Init attention backend for Intel XPU (#10656)

Co-authored-by: guangyey <guangye.yu@intel.com>
Co-authored-by: DiweiSun <105627594+DiweiSun@users.noreply.github.com>

python/sglang/srt/server_args.py

@@ -114,6 +114,7 @@ ATTENTION_BACKEND_CHOICES = [
     # Other platforms
     "intel_amx",
     "ascend",
+    "intel_xpu",
 ]
 
 LORA_BACKEND_CHOICES = ["triton", "csgmv"]
@@ -1098,6 +1099,12 @@ class ServerArgs:
             self.enable_mixed_chunk = False
             self.disable_radix_cache = True
 
+        if self.attention_backend == "intel_xpu":
+            if self.page_size not in [32, 64, 128]:
+                logger.warning(
+                    f"Intel XPU attention backend only supports page_size of 32, 64 or 128, changing page_size from {self.page_size} to 128."
+                )
+                self.page_size = 128
         if self.attention_backend == "fa4" or self.decode_attention_backend == "fa4":
             raise ValueError(
                 "FA4 backend is only supported for prefill. Please use `--prefill-attention-backend fa4` instead."