We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent cda8541 commit 8a82ac4Copy full SHA for 8a82ac4
xllm/core/scheduler/continuous_scheduler.cpp
@@ -1025,8 +1025,13 @@ void ContinuousScheduler::process_batch_output(bool enable_schedule_overlap) {
1025
for (auto request : to_be_processed_requests) {
1026
// ignore cancelled/finished requests when enable_schedule_overlap.
1027
if (options_.enable_schedule_overlap() && request->state().stream) {
1028
- if (!request->finished() && !request->cancelled()) {
+ // skip cancelled request
1029
+ if (request->cancelled()) {
1030
+ continue;
1031
+ }
1032
+ if (!request->finished()) {
1033
stream_requests.emplace_back(request);
1034
1035
}
1036
// handle token when last token not be handled.
1037
if (request->finished() && !request->last_token_handled()) {
0 commit comments