Decouple custom ops in llama_transformer.py Part 2/N (#3007)

Summary:
Pull Request resolved: #3007

Keep llama_transformer.py to look like stock implementation, so that it can be reused everywhere.

Do module swap

Reviewed By: cccclai

Differential Revision: D56048640

fbshipit-source-id: 76de1b09b7f5d79422bb3b32bc830a9a7ecd935c

Author: mergennachin

examples/models/llama2/TARGETS

@@ -18,7 +18,6 @@ runtime.python_library(
     ],
     deps = [
         "//caffe2:torch",
-        "//executorch/examples/models/llama2/custom_ops:custom_ops_aot_py",
     ],
 )
 
@@ -86,6 +85,7 @@ runtime.python_library(
         "//executorch/backends/vulkan/partitioner:vulkan_partitioner",
         "//executorch/examples/models:model_base",
         "//executorch/examples/models:models",
+        "//executorch/examples/models/llama2/custom_ops:custom_ops_aot_py",
         "//executorch/examples/portable:utils",
         "//executorch/exir:lib",
         "//executorch/sdk/etrecord:etrecord",

examples/models/llama2/export_llama_lib.py

@@ -23,7 +23,11 @@
     XnnpackDynamicallyQuantizedPartitioner,
 )
 
-from executorch.examples.models.llama2.llama_transformer import Transformer
+from executorch.examples.models.llama2.llama_transformer import (
+    KVCache,
+    SDPA,
+    Transformer,
+)
 from executorch.exir.backend.backend_details import CompileSpec
 
 from executorch.sdk.etrecord import generate_etrecord
@@ -88,6 +92,58 @@ def materialze_broadcast_of_rope_freq_cis(
     return module
 
 
+class SDPACustom(torch.nn.Module):
+    def __init__(
+        self,
+        kv_cache: KVCache,
+        mask,
+        dim: int,
+    ):
+        super().__init__()
+        self.kv_cache = kv_cache
+        self.mask = mask
+        self.dim = dim
+
+    def forward(
+        self,
+        input_pos: torch.Tensor,
+        q: torch.Tensor,
+        k: torch.Tensor,
+        v: torch.Tensor,
+        bsz,
+        seqlen,
+    ):
+        output = torch.ops.llama.sdpa_with_kv_cache(
+            q,
+            k,
+            v,
+            self.kv_cache.k_cache,
+            self.kv_cache.v_cache,
+            input_pos[-1].item(),
+            seqlen,
+        )
+        return output.view(bsz, seqlen, self.dim)
+
+
+def _replace_sdpa_with_custom_op(module: torch.nn.Module):
+    for name, child in module.named_children():
+        if isinstance(child, SDPA):
+            setattr(
+                module,
+                name,
+                SDPACustom(child.kv_cache, child.mask, child.dim),
+            )
+        else:
+            _replace_sdpa_with_custom_op(child)
+
+
+def replace_sdpa_with_custom_op(module: torch.nn.Module) -> torch.nn.Module:
+    from executorch.examples.models.llama2.custom_ops import sdpa_with_kv_cache  # noqa
+
+    _replace_sdpa_with_custom_op(module)
+    return module
+
+
 def quantize(
     model: torch.nn.Module,
     qmode: str,
@@ -493,8 +549,7 @@ def _prepare_for_llama_export(modelname: str, args) -> LlamaEdgeManager:
         transforms.append(materialze_broadcast_of_rope_freq_cis)
 
     if args.use_sdpa_with_kv_cache:
-        pass
-        # TODO: Next diff transforms.append()
+        transforms.append(replace_sdpa_with_custom_op)
 
     return (
         load_llama_model(

examples/models/llama2/llama_transformer.py

@@ -214,14 +214,12 @@ def __init__(
         self,
         kv_cache: KVCache,
         mask,
-        use_sdpa_with_kv_cache_op: bool,
         dim: int,
         n_rep: int,
     ):
         super().__init__()
         self.kv_cache = kv_cache
         self.mask = mask
-        self.use_sdpa_with_kv_cache_op = use_sdpa_with_kv_cache_op
         self.dim = dim
         self.n_rep = n_rep
 
@@ -233,56 +231,6 @@ def forward(
         v: torch.Tensor,
         bsz,
         seqlen,
-    ) -> torch.Tensor:
-        if not self.use_sdpa_with_kv_cache_op:
-            return self._forward_default(
-                input_pos,
-                q,
-                k,
-                v,
-                bsz,
-                seqlen,
-            )
-        else:
-            return self._forward_custom(
-                input_pos,
-                q,
-                k,
-                v,
-                bsz,
-                seqlen,
-            )
-
-    def _forward_custom(
-        self,
-        input_pos: torch.Tensor,
-        q: torch.Tensor,
-        k: torch.Tensor,
-        v: torch.Tensor,
-        bsz,
-        seqlen,
-    ):
-        from .custom_ops import sdpa_with_kv_cache  # noqa
-
-        output = torch.ops.llama.sdpa_with_kv_cache(
-            q,
-            k,
-            v,
-            self.kv_cache.k_cache,
-            self.kv_cache.v_cache,
-            input_pos[-1].item(),
-            seqlen,
-        )
-        return output.view(bsz, seqlen, self.dim)
-
-    def _forward_default(
-        self,
-        input_pos: torch.Tensor,
-        q: torch.Tensor,
-        k: torch.Tensor,
-        v: torch.Tensor,
-        bsz,
-        seqlen,
     ) -> torch.Tensor:
         q = q.transpose(1, 2)  # (bs, n_local_heads, seqlen, head_dim)
         k = k.transpose(1, 2)
@@ -341,7 +289,6 @@ def __init__(self, args: ModelArgs, layer_id: int):
             self.SDPA = SDPA(
                 self.kv_cache,
                 self.mask,
-                args.use_sdpa_with_kv_cache_op,
                 self.dim,
                 self.n_rep,
             )