Skip to content
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.

Sign up for GitHub

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

Cleanup CUDA, Reuse Memory, Add Serial Model, Cleaup Std Parallelism #202

Open
wants to merge 14 commits into
base: develop
Choose a base branch
from
+1,412 −1,781
Open

Cleanup CUDA, Reuse Memory, Add Serial Model, Cleaup Std Parallelism #202

Changes from 1 commit
Commits
Show all changes
14 commits
Select commit Hold shift + click to select a range
759f7b1
Refactor CUDA memory allocation
gonzalobg May 27, 2024
293ed77
Refactor CUDA kernels and support block-stride loops
gonzalobg May 27, 2024
13e870f
Bugfix: division by zero in solution-check for individual benchmarks
gonzalobg May 28, 2024
46b6d41
Print order used
gonzalobg May 30, 2024
51231ac
Print device peak BW using NVML
gonzalobg May 30, 2024
321ba62
Capitalize order options for consistency with benchmarks
gonzalobg May 30, 2024
616b3b2
Move host-side allocation to benchmarks and reuse device with UVM
gonzalobg May 30, 2024
2929a6c
Add serial model
tom91136 Apr 16, 2024
9ff46ec
Update Serial
gonzalobg Jun 5, 2024
b3786f6
Merge C++ Standard Parallelism and SYCL2020 implementations
gonzalobg Jun 5, 2024
62a5051
Add support for AdaptiveCpp to C++ Standard Parallelism
gonzalobg Jun 5, 2024
6f0c2a1
Add support for hipstdpar
gsitaram Apr 29, 2024
084ef3b
Arm toolchains prefer -mcpu=native
gonzalobg Aug 14, 2024
4cebcde
Support NVHPC stdpar=multicore and AMDGPU interpose-alloc
gonzalobg Aug 14, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Arm toolchains prefer -mcpu=native
@gonzalobg
gonzalobg committed Aug 14, 2024

Unverified

No user is associated with the committer email.
Learn about vigilant mode
commit 084ef3bf724231e97b5d143d93981441f4c5efef
4 changes: 4 additions & 0 deletions CMakeLists.txt
View file
Original file line number Diff line number Diff line change
@@ -50,7 +50,11 @@ option(BUILD_NATIVE "Builds for the current systems CPU and GPU architecture." O
set(DEFAULT_DEBUG_FLAGS -O2 -fno-omit-frame-pointer)
set(DEFAULT_RELEASE_FLAGS -O3)
if (BUILD_NATIVE)
if(CMAKE_SYSTEM_PROCESSOR STREQUAL aarch64)
set(DEFAULT_RELEASE_FLAGS ${DEFAULT_RELEASE_FLAGS} -mcpu=native)
else()
set(DEFAULT_RELEASE_FLAGS ${DEFAULT_RELEASE_FLAGS} -march=native)
endif()
endif()

macro(hint_flag FLAG DESCRIPTION)