[Iluvatar GPU] Optimze attention and moe performance#3234
Merged
Conversation
|
|
wuyujiji
force-pushed
the
iluvatar_optim
branch
2 times, most recently
from
863a582 to
9516b2d
Compare
|
Thanks for your contribution! |
wuyujiji
force-pushed
the
iluvatar_optim
branch
3 times, most recently
from
a315aa2 to
9748bd1
Compare
yongqiangma
previously approved these changes
Aug 6, 2025
wuyujiji
force-pushed
the
iluvatar_optim
branch
3 times, most recently
from
1a8a265 to
5cb095b
Compare
wuyujiji
force-pushed
the
iluvatar_optim
branch
from
5cb095b to
c9a0ffd
Compare
phlrain
approved these changes
Aug 8, 2025
gongshaotian
reviewed
Aug 8, 2025
|
|
||
|
|
||
| class IluvatarWorker(WorkerBase): | ||
| class IluvatarWorker(GpuWorker): |
Collaborator
There was a problem hiding this comment.
iluvatar 的执行流程为什么要跟 gpu 耦合在一起？
Contributor
Author
There was a problem hiding this comment.
iluvatar 的执行流程为什么要跟 gpu 耦合在一起？
这次从0630升级到最新commit发现gpu_model_runner改变很大，而之前适配的iluvatar_model_runner就是从gpu_model_runner copy的，除了import部分算子有区别，其他都是一样的。这次这样改动，以后在升级适配可以直接服用gpu_model_runner的流程，避免在做copy的重复工作，如果出现了不兼容的流程，我们会在iluvatar_model_runner重新覆盖该成员函数保证可以work
Collaborator
There was a problem hiding this comment.
iluvatar 的执行流程为什么要跟 gpu 耦合在一起？
这次从0630升级到最新commit发现gpu_model_runner改变很大，而之前适配的iluvatar_model_runner就是从gpu_model_runner copy的，除了import部分算子有区别，其他都是一样的。这次这样改动，以后在升级适配可以直接服用gpu_model_runner的流程，避免在做copy的重复工作，如果出现了不兼容的流程，我们会在iluvatar_model_runner重新覆盖该成员函数保证可以work
好的，6.30之后执行流程一直在快速迭代，等执行器稳定后不同硬件的执行器需要重新隔离开
gongshaotian
reviewed
Aug 8, 2025
| def initialize_cache(self, num_gpu_blocks: int) -> None: | ||
| """ """ | ||
| self.model_runner.update_share_input_block_num(num_gpu_blocks=num_gpu_blocks) | ||
| class IluvatarPaddleDisWorkerProc(PaddleDisWorkerProc): |
Collaborator
There was a problem hiding this comment.
worker proc 不应该放在 worker 层级，这里留个 TODO 项吧。之前的没有预留支持多个 worker proc 的接口， 待FastDeploy 的执行器重构后能架构会更合理一些。
yuanlehome
reviewed
Aug 8, 2025
Comment on lines
-81
to
-118
| up_gate_proj_weight: paddle.Tensor, | ||
| down_proj_weight: paddle.Tensor, | ||
| up_gate_proj_bias: Optional[paddle.Tensor], | ||
| up_gate_proj_scale: Optional[paddle.Tensor], | ||
| down_proj_scale: Optional[paddle.Tensor], | ||
| down_proj_in_scale: Optional[paddle.Tensor], | ||
| ffn1_weight: paddle.Tensor, | ||
| ffn2_weight: paddle.Tensor, | ||
| ffn1_bias: Optional[paddle.Tensor], | ||
| ffn1_scale: Optional[paddle.Tensor], | ||
| ffn2_scale: Optional[paddle.Tensor], | ||
| ffn2_in_scale: Optional[paddle.Tensor], | ||
| expert_idx_per_token: Optional[paddle.Tensor], | ||
| quant_method: str, | ||
| used_in_ep_low_latency: bool, | ||
| ): | ||
| assert up_gate_proj_bias is None | ||
| assert up_gate_proj_scale is not None | ||
| assert down_proj_scale is not None | ||
| assert down_proj_in_scale is None | ||
| assert ffn1_bias is None | ||
| assert ffn1_scale is not None | ||
| assert ffn2_scale is not None | ||
| assert ffn2_in_scale is None | ||
| assert expert_idx_per_token is None | ||
| assert quant_method in ("weight_only_int8") | ||
| assert not used_in_ep_low_latency | ||
| tokens_expert_prefix_sum_cpu = tokens_expert_prefix_sum.to("cpu") | ||
| up_gate_proj_output = paddle.empty( | ||
| [permute_input.shape[0], up_gate_proj_weight.shape[1]], | ||
| dtype=permute_input.dtype, | ||
| ) | ||
| group_gemm( | ||
| permute_input, | ||
| tokens_expert_prefix_sum_cpu, | ||
| up_gate_proj_weight, | ||
| up_gate_proj_scale, | ||
| up_gate_proj_output, | ||
| ) | ||
| act_out = swiglu(up_gate_proj_output) | ||
| output = paddle.empty([act_out.shape[0], down_proj_weight.shape[1]], dtype=act_out.dtype) | ||
| group_gemm( | ||
| act_out, | ||
| tokens_expert_prefix_sum_cpu, | ||
| down_proj_weight, | ||
| down_proj_scale, | ||
| output, | ||
| ) |
Collaborator
There was a problem hiding this comment.
这些建议保持原命名，FD中不建议出现ffn1/ffn2字样
xiaoguoguo626807
pushed a commit
to xiaoguoguo626807/FastDeploy
that referenced
this pull request
May 7, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
9 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
FD在天数硬件上的第一版性能优化，具体优化策略有：
该版基于GSM8K数据集跑erine45 300B模型总体耗时约6.3h，精度0.964