Does *&s1 refer to the first member of structure variable s1

[lovecreatesbea...]

Does the expression *(int *)&s1 below inside the printf() statement
guarantee to refer to the first member of the structure variable s1?
I've tried the code and it seems that it works that way. The C
standard states this? Thank you very much.

#include <stdio.h>

struct S{
int m1, m2, m3;
};

int main(void)
{
struct S s1 = {3, 4, 5};
printf("%d\n", *(int *)&s1);
return 0;
}

/*
$ gcc -ansi -pedantic -W -Wall a.c
$ ./a.exe

3

$
*/

[Jens Thoms Toerring]

[e-mail] <lo***************@gmail.comwrote:

Does the expression *(int *)&s1 below inside the printf() statement
guarantee to refer to the first member of the structure variable s1?
I've tried the code and it seems that it works that way. The C
standard states this? Thank you very much.

#include <stdio.h>

struct S{
int m1, m2, m3;
};

int main(void)
{
struct S s1 = {3, 4, 5};
printf("%d\n", *(int *)&s1);
return 0;
}

Yes, this is ok since it's guaranteed that the first member of a
structure always is at the very start of the structure without
any padding bytes before the first member.

$ gcc -ansi -pedantic -W -Wall a.c
$ ./a.exe

The compiler warning flags don't help you much if you use casts -
they tell the compiler that you know what you're doing, so don't
expect too many helpful warnings...

Regards, Jens
--
\ Jens Thoms Toerring ___ [e-mail]
\__________________________ http://toerring.de

[Eric Sosman]

[e-mail] wrote On 07/05/07 10:24,:

Does the expression *(int *)&s1 below inside the printf() statement
guarantee to refer to the first member of the structure variable s1?
I've tried the code and it seems that it works that way. The C
standard states this? Thank you very much.

#include <stdio.h>

struct S{
int m1, m2, m3;
};

int main(void)
{
struct S s1 = {3, 4, 5};
printf("%d\n", *(int *)&s1);
return 0;
}

Yes. A pointer to a struct can be converted to a
pointer to the struct's first element (if it is possible
to make such a pointer at all), and vice versa.

This guarantee applies only to the *first* element,
not to any subsequent elements. In your sample, it is
*not* guaranteed that (int*)&s1+1 == &s1.m2.

--
Er*********@sun.com

[Christopher Benson-Manica]

Eric Sosman <Er*********@sun.comwrote:

Yes. A pointer to a struct can be converted to a
pointer to the struct's first element (if it is possible
to make such a pointer at all), and vice versa.

Can you cite C&V on this? I know it's there somewhere, but my first
guess (6.7.2.1) proved to be wrong.

--
C. Benson Manica | I *should* know what I'm talking about - if I
cbmanica(at)gmail.com | don't, I need to know. Flames welcome.

[Ralf Damaschke]

Christopher Benson-Manica wrote:

Eric Sosman <Er*********@sun.comwrote:

> Yes. A pointer to a struct can be converted to a
pointer to the struct's first element (if it is possible
to make such a pointer at all), and vice versa.

Can you cite C&V on this? I know it's there somewhere, but my
first guess (6.7.2.1) proved to be wrong.

It's inside 6.7.2.1 p13:

[...] A pointer to a structure object, suitably converted, points
to its initial member [...], and vice versa.

Ralf

[Christopher Benson-Manica]

Ralf Damaschke <rw****@gmx.dewrote:

It's inside 6.7.2.1 p13:

[...] A pointer to a structure object, suitably converted, points
to its initial member [...], and vice versa.

Darn, so it is, albeit buried in a paragraph that seems initially to
be about something else. Thanks.

--
C. Benson Manica | I *should* know what I'm talking about - if I
cbmanica(at)gmail.com | don't, I need to know. Flames welcome.