[Bug middle-end/44984] New: gcc passes unsigned instead of int for printf width/precision (warnings generated)

[Bug middle-end/44984] New: gcc passes unsigned instead of int for printf width/precision (warnings generated)

[jay dot krell at cornell dot edu]

4.3.5 but 4.5.0 looks about the same, but it has an additional instance:


mber/dpd -I../libdecnumber -I/usr/local/include   ../../gcc/gcc/opts.c -o
opts.o
../../gcc/gcc/opts.c: In function `wrap_help':
../../gcc/gcc/opts.c:1037: warning: field width is not type int (arg 3)
../../gcc/gcc/opts.c:1037: warning: field width is not type int (arg 5)
../../gcc/gcc/opts.c: In function `common_handle_option':
../../gcc/gcc/opts.c:1409: warning: field width is not type int (arg 3)
gcc -c   -g -O2 -DIN_GCC   -W -Wall -Wwrite-strings -Wstrict-prototypes
-Wmissin


jbook2:gcc jay$ diff -uw opts.c.orig opts.c
--- opts.c.orig    2010-07-19 00:07:18.000000000 -0700
+++ opts.c    2010-07-19 00:16:27.000000000 -0700
@@ -1034,7 +1034,10 @@
         }
     }

-      printf( "  %-*.*s %.*s\n", col_width, item_width, item, len, help);
+      gcc_assert(col_width <= INT_MAX);
+      gcc_assert(item_width <= INT_MAX);
+      gcc_assert(len <= INT_MAX);
+      printf ("  %-*.*s %.*s\n", (int)col_width, (int)item_width, item,
(int)len, help);
       item_width = 0;
       while (help[len] == ' ')
     len++;
@@ -1404,9 +1407,12 @@
             }

         if (i == cl_lang_count)
+          {
+            gcc_assert (len <= INT_MAX);
           fnotice (stderr,
                "warning: unrecognized argument to --help= switch: %.*s\n",
-               len, a);
+                 (int)len, a);
+          }
           }

         if (comma == NULL)


This makes a line go a little over 80 columns, but one of the "nearby" lines
does already.
 (nearby as presented here, not really in the code)

This probably isn't the best fix.
  if the output is "not important", and someone manages to create huge data,
should still work.
  Yes, I'm aware of that "huge" is 2GB.

Maybe:

int to_reasonable_diagnostic_length(unsigned i)
{
  return (i <= INT_MAX) ? i : 255;
}

or

int unint_to_int_pin(unsigned i)
{
  return (i <= INT_MAX) ? i : INT_MAX;
}

and then apply those.

 - Jay


-- 
           Summary: gcc passes unsigned instead of int for printf
                    width/precision (warnings generated)
           Product: gcc
           Version: 4.5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: jay dot krell at cornell dot edu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44984

[Bug middle-end/44984] gcc passes unsigned instead of int for printf width/precision (warnings generated)

[pinskia at gcc dot gnu dot org]

------- Comment #1 from pinskia at gcc dot gnu dot org  2010-07-20 22:50 -------
unsigned int and int are passed exactly the same for varargs and in fact the C
standard says getting an unsigned version of the signed type for varargs is
valid.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44984

[Bug middle-end/44984] gcc passes unsigned instead of int for printf width/precision (warnings generated)

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2010-07-20 22:52 -------
What compiler version is giving a warning for this case?  That is can you
provide the output of "gcc -v" ?


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |WAITING


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44984

[Bug middle-end/44984] gcc passes unsigned instead of int for printf width/precision (warnings generated)

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44984

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |RESOLVED
         Resolution|---                         |INVALID

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
No feedback in 3 years so closing.

[Bug middle-end/44984] gcc passes unsigned instead of int for printf width/precision (warnings generated)

[jay.krell at cornell dot edu]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44984

--- Comment #3 from Jay <jay.krell at cornell dot edu> 2010-10-02 10:27:53 UTC ---
> which compiler produces this


I'm afraid I'm not sure and can't quickly/easily make it happen again. Sorry.