Skip to content

Commit 2a1fb38

Browse files
committed
Latency calculation optimization
1 parent 61f27bb commit 2a1fb38

File tree

1 file changed

+4
-3
lines changed

1 file changed

+4
-3
lines changed

src/scheduler/SessionElement.cpp

Lines changed: 4 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -328,9 +328,10 @@ int SessionElement::calculate_inference_caused_latency(float max_possible_infere
328328
wait_time_left -= m_inference_config.m_max_inference_time;
329329
}
330330

331-
while (inference_time_left >= host_buffer_time_int && std::ceil(host_buffer_size_int) > 0) {
332-
inference_caused_latency += host_buffer_size_int;
333-
inference_time_left -= host_buffer_time_int;
331+
if (host_buffer_time_int > 0) {
332+
int iterations = inference_time_left / host_buffer_time_int;
333+
inference_caused_latency += iterations * host_buffer_size_int;
334+
inference_time_left -= iterations * host_buffer_time_int;
334335
}
335336
}
336337

0 commit comments

Comments
 (0)