From: Georgi Gerganov <redacted>
Date: Wed, 2 Jul 2025 11:12:07 +0000 (+0300)
Subject: simple-chat : fix context-exceeded condition (#14494)
X-Git-Tag: upstream/0.0.5882~79
X-Git-Url: https://git.djapps.eu/?a=commitdiff_plain;h=d7f5f4e578d1f60b0835d1734a50438c309b3e5c;p=pkg%2Fggml%2Fsources%2Fllama.cpp

simple-chat : fix context-exceeded condition (#14494)

* simple-chat : fix context-exceeded condition

ggml-ci

* cont : fix n_ctx_used computation

ggml-ci
---

diff --git a/examples/simple-chat/simple-chat.cpp b/examples/simple-chat/simple-chat.cpp
index cf117804..57195df3 100644
--- a/examples/simple-chat/simple-chat.cpp
+++ b/examples/simple-chat/simple-chat.cpp
@@ -113,15 +113,16 @@ int main(int argc, char ** argv) {
         while (true) {
             // check if we have enough space in the context to evaluate this batch
             int n_ctx = llama_n_ctx(ctx);
-            int n_ctx_used = llama_memory_seq_pos_max(llama_get_memory(ctx), 0);
+            int n_ctx_used = llama_memory_seq_pos_max(llama_get_memory(ctx), 0) + 1;
             if (n_ctx_used + batch.n_tokens > n_ctx) {
                 printf("\033[0m\n");
                 fprintf(stderr, "context size exceeded\n");
                 exit(0);
             }
 
-            if (llama_decode(ctx, batch)) {
-                GGML_ABORT("failed to decode\n");
+            int ret = llama_decode(ctx, batch);
+            if (ret != 0) {
+                GGML_ABORT("failed to decode, ret = %d\n", ret);
             }
 
             // sample the next token