowelsnl 393504b
Fixed timing code
DavidPoliakoff 201d7ad
Merge pull request #1 from DavidPoliakoff/feature/kokkos
ajpowelsnl b7de26a
rm name passed to pop.region()
ajpowelsnl 3dc1751
First major clean up for David B.: does not include commented Executo…
ajpowelsnl 18cca87
First pass clean up of Executor files
ajpowelsnl 487bce9
Addressing PR comments from David Beckingsale & polybench-kokkos stubs
ajpowelsnl File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
|
@@ -7,11 +7,11 @@ | |
############################################################################### | ||
|
||
project(RAJAPerfSuite CXX) | ||
|
||
cmake_minimum_required(VERSION 3.9) | ||
|
||
option(ENABLE_RAJA_SEQUENTIAL "Run sequential variants of RAJA kernels. Disable | ||
this, and all other variants, to run _only_ raw C loops." On) | ||
option(ENABLE_KOKKOS "Include Kokkos implementations of the kernels in the RAJA Perfsuite" Off) | ||
|
||
# | ||
# Initialize the BLT build system | ||
|
@@ -25,14 +25,14 @@ set(ENABLE_TESTS Off CACHE BOOL "Enable BLT and RAJA tests") | |
set(ENABLE_EXAMPLES Off CACHE BOOL "Enable RAJA examples") | ||
set(RAJA_ENABLE_EXERCISES Off CACHE BOOL "Enable RAJA exercises") | ||
|
||
set(CMAKE_CXX_STANDARD 11) | ||
set(BLT_CXX_STANDARD 11) | ||
|
||
if(ENABLE_KOKKOS) | ||
set(CMAKE_CXX_STANDARD 14) | ||
set(BLT_CXX_STANDARD 14) | ||
endif() | ||
include(blt/SetupBLT.cmake) | ||
|
||
# | ||
# Define RAJA settings... | ||
# | ||
|
||
set(ENABLE_TESTS Off CACHE BOOL "") | ||
set(ENABLE_EXAMPLES Off CACHE BOOL "") | ||
|
@@ -53,11 +53,8 @@ add_subdirectory(tpl/RAJA) | |
get_property(RAJA_INCLUDE_DIRS DIRECTORY tpl/RAJA PROPERTY INCLUDE_DIRECTORIES) | ||
include_directories(${RAJA_INCLUDE_DIRS}) | ||
|
||
|
||
# | ||
# Setup variables to pass to Perf suite | ||
# | ||
|
||
# | ||
# These (hopefully temporary) macro constants are needed to work-around | ||
# performance issues in the xl compiler. | ||
|
@@ -80,18 +77,24 @@ if (ENABLE_OPENMP) | |
endif() | ||
if (ENABLE_CUDA) | ||
list(APPEND RAJA_PERFSUITE_DEPENDS cuda) | ||
endif() | ||
if (ENABLE_HIP) | ||
endif() | ||
|
||
# Kokkos requires hipcc as the CMAKE_CXX_COMPILER for HIP AMD/VEGA GPU | ||
# platforms, whereas RAJAPerf Suite uses blt/CMake FindHIP to set HIP compiler | ||
# Separate RAJAPerf Suite and Kokkos handling of HIP compilers | ||
|
||
if ((ENABLE_HIP) AND (NOT ENABLE_KOKKOS)) | ||
list(APPEND RAJA_PERFSUITE_DEPENDS hip) | ||
endif() | ||
|
||
set(RAJAPERF_BUILD_SYSTYPE $ENV{SYS_TYPE}) | ||
set(RAJAPERF_BUILD_HOST $ENV{HOSTNAME}) | ||
|
||
if (ENABLE_CUDA) | ||
set(CMAKE_CUDA_STANDARD 11) | ||
if (ENABLE_CUDA AND ENABLE_KOKKOS) | ||
set(CMAKE_CUDA_STANDARD 14) | ||
endif() | ||
set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -restrict -arch ${CUDA_ARCH} --expt-extended-lambda --expt-relaxed-constexpr") | ||
|
||
set(RAJAPERF_COMPILER "${CUDA_NVCC_EXECUTABLE}") | ||
list(APPEND RAJAPERF_COMPILER ${CMAKE_CXX_COMPILER}) | ||
set(RAJAPERF_COMPILER_OPTIONS "${CUDA_NVCC_FLAGS}") | ||
|
@@ -106,14 +109,58 @@ else() | |
list(APPEND RAJAPERF_COMPILER_OPTIONS ${CMAKE_CXX_FLAGS}) | ||
endif() | ||
|
||
configure_file(${CMAKE_SOURCE_DIR}/src/rajaperf_config.hpp.in | ||
configure_file(${CMAKE_CURRENT_SOURCE_DIR}/src/rajaperf_config.hpp.in | ||
${CMAKE_CURRENT_BINARY_DIR}/bin/rajaperf_config.hpp) | ||
|
||
# Make sure RAJA flag propagate (we need to do some house cleaning to | ||
# Make sure RAJA flags propagate (we need to do some tidying to | ||
# remove project-specific CMake variables that are no longer needed) | ||
set (CUDA_NVCC_FLAGS ${RAJA_NVCC_FLAGS}) | ||
|
||
# | ||
# Each directory in the perf suite has its own CMakeLists.txt file. | ||
# | ||
# The statement below is required for Kokkos compilation. | ||
if(ENABLE_KOKKOS) | ||
include_directories(SYSTEM ${CMAKE_CURRENT_SOURCE_DIR}/tpl/RAJA/include/) | ||
endif() | ||
|
||
|
||
# ENABLE_KOKKOS is A RAJAPerf Suite Option | ||
if(ENABLE_KOKKOS) | ||
add_definitions(-DRUN_KOKKOS) | ||
if(ENABLE_HIP) | ||
set(Kokkos_ENABLE_HIP ON CACHE BOOL "Kokkos builds for AMD HIP set the | ||
Kokkos_ENABLE_HIP variable to ON") | ||
#set(Kokkos_ARCH_VEGA900 ON CACHE BOOL "Docstring") #TODO: better | ||
endif() | ||
if(ENABLE_TARGET_OPENMP) | ||
set(Kokkos_ENABLE_OPENMPTARGET ON CACHE BOOL "Docstring") | ||
set(Kokkos_ARCH_VOLTA70 ON CACHE BOOL "Docstring") #TODO: better | ||
set(CMAKE_CXX_STANDARD 17) | ||
set(BLT_CXX_STANDARD 17) | ||
set(RAJA_ENABLE_TARGET_OPENMP ON CACHE BOOL "Docstring") | ||
if(NOT CMAKE_BUILD_TYPE MATCHES Debug) | ||
if(NOT EXPERIMENTAL_BUILD) | ||
message(FATAL_ERROR "Kokkos builds with OpenMPTarget require a Debug build to succeed at the moment. Rebuild with CMAKE_BUILD_TYPE=Debug. If you're a compiler developer, rebuild with -DEXPERIMENTAL_BUILD=ON") | ||
endif() | ||
endif() | ||
|
||
#add_definitions(-DRAJA_ENABLE_TARGET_OPENMP) | ||
endif() | ||
|
||
# ENABLE_CUDA IS A RAJA PERFSUITE OPTION | ||
if(ENABLE_CUDA) | ||
set(Kokkos_ENABLE_CUDA ON CACHE BOOL "Docstring") | ||
set(Kokkos_ENABLE_CUDA_LAMBDA ON CACHE BOOL "Docstring") | ||
set(Kokkos_ARCH_VOLTA70 ON CACHE BOOL "Docstring") #TODO: better | ||
enable_language(CUDA) | ||
endif() | ||
if(ENABLE_OPENMP) | ||
set(Kokkos_ENABLE_OPENMP ON CACHE BOOL "Docstring") | ||
endif() | ||
|
||
add_subdirectory(tpl/kokkos) | ||
get_property(KOKKOS_INCLUDE_DIRS DIRECTORY tpl/kokkos PROPERTY INCLUDE_DIRECTORIES) | ||
include_directories(${KOKKOS_INCLUDE_DIRS}) | ||
There was a problem hiding this comment. Choose a reason for hiding this commentThe reason will be displayed to describe this comment to others. Learn more. Is this required? I think that having |
||
|
||
list(APPEND RAJA_PERFSUITE_DEPENDS kokkos) | ||
endif() | ||
|
||
add_subdirectory(src) |
Submodule blt
updated
26 files
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,132 @@ | ||
{ | ||
"plots" : { | ||
"files" : { | ||
"fileName": "RAJAPerf*", | ||
"type" : "xml", | ||
"ignoreOldFiles" : true, | ||
"recurseDirectories" : true | ||
}, | ||
"categories": [ | ||
"kokkos_lambda", | ||
"lambda_cuda", | ||
"base_cuda", | ||
"base_seq", | ||
"lambda_seq", | ||
"raja_cuda", | ||
"raja_seq" | ||
], | ||
"plot" : [ | ||
{ | ||
"autoname" : { | ||
"useProperty" : "y/path", | ||
"formatByRemovingPrefix" : "\\/RAJAPerf\\d{4}\\-\\d{2}\\-\\d{2}T\\d{2}:\\d{2}:\\d{2}\\.xml\\/" | ||
}, | ||
"category" : "kokkos_lambda", | ||
"template" : "kokkos_template", | ||
"dataLines" : [ | ||
{ | ||
"name" : "Data Line", | ||
"template" : "Line_Template", | ||
"x" : { | ||
"getPath": "*", | ||
"getElement" : "performance-report", | ||
"getKey" : "date", | ||
"unit" : "timestamp" | ||
}, | ||
"y" : { | ||
"getElement" : "performance-report|timing", | ||
"getPath": "*/kokkos_perf_suite/*", | ||
"getPathAttribute": "name", | ||
"getKey" : "kokkos_lambda", | ||
"unit" : "seconds", | ||
"strategy" : { | ||
"getFirstMatchOnly" : "false", | ||
"recurseChildGraphs" : "true" | ||
} | ||
}, | ||
"color" : "202,77,77" | ||
} | ||
] | ||
},{ | ||
"inherit" : "kokkos_template", | ||
"category" : "base_cuda", | ||
"dataLines" : [ | ||
{ | ||
"inherit" : "Line_Template", | ||
"y" : { | ||
"getKey" : "base_cuda" | ||
} | ||
} | ||
] | ||
}, { | ||
"inherit" : "kokkos_template", | ||
"category" : "base_seq", | ||
"dataLines" : [ | ||
{ | ||
"inherit" : "Line_Template", | ||
"y" : { | ||
"getKey" : "base_seq" | ||
} | ||
} | ||
] | ||
}, { | ||
|
||
"inherit" : "kokkos_template", | ||
"category" : "lambda_seq", | ||
"dataLines" : [ | ||
{ | ||
"inherit" : "Line_Template", | ||
"y" : { | ||
"getKey" : "lambda_seq" | ||
} | ||
} | ||
] | ||
|
||
},{ | ||
"inherit" : "kokkos_template", | ||
"category" : "lambda_cuda", | ||
"dataLines" : [ | ||
{ | ||
"inherit" : "Line_Template", | ||
"y" : { | ||
"getKey" : "lambda_cuda" | ||
} | ||
} | ||
] | ||
},{ | ||
"inherit" : "kokkos_template", | ||
"category" : "raja_cuda", | ||
"dataLines" : [ | ||
{ | ||
"inherit" : "Line_Template", | ||
"y" : { | ||
"getKey" : "raja_cuda" | ||
} | ||
} | ||
] | ||
}, { | ||
"inherit" : "kokkos_template", | ||
"category" : "raja_seq", | ||
"dataLines" : [ | ||
{ | ||
"inherit" : "Line_Template", | ||
"y" : { | ||
"getKey" : "raja_seq" | ||
} | ||
} | ||
] | ||
} | ||
] | ||
}, | ||
"graphDisplay": { | ||
"dbLocation" : "root", | ||
"page" : 1, | ||
"displayCategory" : "kokkos_lambda", | ||
"displayRange" : 30, | ||
"graphWidth" : 450, | ||
"graphHeight" : 450, | ||
"graphsPerRow" : 3, | ||
"graphsPerPage" : 15, | ||
"displayedDecimalPlaces" : 3 | ||
} | ||
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,48 @@ | ||
#!/usr/bin/env bash | ||
|
||
############################################################################### | ||
# Copyright (c) 2017-20, Lawrence Livermore National Security, LLC | ||
# and RAJA Performance Suite project contributors. | ||
# See the RAJAPerf/COPYRIGHT file for details. | ||
# | ||
# SPDX-License-Identifier: (BSD-3-Clause) | ||
################################################################################# | ||
|
||
BUILD_SUFFIX=snl_rhel7-hipcc-4.0.0 | ||
|
||
rm -rf build_${BUILD_SUFFIX} 2>/dev/null | ||
mkdir build_${BUILD_SUFFIX} && cd build_${BUILD_SUFFIX} | ||
|
||
################################## | ||
#Caraway Build (AMD) | ||
################################# | ||
module purge | ||
|
||
module load cmake/3.19.3 | ||
|
||
module load git/2.9.4 | ||
|
||
################################## | ||
# FOR COMPUTE NODE (caraway04 GPU): | ||
|
||
module load rocm/4.0.0 | ||
|
||
module load python/3.7.3 | ||
|
||
cmake \ | ||
-DCMAKE_BUILD_TYPE=Release \ | ||
-DENABLE_KOKKOS=ON \ | ||
-DENABLE_HIP=ON \ | ||
-DKokkos_ARCH_VEGA900=ON \ | ||
-DCMAKE_CXX_FLAGS="--gcc-toolchain=/home/projects/x86-64/gcc/8.2.0/" \ | ||
-DHIP_HIPCC_FLAGS="--gcc-toolchain=/home/projects/x86-64/gcc/8.2.0/ -std=c++17" \ | ||
-DCMAKE_CXX_STANDARD=17 \ | ||
-DCMAKE_CXX_COMPILER=hipcc .. \ | ||
|
||
make -j24;make | ||
|
||
cd bin/ | ||
./raja-perf.exe | ||
|
||
|
||
|
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
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.
The ENABLE_CUDA here is redundant (since we are already inside an
if (ENABLE_CUDA)
block.