updated clone command to use http

Author: gregbolet

Dockerfile

@@ -22,7 +22,7 @@ RUN update-alternatives --install /usr/bin/clang++ clang++ /usr/bin/clang++-18 1
     update-alternatives --install /usr/bin/llvm-cxxfilt llvm-cxxfilt /usr/bin/llvm-cxxfilt-18 100
 
 # clone the repo into the container
-RUN git clone git@github.com:Scientific-Computing-Lab/gpuFLOPBench.git /gpu-flopbench
+RUN git clone https://github.com/Scientific-Computing-Lab/gpuFLOPBench.git /gpu-flopbench
 
 # get the LFS files
 RUN cd /gpu-flopbench && git checkout main && git pull && git lfs pull && git lfs fetch --all && git lfs checkout
@@ -59,4 +59,4 @@ RUN echo 'conda activate gpu-flopbench' >> ~/.bashrc
 ENV GPU_FLOPBENCH_ROOT=/gpu-flopbench
 
 # expose the Jupyter notebook port
-EXPOSE 8888
+EXPOSE 8888

README.md

@@ -91,7 +91,7 @@ git clone [email protected]:Scientific-Computing-Lab/gpuFLOPBench.git ./gpu-flopben
 # we only really need the Dockerfile from the repo
 cd ./gpu-flopbench
 
-# this takes about 2 minutes on my Macbook Air M4
+# this takes about 10 minutes on my Macbook Air M4
 docker build --platform=linux/amd64 --progress=plain -t 'gpu-flopbench' .
 
 # please make sure that the Settings > Resources > Network > 'Enable Host Networking' option is enabled on Docker Desktop
@@ -460,4 +460,4 @@ These also include directions we would like to move into though.
 - Only a few codes don't have their `main(...)` included in the source scrape due to us not including source files from other source directories.
 - Our version of LangGraph and LangChain libraries became quickly outdated in the process of this research; we opted to keep our original versions as everything was working, therefore we need to update our scripts to better match the 1.0 release of LangGraph/LangChain.
 - Extend the work to predict better on hidden FLOP counts.
-- Extend the work to predict OpenMP FLOP counts (this is partly why we built with `clang`, to have one compiler do it all).
+- Extend the work to predict OpenMP FLOP counts (this is partly why we built with `clang`, to have one compiler do it all).