transpose8()

template<int m, int n>

FASTLED_FORCE_INLINE void fl::transpose8	(	unsigned char *	A,
		unsigned char *	B )

Templated 8x8 bit transpose with custom stride.

Based on: https://web.archive.org/web/20190108225554/http://www.hackersdelight.org/hdcodetxt/transpose8.c.txt

Template Parameters

m	Input stride (1 for sequential bytes)
n	Output stride (1 for sequential bytes)

Parameters

A	Input array (8 bytes at stride m)
B	Output array (8 bytes at stride n, transposed)

Examples

/home/runner/work/FastLED/FastLED/src/fl/math/transposition.h.

Definition at line 186 of file transposition.h.

                                                                                     {
    fl::u32 x, y, t;
 
    // Load the array and pack it into x and y.
    if(m == 1) {
        y = *(fl::u32*)(A);
        x = *(fl::u32*)(A+4);
    } else {
        x = (fl::u32(A[0])<<24)   | (fl::u32(A[m])<<16)   | (fl::u32(A[2*m])<<8) | A[3*m];
        y = (fl::u32(A[4*m])<<24) | (fl::u32(A[5*m])<<16) | (fl::u32(A[6*m])<<8) | A[7*m];
    }
 
    // pre-transform x
    t = (x ^ (x >> 7)) & 0x00AA00AA;  x = x ^ t ^ (t << 7);
    t = (x ^ (x >>14)) & 0x0000CCCC;  x = x ^ t ^ (t <<14);
 
    // pre-transform y
    t = (y ^ (y >> 7)) & 0x00AA00AA;  y = y ^ t ^ (t << 7);
    t = (y ^ (y >>14)) & 0x0000CCCC;  y = y ^ t ^ (t <<14);
 
    // final transform
    t = (x & 0xF0F0F0F0) | ((y >> 4) & 0x0F0F0F0F);
    y = ((x << 4) & 0xF0F0F0F0) | (y & 0x0F0F0F0F);
    x = t;
 
    B[7*n] = y; y >>= 8;
    B[6*n] = y; y >>= 8;
    B[5*n] = y; y >>= 8;
    B[4*n] = y;
 
    B[3*n] = x; x >>= 8;
    B[2*n] = x; x >>= 8;
    B[n] = x; x >>= 8;
    B[0] = x;
}

References B, FASTLED_FORCE_INLINE, FL_NOEXCEPT, t, x, and y.