[Bug c/28437] New: multiple fno-builtin-* flags broken

[Bug c/28437] New: multiple fno-builtin-* flags broken

[mike at codeweavers dot com]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1601 bytes --]

If more than one -fno-builtin-* flag is specified, the first one is ignored.

I've seen other reports of this on Wine the mailing lists.  We use -fno-builtin
to supress the gcc builtins that are incompatible with msvcrt's string
functions.

mike@black:~/wine$ cat b.c
int iswalpha(unsigned short);
mike@black:~/wine$ gcc -fno-builtin-iswalpha -c b.c
mike@black:~/wine$ gcc -fno-builtin-iswalpha -fno-builtin-iswalnum -c b.c
b.c:1: warning: conflicting types for built-in function ‘iswalpha’

mike@black:~/wine$ gcc -v
Using built-in specs.
Target: x86_64-linux-gnu
Configured with: ../src/configure -v
--enable-languages=c,c++,fortran,objc,obj-c++,ada,treelang --prefix=/usr
--enable-shared --with-system-zlib --libexecdir=/usr/lib
--without-included-gettext --enable-threads=posix --enable-nls
--program-suffix=-4.1 --enable-__cxa_atexit --enable-clocale=gnu
--enable-libstdcxx-debug --enable-mpfr --enable-checking=release
x86_64-linux-gnu
Thread model: posix
gcc version 4.1.2 20060715 (prerelease) (Debian 4.1.1-9)
mike@black:~/wine$ uname -a
Linux black 2.6.16.11 #1 Mon May 1 16:44:30 KST 2006 x86_64 GNU/Linux


-- 
           Summary: multiple fno-builtin-* flags broken
           Product: gcc
           Version: 4.1.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: mike at codeweavers dot com
 GCC build triplet: x86_64-linux-gnu
  GCC host triplet: x86_64-linux-gnu
GCC target triplet: x86_64-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug c/28437] [4.1/4.2 Regression] multiple fno-builtin-* flags broken

[pinskia at gcc dot gnu dot org]

------- Comment #1 from pinskia at gcc dot gnu dot org  2006-07-19 15:31 -------
This is a regression, 4.0.0 worked so did some prerelease versions of 4.1.0.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pinskia at gcc dot gnu dot
                   |                            |org
           Keywords|                            |diagnostic
      Known to work|                            |4.0.0
            Summary|multiple fno-builtin-* flags|[4.1/4.2 Regression]
                   |broken                      |multiple fno-builtin-* flags
                   |                            |broken
   Target Milestone|---                         |4.1.3


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug c/28437] [4.1/4.2 Regression] multiple fno-builtin-* flags broken

[pinskia at gcc dot gnu dot org]

-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.1.3                       |4.1.2


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug c/28437] [4.1/4.2 Regression] multiple fno-builtin-* flags broken

[steven at gcc dot gnu dot org]

------- Comment #2 from steven at gcc dot gnu dot org  2006-07-24 05:46 -------
gcc -v shows that only the last -fno-builtin-* is passed to cc1:
$ /xgcc -B. -fno-builtin-iswalpha -fno-builtin-iswalnum -c b.c -v
Reading specs from ./specs
Target: x86_64-unknown-linux-gnu
Configured with: ../trunk/configure --enable-languages=c,c++,objc,fortran
Thread model: posix
gcc version 4.2.0 20060723 (experimental)
 ./cc1 -quiet -v -iprefix
/home/steven/devel/build-trunk-test/gcc/../lib/gcc/x86_64-unknown-linux-gnu/4.2.0/
-isystem ./include t.c -quiet -dumpbase t.c -mtune=generic -auxbase t -version
-fno-builtin-iswalnum -o /tmp/cc6JyivL.s
(etc.)

Likewise for cc1plus.


-- 

steven at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2006-07-24 05:46:26
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.1/4.2 Regression] multiple fno-builtin-* flags broken

[pinskia at gcc dot gnu dot org]

------- Comment #3 from pinskia at gcc dot gnu dot org  2006-07-24 05:55 -------
I bet a beer or two it was caused by:
2006-05-16  H.J. Lu  <hongjiu.lu@intel.com>

        PR driver/26885


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |hjl at gcc dot gnu dot org
          Component|c                           |driver


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.1/4.2 Regression] multiple fno-builtin-* flags broken

[pinskia at gcc dot gnu dot org]

-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |blocker


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.1/4.2 Regression] multiple fno-builtin-* flags broken

[hjl at lucon dot org]

------- Comment #4 from hjl at lucon dot org  2006-07-24 15:42 -------
Is my patch in 4.1?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.1/4.2 Regression] multiple fno-builtin-* flags broken

[steven at gcc dot gnu dot org]

------- Comment #5 from steven at gcc dot gnu dot org  2006-07-24 20:52 -------
Works with "GCC: (GNU) 4.1.1 20060518 (prerelease)".


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.1/4.2 Regression] multiple fno-builtin-* flags broken

[steven at gcc dot gnu dot org]

------- Comment #6 from steven at gcc dot gnu dot org  2006-07-24 21:46 -------
Also works with "GCC: (GNU) 4.1.2 20060724 (prerelease)", but fails with "GCC:
(GNU) 4.2.0 20060723 (experimental)".

So my guess is: HJ, no your patch is not in the FSF GCC 4.1 but it is in the
Debian GCC 4.1.  And I agree with Andrew that your patch is the first suspect
for causing the bug.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.2 Regression] multiple fno-builtin-* flags broken

[steven at gcc dot gnu dot org]

------- Comment #7 from steven at gcc dot gnu dot org  2006-07-24 21:47 -------
Not an FSF GCC 4.1 bug.


-- 

steven at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
      Known to work|4.0.0                       |4.0.0 4.1.1
            Summary|[4.1/4.2 Regression]        |[4.2 Regression] multiple
                   |multiple fno-builtin-* flags|fno-builtin-* flags broken
                   |broken                      |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.2 Regression] multiple fno-builtin-* flags broken

[steven at gcc dot gnu dot org]

------- Comment #8 from steven at gcc dot gnu dot org  2006-07-24 21:51 -------
HJ definitely caused this bug:

Starting program: /home/steven/devel/build-trunk-test/gcc/xgcc -B.
-fno-builtin-iswalpha -fno-builtin-iswalnum -S b.c

Breakpoint 2, main (argc=6, argv=0x7fffffc7bbb8) at gcc.c:6096
6096      prune_options (&argc, &argv);
(gdb) p argc
$10 = 6
(gdb) p argv[0]
$11 = 0x7fffffc7def0 "/home/steven/devel/build-trunk-test/gcc/xgcc"
(gdb) p argv[1]
$12 = 0x7fffffc7df1d "-B."
(gdb) p argv[2]
$13 = 0x7fffffc7df21 "-fno-builtin-iswalpha"
(gdb) p argv[3]
$14 = 0x7fffffc7df37 "-fno-builtin-iswalnum"
(gdb) p argv[4]
$15 = 0x7fffffc7df4d "-S"
(gdb) p argv[5]
$16 = 0x7fffffc7df50 "b.c"
(gdb) next
6104      unlock_std_streams ();
(gdb) p argc
$17 = 5
(gdb) p argv[0]
$18 = 0x7fffffc7def0 "/home/steven/devel/build-trunk-test/gcc/xgcc"
(gdb) p argv[1]
$19 = 0x7fffffc7df1d "-B."
(gdb) p argv[2]
$20 = 0x7fffffc7df37 "-fno-builtin-iswalnum"
(gdb) p argv[3]
$21 = 0x7fffffc7df4d "-S"
(gdb) p argv[4]
$22 = 0x7fffffc7df50 "b.c"
(gdb) p argv[5]
$23 = 0x0
(gdb)  

So prune_options prunes too much.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.2 Regression] multiple fno-builtin-* flags broken

[steven at gcc dot gnu dot org]

------- Comment #9 from steven at gcc dot gnu dot org  2006-07-24 22:13 -------
Created an attachment (id=11932)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11932&action=view)
Don't prune options that can be "Joined"

Seems to me like pruning Joined options is always wrong.

But, HJ, since you're more into this option stuff and you caused the bug, I
suppose you can tell whether my patch makes sense...


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.2 Regression] multiple fno-builtin-* flags broken

[pinskia at gcc dot gnu dot org]

-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.1.2                       |4.2.0
            Version|4.1.2                       |4.2.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.2 Regression] multiple fno-builtin-* flags broken

[hjl at lucon dot org]

------- Comment #10 from hjl at lucon dot org  2006-07-25 00:36 -------
A patch is posted at

http://gcc.gnu.org/ml/gcc-patches/2006-07/msg01048.html


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.2 Regression] multiple fno-builtin-* flags broken

[hjl at gcc dot gnu dot org]

------- Comment #11 from hjl at gcc dot gnu dot org  2006-07-27 21:27 -------
Subject: Bug 28437

Author: hjl
Date: Thu Jul 27 21:26:55 2006
New Revision: 115780

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=115780
Log:
2006-07-27  H.J. Lu  <hongjiu.lu@intel.com>

        PR driver/28437
        * opts-common.c (prune_options): Skip joined switches.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/opts-common.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.2 Regression] multiple fno-builtin-* flags broken

[mmitchel at gcc dot gnu dot org]

------- Comment #12 from mmitchel at gcc dot gnu dot org  2006-07-31 23:32 -------
Shouldn't this issue be closed now?


-- 

mmitchel at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437

[Bug driver/28437] [4.2 Regression] multiple fno-builtin-* flags broken

[pinskia at gcc dot gnu dot org]

------- Comment #13 from pinskia at gcc dot gnu dot org  2006-07-31 23:33 -------
Fixed, though really this is just a workaround as mentioned on the mailing
list.


-- 

pinskia at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28437