Skip to content

Rename kernels_custom to kernels_llm #12389

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
merged 2 commits into from
Jul 11, 2025
Merged

Rename kernels_custom to kernels_llm #12389

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
a506c2b
Rename kernels_custom to kernels_llm
shoumikhin Jul 11, 2025
d1a924e
Merge branch 'main' into rename-kernels-custom
shoumikhin Jul 11, 2025
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
0 .Package.swift/kernels_custom/dummy.swift → .Package.swift/kernels_llm/dummy.swift
View file
Open in desktop
File renamed without changes.
0 ...ge.swift/kernels_custom_debug/dummy.swift → .Package.swift/kernels_llm_debug/dummy.swift
View file
Open in desktop
File renamed without changes.
4 changes: 2 additions & 2 deletions .ci/scripts/test_llama_torchao_lowbit.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,15 +35,15 @@ cmake -DPYTHON_EXECUTABLE=python \
-DEXECUTORCH_BUILD_XNNPACK=OFF \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-Bcmake-out .
cmake --build cmake-out -j16 --target install --config Release

# Install llama runner with torchao
cmake -DPYTHON_EXECUTABLE=python \
-DBUILD_TESTING=OFF \
-DCMAKE_BUILD_TYPE=Release \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_XNNPACK=OFF \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
Expand Down
4 changes: 2 additions & 2 deletions .ci/scripts/test_llava.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -39,7 +39,7 @@ EXECUTORCH_COMMON_CMAKE_ARGS=" \
-DEXECUTORCH_BUILD_EXTENSION_DATA_LOADER=ON \
-DEXECUTORCH_BUILD_EXTENSION_FLAT_TENSOR=ON \
-DEXECUTORCH_BUILD_EXTENSION_TENSOR=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_XNNPACK=ON \
Expand Down Expand Up @@ -69,7 +69,7 @@ LLAVA_COMMON_CMAKE_ARGS=" \
-DPYTHON_EXECUTABLE="$PYTHON_EXECUTABLE" \
-DCMAKE_INSTALL_PREFIX=${BUILD_DIR} \
-DCMAKE_BUILD_TYPE=${CMAKE_BUILD_TYPE} \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_XNNPACK=ON"

Expand Down
4 changes: 2 additions & 2 deletions .ci/scripts/test_phi_3_mini.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,7 +33,7 @@ cmake_install_executorch_libraries() {
-DEXECUTORCH_BUILD_XNNPACK=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-B${BUILD_DIR} .

cmake --build ${BUILD_DIR} -j${NPROC} --target install --config ${BUILD_TYPE}
Expand All @@ -43,7 +43,7 @@ cmake_build_phi_3_mini() {
cmake -DPYTHON_EXECUTABLE=$PYTHON_EXECUTABLE \
-DCMAKE_INSTALL_PREFIX=${BUILD_DIR} \
-DCMAKE_BUILD_TYPE=${BUILD_TYPE} \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_XNNPACK=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/apple.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -152,7 +152,7 @@ jobs:
"backend_coreml"
"backend_mps"
"backend_xnnpack"
"kernels_custom"
"kernels_llm"
"kernels_optimized"
"kernels_quantized"
"threadpool"
Expand Down
2 changes: 1 addition & 1 deletion .github/workflows/trunk.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -645,7 +645,7 @@ jobs:
-DEXECUTORCH_BUILD_XNNPACK=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DEXECUTORCH_BUILD_DEVTOOLS=ON \
-DEXECUTORCH_ENABLE_EVENT_TRACER=ON \
-Bcmake-out .
Expand Down
4 changes: 2 additions & 2 deletions CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -677,7 +677,7 @@ if(EXECUTORCH_BUILD_EXTENSION_TRAINING)
add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/extension/training)
endif()

if(EXECUTORCH_BUILD_KERNELS_CUSTOM)
if(EXECUTORCH_BUILD_KERNELS_LLM)
# TODO: move all custom kernels to ${CMAKE_CURRENT_SOURCE_DIR}/kernels/custom
add_subdirectory(${CMAKE_CURRENT_SOURCE_DIR}/extension/llm/custom_ops)
endif()
Expand All @@ -704,7 +704,7 @@ if(EXECUTORCH_BUILD_EXECUTOR_RUNNER)
list(APPEND _executor_runner_libs quantized_ops_lib)
endif()

if(EXECUTORCH_BUILD_KERNELS_CUSTOM)
if(EXECUTORCH_BUILD_KERNELS_LLM)
list(APPEND _executor_runner_libs $<LINK_LIBRARY:WHOLE_ARCHIVE,custom_ops>)
endif()

Expand Down
2 changes: 1 addition & 1 deletion Package.swift
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,7 +69,7 @@ let products = deliverables([
"c++",
],
],
"kernels_custom": [:],
"kernels_llm": [:],
"kernels_optimized": [
"frameworks": [
"Accelerate",
Expand Down
4 changes: 2 additions & 2 deletions backends/cadence/build_cadence_fusionG3.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,7 +47,7 @@ if $STEPWISE_BUILD; then
-DEXECUTORCH_ENABLE_PROGRAM_VERIFICATION=ON \
-DEXECUTORCH_USE_DL=OFF \
-DEXECUTORCH_BUILD_PORTABLE_OPS=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=OFF \
-DEXECUTORCH_BUILD_KERNELS_LLM=OFF \
-DPYTHON_EXECUTABLE=python3 \
-DEXECUTORCH_FUSION_G3_OPT=ON \
-DHAVE_FNMATCH_H=OFF \
Expand All @@ -72,7 +72,7 @@ else
-DEXECUTORCH_ENABLE_PROGRAM_VERIFICATION=ON \
-DEXECUTORCH_USE_DL=OFF \
-DEXECUTORCH_BUILD_PORTABLE_OPS=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=OFF \
-DEXECUTORCH_BUILD_KERNELS_LLM=OFF \
-DPYTHON_EXECUTABLE=python3 \
-DEXECUTORCH_FUSION_G3_OPT=ON \
-DHAVE_FNMATCH_H=OFF \
Expand Down
4 changes: 2 additions & 2 deletions backends/cadence/build_cadence_hifi4.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ if $STEPWISE_BUILD; then
-DEXECUTORCH_ENABLE_PROGRAM_VERIFICATION=ON \
-DEXECUTORCH_USE_DL=OFF \
-DEXECUTORCH_BUILD_PORTABLE_OPS=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=OFF \
-DEXECUTORCH_BUILD_KERNELS_LLM=OFF \
-DPYTHON_EXECUTABLE=python3 \
-DEXECUTORCH_NNLIB_OPT=ON \
-DHAVE_FNMATCH_H=OFF \
Expand All @@ -70,7 +70,7 @@ else
-DEXECUTORCH_ENABLE_PROGRAM_VERIFICATION=ON \
-DEXECUTORCH_USE_DL=OFF \
-DEXECUTORCH_BUILD_PORTABLE_OPS=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=OFF \
-DEXECUTORCH_BUILD_KERNELS_LLM=OFF \
-DPYTHON_EXECUTABLE=python3 \
-DEXECUTORCH_NNLIB_OPT=ON \
-DHAVE_FNMATCH_H=OFF \
Expand Down
4 changes: 2 additions & 2 deletions backends/vulkan/docs/android_demo.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -91,7 +91,7 @@ binary using the Android NDK toolchain.
-DEXECUTORCH_BUILD_EXTENSION_TENSOR=ON \
-DEXECUTORCH_BUILD_VULKAN=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DPYTHON_EXECUTABLE=python \
-Bcmake-android-out && \
cmake --build cmake-android-out -j16 --target install)
Expand All @@ -102,7 +102,7 @@ binary using the Android NDK toolchain.
-DCMAKE_TOOLCHAIN_FILE=$ANDROID_NDK/build/cmake/android.toolchain.cmake \
-DANDROID_ABI=$ANDROID_ABI \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DCMAKE_INSTALL_PREFIX=cmake-android-out \
-DPYTHON_EXECUTABLE=python \
-Bcmake-android-out/examples/models/llama && \
Expand Down
4 changes: 2 additions & 2 deletions docs/source/llm/build-run-llama3-qualcomm-ai-engine-direct-backend.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -58,7 +58,7 @@ llama3/Meta-Llama-3-8B-Instruct/tokenizer.model -p <path_to_params.json> -c <pat
-DQNN_SDK_ROOT=${QNN_SDK_ROOT} \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-Bcmake-android-out .

cmake --build cmake-android-out -j16 --target install --config Release
Expand All @@ -73,7 +73,7 @@ llama3/Meta-Llama-3-8B-Instruct/tokenizer.model -p <path_to_params.json> -c <pat
-DEXECUTORCH_BUILD_QNN=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-Bcmake-android-out/examples/models/llama examples/models/llama

cmake --build cmake-android-out/examples/models/llama -j16 --config Release
Expand Down
2 changes: 1 addition & 1 deletion docs/source/using-executorch-building-from-source.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -376,7 +376,7 @@ cmake . `
-DEXECUTORCH_BUILD_FLATC=ON `
-DEXECUTORCH_BUILD_PYBIND=OFF `
-DEXECUTORCH_BUILD_XNNPACK=ON `
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON `
-DEXECUTORCH_BUILD_KERNELS_LLM=ON `
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON `
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON `
-DEXECUTORCH_ENABLE_LOGGING=ON `
Expand Down
2 changes: 1 addition & 1 deletion docs/source/using-executorch-ios.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@ The ExecuTorch Runtime for iOS and macOS (ARM64) is distributed as a collection
* `backend_coreml` - Core ML backend
* `backend_mps` - MPS backend
* `backend_xnnpack` - XNNPACK backend
* `kernels_custom` - Custom kernels for LLMs
* `kernels_llm` - Custom kernels for LLMs
* `kernels_optimized` - Accelerated generic CPU kernels
* `kernels_quantized` - Quantized kernels

Expand Down
6 changes: 3 additions & 3 deletions examples/demo-apps/android/LlamaDemo/docs/delegates/qualcomm_README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,7 +69,7 @@ cmake -DPYTHON_EXECUTABLE=python \
-DQNN_SDK_ROOT=${QNN_SDK_ROOT} \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-Bcmake-out .
cmake --build cmake-out -j16 --target install --config Release
```
Expand All @@ -86,7 +86,7 @@ cmake -DPYTHON_EXECUTABLE=python \
-DCMAKE_BUILD_TYPE=Release \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DEXECUTORCH_BUILD_EXTENSION_TENSOR=ON \
-DEXECUTORCH_BUILD_QNN=ON \
-Bcmake-out/examples/models/llama \
Expand Down Expand Up @@ -238,4 +238,4 @@ If the app successfully run on your device, you should see something like below:
</p>

## Reporting Issues
If you encountered any bugs or issues following this tutorial please file a bug/issue here on Github.
If you encountered any bugs or issues following this tutorial please file a bug/issue here on Github.
22 changes: 11 additions & 11 deletions examples/demo-apps/apple_ios/LLaMA/LLaMA.xcodeproj/project.pbxproj
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -30,8 +30,8 @@
03CF439E2CEC5CEC00C7113B /* backend_xnnpack in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF439D2CEC5CEC00C7113B /* backend_xnnpack */; };
03CF43A02CEC5CEC00C7113B /* backend_xnnpack_debug in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF439F2CEC5CEC00C7113B /* backend_xnnpack_debug */; };
03CF43A22CEC5CEC00C7113B /* executorch_debug in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF43A12CEC5CEC00C7113B /* executorch_debug */; };
03CF43A42CEC5CEC00C7113B /* kernels_custom in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF43A32CEC5CEC00C7113B /* kernels_custom */; };
03CF43A62CEC5CEC00C7113B /* kernels_custom_debug in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF43A52CEC5CEC00C7113B /* kernels_custom_debug */; };
03CF43A42CEC5CEC00C7113B /* kernels_llm in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF43A32CEC5CEC00C7113B /* kernels_llm */; };
03CF43A62CEC5CEC00C7113B /* kernels_llm_debug in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF43A52CEC5CEC00C7113B /* kernels_llm_debug */; };
03CF43A82CEC5CEC00C7113B /* kernels_optimized in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF43A72CEC5CEC00C7113B /* kernels_optimized */; };
03CF43AA2CEC5CEC00C7113B /* kernels_optimized_debug in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF43A92CEC5CEC00C7113B /* kernels_optimized_debug */; };
03CF43B02CEC5CEC00C7113B /* kernels_quantized in Frameworks */ = {isa = PBXBuildFile; productRef = 03CF43AF2CEC5CEC00C7113B /* kernels_quantized */; };
Expand Down Expand Up @@ -185,8 +185,8 @@
03CF439E2CEC5CEC00C7113B /* backend_xnnpack in Frameworks */,
03CF43A02CEC5CEC00C7113B /* backend_xnnpack_debug in Frameworks */,
03CF43A22CEC5CEC00C7113B /* executorch_debug in Frameworks */,
03CF43A42CEC5CEC00C7113B /* kernels_custom in Frameworks */,
03CF43A62CEC5CEC00C7113B /* kernels_custom_debug in Frameworks */,
03CF43A42CEC5CEC00C7113B /* kernels_llm in Frameworks */,
03CF43A62CEC5CEC00C7113B /* kernels_llm_debug in Frameworks */,
03CF43A82CEC5CEC00C7113B /* kernels_optimized in Frameworks */,
03CF43AA2CEC5CEC00C7113B /* kernels_optimized_debug in Frameworks */,
03CF43B02CEC5CEC00C7113B /* kernels_quantized in Frameworks */,
Expand Down Expand Up @@ -481,8 +481,8 @@
03CF439D2CEC5CEC00C7113B /* backend_xnnpack */,
03CF439F2CEC5CEC00C7113B /* backend_xnnpack_debug */,
03CF43A12CEC5CEC00C7113B /* executorch_debug */,
03CF43A32CEC5CEC00C7113B /* kernels_custom */,
03CF43A52CEC5CEC00C7113B /* kernels_custom_debug */,
03CF43A32CEC5CEC00C7113B /* kernels_llm */,
03CF43A52CEC5CEC00C7113B /* kernels_llm_debug */,
03CF43A72CEC5CEC00C7113B /* kernels_optimized */,
03CF43A92CEC5CEC00C7113B /* kernels_optimized_debug */,
03CF43AF2CEC5CEC00C7113B /* kernels_quantized */,
Expand Down Expand Up @@ -948,7 +948,7 @@
isa = XCRemoteSwiftPackageReference;
repositoryURL = "https://github.com/pytorch/executorch";
requirement = {
branch = "swiftpm-0.8.0.20250702";
branch = "swiftpm-0.8.0.20250711";
kind = branch;
};
};
Expand Down Expand Up @@ -990,15 +990,15 @@
package = 03CF43942CEC5CEC00C7113B /* XCRemoteSwiftPackageReference "executorch" */;
productName = executorch_debug;
};
03CF43A32CEC5CEC00C7113B /* kernels_custom */ = {
03CF43A32CEC5CEC00C7113B /* kernels_llm */ = {
isa = XCSwiftPackageProductDependency;
package = 03CF43942CEC5CEC00C7113B /* XCRemoteSwiftPackageReference "executorch" */;
productName = kernels_custom;
productName = kernels_llm;
};
03CF43A52CEC5CEC00C7113B /* kernels_custom_debug */ = {
03CF43A52CEC5CEC00C7113B /* kernels_llm_debug */ = {
isa = XCSwiftPackageProductDependency;
package = 03CF43942CEC5CEC00C7113B /* XCRemoteSwiftPackageReference "executorch" */;
productName = kernels_custom_debug;
productName = kernels_llm_debug;
};
03CF43A72CEC5CEC00C7113B /* kernels_optimized */ = {
isa = XCSwiftPackageProductDependency;
Expand Down
2 changes: 1 addition & 1 deletion examples/demo-apps/apple_ios/LLaMA/LLaMA/SupportingFiles/Debug.xcconfig
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ OTHER_LDFLAGS = $(inherited) \
-force_load $(BUILT_PRODUCTS_DIR)/libbackend_coreml_debug_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libbackend_mps_debug_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libbackend_xnnpack_debug_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libkernels_custom_debug_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libkernels_llm_debug_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libkernels_optimized_debug_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libkernels_quantized_debug_$(ET_PLATFORM).a \
@$(TEMP_DIR)/cmake/linker_flags
Expand Down
2 changes: 1 addition & 1 deletion examples/demo-apps/apple_ios/LLaMA/LLaMA/SupportingFiles/Release.xcconfig
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,7 +9,7 @@ OTHER_LDFLAGS = $(inherited) \
-force_load $(BUILT_PRODUCTS_DIR)/libbackend_coreml_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libbackend_mps_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libbackend_xnnpack_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libkernels_custom_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libkernels_llm_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libkernels_optimized_$(ET_PLATFORM).a \
-force_load $(BUILT_PRODUCTS_DIR)/libkernels_quantized_$(ET_PLATFORM).a \
@$(TEMP_DIR)/cmake/linker_flags
Expand Down
2 changes: 1 addition & 1 deletion examples/demo-apps/apple_ios/LLaMA/docs/delegates/xnnpack_README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -147,7 +147,7 @@ The following packages should be linked in your app target `LLaMA` (left side, L
- backend_coreml
- backend_mps
- backend_xnnpack
- kernels_custom
- kernels_llm
- kernels_optimized
- kernels_portable
- kernels_quantized
Expand Down
8 changes: 4 additions & 4 deletions examples/models/llama/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -284,7 +284,7 @@ cmake -DCMAKE_TOOLCHAIN_FILE=$ANDROID_NDK/build/cmake/android.toolchain.cmake \
-DEXECUTORCH_BUILD_XNNPACK=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-Bcmake-out-android .

cmake --build cmake-out-android -j16 --target install --config Release
Expand All @@ -301,7 +301,7 @@ cmake -DCMAKE_TOOLCHAIN_FILE=$ANDROID_NDK/build/cmake/android.toolchain.cmake \
-DEXECUTORCH_BUILD_XNNPACK=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DSUPPORT_REGEX_LOOKAHEAD=ON
-Bcmake-out-android/examples/models/llama \
examples/models/llama
Expand Down Expand Up @@ -386,7 +386,7 @@ cmake -DPYTHON_EXECUTABLE=python \
-DEXECUTORCH_BUILD_XNNPACK=OFF \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-Bcmake-out .
cmake --build cmake-out -j16 --target install --config Release
```
Expand All @@ -396,7 +396,7 @@ Next install the llama runner with torchao kernels enabled (similar to step 3.2
```
cmake -DPYTHON_EXECUTABLE=python \
-DCMAKE_BUILD_TYPE=Release \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_XNNPACK=OFF \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
Expand Down
2 changes: 1 addition & 1 deletion examples/models/llava/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,7 +107,7 @@ endif()
target_link_options_shared_lib(quantized_ops_lib)
list(APPEND link_libraries quantized_kernels quantized_ops_lib)

if(EXECUTORCH_BUILD_KERNELS_CUSTOM)
if(EXECUTORCH_BUILD_KERNELS_LLM)
list(APPEND link_libraries $<LINK_LIBRARY:WHOLE_ARCHIVE,custom_ops>)
endif()

Expand Down
4 changes: 2 additions & 2 deletions examples/models/phi-3-mini/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -32,7 +32,7 @@ python -m examples.models.phi-3-mini.export_phi-3-mini -c "4k" -s 128 -o phi-3-m
-DEXECUTORCH_BUILD_XNNPACK=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-Bcmake-out .

cmake --build cmake-out -j16 --target install --config Release
Expand All @@ -42,7 +42,7 @@ python -m examples.models.phi-3-mini.export_phi-3-mini -c "4k" -s 128 -o phi-3-m
cmake -DPYTHON_EXECUTABLE=python \
-DCMAKE_INSTALL_PREFIX=cmake-out \
-DCMAKE_BUILD_TYPE=Release \
-DEXECUTORCH_BUILD_KERNELS_CUSTOM=ON \
-DEXECUTORCH_BUILD_KERNELS_LLM=ON \
-DEXECUTORCH_BUILD_KERNELS_OPTIMIZED=ON \
-DEXECUTORCH_BUILD_XNNPACK=ON \
-DEXECUTORCH_BUILD_KERNELS_QUANTIZED=ON \
Expand Down
2 changes: 1 addition & 1 deletion extension/android/CMakeLists.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -120,7 +120,7 @@ if(TARGET vulkan_backend)
list(APPEND link_libraries vulkan_backend)
endif()

if(EXECUTORCH_BUILD_KERNELS_CUSTOM)
if(EXECUTORCH_BUILD_KERNELS_LLM)
list(APPEND link_libraries $<LINK_LIBRARY:WHOLE_ARCHIVE,custom_ops>)
endif()

Expand Down
Loading
Loading