Add buck file for static llama

Summary: As title, moves the buck file related changes in D66107964 to here.

Differential Revision: D67057242

Author: cccclai

examples/qualcomm/TARGETS

@@ -20,3 +20,12 @@ runtime.python_binary(
         "//executorch/extension/export_util:export_util",
     ],
 )
+
+python_library(
+    name = "utils",
+    srcs = ["utils.py"],
+    deps = [
+        "//executorch/backends/qualcomm/partition:partition",
+        "//executorch/backends/qualcomm/quantizer:quantizer",
+    ],
+)

examples/qualcomm/oss_scripts/llama2/TARGETS

@@ -0,0 +1,35 @@
+# Any targets that should be shared between fbcode and xplat must be defined in
+# targets.bzl. This file can contain xplat-only targets.
+
+load("@fbsource//xplat/executorch/backends/qualcomm/qnn_version.bzl", "get_qnn_library_verision")
+load("@fbcode_macros//build_defs:python_binary.bzl", "python_binary")
+load("@fbsource//xplat/executorch/build:runtime_wrapper.bzl", "runtime")
+
+python_binary(
+    name = "llama",
+    srcs = ["llama.py"],
+    main_function = "executorch.examples.qualcomm.oss_scripts.llama3_2.llama.main",
+    preload_deps = [
+        "//executorch/extension/llm/custom_ops:model_sharding_py",
+    ],
+    deps = [
+        "//executorch/examples/qualcomm/oss_scripts/llama2:static_llama",
+        "//caffe2:torch",
+        "//executorch/extension/pybindings:aten_lib",
+        "//executorch/backends/qualcomm/partition:partition",
+        "//executorch/backends/qualcomm/quantizer:quantizer",
+        "//executorch/devtools:lib",
+        "//executorch/examples/models:models",
+        "//executorch/examples/qualcomm:utils",
+        "//executorch/extension/export_util:export_util",
+        "//executorch/extension/llm/export:export_lib",
+    ],
+)
+
+runtime.command_alias(
+    name = "llama_qnn",
+    env = {
+        "LD_LIBRARY_PATH": "$(location fbsource//third-party/qualcomm/qnn/qnn-{0}:qnn_offline_compile_libs)".format(get_qnn_library_verision()),
+    },
+    exe = ":llama",
+)

examples/qualcomm/oss_scripts/llama3_2/TARGETS

@@ -1,8 +1,44 @@
-# Any targets that should be shared between fbcode and xplat must be defined in
-# targets.bzl. This file can contain xplat-only targets.
-
-load(":targets.bzl", "define_common_targets")
+load("@fbcode_macros//build_defs:python_library.bzl", "python_library")
+load("@fbcode_macros//build_defs:python_unittest.bzl", "python_unittest")
+load("@fbsource//xplat/executorch/backends/qualcomm/qnn_version.bzl", "get_qnn_library_verision")
+load("@fbcode_macros//build_defs:python_binary.bzl", "python_binary")
+load("@fbsource//xplat/executorch/build:runtime_wrapper.bzl", "runtime")
 
 oncall("executorch")
 
-define_common_targets()
+
+python_library(
+    name = "static_llama",
+    srcs = [
+        "model/static_llama.py",
+    ],
+    deps = [
+        "//caffe2:torch",
+    ],
+)
+
+python_binary(
+    name = "llama",
+    srcs = ["llama.py"],
+    main_function = "executorch.examples.qualcomm.oss_scripts.llama2.llama.main",
+    deps = [
+        ":static_llama",
+        "//caffe2:torch",
+        "//executorch/extension/pybindings:aten_lib",
+        "//executorch/backends/qualcomm/partition:partition",
+        "//executorch/backends/qualcomm/quantizer:quantizer",
+        "//executorch/devtools:lib",
+        "//executorch/examples/models:models",
+        "//executorch/examples/qualcomm:utils",
+        "//executorch/extension/export_util:export_util",
+        "//executorch/extension/llm/export:export_lib",
+    ],
+)
+
+runtime.command_alias(
+    name = "llama_qnn",
+    env = {
+        "LD_LIBRARY_PATH": "$(location fbsource//third-party/qualcomm/qnn/qnn-{0}:qnn_offline_compile_libs)".format(get_qnn_library_verision()),
+    },
+    exe = ":llama",
+)