minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[Feat]Add scheduler recv skipper weights to environment configuration (#13855)

Browse Source
This commit is contained in:
Jimmy
2025-11-27 18:16:11 +08:00
committed by GitHub
parent 6edffc6391
commit ab843ced31
3 changed files with 23 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/references/environment_variables.md
+5
View File
@@ -31,6 +31,11 @@ SGLang supports various environment variables that can be used to configure its
| `SGLANG_FUSED_MLA_ENABLE_ROPE_FUSION` | Enable RoPE fusion in Fused Multi-Layer Attention | `1` |
| `SGLANG_DISABLE_CONSECUTIVE_PREFILL_OVERLAP` | Disable overlap schedule for consecutive prefill batches | `false` |
| `SGLANG_DISABLE_FA4_WARMUP` | Disable Flash Attention 4 warmup passes (set to `1`, `true`, `yes`, or `on` to disable) | `false` |
| `SGLANG_SCHEDULER_RECV_SKIPPER_WEIGHT_DEFAULT` | Default weight value for scheduler recv skipper counter (used when forward mode doesn't match specific modes). Only active when `--scheduler-recv-interval > 1`. The counter accumulates weights and triggers request polling when reaching the interval threshold. | `1000` |
| `SGLANG_SCHEDULER_RECV_SKIPPER_WEIGHT_DECODE` | Weight increment for decode forward mode in scheduler recv skipper. Works with `--scheduler-recv-interval` to control polling frequency during decode phase. | `1` |
| `SGLANG_SCHEDULER_RECV_SKIPPER_WEIGHT_VERIFY` | Weight increment for target verify forward mode in scheduler recv skipper. Works with `--scheduler-recv-interval` to control polling frequency during verification phase. | `1` |
| `SGLANG_SCHEDULER_RECV_SKIPPER_WEIGHT_NONE` | Weight increment when forward mode is None in scheduler recv skipper. Works with `--scheduler-recv-interval` to control polling frequency when no specific forward mode is active. | `1` |
## DeepGEMM Configuration (Advanced Optimization)