结构体内存对齐问题(国外英语资料).docVIP

结构体内存对齐问题(国外英语资料) When a structure is defined in C, its size is equal to the size of the field (field) and? How will the compiler in memory placed in these fields? ANSI C on the structure of the memory layout of what are the requirements? Our program can rely on this layout? These problems may be a lot of friends and a little fuzzy, so this paper will try to explore the secrets behind them. First of all, at least one thing for sure, it is ANSI C to ensure the field structure in memory occurs with their declaration order increasing, and the first field is equal to the first address of the entire structure, examples of the first address. For example, such a structure: Struct vector{int x, y, Z; s}; Int *p, *q, *r; Struct vector *ps; P = s.x; Q = s.y; R = s.z; PS = s; Assert (P Q); Assert (P R); Assert (Q R); Assert ((int*) PS = = P); The above assertion / fail At this time, a friend might ask: the standard whether the provisions of the adjacent fields are also adjacent in memory?. Oh, sorry, ANSI C did not make that your program at any time should not rely on this assumption. Does this mean we will never be able to outline a clearer structure memory layout more accurate? Oh, of course not. But let us temporarily retreat from this problem, look at another important issue -- memory alignment. The computer system has many practical restrictions on the basic types of data stored in the memory location, they will require these data the first address value is a number k (usually it is 4 or 8 times), this is the so-called memory alignment, and this k is called alignment modulus of this data type (alignment modulus). When the ratio of modulus of alignment alignment modulus a type of S with another type of T is an integer greater than 1, we call type S alignment is stronger than T (Yan Ge), and that T is weaker than S (Kuan Song). This mandatory requirement way simplifies the design of transmission system between the processor and memory, and can improve the speed of readi