Public/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-06 13:01:08 +02:00
Code Issues Packages Projects Releases Wiki Activity

fix stream case

Browse Source
This commit is contained in:
Xuan Son Nguyen
2026-08-04 17:54:18 +02:00
committed by Xuan-Son Nguyen
parent 07491664e2
commit 3954a9c39b
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/server/server-models.cpp
+3 -1
View File
@@ -1882,7 +1882,9 @@ void server_models_routes::init_routes() {
// this request instead of leaving an orphan generation
std::string conv_id = server_stream_conv_id_from_headers(req.headers);
uint64_t ticket = models.conv_models.remember(conv_id, name);
bool waited = autoload && models.ensure_model_ready(name, req.should_stop);
// a dead socket must not cancel a session request, only a stop does (checked right below)
auto should_stop = ticket == 0 ? req.should_stop : nullptr;
bool waited = autoload && models.ensure_model_ready(name, should_stop);
if (ticket != 0 && !models.conv_models.alive(conv_id, ticket)) {
SRV_INF("request for conv_id=%s cancelled while model name=%s was loading\n",
conv_id.c_str(), name.c_str());