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

Jump to bottom

[Hold for pipeline test] Support control-flow #353

Draft
wants to merge 19 commits into
base: master
Choose a base branch
from
Draft

[Hold for pipeline test] Support control-flow #353

wants to merge 19 commits into from

Conversation

xysmlx
Copy link
Contributor

@xysmlx xysmlx commented Nov 18, 2021 •
edited
Loading

  • ONNX frontend
    • Refactor GraphConvert
    • Loop op convert
    • If op convert
    • Refactor
  • Fix benchmark convert problems
    • GatherV2 check failure (temporally comment generic_op translate function to bypass this problem)
    • Support ScatterND (TODO: cuda kernel emitter)
  • NNFusion graph and graph passes
    • Build graph with control-flow ops
      • Loop op define
      • If op define
    • Refactor graph passes for control-flow ops
      • Memory allocator
  • Backend codegen

xysmlx and others added 19 commits July 15, 2021 12:30
@xysmlx
some util functions for onnx frontend
a488ac2
@xysmlx
more datatype support in constant op
ae261c7
@xysmlx
draft: onnx frontend support for if and loop op
73d3e6f
@xysmlx
draft: if and loop op_define
6cff7e7
@xysmlx
refactor GraphConvert of the ONNX frontend, support if and loop convert
ed0221a
@xysmlx
draft: cuda kernel emitter for if and loop op (placeholder)
1a30651
@xysmlx
update onnx frontend convert and shape inference for loop op
436d7d3
@xysmlx
fix output size bug in ONNX Loop op convert
a2863d7
@xysmlx
Generic_op_define and ONNX opset_11 frontend converter for ScatterND op
b7b66b2
@xysmlx
Comment m_expression construction of generic_op to bypass translate f…
ffae3f7 
…unction problem
@xysmlx
Merge branch 'master' into control_flow
f771004
@xysmlx
Merge master branch into control_flow branch
811a126
@xysmlx
disable ORT optimizations
a77ee7b
@xysmlx
fix bug for disabling ORT optimizations
d42be30
@xysmlx
fix bug for disabling ORT optimizations
0a29cc1
@xysmlx
Support -ftuning_list in kernel tuning pass
b0d4c56
@xysmlx
add mod operator
324f2cf
@xysmlx
ad-hoc support tuning_list in AntaresCudaKernelEmitter
bcf5b7d
@xysmlx
fix bug in tuning_list
32ebc44
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@xysmlx