[AutoScheduler] Improve SearchTask and ComputeDAG serialization #7145
Conversation
This file contains 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
jcf94
reviewed
Dec 22, 2020
e88dafa to
dfecfa7
Compare
dfecfa7 to
dda0ea2
Compare
|
Per offline discussion, now we only support (de)serialization of ComputeDAG constructed by compute, because this limitation can largely simplify the design. @merrymercy @jcf94 PTAL. |
merrymercy
approved these changes
Dec 23, 2020
masahi
pushed a commit
to masahi/tvm
that referenced
this pull request
Dec 24, 2020
…he#7145) * Use self.dag in Python object * Add sch to ComputeDAG * address comment
tkonolige
pushed a commit
to tkonolige/incubator-tvm
that referenced
this pull request
Jan 11, 2021
…he#7145) * Use self.dag in Python object * Add sch to ComputeDAG * address comment
TusharKanekiDey
pushed a commit
to TusharKanekiDey/tvm
that referenced
this pull request
Jan 20, 2021
…he#7145) * Use self.dag in Python object * Add sch to ComputeDAG * address comment
trevor-m
pushed a commit
to neo-ai/tvm
that referenced
this pull request
Jan 21, 2021
…he#7145) * Use self.dag in Python object * Add sch to ComputeDAG * address comment
electriclilies
pushed a commit
to electriclilies/tvm
that referenced
this pull request
Feb 18, 2021
…he#7145) * Use self.dag in Python object * Add sch to ComputeDAG * address comment
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
The previous implementation of task serialization mechanism has a potential problem, which is a bit tricky so I didn't notice that before.
Previously, I defined a set of attributes in the Python object (e.g.,
hardware_params) and use them to be the state when serializing a task object in Python. However, in the case that users don't providehardware_params, the SearchTask constructor in C++ invokesGetDefaultHardwareParamsto get the default hardware parameters. These default hardware parameters aren't exposed to the Python object because of the same name attribute.It wouldn't be a bit deal because when a task is deserialized, we follow the same process to get the default hardware parameters. However, since
GetDefaultHardwareParamsmay access hardware context (e.g., CUDA context), this prevents tasks from being deserialized in parallel (with multiprocessing).This PR removes all attributes in
SearchTaskPython object to make sure we access the up-to-date C++ object attributes when serializing a task. The only exception isself.dagbecause we need it to reconstruct a ComputeDAG.cc @merrymercy @jcf94