apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-05 16:48:03 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[feat] support prefix cache clearing when /clear_load_weight is called (#4091)
Some checks failed
CE Compile Job / ce_job_pre_check (push) Has been cancelled
Details
CE Compile Job / print_ce_job_pre_check_outputs (push) Has been cancelled
Details
CE Compile Job / FD-Clone-Linux (push) Has been cancelled
Details
CE Compile Job / Show Code Archive Output (push) Has been cancelled
Details
CE Compile Job / BUILD_SM8090 (push) Has been cancelled
Details
CE Compile Job / BUILD_SM8689 (push) Has been cancelled
Details
CE Compile Job / CE_UPLOAD (push) Has been cancelled
Details

Browse Source 
* [feat] support clearing prefix cache (cherry-picked from release/2.1)

* [fix] fix ipc suffix, use port instead

* [fix] fix prefix caching not enabled

* [fix] fix code style

* [fix] wait for rank0 to update weight status
This commit is contained in:
李泳桦
2025-09-16 11:11:20 +08:00
committed by GitHub
parent fbb4e0f8d1
commit 7ccbcc5a62
17 changed files with 624 additions and 181 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
fastdeploy/cache_manager/cache_messager.py
View File

@@ -98,8 +98,8 @@ class CacheMessager:
cache_v = []
self.messager = {}
for layer_idx in range(self.num_layers):
key_cache = self.gpu_cache_kvs[f"key_caches_{layer_idx}_rank{self.rank}_device{gpu_id}"]
val_cache = self.gpu_cache_kvs[f"value_caches_{layer_idx}_rank{self.rank}_device{gpu_id}"]
key_cache = self.gpu_cache_kvs[f"key_caches_{layer_idx}_rank{self.rank}.device{gpu_id}"]
val_cache = self.gpu_cache_kvs[f"value_caches_{layer_idx}_rank{self.rank}.device{gpu_id}"]
cache_k.append(key_cache)
cache_v.append(val_cache)
cache_k_ptr_list.append(key_cache.data_ptr())