thek0tyara/llama-cpp-turboquant
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 4

server: Fix undefined timing measurement errors in server context (#21201)

Browse source 
Co-authored-by: Dan Hoffman <dhoffman@cyket.net>
This commit is contained in:
Dan Hoffman 2026-04-04 07:11:19 -07:00 • committed by GitHub
parent d01f6274c0
commit 9c699074c9
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 8 additions and 8 deletions
Download patch file Download diff file Expand all files Collapse all files

12
tools/server/server-task.h
View file

@ -261,14 +261,14 @@ struct result_timings {
int32_t cache_n = -1;
int32_t prompt_n = -1;
double prompt_ms;
double prompt_per_token_ms;
double prompt_per_second;
double prompt_ms = 0.0;
double prompt_per_token_ms = 0.0;
double prompt_per_second = 0.0;
int32_t predicted_n = -1;
double predicted_ms;
double predicted_per_token_ms;
double predicted_per_second;
double predicted_ms = 0.0;
double predicted_per_token_ms = 0.0;
double predicted_per_second = 0.0;
// Optional speculative metrics - only included when > 0
int32_t draft_n = 0;