CANN: refactor ACL graph cache matching #18
Conversation
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
* CANN: implement LRU cache for ACL graphs in CANN backend - Introduce ggml_cann_graph_lru_cache to store multiple ggml_cann_graph objects. - Graphs are loaded on demand and evicted using LRU policy when capacity is exceeded. - Updated push, move_to_front, and clear methods to manage cached graphs efficiently. - Ensures reuse of graphs, reducing graph reconstruction overhead in CANN backend. * fix typo * The LRU cache capacity can be configured via an env variable Signed-off-by: noemotiovon <[email protected]> * refactory acl graph * refactory && fix review comments Signed-off-by: noemotiovon <[email protected]> --------- Signed-off-by: noemotiovon <[email protected]>
* CANN: improve ACL graph matching Record `ne` and `nb` information for src tensors and include them in the graph matching check. This enhances the robustness of ACL graph matching by preventing incorrect matches when src tensors share the same data address but differ in shape or stride. * CANN: add op_params match
* CANN: Refactor `evaluate_and_capture_cann_graph` **Description of the problem** * `matched_graph` is obtained even if graph mode is disabled. * End of graph capture and graph replay are unnecessarily placed in different `if` blocks. **Proposed solution** * Obtain `matched_graph` only if graph mode is enabled. * Place end of graph capture and graph reply inside the same `if` block. * Unify graph related comments. * Remove trailing whitespace
**Description of the problem** `cann_graph_update_required` is redundantly defined and initialized as `false` inside two mutually exclusive macro branches. **Proposed solution** Define it right before the macro so that it could serve both branches.
wangweixuan
force-pushed
the
acl_graph_cache_refactor
branch
from
1c8ddc8 to
79d3f3d
Compare
Move the graph property checking code into methods of LRU cache. Signed-off-by: Wang Weixuan <[email protected]>
wangweixuan
force-pushed
the
acl_graph_cache_refactor
branch
from
79d3f3d to
c406a52
Compare
Owner
|
LGTM，我的master落后上游很多，可以直接贡献到上游社区嘛 |
Author
I have opened new PR: ggml-org#17752 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
Description
Motivation: This PR refactors the LRU cache responsible for mapping GGML graph properties to CANN graphs. We have reorganized the graph property-related code from the lengthy
ggml-cann.cppfile into dedicated structuresggml_graph_node_properties,ggml_cann_graph, andggml_cann_graph_lru_cache.Change summary: This change aims to improve code clarity by separating concerns. There are no functional changes. The code has been formatted with clang-format.
Testing
Notes
This PR is based on a cherry-pick of existing LRU cache implementation ggml-org#15814. The original changes are in the last commit.
CC @noemotiovon