https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90036
Martin Sebor <msebor at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Known to work| |7.3.0
Keywords| |diagnostic
Last reconfirmed| |2019-04-10
CC| |msebor at gcc dot gnu.org
Blocks| |85741
Ever confirmed|0 |1
Summary|False positive: directive |[8/9 Regression] false
|argument is null |positive: directive
|[-Werror=format-overflow=] |argument is null
| |[-Werror=format-overflow=]
Known to fail| |8.3.0, 9.0
--- Comment #1 from Martin Sebor <msebor at gcc dot gnu.org> ---
(When reporting bugs we ask for a test case. Please see
https://www.gnu.org/software/gcc/bugs).
That said, I can reproduce the warning with the top of trunk and with GCC 8.
The reason why the warning is issued for sprintf and not for strlen is because
it is implemented differently between the two functions (it runs on different
IL).
The null in the IL is the result of the jump threading optimization. The
warning can be suppressed by adding 'assert (vstring)' just above the sprintf
call.
Here's the IL the warning code sees:
stab_start_class_type (void * p, const char * tag, unsigned int id, bfd_boolean
structp, unsigned int size, bfd_boolean vptr, bfd_boolean ownvptr)
{
...
<bb 8> [local count: 237404318]:
if (ownvptr_24(D) != 0)
goto <bb 9>; [100.00%]
else
goto <bb 15>; [0.00%]
...
<bb 15> [local count: 0]:
_51 = strlen (0B);
_59 = _51 + 3;
vtable_16 = xmalloc (_59);
sprintf (vtable_16, "~%%%s", 0B);
_18 = MEM[(struct stab_write_handle *)p_22(D)].type_stack;
_18->vtable = vtable_16;
goto <bb 14>; [100.00%]
}
Referenced Bugs:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85741
[Bug 85741] [meta-bug] bogus/missing -Wformat-overflow
