Skip to content

add mini cutlass build target for faster compile #78

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

Merged
JRPan merged 4 commits into from
Jan 26, 2026
Merged

add mini cutlass build target for faster compile #78

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
3e8b548
Update Makefile and setup_environment for enhanced Cutlass support an…
JRPan Jan 26, 2026
c629fdc
Enhance Cutlass build process in Makefile with submodule initializati…
JRPan Jan 26, 2026
6867e16
Update Makefile to include cutlass_mini in CI targets
JRPan Jan 26, 2026
9219d74
Update Makefile to change cutlass build directory and adjust submodul…
JRPan Jan 26, 2026
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
22 changes: 17 additions & 5 deletions src/Makefile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ else
all: pannotia rodinia_2.0-ft proxy-apps microbench rodinia-3.1 ispass-2009 polybench parboil shoc custom_apps
endif
endif
ci: rodinia_2.0-ft rodinia-3.1 GPU_Microbenchmark cutlass cuda_samples
ci: rodinia_2.0-ft rodinia-3.1 GPU_Microbenchmark cutlass_mini cuda_samples
accelwattch: accelwattch_validation accelwattch_hw_power accelwattch_ubench
accelwattch_validation: rodinia-3.1_accelwattch_validation parboil_accelwattch_validation cutlass cuda_samples-11.0_accelwattch_validation
accelwattch_hw_power: rodinia-3.1_hw_power parboil_hw_power cuda_samples-11.0_hw_power
Expand Down Expand Up @@ -430,15 +430,27 @@ clean_accelwattch_ubench:

cutlass:
mkdir -p $(BINDIR)/$(BINSUBDIR)/
echo $(BINDIR)
git submodule init && git submodule update
$(SETENV) mkdir -p cuda/cutlass-bench/build && cd cuda/cutlass-bench/build && cmake .. -DUSE_GPGPUSIM=1 -DCUTLASS_NVCC_ARCHS=80 -DCMAKE_CUDA_ARCHITECTURES=80 -DCMAKE_CUDA_COMPILER=$(CUDA_INSTALL_PATH)/bin/nvcc && $(MAKE) cutlass_profiler && $(MAKE) cutlass_examples
echo $(BINDIR)
cp cuda/cutlass-bench/build/tools/profiler/cutlass_profiler $(BINDIR)/$(BINSUBDIR)/cutlass_profiler
mkdir -p cuda/cutlass/build
$(SETENV) cmake -S cuda/cutlass-bench -B cuda/cutlass-bench/build -DCUTLASS_NVCC_ARCHS="80;90;90a" -DCMAKE_CUDA_ARCHITECTURES="80;90;90a" -DCMAKE_CUDA_COMPILER=$(CUDA_INSTALL_PATH)/bin/nvcc
$(SETENV) $(MAKE) -C cuda/cutlass-bench/build/ cutlass_examples cutlass_profiler
find cuda/cutlass-bench/build/examples -type f -executable -exec sh -c 'cp "$$1" "$(BINDIR)/$(BINSUBDIR)/cutlass_$$(basename "$$1")"' sh {} \;

cutlass_mini:
mkdir -p $(BINDIR)/$(BINSUBDIR)/
git submodule init && git submodule update
mkdir -p cuda/cutlass-bench/build
$(SETENV) cmake -S cuda/cutlass-bench -B cuda/cutlass-bench/build -DCUTLASS_NVCC_ARCHS="80;90;90a" -DCMAKE_CUDA_ARCHITECTURES="80;90;90a" -DCMAKE_CUDA_COMPILER=$(CUDA_INSTALL_PATH)/bin/nvcc
# add any other cutlass examples you want to build here
$(SETENV) $(MAKE) -C cuda/cutlass-bench/build/ 00_basic_gemm 48_hopper_warp_specialized_gemm 52_hopper_gather_scatter_fusion 57_hopper_grouped_gemm

find cuda/cutlass-bench/build/examples -type f -executable -exec sh -c 'cp "$$1" "$(BINDIR)/$(BINSUBDIR)/cutlass_$$(basename "$$1")"' sh {} \;






cutlass_examples_turing:
mkdir -p $(BINDIR)/$(BINSUBDIR)/
git submodule init && git submodule update
Expand Down
1 change: 1 addition & 0 deletions src/setup_environment
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -97,6 +97,7 @@ if [ $CUDA_VERSION_MAJOR -eq 12 ]; then
export CUDA_CPPFLAGS="$CUDA_CPPFLAGS -gencode=arch=compute_86,code=compute_86"
if [ $CUDA_VERSION_MINOR -ge 8 ]; then
export CUDA_CPPFLAGS="$CUDA_CPPFLAGS -gencode=arch=compute_90,code=compute_90"
export CUDA_CPPFLAGS="$CUDA_CPPFLAGS -gencode=arch=compute_90a,code=compute_90a"
export CUDA_CPPFLAGS="$CUDA_CPPFLAGS -gencode=arch=compute_100,code=compute_100"
fi
export CUDA_CPPFLAGS="$CUDA_CPPFLAGS -std=c++17"
Expand Down