doc updates

docs/executorch_setup.md

@@ -5,9 +5,8 @@ Before running any commands in torchchat that require ExecuTorch, you must first
 To install ExecuTorch, run the following commands *from the torchchat root directory*.
 
 ```
-export TORCHCHAT_ROOT=${PWD}
-export ENABLE_ET_PYBIND=true
-./scripts/install_et.sh $ENABLE_ET_PYBIND
+export TORCHCHAT_ROOT=$PWD
+./scripts/install_et.sh
 ```
 
 This will download the ExecuTorch repo to ./et-build/src and install various ExecuTorch libraries to ./et-build/install.

docs/runner_build.md

@@ -47,12 +47,7 @@ Options:
 To build runner-aoti, run the following commands *from the torchchat root directory*
 
 ```
-# Pull submodules (re2, abseil) for Tiktoken
-git submodule sync
-git submodule update --init
-
-cmake -S . -B ./cmake-out -G Ninja -DCMAKE_PREFIX_PATH=`python3 -c 'import torch;print(torch.utils.cmake_prefix_path)'`
-cmake --build ./cmake-out --target aoti_run
+./scripts/build_native.sh aoti
 ```
 
 After running these, the runner-aoti binary is located at ./cmake-out/aoti_run.
@@ -80,6 +75,12 @@ Before building runner-et, you must first setup ExecuTorch by following [setup E
 
 To build runner-et, run the following commands *from the torchchat root directory*
 
+```
+./scripts/build_native.sh et
+```
+
+Note: the above script will wipe ./et-build if present and re-install ExecuTorch to ./et-build, which can take a while.  If you already installed ExecuTorch, running the commands below will build the runner, without re-installing ExecuTorch from source:
+
 ```
 # Pull submodules (re2, abseil) for Tiktoken
 git submodule sync

scripts/build_native.sh

@@ -76,4 +76,4 @@ fi
 cmake -S . -B ./cmake-out -DCMAKE_PREFIX_PATH=`python -c 'import torch;print(torch.utils.cmake_prefix_path)'` -G Ninja
 cmake --build ./cmake-out --target "${TARGET}"_run
 
-printf "Build finished. Please run: \n./cmake-out/${TARGET}_run model.<pte|so> -z tokenizer.model -l <llama version (2 or 3)> -i <prompt>"
+printf "Build finished. Please run: \n./cmake-out/${TARGET}_run model.<pte|so> -z tokenizer.model -l <llama version (2 or 3)> -i <prompt>\n"