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

Jump to bottom

cutlass update #18

Merged
merged 11 commits into from
May 7, 2024
Merged

cutlass update #18

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
bff309d
Cutlass Version update
Shreya-gaur Jul 10, 2023
fa65ff6
Merge with the dev branch
Shreya-gaur Jul 10, 2023
b7d0ed3
Merge branch 'dev' into temp
Shreya-gaur Jul 20, 2023
b95cecc
Removing the .github folder from cutlass-bench
Shreya-gaur Jul 20, 2023
50f4eae
Merge branch 'temp' of https://github.com/Shreya-gaur/gpu-app-collect…
Shreya-gaur Jul 20, 2023
e855843
Merge branch 'dev' into temp
Shreya-gaur Nov 29, 2023
7db9564
Updating cutlass to submodule format
Shreya-gaur Feb 7, 2024
e00d15b
Merge branch 'temp' of https://github.com/Shreya-gaur/gpu-app-collect…
Shreya-gaur Feb 7, 2024
60e781e
changes to the gitmodule
Shreya-gaur Feb 7, 2024
f662582
retesting
Shreya-gaur Feb 7, 2024
0c9dc3a
cutlass-bench readded
Shreya-gaur Feb 7, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
The table of contents is too big for display.
Diff view
Diff view
  •  
  •  
  •  
The diff you're trying to view is too large. We only load the first 3000 changed files.
3 changes: 3 additions & 0 deletions .gitmodules
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,6 @@
[submodule "src/cuda/cutlass-bench/tools/external/googletest"]
path = src/cuda/cutlass-bench/tools/external/googletest
url = https://github.com/google/googletest.git
[submodule "src/cuda/cutlass-bench"]
path = src/cuda/cutlass-bench
url = https://github.com/NVIDIA/cutlass.git
23 changes: 17 additions & 6 deletions src/Makefile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -426,13 +426,24 @@ deeplearning:

cutlass:
mkdir -p $(BINDIR)/$(BINSUBDIR)/
cd ../ && 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=70 && make cutlass_perf_test
cd cuda/cutlass-bench/build/tools/test/perf && ln -s -f ../../../../binary.sh . && ./binary.sh
cp cuda/cutlass-bench/build/tools/test/perf/cutlass_perf_test $(BINDIR)/$(BINSUBDIR)/
cp cuda/cutlass-bench/build/tools/test/perf/cutlass_perf_test $(BINDIR)/$(BINSUBDIR)/cutlass_perf_test_k1
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=70 && make cutlass_profiler -j12
$(SETENV) mkdir -p cuda/cutlass-bench/build && cd cuda/cutlass-bench/build && cmake .. -DUSE_GPGPUSIM=1 -DCUTLASS_NVCC_ARCHS=70
# cd cuda/cutlass-bench/build/tools/profiler && ln -s -f ../../../../binary.sh . && ./binary.sh
echo $(BINDIR)
cp cuda/cutlass-bench/build/tools/profiler/cutlass_profiler $(BINDIR)/$(BINSUBDIR)/cutlass_perf_test_k1
cp $(BINDIR)/$(BINSUBDIR)/cutlass_perf_test_k1 $(BINDIR)/$(BINSUBDIR)/cutlass_perf_test_k2
cp $(BINDIR)/$(BINSUBDIR)/cutlass_perf_test_k1 $(BINDIR)/$(BINSUBDIR)/cutlass_perf_test_k3


cutlass_examples_turing:
mkdir -p $(BINDIR)/$(BINSUBDIR)/
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=75 && cd ./examples/09_turing_tensorop_conv2dfprop && make 09_turing_tensorop_conv2dfprop
echo $(BINDIR)
cp cuda/cutlass-bench/build/examples/09_turing_tensorop_conv2dfprop/09_turing_tensorop_conv2dfprop $(BINDIR)/$(BINSUBDIR)/turing_tensorop_conv2dfrop

# Maybe we should use submodules for this - but I have heard a lot of horor stories about these..
# For now - lets just clone if we don't have it and set the SHA we want.
heterosync:
Expand Down Expand Up @@ -484,7 +495,7 @@ clean_heterosync:
rm -rf cuda/heterosync

clean_cutlass:
rm -fr cuda/cutlass-bench/build
rm -rf cuda/cutlass-bench/build

clean_deeplearning:
$(SETENV) make $(MAKE_ARGS) noinline=$(noinline) -C cuda/cudnn/mnist clean
Expand Down
1 change: 1 addition & 0 deletions src/cuda/cutlass-bench
Open in desktop
Submodule cutlass-bench added at 47a3eb
Loading