Fixed CI and doc

Author: shewu-quic

.ci/scripts/build-qnn-sdk.sh

@@ -11,7 +11,7 @@ set -o xtrace
 build_qnn_backend() {
   echo "Start building qnn backend."
   export ANDROID_NDK_ROOT=/opt/ndk
-  export QNN_SDK_ROOT=/tmp/qnn/2.25.0.240728
+  export QNN_SDK_ROOT=/tmp/qnn/2.28.0.241029
   export EXECUTORCH_ROOT="$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")/../.." && pwd)"
 
   bash backends/qualcomm/scripts/build.sh --skip_aarch64 --job_number 2 --release

.ci/scripts/setup-qnn-deps.sh

@@ -16,9 +16,9 @@ install_qnn() {
   QNN_INSTALLATION_DIR=/tmp/qnn
   mkdir -p "${QNN_INSTALLATION_DIR}"
 
-  curl -Lo /tmp/v2.25.0.24.07.28.zip "https://softwarecenter.qualcomm.com/api/download/software/qualcomm_neural_processing_sdk/v2.25.0.240728.zip"
+  curl -Lo /tmp/v2.28.0.24.10.29.zip "https://softwarecenter.qualcomm.com/api/download/software/qualcomm_neural_processing_sdk/v2.28.0.241029.zip"
   echo "Finishing downloading qnn sdk."
-  unzip -qo /tmp/v2.25.0.24.07.28.zip -d /tmp
+  unzip -qo /tmp/v2.28.0.24.10.29.zip -d /tmp
   echo "Finishing unzip qnn sdk."
 
 

.ci/scripts/test_llama.sh

@@ -75,7 +75,7 @@ echo "COREML option ${COREML}"
 if [[ "${MODE}" =~ .*qnn.* ]]; then
   QNN=ON
   export EXECUTORCH_ROOT="$(cd -- "$(dirname -- "${BASH_SOURCE[0]}")/.." && pwd)"
-  export QNN_SDK_ROOT=/tmp/qnn/2.25.0.240728
+  export QNN_SDK_ROOT=/tmp/qnn/2.28.0.241029
   export LD_LIBRARY_PATH="${QNN_SDK_ROOT}/lib/x86_64-linux-clang"
   export PYTHONPATH=".."
   cp schema/program.fbs exir/_serialize/program.fbs

docs/source/build-run-qualcomm-ai-engine-direct-backend.md

@@ -59,7 +59,7 @@ This example is verified with SM8550 and SM8450.
    - Click the "Get Software" button to download a version of QNN SDK.
    - However, at the moment of updating this tutorial, the above website doesn't provide QNN SDK newer than 2.22.6.
    - The below is public links to download various QNN versions. Hope they can be publicly discoverable soon.
-   - [QNN 2.26.0](https://softwarecenter.qualcomm.com/api/download/software/qualcomm_neural_processing_sdk/v2.26.0.240828.zip)
+   - [QNN 2.28.0](https://softwarecenter.qualcomm.com/api/download/software/qualcomm_neural_processing_sdk/v2.28.0.241029.zip)
 
 The directory with installed Qualcomm AI Engine Direct SDK looks like:
 ```

docs/source/llm/build-run-llama3-qualcomm-ai-engine-direct-backend.md

@@ -9,7 +9,7 @@ This tutorial demonstrates how to export Llama 3 8B Instruct for Qualcomm AI Eng
 - Follow [the README for executorch llama](https://github.com/pytorch/executorch/tree/main/examples/models/llama) to know how to run a llama model on mobile via ExecuTorch.
 - A Qualcomm device with 16GB RAM
   - We are continuing to optimize our memory usage to ensure compatibility with lower memory devices.
-- The version of [Qualcomm AI Engine Direct SDK](https://developer.qualcomm.com/software/qualcomm-ai-engine-direct-sdk) is 2.26.0 or above.
+- The version of [Qualcomm AI Engine Direct SDK](https://developer.qualcomm.com/software/qualcomm-ai-engine-direct-sdk) is 2.28.0 or above.
 
 ## Instructions