c - printf redifines UINT32_MAX - Stack Overflow

Collectives™ on Stack Overflow

Find centralized, trusted content and collaborate around the technologies you use most.

Learn more about Collectives

Teams

Q&A for work

Connect and share knowledge within a single location that is structured and easy to search.

Learn more about Teams int main(int argc, char *argv[]) { printf("%lu\n%lu\n%lu\n%lu\n%lu\n%lu\n%lu\n", UINT32_MAX, UINT32_MAX, UINT32_MAX, UINT32_MAX, UINT32_MAX, UINT32_MAX, UINT32_MAX); return 0;

output:

4294967295
4294967295
4294967295
4294967295
4294967295
18446744073709551615
18446744073709551615
bug or intentional? and why?
i think it speaks for itself but the interface needs me to add more lines before i can post it (to much code for to less text lol)
                Your printf formats don't match the values you're passing.  Your compiler should have warned about this.  Because you're passing unexpected argument types, they are being unpacked incorrectly by printf.
– Tom Karzes
                Feb 17, 2016 at 13:04
                If you're using GCC or Clang, it's a good practice to add these flags when compiling -Wall -Wextra. Make this an habit.
– jweyrich
                Feb 17, 2016 at 13:05
You are using %lu, this is incorrect.  The %lu specifier is for unsigned long and only unsigned long, not uint32_t.  This is why the output is incorrect.  Use PRIu32 instead.
Your compiler should have caught this error for you, if you were compiling with warnings enabled.
#include <stdio.h>
#include <stdint.h>
#include <inttypes.h> // defines PRIu32
printf(
    "%" PRIu32 "\n"
    "%" PRIu32 "\n"
    "%" PRIu32 "\n"
    "%" PRIu32 "\n"
    "%" PRIu32 "\n"
    "%" PRIu32 "\n"
    "%" PRIu32 "\n",
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX);
In practice, PRIu32 is defined to "u" on most systems.
                An alternative would be to continue using %lu, and to use (unsigned long)UINT32_MAX. This may be useful if you cannot use <inttypes.h> for some reason.
– Ian Abbott
                Feb 17, 2016 at 15:32
                @Ian Abbott Need to add: casting to (unsigned long) is a good alternative because unsigned long is at least the width of uint32_t.
– chux - Reinstate Monica
                Feb 17, 2016 at 16:22
This is undefined behavior: you are passing unsigned int, but your %lu format tells printf that you are passing long unsigned int. This leads to undefined behavior, because printf goes by what the format specifier says (in fact, it has no other way to find out, because type information is not passed to printf).
Once you fix your format issue, the problem goes away:
printf("%u\n%u\n%u\n%u\n%u\n%u\n%u\n",
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX,
    UINT32_MAX);
This prints
4294967295
4294967295
4294967295
4294967295
4294967295
4294967295
4294967295
Demo.
                This makes the reasonable assumption that on OP's system, uint32_t and unsigned are the same type.  This is not a portable solution in general though.
– chux - Reinstate Monica
                Feb 17, 2016 at 16:18
$ gcc -Wall -Wextra your_code.c
warning: format ‘%lu’ expects type ‘long unsigned int’, but argument 2 has type ‘unsigned int’
It's not a long, don't use the l specifier.
Your code has Undefined Behaviour, and in practice it essentially reads garabage.
                It's not an unsigned long, but it's not an unsigned int either, so merely omitting the l is incorrect.
– Ian Abbott
                Feb 18, 2016 at 11:22
                Yes, technically the minimum size for int is 2 bytes. Finding such a system can be tricky though ;)
– Karoly Horvath
                Feb 18, 2016 at 11:31
                @IanAbbott (replying to myself) - However, I think the type of the constant UINT32_MAX will always be one of unsigned int or unsigned long, rather than something inbetween, so one of %u or %lu would work, but the problem is knowing which one. Hence the need for the PRIu32 macro from <inttypes.h>.
– Ian Abbott
                Feb 18, 2016 at 11:34
                I think int is typically 2 bytes on C compilers for 16-bit systems, such as MS-DOS (without any 32-bit extensions).
– Ian Abbott
                Feb 18, 2016 at 11:47
                @IanAbbott: pretty much what I was saying... but who knows, perhaps MS-DOS is popular in the area you live in.
– Karoly Horvath
                Feb 18, 2016 at 11:53
        Thanks for contributing an answer to Stack Overflow!
Please be sure to answer the question. Provide details and share your research!
But avoid …
Asking for help, clarification, or responding to other answers.
Making statements based on opinion; back them up with references or personal experience.
To learn more, see our tips on writing great answers.