llama : fix integer overflow during quantization #9645
| Job | Run time |
|---|---|
| 5m 41s | |
| 1m 40s | |
| 1m 53s | |
| 1m 56s | |
| 2m 3s | |
| 2m 14s | |
| 34m 45s | |
| 4m 13s | |
| 20s | |
| 34m 50s | |
| 1m 5s | |
| 14s | |
| 2m 39s | |
| 1m 31s | |
| 1s | |
| 4m 20s | |
| 5m 28s | |
| 4m 43s | |
| 20s | |
| 3m 50s | |
| 4m 36s | |
| 18m 47s | |
| 1s | |
| 3m 57s | |
| 13m 29s | |
| 18m 56s | |
| 7m 40s | |
| 4m 8s | |
| 4m 21s | |
| 4m 53s | |
| 4m 3s | |
| 4m 8s | |
| 5m 29s | |
| 1m 47s | |
| 0s | |
| 3h 30m 1s |