Skip to content
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

[CI] Upgrade to newest pta.(MLA and FusedMoE) #189

Merged
merged 2 commits into from
Feb 27, 2025
+26 −56
Merged

[CI] Upgrade to newest pta.(MLA and FusedMoE) #189

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
6d6c5f8
upgrade to newest pta
SidaoY Feb 27, 2025
bb619f2
Clean code
SidaoY Feb 27, 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
75 changes: 23 additions & 52 deletions vllm_ascend/attention.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -770,13 +770,9 @@ def forward(
num_blocks, block_size, self.num_kv_heads,
self.qk_rope_head_dim + self.kv_lora_rank)
slots = attn_metadata.slot_mapping
torch_npu.npu_reshapecache(key=k_cache,
value=None,
keyCache=key_cache,
valueCache=None,
slotMapping=slots,
compressType=0,
kvCacheCfg=1)
torch_npu._npu_reshape_and_cache_siso(key=k_cache,
key_cache=key_cache,
Copy link
Contributor

Choose a reason for hiding this comment

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

The default value of valueCache is None, right?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

The SISO interface does not transmit the valueCache.

slot_indices=slots)

if attn_metadata.num_prefills > 0:
attn_output = torch.empty(num_tokens,
Expand All @@ -793,32 +789,16 @@ def forward(
self.seq_lens_tensor_cpu = torch.from_numpy(
np.array(attn_metadata.prefill_metadata.seq_lens).astype(
np.int32))
torch_npu.npu_selfattention(query=query,
key=key,
value=value,
kvcacheCfg=0,
mask=mask,
maskType=1,
isTriuMask=0,
seqLen=self.seq_lens_tensor_cpu,
scale=self.scale,
qScale=1,
scaleType=0,
headNum=self.num_heads,
kvHeadNum=self.num_heads,
mlaVHeadSize=0,
calcType=3,
kernelType=0,
clampType=0,
quantType=0,
cacheType=0,
windowSize=0,
clampMin=0,
clampMax=0,
batchRunStatusEnable=False,
inputLayout=0,
outDataType=0,
out=attn_output)
torch_npu._npu_flash_attention(
query=query,
key=key,
value=value,
mask=mask,
seq_len=self.seq_lens_tensor_cpu,
scale_value=self.scale,
num_heads=self.num_heads,
num_kv_heads=self.num_heads,
Copy link
Collaborator

Choose a reason for hiding this comment

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

should this be self.num_kv_heads?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

The two are consistent during the prefill phase, and the distinction is only made during the decode phase.

out=attn_output)
else:
# TODO: Will support prefix cache and chunked prefill soon.
raise RuntimeError(
Expand All @@ -835,25 +815,16 @@ def forward(
np.array(attn_metadata.decode_metadata.seq_lens).astype(
np.int32))
block_tables = attn_metadata.decode_metadata.block_tables
torch_npu.npu_pagedattention(query=query,
keyCache=key_cache,
valueCache=None,
contextLens=self.seq_lens_tensor_cpu,
maskType=0,
kvHeadNum=self.num_kv_heads,
headNum=self.num_heads,
mlaVHeadSize=self.kv_lora_rank,
qkScale=self.scale,
blockTables=block_tables,
batchRunStatusEnable=False,
hasQuantOffset=False,
compressType=0,
calcType=0,
scaleType=0,
quantType=0,
inputLayout=0,
outDataType=-1,
attnOut=attn_output)
torch_npu._npu_paged_attention_mla(
query=query,
key_cache=key_cache,
num_kv_heads=self.num_kv_heads,
num_heads=self.num_heads,
scale_value=self.scale,
block_table=block_tables,
context_lens=self.seq_lens_tensor_cpu,
mla_vheadsize=self.kv_lora_rank,
out=attn_output)
attn_output_t = torch.transpose(attn_output, 0, 1)
attn_output_t = torch.bmm(attn_output_t, self.w_vc)
attn_output = torch.transpose(attn_output_t, 0, 1)
Expand Down
7 changes: 3 additions & 4 deletions vllm_ascend/ops/fused_moe.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -50,10 +50,9 @@ def group_topk(hidden_states: torch.Tensor,

topk_group = 0 if topk_group is None else topk_group
num_expert_group = 0 if num_expert_group is None else num_expert_group
torch_npu.npu_group_topk(input=scores,
out=scores,
group_num=num_expert_group,
k=topk_group)
torch_npu._npu_group_topk(self=scores,
k=topk_group,
group_num=num_expert_group)
if e_score_correction_bias is not None:
topk_ids = torch.topk(scores, k=topk, dim=-1, sorted=False)[1]
# Use original unbiased scores for the routing weights
Expand Down
Loading