[Bug math/3439] New: math.h, fenv.h macros do not conform to C99

[Bug math/3439] New: math.h, fenv.h macros do not conform to C99

[eggert at gnu dot org]

In
<https://www.opengroup.org/sophocles/show_mail.tpl?CALLER=index.tpl&source=L&listname=austin-group-l&id=9979>
Geoff Clare reported that the glibc implementations of <math.h> and
<fenv.h> do not conform to C99 or to POSIX, because they define some
macros using techniques that do not work inside #if, but C99 and POSIX
require that the macros work in #if.

I'll attach a patch against glibc head.

-- 
           Summary: math.h, fenv.h macros do not conform to C99
           Product: glibc
           Version: unspecified
            Status: NEW
          Severity: normal
          Priority: P2
         Component: math
        AssignedTo: aj at suse dot de
        ReportedBy: eggert at gnu dot org
                CC: glibc-bugs at sources dot redhat dot com


http://sourceware.org/bugzilla/show_bug.cgi?id=3439

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[eggert at gnu dot org]

------- Additional Comments From eggert at gnu dot org  2006-10-31 19:43 -------
Created an attachment (id=1395)
 --> (http://sourceware.org/bugzilla/attachment.cgi?id=1395&action=view)
fix math.h, fenv.h macros to be usable in #if


-- 


http://sourceware.org/bugzilla/show_bug.cgi?id=3439

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[drepper at redhat dot com]

------- Additional Comments From drepper at redhat dot com  2006-11-10 16:24 -------
I don't want to change this.  It's a pointless requirement which just makes it
harder to debug programs since the macros might not be available in the debug info.

-- 


http://sourceware.org/bugzilla/show_bug.cgi?id=3439

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[eggert at gnu dot org]

------- Additional Comments From eggert at gnu dot org  2006-11-11 01:01 -------
Created an attachment (id=1411)
 --> (http://sourceware.org/bugzilla/attachment.cgi?id=1411&action=view)
fix math.h, fenv.h macros to be usable in #if (without removing enums)

Here is a revised patch.  This keeps the enums, so ordinary gcc -g compiles
will have the names in the debug symbol table.	This should overcome the
objection to the original patch.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
Attachment #1395 is|0                           |1
           obsolete|                            |


http://sourceware.org/bugzilla/show_bug.cgi?id=3439

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[drepper at redhat dot com]

------- Additional Comments From drepper at redhat dot com  2007-02-17 08:59 -------
That is just calling for trouble.  The requirement is unnecessary and most
probably just an oversight.  It makes no sense to have these as macros.  The
best way is to change the requirement in the spec so that this desease isn't
spreading  further.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |WONTFIX


http://sourceware.org/bugzilla/show_bug.cgi?id=3439

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[jsm28 at gcc dot gnu.org]

http://sourceware.org/bugzilla/show_bug.cgi?id=3439

Joseph Myers <jsm28 at gcc dot gnu.org> changed:

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

--- Comment #11 from Joseph Myers <jsm28 at gcc dot gnu.org> 2012-11-03 17:09:45 UTC ---
fenv.h fixed for 2.17 by:

commit fbeafedeea37e0af1984a6511018d159f5ceed6a
Author: Joseph Myers <joseph@codesourcery.com>
Date:   Sat Nov 3 17:07:56 2012 +0000

    Make fenv.h FE_* macros usable in #if (bug 3439).

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[jsm28 at gcc dot gnu.org]

http://sourceware.org/bugzilla/show_bug.cgi?id=3439

--- Comment #10 from Joseph Myers <jsm28 at gcc dot gnu.org> 2012-11-02 22:15:11 UTC ---
math.h fixed, following the approach suggested in comment 8, by:

commit 0155d5b26e32b2012d9f1842c3b06324746d3087
Author: Joseph Myers <joseph@codesourcery.com>
Date:   Fri Nov 2 22:11:21 2012 +0000

    Make math.h FP_* macros usable in #if (bug 3439).

fenv.h still needs fixing.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[joseph at codesourcery dot com]

http://sourceware.org/bugzilla/show_bug.cgi?id=3439

--- Comment #9 from joseph at codesourcery dot com <joseph at codesourcery dot com> 2012-04-29 00:12:48 UTC ---
I think that would be fine as well.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[eggert at gnu dot org]

http://sourceware.org/bugzilla/show_bug.cgi?id=3439

--- Comment #8 from Paul Eggert <eggert at gnu dot org> 2012-04-28 23:15:42 UTC ---
How about this instead?

          FE_WHATEVER =
  #define FE_WHATEVER 0x100
          FE_WHATEVER,

This repeats the number just once and the identifier just thrice,
with no need for __FE_WHATEVER.  It's odd, but I've seen odder.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[aj at suse dot de]

http://sourceware.org/bugzilla/show_bug.cgi?id=3439

--- Comment #7 from Andreas Jaeger <aj at suse dot de> 2012-04-27 11:41:30 UTC ---
ignore my last comment, I see the difference...

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[aj at suse dot de]

http://sourceware.org/bugzilla/show_bug.cgi?id=3439

Andreas Jaeger <aj at suse dot de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |aj at suse dot de

--- Comment #6 from Andreas Jaeger <aj at suse dot de> 2012-04-27 11:23:09 UTC ---
Check <bits/confname.h> as well, it does the enum like:

    _PC_MAX_CANON,
#define _PC_MAX_CANON                   _PC_MAX_CANON

this should work similar for fenv.h

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.

[Bug math/3439] math.h, fenv.h macros do not conform to C99

[jsm28 at gcc dot gnu.org]

http://sourceware.org/bugzilla/show_bug.cgi?id=3439

Joseph Myers <jsm28 at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
         Resolution|WONTFIX                     |
         AssignedTo|aj at suse dot de           |unassigned at sourceware
                   |                            |dot org

--- Comment #5 from Joseph Myers <jsm28 at gcc dot gnu.org> 2012-04-27 10:47:50 UTC ---
Reopening.  We discussed this on #glibc last night.  We want a fix that avoids
duplicating the numeric value of each macro; that is, something along the lines
of (inside the enum):

#define __FE_WHATEVER 0x100
  FE_WHATEVER = __FE_WHATEVER,
#define FE_WHATEVER __FE_WHATEVER

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.