|
|
@@ -100,7 +100,7 @@ static inline int32_t multiply_32x32_rshift32(int32_t a, int32_t b) |
|
|
|
asm volatile("smmul %0, %1, %2" : "=r" (out) : "r" (a), "r" (b)); |
|
|
|
return out; |
|
|
|
#elif defined(KINETISL) |
|
|
|
return 0; // TODO.... |
|
|
|
return ((int64_t)a * (int64_t)b) >> 32; |
|
|
|
#endif |
|
|
|
} |
|
|
|
|
|
|
@@ -113,7 +113,7 @@ static inline int32_t multiply_32x32_rshift32_rounded(int32_t a, int32_t b) |
|
|
|
asm volatile("smmulr %0, %1, %2" : "=r" (out) : "r" (a), "r" (b)); |
|
|
|
return out; |
|
|
|
#elif defined(KINETISL) |
|
|
|
return 0; // TODO.... |
|
|
|
return (((int64_t)a * (int64_t)b) + 0x8000000) >> 32; |
|
|
|
#endif |
|
|
|
} |
|
|
|
|
|
|
@@ -126,7 +126,7 @@ static inline int32_t multiply_accumulate_32x32_rshift32_rounded(int32_t sum, in |
|
|
|
asm volatile("smmlar %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b)); |
|
|
|
return out; |
|
|
|
#elif defined(KINETISL) |
|
|
|
return 0; // TODO.... |
|
|
|
return sum + ((((int64_t)a * (int64_t)b) + 0x8000000) >> 32); |
|
|
|
#endif |
|
|
|
} |
|
|
|
|
|
|
@@ -139,7 +139,7 @@ static inline int32_t multiply_subtract_32x32_rshift32_rounded(int32_t sum, int3 |
|
|
|
asm volatile("smmlsr %0, %2, %3, %1" : "=r" (out) : "r" (sum), "r" (a), "r" (b)); |
|
|
|
return out; |
|
|
|
#elif defined(KINETISL) |
|
|
|
return 0; // TODO.... |
|
|
|
return sum - ((((int64_t)a * (int64_t)b) + 0x8000000) >> 32); |
|
|
|
#endif |
|
|
|
} |
|
|
|
|