From 6ce21345be1f447063adfd223916b0f2f2962a7a Mon Sep 17 00:00:00 2001
From: Gahow Wang <yuanqu.wjh@alibaba-inc.com>
Date: Sat, 30 May 2026 12:50:04 +0800
Subject: [PATCH] cuda: add cached_trim() to release pooled GPU buffers

Exposes the caching allocator's trim() through a public free function.
Called after weight fusion during model loading to free temporary buffers
that would otherwise sit in the pool and cause OOM.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---
 crates/xserv-cuda/src/allocator.rs | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/crates/xserv-cuda/src/allocator.rs b/crates/xserv-cuda/src/allocator.rs
index f007d22..3bd9e31 100644
--- a/crates/xserv-cuda/src/allocator.rs
+++ b/crates/xserv-cuda/src/allocator.rs
@@ -100,6 +100,13 @@ pub fn cached_alloc(size: usize) -> Result<GpuBuffer> {
     })
 }
 
+/// Free all cached (unused) GPU buffers back to the driver.
+pub fn cached_trim() {
+    ALLOCATOR.with(|cell| {
+        cell.borrow_mut().trim();
+    });
+}
+
 /// Return a raw GPU pointer to the caching allocator's free list.
 /// Called from `GpuBuffer::Drop` for pooled buffers. Takes raw pointer
 /// and size to avoid re-triggering Drop.