K1 IME拓展 矩阵乘 程序执行错误

K1
1

我使用 vmdot 指定实现了一个简单的矩阵乘算法，验证功能。在运行时出现了 segmentfault 问题，求大佬解答。以及是否有编程手册及程序示例

#include <stdio.h>
#include <stdint.h>

int main() {
int M = 4;
int N = 4;
int K = 8;

int8_t A[M][K];
int8_t B[N][K];
int32_t C[M][N];
int32_t C_baseline[M][N];

// init A,B,C
for (int i = 0; i < M; i++) {
	for (int j = 0; j < K; j++) {
		A[i][j] = i;
	}
}
for (int i = 0; i < N; i++) {
	for (int j = 0; j < K; j++) {
		B[i][j] = i;
	}
}
for (int i = 0; i < M; i++) {
	for (int j = 0; j < N; j++) {
		C[i][j] = 0;
		C_baseline[i][j] = 0;
	}
}

	call_vmadot(A, B, C);

// after vmadot
printf("after vmadot C:\n");
for (int i = 0; i < M; i++) {
	for (int j = 0; j < N; j++) {
		printf("%d ", C[i][j]);
	}
	printf("\n");
}
printf("\n");

return 0;
}

.align 4

; riscv vmadot martrix mac unit [4,4,8]
; implement call_vmadot(a_base, b_base, c_base) function

.global call_vmadot
call_vmadot:
  ; save registers
  addi sp, sp, -16
  sw ra, 0(sp)
  sw a0, 4(sp)
  sw a1, 8(sp)
  sw a2, 12(sp)

  ; load arguments
  lw a0, 0(a0) ; a_base
  lw a1, 0(a1) ; b_base
  lw a2, 0(a2) ; c_base

  ; set vector length
	vsetvli t0, x0, e8, m1, ta, ma

  ; load a_base to v0
  vle8.v v0, (a0)

  ; load b_base to v1
  vle8.v v1, (a1)

  ; vmadot.vv v2, v0, v1
  vmadot v2, v0, v1

  ; store c_base to v2
	vsetvli t0, x0, e32, m2, ta, ma
  vse32.v v2, (a2)

  ; restore registers
  lw ra, 0(sp)
  lw a0, 4(sp)
  lw a1, 8(sp)
  lw a2, 12(sp)
  addi sp, sp, 16
  ret