Update

[ghstack-poisoned]

Author: shoumikhin

backends/vulkan/test/op_tests/conv2d_gemm_dynamic_test.cpp

@@ -247,7 +247,6 @@ std::vector<float> conv2d_ref_xnnpack(
       out_min,
       out_max,
       /*flags=*/0,
-      /*code_cache=*/nullptr,
       /*weights_cache=*/nullptr,
       &op);
   if (create_status != xnn_status_success || op == nullptr) {
@@ -267,7 +266,6 @@ std::vector<float> conv2d_ref_xnnpack(
       static_cast<size_t>(H_in),
       static_cast<size_t>(W_in),
       &workspace_size,
-      &workspace_alignment,
       &out_h,
       &out_w,
       /*threadpool=*/nullptr);

examples/llm_server/python/TARGETS examples/llm_server/python/BUCKexamples/llm_server/python/TARGETS renamed to examples/llm_server/python/BUCK

@@ -1,8 +1,9 @@
 load("@fbsource//xplat/executorch/build:runtime_wrapper.bzl", "runtime")
+load("@fbcode_macros//build_defs:build_file_migration.bzl", "fbcode_target")
 
 oncall("executorch")
 
-runtime.python_library(
+fbcode_target(_kind = runtime.python_library,
     name = "lib",
     srcs = [
         "__init__.py",

examples/models/gemma4_31b/BUCK

@@ -0,0 +1,40 @@
+load("@fbsource//xplat/executorch/build:runtime_wrapper.bzl", "runtime")
+load("@fbcode_macros//build_defs:build_file_migration.bzl", "fbcode_target")
+
+oncall("executorch")
+
+fbcode_target(_kind = runtime.python_library,
+    name = "quant",
+    srcs = [
+        "cuda_packers.py",
+        "quant/quantize.py",
+        "quant/recipe.py",
+    ],
+    base_module = "executorch.examples.models.gemma4_31b",
+    visibility = ["PUBLIC"],
+    deps = [
+        "//caffe2:torch",
+        "//executorch/backends/cuda:coalesced_int4_tensor",
+        "//executorch/backends/cuda:dp4a_planar_int5_tensor",
+        "//executorch/backends/cuda:dp4a_planar_int6_tensor",
+        "//executorch/extension/llm/export:gguf",
+        "//executorch/extension/llm/export:int4",
+        "//executorch/extension/llm/export:quant",
+        "//pytorch/ao:torchao",
+        "fbsource//third-party/pypi/safetensors:safetensors",
+    ],
+)
+
+fbcode_target(_kind = runtime.python_library,
+    name = "quantize_and_save",
+    srcs = ["quantize_and_save.py"],
+    base_module = "executorch.examples.models.gemma4_31b",
+    visibility = ["PUBLIC"],
+    deps = [
+        "//caffe2:torch",
+        "//executorch/extension/llm/export:load",
+        "//executorch/extension/llm/export:quant",
+        "//pytorch/ao:torchao",
+        "fbsource//third-party/pypi/safetensors:safetensors",
+    ],
+)