[Bugfix][LoRA][Operator] Fix LoRA custom operators accuracy issue (#2672)
### What this PR does / why we need it?
Fix the LoRA accuracy issue that introduced by custom AscendC operator
"bgmv_shrink, sgmv_shrink, bgmv_expand, sgmv_epand".
The bug details are:
- In the kernel function, if you want to call GlobalTensor.GetSize
method, you have to pass the second parameter of bufferSize when you
call GlobalTensor.SetGlobalBuffer first.
- Or GlobalTensor.GetSize method will return a random value.
- You can refer to [this
doc](https://www.hiascend.com/document/detail/zh/CANNCommunityEdition/81RC1alpha002/apiref/ascendcopapi/atlasascendc_api_07_00024.html).
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
pytest -sv tests/e2e/singlecard/test_ilama_lora.py
pytest -sv tests/e2e/multicard/test_ilama_lora_tp2.py
- vLLM version: v0.10.1.1
- vLLM main:
a344a5aa0a
---------
Signed-off-by: paulyu12 <paulyu0307@gmail.com>
Signed-off-by: paulyu12 <507435917@qq.com>
Co-authored-by: paulyu12 <paulyu0307@gmail.com>
This commit is contained in:
4
.github/workflows/vllm_ascend_test.yaml
vendored
4
.github/workflows/vllm_ascend_test.yaml
vendored
@@ -201,7 +201,7 @@ jobs:
|
||||
pytest -sv tests/e2e/singlecard/test_embedding.py
|
||||
pytest -sv tests/e2e/singlecard/test_guided_decoding.py
|
||||
# TODO: Fix lora accuracy error
|
||||
# pytest -sv tests/e2e/singlecard/test_ilama_lora.py
|
||||
pytest -sv tests/e2e/singlecard/test_ilama_lora.py
|
||||
pytest -sv tests/e2e/singlecard/test_profile_execute_duration.py
|
||||
pytest -sv tests/e2e/singlecard/test_quantization.py
|
||||
pytest -sv tests/e2e/singlecard/test_sampler.py
|
||||
@@ -280,7 +280,7 @@ jobs:
|
||||
# external_launcher test is not stable enough. Fix it later
|
||||
# pytest -sv tests/e2e/multicard/test_external_launcher.py
|
||||
pytest -sv tests/e2e/multicard/test_fused_moe_allgather_ep.py
|
||||
# pytest -sv tests/e2e/multicard/test_ilama_lora_tp2.py
|
||||
pytest -sv tests/e2e/multicard/test_ilama_lora_tp2.py
|
||||
|
||||
# To avoid oom, we need to run the test in a single process.
|
||||
pytest -sv tests/e2e/multicard/test_offline_inference_distributed.py::test_models_distributed_QwQ
|
||||
|
||||
Reference in New Issue
Block a user