From: Anthony Umfer <redacted>
Date: Sun, 11 May 2025 15:08:26 +0000 (-0400)
Subject: tools : fix uninitialized llama_batch in server (#13436)
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=9a390c4829cd3058d26a2e2c09d16e3fd12bf1b1;p=pkg%2Fggml%2Fsources%2Fllama.cpp

tools : fix uninitialized llama_batch in server (#13436)

* add constructor to initialize server_context::batch, preventing destructor's call to llama_batch_free from causing an invalid free()

* Update tools/server/server.cpp

Co-authored-by: Xuan-Son Nguyen <redacted>

* use C++11 initializer syntax

* switch from Copy-list-initialization to Direct-list-initialization

---------

Co-authored-by: Xuan-Son Nguyen <redacted>
---

diff --git a/tools/server/server.cpp b/tools/server/server.cpp
index de8ded71..7169ffdc 100644
--- a/tools/server/server.cpp
+++ b/tools/server/server.cpp
@@ -1862,7 +1862,7 @@ struct server_context {
 
     llama_context_params cparams_dft;
 
-    llama_batch batch;
+    llama_batch batch {};
 
     bool clean_kv_cache = true;
     bool add_bos_token  = true;