Could you provide a more complex example like a GEMM with AMX ?

[kzjeef]

Could you provide a more complex example like a larger gemm (more than 16x64, like 1024x1024) to show how to use change the load tile register while moving through different axis ?

For more clear explanation, like this figure, the M is larger than 16 and K is larger than 64, and N is also larger than 16, so the tile calculation should be moving, but from the tile config, I cannot figure out how to moving in N axis(B matrix).

A more practical gemm example will be help, thanks.