Public/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-19 17:24:57 +02:00
Code Issues Packages Projects Releases Wiki Activity

also handle process_mtmd_chunk

Browse Source
This commit is contained in:
Xuan Son Nguyen
2026-08-13 23:54:19 +02:00
parent 0b42458e10
commit a35d2af15f
1 changed files with 9 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/server/server-context.cpp
+9 -2
View File
@@ -2291,7 +2291,7 @@ private:
// returns false to decline the task, it is offered again after the decode is done
bool process_single_task(server_task && task) {
// only metrics is safe while decoding, it touches neither ctx_tgt nor the slots
// during encoding / decoding, only accessing metrics is safe
if (is_decoding && task.type != SERVER_TASK_TYPE_METRICS) {
SRV_DBG("decoding, decline task, id_task = %d\n", task.id);
return false;
@@ -3394,8 +3394,15 @@ private:
// so the timing is queued and flushed on the next sync
metrics_pre_decode();
// encode on the worker thread, so we can still handle metrics tasks
size_t n_tokens_out = 0;
int32_t res = slot.process_mtmd_chunk(cur_token_idx, n_tokens_out);
int32_t res = 0;
is_decoding = true;
queue_tasks.yield_to_queue([&]() {
res = slot.process_mtmd_chunk(cur_token_idx, n_tokens_out);
});
is_decoding = false;
if (res != 0) {
SLT_ERR(slot, "failed to process mtmd chunk, res = %d\n", res);
send_error(slot, "failed to process mtmd chunk", ERROR_TYPE_SERVER);