server: free sampling contexts on exit (#7264)

* server: free sampling contexts on exit

This cleans up last leak found by the address sanitizer.

* fix whitespace

* fix whitespace

diff --git a/examples/server/server.cpp b/examples/server/server.cpp
index ceaeb1f76dc3dec4cc1321dc0dc858663f2bbb16..7e0d068f830abb2659d7696f8d89e663eec85365 100644 (file)
--- a/examples/server/server.cpp
+++ b/examples/server/server.cpp
@@ -671,6 +671,13 @@ struct server_context {
             model = nullptr;
         }
 
+        // Clear any sampling context
+        for (server_slot & slot : slots) {
+            if (slot.ctx_sampling != nullptr) {
+                llama_sampling_free(slot.ctx_sampling);
+            }
+        }
+
         llama_batch_free(batch);
     }