Thanks for sending an issue! Please fill in the following template to help quickly solve your problem.

Describe the current behavior / 问题描述 (Mandatory / 必填)

WeightQuantBatchMatmulV2算子在N非16对齐场景的某些case情况下存在内存越界访问问题：

问题场景：

• W4A16 PerGroup-ND场景
• x, w 非转置

问题用例

• case1: x=[80704, 66915], w=[66915, 276], antiquantScale=[131, 276], bias=[1, 276]
• case2: x=[26422, 3460], w=[3460, 1208], antiquantScale=[8, 1208], antiquantOffset=[8, 1208], bias=[1, 1208]

Environment / 环境信息 (Mandatory / 必填)

Ascend950

Steps to reproduce the issue / 重现步骤 (Mandatory / 必填)

复现该问题需开启OOM检测，输入上述用例即可。

Describe the expected behavior / 预期结果 (Mandatory / 必填)

预期精度正常，无内存越界访问，OOM检测通过。

Related log / screenshot / 日志 / 截图 (Mandatory / 必填)

无

Special notes for this issue/备注 (Optional / 选填)