ver4a/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

server : fix divide-by-zero in metrics reporting (#11915)

Browse source
This commit is contained in:
Antoine Viallon 2025-02-17 11:25:12 +01:00 • committed by GitHub
parent 2eea03d86a
commit c4d29baf32
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

2
examples/server/server.cpp
View file

@ -3656,7 +3656,7 @@ int main(int argc, char ** argv) {
}, { }, {
{"name", "n_busy_slots_per_decode"}, {"name", "n_busy_slots_per_decode"},
{"help", "Average number of busy slots per llama_decode() call"}, {"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", {{ {"gauge", {{
{"name", "prompt_tokens_seconds"}, {"name", "prompt_tokens_seconds"},