Yes, I think it was designed in particular for determining addresses of fields within structures and variables on the stack, for example if you had: struct

Yes, I think it was designed in particular for determining addresses of fields within structures and variables on the stack, for example if you had: struct foo { int field1; int field2[10]; } Then an access like: fooptr->field2[index] Would compile to: fooptr + sizeof(int) * index + offsetof(struct foo, field2) Which is: lea $4(fooptr,index,4)