antirez / antirez/ds4

failed to run on R6000 prof

Abierto
#563 0 comentarios 0 reacciones 0 asignados Ver en GitHub
Lenguaje dominante
C
Estrellas
22.3k
Forks
2.1k
Merge medio
1 d 3 h
PR fusionados (30 d)
4

Descripción

hello author
I run the same setup that works even well on 1xL40s, but failed to run R6000 server with error below
```
(ds4redis1) root@salab-hpedl380g11-05:/mnt/tlc/wayne/ds4redis1/ds4# ./ds4-server --ctx 1000000 --kv-disk-dir /mnt/qlc1/nfsrdma/dsv4-kv --kv-disk-space-mb 2621440 --debug
ds4: Linux cuda backend set oom_score_adj=1000
ds4: CUDA backend initialized on NVIDIA RTX PRO 6000 Blackwell Server Edition (sm_120)
ds4: CUDA host registration skipped: operation not supported
ds4: CUDA preparing model tensor mappings: 80.24 GiB
ds4: CUDA q8 fp16 cache budget exhausted; using q8 kernels (request=64.00 MiB cached=5.66 GiB limit=12.00 GiB free=4.75 GiB reserve=4.75 GiB total=94.97 GiB)
ds4: CUDA startup model preparation covered 80.76 GiB of tensor spans in 7.835s
ds4: cuda backend initialized for graph diagnostics
0714 06:04:00 ds4-server: context buffers 21302.79 MiB (ctx=1000000, backend=cuda, prefill_chunk=4096, raw_kv_rows=4352, compressed_kv_rows=250002)
ds4: CUDA using managed KV cache for ctx=1000000 (kv cache 13.17 GiB, context buffers 20.80 GiB); this may degrade performance but is needed for very large contexts
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
ds4: CUDA tensor alloc failed: out of memory
0714 06:04:00 ds4-server: failed to create cuda session

(ds4redis1) root@salab-hpedl380g11-05:/mnt/tlc/wayne/ds4redis1/ds4# nvidia-smi
Tue Jul 14 06:04:13 2026
+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 580.159.03 Driver Version: 580.159.03 CUDA Version: 13.0 |
+-----------------------------------------+------------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+========================+======================|
| 0 NVIDIA RTX PRO 6000 Blac... Off | 00000000:8A:00.0 Off | 0 |
| N/A 42C P8 35W / 450W | 0MiB / 97887MiB | 0% Default |
| | | Disabled |
+-----------------------------------------+------------------------+----------------------+

+-----------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=========================================================================================|
| No running processes found |
+-----------------------------------------------------------------------------------------+
(ds4redis1) root@salab-hpedl380g11-05:/mnt/tlc/wayne/ds4redis1/ds4#

````

Guía de contribución

Abrir la guía de contribución

Evaluación

Este issue todavía no se ha evaluado.

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.