the code comes from a Qt library that helps produce buttons with the shape of an image; it scans through all lines y and all the width x, generating the following change when the rgb part of the pixel coincides with the masking one (mp is the pointer at the start of the line and it is prefilled with 0xff):
*(mp + (x >> 3)) &= ~(1 << (x & 7));
I can’t really interpret it; anyone with background to give a hand?
From the looks of the code, mp points to the current line of a 1 bit per pixel image. The code clears the bit representing the pixel at X. It converts the X offset into a byte offset (x >> 3) and then logical AND’s the byte with a mask created from the inverse 1 shifted left by the X position within the byte.