We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent 61f27bb commit 2a1fb38Copy full SHA for 2a1fb38
src/scheduler/SessionElement.cpp
@@ -328,9 +328,10 @@ int SessionElement::calculate_inference_caused_latency(float max_possible_infere
328
wait_time_left -= m_inference_config.m_max_inference_time;
329
}
330
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;
+ if (host_buffer_time_int > 0) {
+ int iterations = inference_time_left / host_buffer_time_int;
+ inference_caused_latency += iterations * host_buffer_size_int;
334
+ inference_time_left -= iterations * host_buffer_time_int;
335
336
337
0 commit comments