mirror of
https://github.com/RYDE-WORK/llama.cpp.git
synced 2026-01-19 13:13:38 +08:00
server : fix divide-by-zero in metrics reporting (#11915)
This commit is contained in:
parent
2eea03d86a
commit
c4d29baf32
@ -3656,7 +3656,7 @@ int main(int argc, char ** argv) {
|
||||
}, {
|
||||
{"name", "n_busy_slots_per_decode"},
|
||||
{"help", "Average number of busy slots per llama_decode() call"},
|
||||
{"value", (float) res_metrics->n_busy_slots_total / (float) res_metrics->n_decode_total}
|
||||
{"value", (float) res_metrics->n_busy_slots_total / std::max((float) res_metrics->n_decode_total, 1.f)}
|
||||
}}},
|
||||
{"gauge", {{
|
||||
{"name", "prompt_tokens_seconds"},
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user