EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

[Platform] Add initial experimental support for Altlas 300I series (#1333)

Browse Source 
### What this PR does / why we need it?
Add initial experimental support for Ascend 310P, this patch squash
below PR into one to help validation:

- https://github.com/vllm-project/vllm-ascend/pull/914
- https://github.com/vllm-project/vllm-ascend/pull/1318
- https://github.com/vllm-project/vllm-ascend/pull/1327


### Does this PR introduce _any_ user-facing change?
User can run vLLM on Altlas 300I DUO series

### How was this patch tested?
CI passed with:
- E2E image build for 310P
- CI test on A2 with e2e test and longterm test
- Unit test missing because need a real 310P image to have the test,
will add in a separate PR later.
- Manually e2e test:
- Qwen2.5-7b-instruct, Qwen2.5-0.5b, Qwen3-0.6B, Qwen3-4B, Qwen3-8B:
https://github.com/vllm-project/vllm-ascend/pull/914#issuecomment-2942989322
  - Pangu MGoE 72B


The patch has been tested locally on Ascend 310P hardware to ensure that
the changes do not break existing functionality and that the new
features work as intended.

#### ENV information

CANN, NNAL version: 8.1.RC1
> [!IMPORTANT]  
> PTA 2.5.1 version >= torch_npu-2.5.1.post1.dev20250528 to support NZ
format and calling NNAL operators on 310P

#### Code example

##### Build vllm-ascend from source code

```shell
# download source code as vllm-ascend
cd vllm-ascend
export SOC_VERSION=Ascend310P3
pip install -v -e .
cd ..
```

##### Run offline inference

```python
from vllm import LLM, SamplingParams
prompts = ["水的沸点是100摄氏度吗？请回答是或者否。", "若腋下体温为38摄氏度，请问这人是否发烧？请回答是或者否。",
           "水的沸点是100摄氏度吗？请回答是或者否。", "若腋下体温为38摄氏度，请问这人是否发烧？请回答是或者否。"]

# Create a sampling params object.
sampling_params = SamplingParams(temperature=0.0, top_p=0.95, max_tokens=10)
# Create an LLM.
llm = LLM(
    model="Qwen/Qwen2.5-7B-Instruct",
    max_model_len=4096,
    max_num_seqs=4,
    dtype="float16", # IMPORTANT cause some ATB ops cannot support bf16 on 310P
    disable_custom_all_reduce=True,
    trust_remote_code=True,
    tensor_parallel_size=2,
    compilation_config={"custom_ops":['none', "+rms_norm", "+rotary_embedding"]},
)

# Generate texts from the prompts.
outputs = llm.generate(prompts, sampling_params)
for output in outputs:
    prompt = output.prompt
    generated_text = output.outputs[0].text
    print(f"Prompt: {prompt!r}, Generated text: {generated_text!r}")

```

---------

Signed-off-by: Vincent Yuan <farawayboat@gmail.com>
Signed-off-by: Yikun Jiang <yikunkero@gmail.com>
Signed-off-by: angazenn <zengyanjia@huawei.com>
Co-authored-by: Vincent Yuan <farawayboat@gmail.com>
Co-authored-by: angazenn <zengyanjia@huawei.com>
Co-authored-by: wangxiyuan <wangxiyuan1007@gmail.com>
Co-authored-by: leo-pony <nengjunma@outlook.com>
Co-authored-by: shen-shanshan <467638484@qq.com>
This commit is contained in:
Yikun Jiang
2025-06-21 09:00:16 +08:00
committed by GitHub
parent 2009fdb8da
commit 097e7149f7
23 changed files with 839 additions and 62 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
vllm_ascend/worker/worker.py
View File

@@ -51,7 +51,8 @@ from vllm_ascend.ascend_config import init_ascend_config
from vllm_ascend.device_allocator.camem import CaMemAllocator
from vllm_ascend.distributed.parallel_state import init_ascend_model_parallel
from vllm_ascend.platform import NPUPlatform
from vllm_ascend.utils import try_register_lib
from vllm_ascend.utils import (ACL_FORMAT_FRACTAL_ND, ACL_FORMAT_FRACTAL_NZ,
is_310p, try_register_lib)
from vllm_ascend.worker.model_runner import NPUModelRunner
from vllm_ascend.worker.pooling_model_runner import NPUPoolingModelRunner
@@ -342,17 +343,22 @@ class NPUWorker(LocalOrDistributedWorkerBase):
for _ in range(self.parallel_config.pipeline_parallel_size)
]
import torch_npu
acl_format = ACL_FORMAT_FRACTAL_NZ if is_310p(
) else ACL_FORMAT_FRACTAL_ND
for ve in range(self.parallel_config.pipeline_parallel_size):
num_layers = len(self.cache_engine[ve].gpu_cache)
for i in range(num_layers):
if torch.is_tensor(self.cache_engine[ve].gpu_cache[i]):
torch_npu.npu_format_cast(
self.cache_engine[ve].gpu_cache[i], 2)
self.cache_engine[ve].gpu_cache[
i] = torch_npu.npu_format_cast(
self.cache_engine[ve].gpu_cache[i], acl_format)
else:
torch_npu.npu_format_cast(
self.cache_engine[ve].gpu_cache[i][0], 2)
torch_npu.npu_format_cast(
self.cache_engine[ve].gpu_cache[i][1], 2)
self.cache_engine[ve].gpu_cache[i][
0] = torch_npu.npu_format_cast(
self.cache_engine[ve].gpu_cache[i][0], acl_format)
self.cache_engine[ve].gpu_cache[i][
1] = torch_npu.npu_format_cast(
self.cache_engine[ve].gpu_cache[i][1], acl_format)
self.gpu_cache = [
self.cache_engine[ve].gpu_cache
for ve in range(self.parallel_config.pipeline_parallel_size)