bugfix:DBO ubatch NPUGraph capture passes per-stage metadata to the standard full-graph updater - #178
bugfix:DBO ubatch NPUGraph capture passes per-stage metadata to the standard full-graph updater#178yujuancao07 wants to merge 1 commit into
Conversation
Signed-off-by: yujuancao07 <yujuancao07@gmail.com>
|
Thanks for the investigation. The root cause is not the shape of In the previous AFD v0.13-based implementation, MLA had a dedicated mechanism to update the full-graph parameters. That mechanism should be ported to the current implementation. Skipping This is why DeepSeek-V2-Lite cannot currently run with DBO + FULL graph. DeepSeek-V3.2 can run with the same configuration because it uses DSA rather than MLA, so it does not hit this MLA-specific gap. Therefore, we will not accept the current workaround as the fix. Please port the previous MLA-specific full-graph update mechanism and validate DeepSeek-V2-Lite with DBO + FULL graph. |
Purpose
When AFD runs with DBO,
ubatch-size > 1, and full decode NPUGraph capture, the Attention side can fail during graph warm-up with:AttributeError: 'list' object has no attribute 'keys'
Issue