[Bug c++/108285] New: [13 Regression] error: conversion from ‘long double’ to ‘double’ may change value [-Werror=float-conversion] since r13-3291-g16ec267063c8ce60

[Bug c++/108285] New: [13 Regression] error: conversion from ‘long double’ to ‘double’ may change value [-Werror=float-conversion] since r13-3291-g16ec267063c8ce60

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108285

            Bug ID: 108285
           Summary: [13 Regression] error: conversion from ‘long double’
                    to ‘double’ may change value
                    [-Werror=float-conversion] since
                    r13-3291-g16ec267063c8ce60
           Product: gcc
           Version: 13.0
            Status: UNCONFIRMED
          Keywords: rejects-valid
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: marxin at gcc dot gnu.org
                CC: jakub at gcc dot gnu.org
  Target Milestone: ---

I noticed libheif package fails due to:

$ cat heif.ii
void lround(double);

void foo(float coeffs) {
  lround(2 * coeffs);
}

$ g++ heif.ii -c -m32 -std=c++14 -Werror=float-conversion 
heif.ii: In function ‘void foo(float)’:
heif.ii:4:12: error: conversion from ‘long double’ to ‘double’ may change value
[-Werror=float-conversion]
    4 |   lround(2 * coeffs);
      |          ~~^~~~~~~~
cc1plus: some warnings being treated as errors

while -m64 is fine:

$ g++ heif.ii -c -std=c++14 -Werror=float-conversion

[Bug c++/108285] [13 Regression] error: conversion from ‘long double’ to ‘double’ may change value [-Werror=float-conversion] since r13-3291-g16ec267063c8ce60

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108285

Martin Liška <marxin at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2023-01-04
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
   Target Milestone|---                         |13.0

[Bug c++/108285] [13 Regression] error: conversion from ‘long double’ to ‘double’ may change value [-Werror=float-conversion] since r13-3291-g16ec267063c8ce60

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108285

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Target|                            |i?86

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Looks like only a C++ front-end issue with -Wfloat-conversion Which looks like
it ignores the conversion from excess precision .

-xc++ -m32 -fexcess-precision=standard -Werror=float-conversion 
is enough to reproduce the warning (converted to error via -Werror).

You might be able to reproduce the issue with 16bit float too on some targets
too ..
(32bit x86 is the only target where 32bit [and 64bit] floating point gets
converted to 80bit float and all).

[Bug c++/108285] [13 Regression] error: conversion from ‘long double’ to ‘double’ may change value [-Werror=float-conversion] since r13-3291-g16ec267063c8ce60

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108285

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P1

[Bug c++/108285] [13 Regression] error: conversion from ‘long double’ to ‘double’ may change value [-Werror=float-conversion] since r13-3291-g16ec267063c8ce60

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108285

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |jakub at gcc dot gnu.org
             Status|NEW                         |ASSIGNED

--- Comment #2 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Created attachment 54239
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=54239&action=edit
gcc13-pr108285.patch

Untested fix.

[Bug c++/108285] [13 Regression] error: conversion from ‘long double’ to ‘double’ may change value [-Werror=float-conversion] since r13-3291-g16ec267063c8ce60

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108285

--- Comment #3 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Jakub Jelinek <jakub@gcc.gnu.org>:

https://gcc.gnu.org/g:3456db4de8940235b303ca38a689353854c9239f

commit r13-5152-g3456db4de8940235b303ca38a689353854c9239f
Author: Jakub Jelinek <jakub@redhat.com>
Date:   Fri Jan 13 18:23:57 2023 +0100

    c++: Avoid some false positive -Wfloat-conversion warnings with extended
precision [PR108285]

    On the following testcase trunk emits a false positive warning on ia32.
    convert_like_internal is there called with type of double and
    expr EXCESS_PRECISION_EXPR with float type with long double operand
    2.L * (long double) x.
    Now, for the code generation we do the right thing, cp_convert
    to double from that 2.L * (long double) x, but we call even
    cp_convert_and_check with that and that emits the -Wfloat-conversion
    warning.  Looking at what the C FE does in this case, it calls
    convert_and_check with the EXCESS_PRECISION_EXPR expression rather
    than its operand, and essentially uses the operand for code generation
    and EXCESS_PRECISION_EXPR itself for warnings.

    The following patch does that too for the C++ FE.

    2023-01-13  Jakub Jelinek  <jakub@redhat.com>

            PR c++/108285
            * cvt.cc (cp_convert_and_check): For EXCESS_PRECISION_EXPR
            use its operand except that for warning purposes use the original
            EXCESS_PRECISION_EXPR.
            * call.cc (convert_like_internal): Only look through
            EXCESS_PRECISION_EXPR when calling cp_convert, not when calling
            cp_convert_and_check.

            * g++.dg/warn/pr108285.C: New test.

[Bug c++/108285] [13 Regression] error: conversion from ‘long double’ to ‘double’ may change value [-Werror=float-conversion] since r13-3291-g16ec267063c8ce60

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108285

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED

--- Comment #4 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
Fixed.