Author
Hello
Does anyone know what''s the problem of the following declarting of struct?
I am getting a redefiniation of struct pcb error
Thanks
#define PCB struct pcb
PCB
{
PCB *ptr;
char *name;
UINT32 id;
UINT32 flag;
UINT32 program_counter;
// DATA REGISTER
UINT32 reg_D0;
UINT32 reg_D1;
UINT32 reg_D2;
UINT32 reg_D3;
UINT32 reg_D4;
UINT32 reg_D5;
UINT32 reg_D6;
UINT32 reg_D7;
// ADDRESS REGISTER
UINT32 reg_A0;
UINT32 reg_A1;
UINT32 reg_A2;
UINT32 reg_A3;
UINT32 reg_A4;
UINT32 reg_A5;
UINT32 reg_A6;
UINT32 reg_A7;
UINT32 process_state;
UINT32 priority;
MSG_ENV *msg_ptr_head;
MSG_ENV *msg_ptr_tail;
};
#define PCB struct pcb
PCB
{
PCB *ptr;
this expands to
struct pcb
{
struct pcb *ptr;
your redefining the struct inside the struct
extended waranty, how can I lose!
[edited by - brass_fish on March 15, 2003 10:50:12 PM]
[edited by - brass_fish on March 15, 2003 10:51:21 PM]
I could be wrong about this (it''s been a while since I looked at the C standard) but I don''t think compilers are required to allow pointers to incomplete types. Try sticking "struct pcb;" before your definition, and see if that fixes it.
But... but that''s what HITLER would say!!
But... but that''s what HITLER would say!!
C compilers are required to support pointers to incomplete types (Section 6.2.5 paragraph 19 in the C99 standard (well the final draft anyway)). btw, inserting struct pcb; before it''s definition doesn''t change the fact that is an incomplete type inside the definition.
As far as I can tell, unless your definitions of UINT32 and MSG_ENV are really screwy, your declaration is fine by itself. Emphasis on by itself . If you have in a header file and forgot inclusion guards it could cause problems.
As far as I can tell, unless your definitions of UINT32 and MSG_ENV are really screwy, your declaration is fine by itself. Emphasis on by itself . If you have in a header file and forgot inclusion guards it could cause problems.
This topic is closed to new replies.
Advertisement
Popular Topics
Advertisement
