CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-27 08:21:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

CANN: implement LRU cache for ACL graphs (#15814)

Browse Source 
* 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 <757486878@qq.com>

* refactory acl graph

* refactory && fix review comments

Signed-off-by: noemotiovon <757486878@qq.com>

---------

Signed-off-by: noemotiovon <757486878@qq.com>
This commit is contained in:
Chenguang Li
2025-09-10 15:29:12 +08:00
committed by GitHub
parent 86587da03b
commit 28b5f190ef
3 changed files with 164 additions and 51 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
docs/backend/CANN.md
View File

@@ -314,3 +314,7 @@ Converting the matmul weight format from ND to NZ to improve performance. Enable
### GGML_CANN_ACL_GRAPH
Operators are executed using ACL graph execution, rather than in op-by-op (eager) mode. Enabled by default.
### GGML_CANN_GRAPH_CACHE_CAPACITY
Maximum number of compiled CANN graphs kept in the LRU cache, default is 12. When the number of cached graphs exceeds this capacity, the least recently used graph will be evicted.