diff --git a/docs/developer_guide/contribution_guide.md b/docs/developer_guide/contribution_guide.md
index e2171f447..f8e6f692d 100644
--- a/docs/developer_guide/contribution_guide.md
+++ b/docs/developer_guide/contribution_guide.md
@@ -72,8 +72,13 @@ If you modify files protected by code owners, their approval is required to merg
 - Avoid code duplication. If the same code snippet (more than five lines) appears multiple times, extract it into a shared function.
 - Minimize device synchronization. Reduce expensive CPU-GPU synchronization operations, such as `tensor.item()` or `tensor.cpu()`, whenever possible. Use vectorized code.
 - Keep files concise. If a file exceeds 2,000 lines of code, split it into multiple smaller files.
-- Prioritize extreme efficiency. SGLang is a runtime, and most of your code runs on the critical path for every request. Optimize every minor overhead as much as possible.
-- Try to make functions as pure as possible. Avoid in-place modification of arguments.
+- Prioritize extreme efficiency. SGLang is a runtime, and most of your code runs on the critical path for every request. Optimize all minor overheads as much as possible, especially in the model forward code.
+  - A common pattern is some runtime checks in the model forward pass (e.g., [this](https://github.com/sgl-project/sglang/blob/f1b0eda55c2c4838e8ab90a0fac7fb1e3d7064ab/python/sglang/srt/models/deepseek_v2.py#L486-L491)). These are very likely the same for every layer. Please cache the result as a single boolean value whenever possible.
+- Strive to make functions as pure as possible. Avoid in-place modification of arguments.
+- When supporting new hardware or features, follow these guidelines:
+  - Do not drastically change existing code.
+  - Always prefer new files to introduce specific components for your new hardware (e.g., `allocator_ascend.py`).
+  - If you write multiple if/else blocks for new features, ensure the common path (e.g., NVIDIA hardware or the existing code path) is the first branch.
 
 ## How to update sgl-kernel
 Since sglang and sgl-kernel are separate Python packages, our current GitHub CI infrastructure does not support updating a kernel and using it immediately within the same pull request (PR).
diff --git a/python/sglang/srt/server_args.py b/python/sglang/srt/server_args.py
index 78515e898..c24c63ce9 100644
--- a/python/sglang/srt/server_args.py
+++ b/python/sglang/srt/server_args.py
@@ -297,9 +297,6 @@ class ServerArgs:
 
     def __post_init__(self):
         # Check deprecated arguments
-        def print_deprecated_warning(message: str):
-            logger.warning(f"\033[33m{message}\033[0m")
-
         if self.enable_ep_moe:
             self.ep_size = self.tp_size
             print_deprecated_warning(
@@ -1955,24 +1952,25 @@ class ServerArgs:
             default=None,
             help="The custom dataloader which used to update the model. Should be set with a valid import path, such as my_package.weight_load_func",
         )
+        parser.add_argument(
+            "--weight-loader-disable-mmap",
+            action="store_true",
+            help="Disable mmap while loading weight using safetensors.",
+        )
+
+        # For PD-Multiplexing
         parser.add_argument(
             "--enable-pdmux",
             action="store_true",
             help="Enable PD-Multiplexing, PD running on greenctx stream.",
         )
 
-        # For PD-Multiplexing
         parser.add_argument(
             "--sm-group-num",
             type=int,
             default=ServerArgs.sm_group_num,
             help="Number of sm partition groups.",
         )
-        parser.add_argument(
-            "--weight-loader-disable-mmap",
-            action="store_true",
-            help="Disable mmap while loading weight using safetensors.",
-        )
 
         # Deprecated arguments
         parser.add_argument(
@@ -2379,6 +2377,10 @@ class DeprecatedAction(argparse.Action):
         raise ValueError(self.help)
 
 
+def print_deprecated_warning(message: str):
+    logger.warning(f"\033[33m{message}\033[0m")
+
+
 def auto_choose_speculative_params(self: ServerArgs):
     """
     Automatically choose the parameters for speculative decoding.