You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The vector mask load instruction does not work correctly. For VLEN=256 and SEW=8, LMUL=8 set previously, mask of 32 bytes should be loaded into mask register, but only 4 bytes are loaded actually. That's because micro_vl is constructed as follows in the Vlm_v's constructor. int32_t micro_vl = (std::min(this->vl, micro_vlmax) + 7) / 8;
Though this->vl is 256, micro_vlmax is calculated to be 32.
The text was updated successfully, but these errors were encountered:
The vector mask load instruction does not work correctly. For VLEN=256 and SEW=8, LMUL=8 set previously, mask of 32 bytes should be loaded into mask register, but only 4 bytes are loaded actually. That's because micro_vl is constructed as follows in the Vlm_v's constructor.
int32_t micro_vl = (std::min(this->vl, micro_vlmax) + 7) / 8;
Though
this->vl
is 256, micro_vlmax is calculated to be 32.The text was updated successfully, but these errors were encountered: