Add symmetric debug mode to print stack trace of comm ops with unregistered tensors (#18569)

This commit is contained in:
Nicolas Castet
2026-04-08 22:34:58 -07:00
committed by GitHub
parent 6b96f8341d
commit e379befbac
4 changed files with 89 additions and 0 deletions
+1
View File
@@ -160,6 +160,7 @@ SGLang supports various environment variables that can be used to configure its
| `SGLANG_TEST_RETRACT_NO_PREFILL_BS` | When SGLANG_TEST_RETRACT is enabled, no prefill is performed if the batch size exceeds SGLANG_TEST_RETRACT_NO_PREFILL_BS. | `2 ** 31` |
| `SGLANG_RECORD_STEP_TIME` | Record step time for profiling | `false` |
| `SGLANG_TEST_REQUEST_TIME_STATS` | Test request time statistics | `false` |
| `SGLANG_DEBUG_SYMM_MEM` | Enable debug checks that verify tensors passed to NCCL communication ops are allocated in the symmetric memory pool. Logs warnings (rank 0 only) with stack traces for any tensor not in the pool. | `false` |
| `SGLANG_KERNEL_API_LOGLEVEL` | Controls crash-debug kernel API logging. `0` disables logging, `1` logs API names, `3` logs tensor metadata, `5` adds tensor statistics, and `10` also writes pre-call dump snapshots. | `0` |
| `SGLANG_KERNEL_API_LOGDEST` | Destination for crash-debug kernel API logs. Use `stdout`, `stderr`, or a file path. `%i` is replaced with the process PID. | `stdout` |
| `SGLANG_KERNEL_API_DUMP_DIR` | Output directory for level-10 kernel API input/output dumps. `%i` is replaced with the process PID. | `sglang_kernel_api_dumps` |