https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116581
Bug ID: 116581
Summary: incorrect -Wfloat-conversion warnings for int to
_Decimal64
Product: gcc
Version: 15.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: c
Assignee: unassigned at gcc dot gnu.org
Reporter: vincent-gcc at vinc17 dot net
Target Milestone: ---
GCC emits incorrect -Wfloat-conversion warnings for int to _Decimal64 (this
makes the configure script of GNU MPFR fail when using
-Werror=float-conversion):
qaa:~> cat tst.c
_Decimal64 f(void)
{
volatile _Decimal64 x = 1;
_Decimal64 y = 1;
return x + y;
}
qaa:~> gcc-snapshot -Wfloat-conversion -c tst.c
tst.c: In function 'f':
tst.c:3:27: warning: conversion from 'int' to '_Decimal64' changes value from
'1' to '1' [-Wfloat-conversion]
3 | volatile _Decimal64 x = 1;
| ^
tst.c:4:18: warning: conversion from 'int' to '_Decimal64' changes value from
'1' to '1dd' [-Wfloat-conversion]
4 | _Decimal64 y = 1;
| ^
Obviously, the value doesn't change. There is also an inconsistency in the
messages: why "to '1'" in the volatile case but "to '1dd'" in the other case?
Note: to get correct output with coloring, do not use a 80-column terminal due
to PR90411.
