[Bug c/113414] New: Incorrent checking for printf format: does not distinguish whether the variable is signed or unsigned

[Bug c/113414] New: Incorrent checking for printf format: does not distinguish whether the variable is signed or unsigned

[tongxiangzheng.route at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113414

            Bug ID: 113414
           Summary: Incorrent checking for printf format: does not
                    distinguish whether the variable is signed or unsigned
           Product: gcc
           Version: 11.4.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: tongxiangzheng.route at gmail dot com
  Target Milestone: ---

When checking the printf() function, gcc does not distinguish whether the
variable is signed or unsigned.
for example:
$ gcc test.cpp

#include<stdio.h>
int main(){
        int n;
        printf("%u",n);
}

The code should be warned, but it did not.

example 2:
$ gcc test.cpp

#include<stdio.h>
int main(){
        unsigned long long int n;
        printf("%d",n);
}

The code be warned, but it's wrong:
test.cpp: In function ‘int main()’:
test.cpp:4:18: warning: format ‘%d’ expects argument of type ‘int’, but
argument 2 has type ‘long long unsigned int’ [-Wformat=]
    4 |         printf("%d",n);
      |                 ~^  ~
      |                  |  |
      |                  |  long long unsigned int
      |                  int
      |                 %lld
gcc suggests using %lld, but in reality, %llu should be used.

[Bug c/113414] Incorrent checking for printf format: does not distinguish whether the variable is signed or unsigned

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113414

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
>The code should be warned, but it did not.


No that is by design. You need to use -Wformat-signedness for that.

But for the second example, the fix it note is wrong only with
-Wformat-signedness .

[Bug c/113414] Incorrent checking for printf format: does not distinguish whether the variable is signed or unsigned

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113414

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |INVALID
           See Also|                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=90205
             Status|UNCONFIRMED                 |RESOLVED

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #1)
> >The code should be warned, but it did not.
> 
> 
> No that is by design. You need to use -Wformat-signedness for that.
> 
> But for the second example, the fix it note is wrong only with
> -Wformat-signedness .

But that is already PR 90205 .

[Bug c/113414] Incorrent checking for printf format: does not distinguish whether the variable is signed or unsigned

[xry111 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113414

Xi Ruoyao <xry111 at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |xry111 at gcc dot gnu.org

--- Comment #3 from Xi Ruoyao <xry111 at gcc dot gnu.org> ---
And IMO it's not really "wrong" to suggest %lld.  Using %lld may not be what
you really want, but it's at least allowed by the standard (while using %d here
is prohibited and leading to an undefined behavior).

[Bug c/113414] Incorrent checking for printf format: does not distinguish whether the variable is signed or unsigned

[egallager at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113414

Eric Gallager <egallager at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |egallager at gcc dot gnu.org
           Keywords|                            |diagnostic

--- Comment #4 from Eric Gallager <egallager at gcc dot gnu.org> ---
I forget which other bug I mentioned this in, but this reminds me that I still
think a -Wformat=3 that includes -Wformat-signedness would be nice, as
currently it only goes up to -Wformat=2, which doesn't include
-Wformat-signedness currently.