[feat] mlapo add bf16 no_quant support (#4852)

### What this PR does / why we need it?
This PR adds mlapo operation support for bf16 no_quant mode.

### Does this PR introduce _any_ user-facing change?
This PR makes quant related parameters optional. 
### How was this patch tested?
CI passed with new added/existing test.

- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c

---------

Signed-off-by: chenjunyi <isjunyi.chen@gmail.com>
This commit is contained in:
chenjunyi
2025-12-11 11:06:56 +08:00
committed by GitHub
parent c95c271538
commit c12eb22cbe
12 changed files with 1510 additions and 81 deletions

View File

@@ -90,6 +90,11 @@ struct MlaTilingData {
uint32_t esqHeadTail{0};
uint32_t esqColLoop{0};
uint32_t esqColTail{0};
// hidden state dimension
uint32_t hiddenStateDim{7168};
uint32_t isWeightQuantized{1};
};
#endif // MLAPREPROCESS_TILING_H