printf() error with long double and null pointer.

**A. Sinan Unur** · Feb 24 '06, 12:55 PM

Re: printf() error with long double and null pointer.

fieldfallow <fieldfallow@gm ail.com> wrote in news:dtmt1n$1g3 $1
@emma.aioe.org:
[color=blue]
> I thought null pointers had a value of zero. Why is gcc's executable
> printing ffffffffh?[/color]

Null Pointers

http://c-faq.com/null/index.html

Sinan
--
A. Sinan Unur <1usa@llenroc.u de.invalid>
(reverse each component and remove .invalid for email address)

**pete** · Feb 24 '06, 12:55 PM

Re: printf() error with long double and null pointer.

fieldfallow wrote:[color=blue]
>
> Hello all,
>
> Before stating my question,
> I should mention that I'm fairly new to C.
>
> Now, I attempted a small demo that prints out
> the values of C's numeric types, both uninitialised[/color]

Accessing uninitialised values like that,
causes undefined behavior in your program.
The undefined behavior can show up in any part
of the program's execution.

--
pete

**fieldfallow** · Feb 24 '06, 01:05 PM

Re: printf() error with long double and null pointer.

pete wrote:[color=blue]
> fieldfallow wrote:[color=green]
>> Hello all,
>>
>> Before stating my question,
>> I should mention that I'm fairly new to C.
>>
>> Now, I attempted a small demo that prints out
>> the values of C's numeric types, both uninitialised[/color]
>
> Accessing uninitialised values like that,
> causes undefined behavior in your program.
> The undefined behavior can show up in any part
> of the program's execution.[/color]

Does even printing out the values cause undefined behaviour? Is any
other operation legal, or must I assign them values before I do anything
with them?

Thanks for your answers. The C book I currently have is not very clear
on this point.

**Robin Haigh** · Feb 24 '06, 01:15 PM

Re: printf() error with long double and null pointer.

"fieldfallo w" <fieldfallow@gm ail.com> wrote in message
news:dtmt1n$1g3 $1@emma.aioe.or g...[color=blue]
> Hello all,
>
> Before stating my question, I should mention that I'm fairly new to C.
>
> Now, I attempted a small demo that prints out the values of C's numeric
> types, both uninitialised and after assigning them their maximum defined
> values. However, the output of printf() for the long double 'ld' and the
> pointer of type void 'v_p', after initialisation don't seem to be right.
>
> The compiler used was gcc (mingw) with '-Wall', '-std=c99' and
> '-pedantic' switches. No warnings were emitted. Incidentally the MS
> Visual C++ 2003 compiler's output seems okay. I give them both below:
>
> gcc's output:
> ld == -1.#QNAN0
> or -1.#QNAN
> v_p == FFFFFFFF
>
> msvc's output:
> ld == 179769313486231 570000000000000 000000000000000 000000000000000 000000
> 000000000000000 000000000000000 000000000000000 000000000000000 000000000000
> 000000000000000 000000000000000 000000000000000 000000000000000 000000000000
> 000000000000000 000000000000000 000000000000000 000000000000000 000000000000
> 000000000000000 000000000000.00 0000
> or 1.79769e+308
> v_p == 00000000
>
> I thought null pointers had a value of zero. Why is gcc's executable
> printing ffffffffh? Also the value of 'ld' seems to be wrong.
>
> The code for the demo is:
> #include <stdio.h>
> #include <limits.h>
> #include <float.h>
>
> int main(void) {
> char c;
> unsigned char uc;
> short s;
> unsigned short us;
> int i;
> unsigned int ui;
> long l;
> unsigned long ul;
> float f;
> double d;
> long double ld;
> void *v_p;
>
> printf("c == %c\n\tor %d\nuc == %c\n\tor %u\ns == %hd\nus == %hu\n"
> "i == %d\nui == %u\nl == %ld\nul == %lu\nf == %f\n\tor %g\n"
> "d == %lf\n\tor %g\nld == %Lf\n\tor %Lg\nv_p == %p\n",
> c,c,uc,uc,s,us, i,ui,l,ul,f,f,d ,d,ld,ld,v_p);
>
> puts("Initialis ing them with their maximum allowed values...");
> c = CHAR_MAX;
> uc = UCHAR_MAX;
> s = SHRT_MAX;
> us = USHRT_MAX;
> i = INT_MAX;
> ui = UINT_MAX;
> l = LONG_MAX;
> ul = ULONG_MAX;
> f = FLT_MAX;
> d = DBL_MAX;
> ld = LDBL_MAX;
> puts("Initialis ing v_p with NULL...");
> v_p = NULL;
>
> printf("c == %c\n\tor %d\nuc == %c\n\tor %u\ns == %hd\nus == %hu\n"
> "i == %d\nui == %u\nl == %ld\nul == %lu\nf == %f\n\tor %g\n"
> "d == %lf\n\tor %g\nld == %Lf\n\tor %Lg\nv_p == %p\n",
> c,c,uc,uc,s,us, i,ui,l,ul,f,f,d ,d,ld,ld,v_p);
> return 0;
> }
>
> Where is the mistake?
> Thanks for all the help.[/color]

Works for me running gcc on Cygwin.

%lf (alternative to %f) for double is a new feature in C99. Does your
printf support it?

If you're not on Linux or Cygwin, you may be using your system's libc, in
which case you get what you get, notwithstanding the C99 flag.

--
RSH

**pete** · Feb 24 '06, 01:25 PM

Re: printf() error with long double and null pointer.

fieldfallow wrote:[color=blue]
>
> pete wrote:[color=green]
> > fieldfallow wrote:[color=darkred]
> >> Hello all,
> >>
> >> Before stating my question,
> >> I should mention that I'm fairly new to C.
> >>
> >> Now, I attempted a small demo that prints out
> >> the values of C's numeric types, both uninitialised[/color]
> >
> > Accessing uninitialised values like that,
> > causes undefined behavior in your program.
> > The undefined behavior can show up in any part
> > of the program's execution.[/color]
>
> Does even printing out the values cause undefined behaviour?[/color]

Yes.
[color=blue]
> Is any
> other operation legal, or must I assign them values before I do
> anything with them?[/color]

For uninitialised objects,
you can only portably do operations which don't access the values.
That's pretty much limited to having the variables be
operands of the address operator and or the sizeof operator.

--
pete

**Richard Tobin** · Feb 24 '06, 01:35 PM

Re: printf() error with long double and null pointer.

In article <dtmvse$km3$1@e mma.aioe.org>,
fieldfallow <fieldfallow@gm ail.com> wrote:
[color=blue][color=green]
>> Accessing uninitialised values like that,
>> causes undefined behavior in your program.
>> The undefined behavior can show up in any part
>> of the program's execution.[/color][/color]
[color=blue]
>Does even printing out the values cause undefined behaviour?[/color]

Yes, but I would be surprised if it caused the behaviour shown.
Try removing the first printf and see if you still get the same
results for the second.

-- Richard

**fieldfallow** · Feb 24 '06, 02:06 PM

Re: printf() error with long double and null pointer.

Richard Tobin wrote:[color=blue]
> In article <dtmvse$km3$1@e mma.aioe.org>,
> fieldfallow <fieldfallow@gm ail.com> wrote:
>[color=green][color=darkred]
>>> Accessing uninitialised values like that,
>>> causes undefined behavior in your program.
>>> The undefined behavior can show up in any part
>>> of the program's execution.[/color][/color]
>[color=green]
>> Does even printing out the values cause undefined behaviour?[/color]
>
> Yes, but I would be surprised if it caused the behaviour shown.
> Try removing the first printf and see if you still get the same
> results for the second.[/color]

Yes. Even after commenting out the first printf(), I still get the same
wrong output. Would this indicate a fault in the particular toolchain,
i.e. gcc (MinGW)?

The output is:
Initialising them with their maximum allowed values...
Initialising v_p with NULL...
c ==
or 127
uc == ÿ
or 255
s == 32767
us == 65535
i == 2147483647
ui == 4294967295
l == 2147483647
ul == 4294967295
f == 340282346638528 860000000000000 000000000.00000 0
or 3.40282e+038
d == 179769313486231 570000000000000 000000000000000 000000000000000 00000000
000000000000000 000000000000000 000000000000000 000000000000000 00000000
000000000000000 000000000000000 000000000000000 000000000000000 00000000
000000000000000 000000000000000 000000000000000 000000000000000 00000000
000000000000000 000000000000000 0000000.000000
or 1.79769e+308
ld == -1.#QNAN0
or -1.#QNAN
v_p == FFFFFFFF

Thanks.

**pete** · Feb 24 '06, 02:17 PM

Re: printf() error with long double and null pointer.

fieldfallow wrote:
[color=blue]
> ld == -1.#QNAN0
> or -1.#QNAN[/color]

That result is also wrong.

Are you using still using %lf to output doubles?

%f is for both doubles and floats.

C99 allows %lf for doubles, but most compilers are C90
where using %lf to output a double is also undefined.

--
pete

**fieldfallow** · Feb 24 '06, 02:17 PM

Re: printf() error with long double and null pointer.

Robin Haigh wrote:[color=blue]
> Works for me running gcc on Cygwin.[/color]

Strange. My compiler is gcc 3.4.5 for MinGW 5.0
[color=blue]
> %lf (alternative to %f) for double is a new feature in C99. Does your
> printf support it?[/color]

I think so. The output for the double seems to be correct. Only those of
the long double and void pointer seem to be wrong. I'm saying this
because the MS Visual C++ 2003 toolkit compiler's output for the double
tallies with gcc. In addition the max. for long double is the same as
double and void pointer prints out zero after assigning NULL to it.
[color=blue]
> If you're not on Linux or Cygwin, you may be using your system's libc, in
> which case you get what you get, notwithstanding the C99 flag.[/color]

In this case, the MSVC compiled executable should also exhibit the same
error, shouldn't it.

Anyway, this seems to be heading a bit off-topic here. Maybe a MinGW
group would be better.

Thanks everyone for your help.

**Richard Tobin** · Feb 24 '06, 02:17 PM

Re: printf() error with long double and null pointer.

In article <43FF1048.10506 09@gmail.com>,
fieldfallow <fieldfallow@gm ail.com> wrote:
[color=blue]
>Yes. Even after commenting out the first printf(), I still get the same
>wrong output. Would this indicate a fault in the particular toolchain,
>i.e. gcc (MinGW)?[/color]

Perhaps a bug in the handling of long double. Take out the long
double values and see if the pointer has the expected value. Or put
in some plain integers after the long doubles and see if they are
wrong. (As far as I know, %p is not required to print any particular
value for null pointers, but it will almost certainly be zero.)

-- Richard

**pete** · Feb 24 '06, 02:26 PM

Re: printf() error with long double and null pointer.

pete wrote:[color=blue]
>
> fieldfallow wrote:
>[color=green]
> > ld == -1.#QNAN0
> > or -1.#QNAN[/color]
>
> That result is also wrong.
>
> Are you using still using %lf to output doubles?
>
> %f is for both doubles and floats.
>
> C99 allows %lf for doubles, but most compilers are C90
> where using %lf to output a double is also undefined.[/color]

I'm suggesting that the error may have occured just
before the long double output.

--
pete

**fieldfallow** · Feb 24 '06, 02:26 PM

Re: printf() error with long double and null pointer.

pete wrote:[color=blue]
> fieldfallow wrote:
>[color=green]
>> ld == -1.#QNAN0
>> or -1.#QNAN[/color]
>
> That result is also wrong.
>
> Are you using still using %lf to output doubles?
>
> %f is for both doubles and floats.
>
> C99 allows %lf for doubles, but most compilers are C90
> where using %lf to output a double is also undefined.[/color]

Okay, I switched to %f for both floats and doubles and kept the first
printf() commented out, and still the same error.

Output:
Initialising them with their maximum allowed values...
Initialising v_p with NULL...
c ==
or 127
uc == ÿ
or 255
s == 32767
us == 65535
i == 2147483647
ui == 4294967295
l == 2147483647
ul == 4294967295
f == 340282346638528 860000000000000 000000000.00000 0
or 3.40282e+038
d == 179769313486231 570000000000000 000000000000000 000000
000000000000000 000000000000000 000000000000000 00000000000
000000000000000 000000000000000 000000000000000 00000000000
000000000000000 000000000000000 000000000000000 00000000000
000000000000000 000000000000000 000000000000000 00000000000
000000000000000 000000000000000 0000.000000
or 1.79769e+308
ld == -1.#QNAN0
or -1.#QNAN
v_p == FFFFFFFF

Code is:
#include <stdio.h>
#include <limits.h>
#include <float.h>

int main(void) {
char c;
unsigned char uc;
short s;
unsigned short us;
int i;
unsigned int ui;
long l;
unsigned long ul;
float f;
double d;
long double ld;
void *v_p;
/*
printf("c == %c\n\tor %d\nuc == %c\n\tor %u\ns == %hd\nus == %hu\n"
"i == %d\nui == %u\nl == %ld\nul == %lu\nf == %f\n\tor %g\n"
"d == %lf\n\tor %g\nld == %Lf\n\tor %Lg\nv_p == %p\n",
c,c,uc,uc,s,us, i,ui,l,ul,f,f,d ,d,ld,ld,v_p);
*/
puts("Initialis ing them with their maximum allowed values...");
c = CHAR_MAX;
uc = UCHAR_MAX;
s = SHRT_MAX;
us = USHRT_MAX;
i = INT_MAX;
ui = UINT_MAX;
l = LONG_MAX;
ul = ULONG_MAX;
f = FLT_MAX;
d = DBL_MAX;
ld = LDBL_MAX;
puts("Initialis ing v_p with NULL...");
v_p = NULL;

printf("c == %c\n\tor %d\nuc == %c\n\tor %u\ns == %hd\nus == %hu\n"
"i == %d\nui == %u\nl == %ld\nul == %lu\nf == %f\n\tor %g\n"
"d == %f\n\tor %g\nld == %Lf\n\tor %Lg\nv_p == %p\n",
c,c,uc,uc,s,us, i,ui,l,ul,f,f,d ,d,ld,ld,v_p);
return 0;
}

Thanks.

**fieldfallow** · Feb 24 '06, 02:36 PM

Re: printf() error with long double and null pointer.

Richard Tobin wrote:[color=blue]
> In article <43FF1048.10506 09@gmail.com>,
> fieldfallow <fieldfallow@gm ail.com> wrote:
>[color=green]
>> Yes. Even after commenting out the first printf(), I still get the same
>> wrong output. Would this indicate a fault in the particular toolchain,
>> i.e. gcc (MinGW)?[/color]
>
> Perhaps a bug in the handling of long double. Take out the long
> double values and see if the pointer has the expected value. Or put
> in some plain integers after the long doubles and see if they are
> wrong. (As far as I know, %p is not required to print any particular
> value for null pointers, but it will almost certainly be zero.)
>
> -- Richard[/color]

Your right Richard. After completely removing the long double from
printf(), the void pointer's value is printed out as 00000000.

Also, declaring and initialising the long double with LDBL_MAX, but not
attempting to pass it to printf() also gives the correct output for the
void pointer.

So I guess it appears to be a bug in the printf()'s handling of long
double arguments.

Maybe it's been corrected by any recent patches, though my MinGW
distribution was installed only two months before.

Thanks for your help.

**Richard Bos** · Feb 24 '06, 02:36 PM

Re: printf() error with long double and null pointer.

fieldfallow <fieldfallow@gm ail.com> wrote:
[color=blue]
> Richard Tobin wrote:[color=green]
> > In article <dtmvse$km3$1@e mma.aioe.org>,
> > fieldfallow <fieldfallow@gm ail.com> wrote:
> >[color=darkred]
> >>> Accessing uninitialised values like that,
> >>> causes undefined behavior in your program.
> >>> The undefined behavior can show up in any part
> >>> of the program's execution.[/color]
> >[color=darkred]
> >> Does even printing out the values cause undefined behaviour?[/color]
> >
> > Yes, but I would be surprised if it caused the behaviour shown.
> > Try removing the first printf and see if you still get the same
> > results for the second.[/color]
>
> Yes. Even after commenting out the first printf(), I still get the same
> wrong output.[/color]

It does that for me, too[0]. When I remove the references to long
doubles in the printf(), it prints the expected[1] value for v_p, too.
It appears at first glance that gcc and the MinGW library are not in
accord over the size of a long double. How pernicious.

Try this: print (using sizeof) the size of your double and long double.
Dev-C++ on this computer says 8 and 12, respectively. But the headers
claim that long double is the same as double, and all values in
<float.h> for LDBL_* are the same as for DBL_*. Presumably, *printf()
assume that sizeof (double) == sizeof (long double), too.
[color=blue]
> Would this indicate a fault in the particular toolchain, i.e. gcc (MinGW)?[/color]

Yes.

Anyone know the right channel to report this to MinGW?

Richard

[0] Even when inserting the l in the second %g...
[1] Although not required.

printf() error with long double and null pointer.

printf() error with long double and null pointer.

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment