C ++ bit alignment operation

See a macro: # define pad_align (x, mask) (x + mask )&~ Mask)

 

First, you must understand the alignment operation, that is, when the number of digits is not multiple of Y, the last y is used to complete. For example, 12 is not a multiple of 8. If it is aligned with 8, the number of digits should be 16. For example, when a 24-Bit Bitmap is stored, the number of rows must be a multiple of 32.

Let's look at this macro again. If one of the following mask (number of 00011111,1) bits of x exists (not all values 0 cannot be rounded out by mask + 1, here the mask is 2 ^ N-1 ), the addition of a mask is the completion, and the number of digits is increased, but it is still not divisible by the mask + 1. Don't worry ~ Mask is followed by 0, so it can be divisible by mask + 1.