Skip to content

[BugFix]This PR aims to fix the precision issue of the LoRA feature i… #4046

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 5 commits into from
Closed

[BugFix]This PR aims to fix the precision issue of the LoRA feature i… #4046

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
b3f16d4
[BugFix]This PR aims to fix the precision issue of the LoRA feature i…
liuchenbing Nov 7, 2025
536431b
[BugFix]This PR aims to fix the precision issue of the LoRA feature i…
liuchenbing Nov 11, 2025
7b2b59d
[BugFix]This PR aims to fix the precision issue of the LoRA feature i…
liuchenbing Nov 11, 2025
af719cf
[BugFix]This PR aims to fix the precision issue of the LoRA feature i…
liuchenbing Nov 11, 2025
9ec244e
[BugFix]This PR aims to fix the precision issue of the LoRA feature i…
liuchenbing Nov 11, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 5 additions & 9 deletions csrc/kernels/bgmv_expand.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -341,10 +341,8 @@ class BGMVExpand {
}

// declare all dtype kernel
BGMV_EXPAND_TYPE_DECLARE(half)
#if (__CCE_AICORE__ >= 220)
BGMV_EXPAND_TYPE_DECLARE(bfloat16_t)
#endif
BGMV_EXPAND_TYPE_DECLARE(half);
BGMV_EXPAND_TYPE_DECLARE(bfloat16_t);

namespace vllm_ascend {
extern void bgmv_expand_impl(AscendType type, void* stream, void* x, void* weight, void* indices, uint32_t indicesSize,
Expand All @@ -356,11 +354,9 @@ extern void bgmv_expand_impl(AscendType type, void* stream, void* x, void* weigh
bgmv_expand_half<<<blockDim, nullptr, stream>>>(x, weight, indices, indicesSize, yIn, yOut, batchSize, numTokensPerCore,
maxLoRARank, outputHiddenDim, sliceOffset, outputFullDim);
} else if (type == AscendType::BF16) {
#if (__CCE_AICORE__ >= 220)
bgmv_expand_bfloat16_t<<<blockDim, nullptr, stream>>>(x, weight, indices, indicesSize, yIn, yOut, batchSize,
numTokensPerCore, maxLoRARank, outputHiddenDim,
sliceOffset, outputFullDim);
#endif
bgmv_expand_bfloat16_t<<<blockDim, nullptr, stream>>>(x, weight, indices, indicesSize, yIn, yOut, batchSize,
numTokensPerCore, maxLoRARank, outputHiddenDim,
sliceOffset, outputFullDim);
Comment on lines +357 to +359
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Nov 7, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

critical

You've removed the conditional compilation directive for the bgmv_expand_bfloat16_t kernel call, making it unconditional. However, the kernel's definition using BGMV_EXPAND_TYPE_DECLARE(bfloat16_t) at line 346 is still inside an #if (__CCE_AICORE__ >= 220) block. This will cause a compilation error on platforms where __CCE_AICORE__ < 220, as the function will be called but not defined. You should also remove the conditional compilation around the kernel declaration to fix this.

} else {
return;
}
Expand Down
12 changes: 4 additions & 8 deletions csrc/kernels/bgmv_shrink.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -225,10 +225,8 @@ class BGMVShrink {
}

// declare all dtype kernel
BGMV_SHRINK_TYPE_DECLARE(half)
#if (__CCE_AICORE__ >= 220)
BGMV_SHRINK_TYPE_DECLARE(bfloat16_t)
#endif
BGMV_SHRINK_TYPE_DECLARE(half);
BGMV_SHRINK_TYPE_DECLARE(bfloat16_t);

namespace vllm_ascend {
extern void bgmv_shrink_impl(AscendType type, void* stream, void* x, void* weight, void* indices, uint32_t indicesSize,
Expand All @@ -240,10 +238,8 @@ extern void bgmv_shrink_impl(AscendType type, void* stream, void* x, void* weigh
bgmv_shrink_half<<<blockDim, nullptr, stream>>>(x, weight, indices, indicesSize, y, batchSize, numTokensPerCore,
inputHiddenDim, maxLoRARank, scale);
} else if (type == AscendType::BF16) {
#if (__CCE_AICORE__ >= 220)
bgmv_shrink_bfloat16_t<<<blockDim, nullptr, stream>>>(x, weight, indices, indicesSize, y, batchSize, numTokensPerCore,
inputHiddenDim, maxLoRARank, scale);
#endif
bgmv_shrink_bfloat16_t<<<blockDim, nullptr, stream>>>(x, weight, indices, indicesSize, y, batchSize, numTokensPerCore,
inputHiddenDim, maxLoRARank, scale);
Comment on lines +241 to +242
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Nov 7, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

critical

You've removed the conditional compilation directive for the bgmv_shrink_bfloat16_t kernel call, making it unconditional. However, the kernel's definition using BGMV_SHRINK_TYPE_DECLARE(bfloat16_t) at line 230 is still inside an #if (__CCE_AICORE__ >= 220) block. This will cause a compilation error on platforms where __CCE_AICORE__ < 220, as the function will be called but not defined. You should also remove the conditional compilation around the kernel declaration to fix this.

} else {
return;
}
Expand Down
16 changes: 6 additions & 10 deletions csrc/kernels/sgmv_expand.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -356,10 +356,8 @@ class SGMVExpand {
}

// declare all dtype kernel
SGMV_EXPAND_TYPE_DECLARE(half)
#if (__CCE_AICORE__ >= 220)
SGMV_EXPAND_TYPE_DECLARE(bfloat16_t)
#endif
SGMV_EXPAND_TYPE_DECLARE(half);
SGMV_EXPAND_TYPE_DECLARE(bfloat16_t);

namespace vllm_ascend {
extern void sgmv_expand_impl(AscendType type, void* stream, void* x, void* weight,
Expand All @@ -375,12 +373,10 @@ extern void sgmv_expand_impl(AscendType type, void* stream, void* x, void* weigh
numTokensPerCore, maxLoRARank, outputHiddenDim, sliceOffset,
outputFullDim);
} else if (type == AscendType::BF16) {
#if (__CCE_AICORE__ >= 220)
sgmv_expand_bfloat16_t<<<blockDim, nullptr, stream>>>(x, weight, loraIndices, loraIndicesSize,
seqLen, seqLenSize, yIn, yOut, batchSize,
numTokensPerCore, maxLoRARank, outputHiddenDim,
sliceOffset, outputFullDim);
#endif
sgmv_expand_bfloat16_t<<<blockDim, nullptr, stream>>>(x, weight, loraIndices, loraIndicesSize,
seqLen, seqLenSize, yIn, yOut, batchSize,
numTokensPerCore, maxLoRARank, outputHiddenDim,
sliceOffset, outputFullDim);
Comment on lines +376 to +379
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Nov 7, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

critical

You've removed the conditional compilation directive for the sgmv_expand_bfloat16_t kernel call, making it unconditional. However, the kernel's definition using SGMV_EXPAND_TYPE_DECLARE(bfloat16_t) at line 361 is still inside an #if (__CCE_AICORE__ >= 220) block. This will cause a compilation error on platforms where __CCE_AICORE__ < 220, as the function will be called but not defined. You should also remove the conditional compilation around the kernel declaration to fix this.

} else {
return;
}
Expand Down
18 changes: 7 additions & 11 deletions csrc/kernels/sgmv_shrink.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -241,10 +241,8 @@ class SGMVShrink {
}

// declare all dtype kernel
SGMV_SHRINK_TYPE_DECLARE(half)
#if (__CCE_AICORE__ >= 220)
SGMV_SHRINK_TYPE_DECLARE(bfloat16_t)
#endif
SGMV_SHRINK_TYPE_DECLARE(half);
SGMV_SHRINK_TYPE_DECLARE(bfloat16_t);

namespace vllm_ascend {
extern void sgmv_shrink_impl(AscendType type, void* stream, void* x, void* weight,
Expand All @@ -260,13 +258,11 @@ extern void sgmv_shrink_impl(AscendType type, void* stream, void* x, void* weigh
numTokensPerCore, inputHiddenDim, maxLoRARank,
scale);
} else if (type == AscendType::BF16) {
#if (__CCE_AICORE__ >= 220)
sgmv_shrink_bfloat16_t<<<blockDim, nullptr, stream>>>(x, weight, loraIndices, loraIndicesSize,
seqLen, seqLenSize,
y, batchSize,
numTokensPerCore, inputHiddenDim, maxLoRARank,
scale);
#endif
sgmv_shrink_bfloat16_t<<<blockDim, nullptr, stream>>>(x, weight, loraIndices, loraIndicesSize,
seqLen, seqLenSize,
y, batchSize,
numTokensPerCore, inputHiddenDim, maxLoRARank,
scale);
Comment on lines +261 to +265
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Nov 7, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

critical

You've removed the conditional compilation directive for the sgmv_shrink_bfloat16_t kernel call, making it unconditional. However, the kernel's definition using SGMV_SHRINK_TYPE_DECLARE(bfloat16_t) at line 246 is still inside an #if (__CCE_AICORE__ >= 220) block. This will cause a compilation error on platforms where __CCE_AICORE__ < 220, as the function will be called but not defined. You should also remove the conditional compilation around the kernel declaration to fix this.

} else {
return;
}
Expand Down
1 change: 1 addition & 0 deletions vllm_ascend/lora/punica_npu.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -255,6 +255,7 @@ def add_lora_embedding(self,
# Embedding layer only need expand op
expand_fun: Callable = (self._expand_prefill
if self.is_prefill else self._expand_decode)
x = x.to(torch.float32)
expand_fun(y, x, lora_b_stacked, add_inputs)

def add_lora_linear(self,
Expand Down
Loading