我目前正在尝试找出如何在ARM中编写矩阵乘法的代码。我已经完成并理解了将在下面粘贴的“ C = AB”情况，但是无法弄清楚如何合并上图中的“ C = BA”情况。我已经研究了几个小时，无法弄清楚该怎么做。
.text
.global _start
_start:
@-- Code for initiation of needed loop
MOV r0 , #0 @-- Using r0 for "i" indexing
MOV r8 , #0 @-- Using a seperate index for arrays compared to above
LDR r2 , =I @-- Getting address for I here
LDR r1 , [r2] @-- Getting the value of I here
MOV r2 , #0 @-- Using r2 for "f" indexing
LDR r3 , =a @-- Loading r3 with the base of "a" in .data
LDR r5 , =b @-- Loading r5 with the base of "b" in .data
LDR r7 , =c @-- Loading r7 with the base of "c" in .data
@-- Loop body which was initiated above
loop:
LDR r4 , [r3 , r8] @-- Getting a[i] here
LDR r6 , [r5 , r8] @-- Getting b[i] here
MUL r9 , r4 , r6 @-- Multiplying a[i] * b[i] from above LDR's
ADD r2 , r2 , r9 @-- Add the above in our running sum
STR r2 , [r7] @-- Store that result in memory for use
ADD r8 , r8 , #4 @-- Adding word offset to our index to prep for add next
ADD r0 , r0 , #1 @-- Adding 1 to i here
CMP r0 , r1 @-- Determining if we should exit
BLT loop @ "if i < I , continue"
MLA r0 , r1 , r2 , r3
SWI 0x11
.data @-- Data Section
I: .word 4 @-- Variable
a: .word 1 , 2 , 3 , 4 @-- This is Array A
b: .word 9 , 8 , 7 , 6 @-- This is Array B
c: .word 0
.end