[Bug target/19333] New: Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[Bug target/19333] New: Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[drab at kepler dot fjfi dot cvut dot cz]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 7554 bytes --]

The following test code

-- test.c -------------------------------------
extern short int aa[][];
extern short int ab[][];
extern short int ac[][];

static void f(short int t[][4], int A, int B, int C, int D)
{
        int i;
        for(i=0; i<256; ++i)
        {
                t[i][0] = i*A;
                t[i][1] = i*B;
                t[i][2] = i*C;
                t[i][3] = i*D;
        }
}

void g()
{
        f(aa, -1, 0, 0, 0);
        f(ab, 3, -1, 0, 0);
        f(ac, 6, 3, -1, 0);
}
-----------------------------------------------

when compiled with

-----------------------------------------------
gcc -O1 -finline-functions -S test.c -o test.s
-----------------------------------------------

using the following GCC

-----------------------------------------------
$ gcc -v
Using built-in specs.
Configured with: ../../../gcc-CVS-20050107/gcc-CVS-20050107/configure
--host=x86_64-pc-linux-gnu --prefix=/usr/local/opt/gcc-4.0
--exec-prefix=/usr/local/opt/gcc-4.0 --sysconfdir=/etc
--libdir=/usr/local/opt/gcc-4.0/lib64
--libexecdir=/usr/local/opt/gcc-4.0/libexec64 --sharedstatedir=/var
--localstatedir=/var --program-suffix=-4.0 --with-x-includes=/usr/X11R6/include
--with-x-libraries=/usr/X11R6/lib64 --enable-shared --enable-static
--with-gnu-as --with-gnu-ld --with-stabs --enable-threads=posix
--enable-version-specific-runtime-libs --disable-coverage
--enable-gather-detailed-mem-stats --disable-libgcj --disable-checking
--enable-multilib --with-x --enable-cmath --enable-libstdcxx-debug
--enable-fast-character --enable-hash-synchronization --with-system-zlib
--with-libbanshee --with-demangler-in-ld --with-arch=athlon64 --disable-libada
--enable-languages=c,c++,f95,objc
Thread model: posix
gcc version 4.0.0 20050107 (experimental)
-----------------------------------------------

(i.e., native gcc for the x86_64 architecture) SEGFAULTs in a following way:

-----------------------------------------------
$ gcc -v -S -O1 -finline-functions test.c -o test.s
Using built-in specs.
Configured with: ../../../gcc-CVS-20050107/gcc-CVS-20050107/configure
--host=x86_64-pc-linux-gnu --prefix=/usr/local/opt/gcc-4.0
--exec-prefix=/usr/local/opt/gcc-4.0 --sysconfdir=/etc
--libdir=/usr/local/opt/gcc-4.0/lib64
--libexecdir=/usr/local/opt/gcc-4.0/libexec64 --sharedstatedir=/var
--localstatedir=/var --program-suffix=-4.0 --with-x-includes=/usr/X11R6/include
--with-x-libraries=/usr/X11R6/lib64 --enable-shared --enable-static
--with-gnu-as --with-gnu-ld --with-stabs --enable-threads=posix
--enable-version-specific-runtime-libs --disable-coverage
--enable-gather-detailed-mem-stats --disable-libgcj --disable-checking
--enable-multilib --with-x --enable-cmath --enable-libstdcxx-debug
--enable-fast-character --enable-hash-synchronization --with-system-zlib
--with-libbanshee --with-demangler-in-ld --with-arch=athlon64 --disable-libada
--enable-languages=c,c++,f95,objc
Thread model: posix
gcc version 4.0.0 20050107 (experimental)
 /usr/local/opt/gcc-4.0/libexec64/gcc/x86_64-pc-linux-gnu/4.0.0/cc1 -quiet -v
-iprefix
/usr/local/opt/gcc-4.0/lib64/gcc/x86_64-pc-linux-gnu/4.0.0/x86_64-pc-linux-gnu/4.0.0/
test.c -quiet -dumpbase test.c -march=athlon64 -auxbase-strip test.s -O1
-version -finline-functions -o test.s
ignoring nonexistent directory
"/usr/local/opt/gcc-4.0/lib64/gcc/x86_64-pc-linux-gnu/4.0.0/x86_64-pc-linux-gnu/4.0.0/include"
ignoring nonexistent directory
"/usr/local/opt/gcc-4.0/lib64/gcc/x86_64-pc-linux-gnu/4.0.0/x86_64-pc-linux-gnu/4.0.0/../../../../x86_64-pc-linux-gnu/include"
ignoring nonexistent directory
"/usr/local/opt/gcc-4.0/lib64/gcc/x86_64-pc-linux-gnu/4.0.0/../../../../x86_64-pc-linux-gnu/include"
ignoring duplicate directory "/usr/local/include"
  as it is a non-system directory that duplicates a system directory
ignoring duplicate directory
"/usr/local/opt/gcc-4.0/lib64/gcc/x86_64-pc-linux-gnu/4.0.0/include"
  as it is a non-system directory that duplicates a system directory
#include "..." search starts here:
#include <...> search starts here:
 /usr/local/include
 /usr/local/opt/gcc-4.0/include
 /usr/local/opt/gcc-4.0/lib64/gcc/x86_64-pc-linux-gnu/4.0.0/include
 /usr/include
End of search list.
GNU C version 4.0.0 20050107 (experimental) (x86_64-pc-linux-gnu)
        compiled by GNU C version 4.0.0 20050107 (experimental).
GGC heuristics: --param ggc-min-expand=64 --param ggc-min-heapsize=63859
test.c: In function âgâ:
test.c:18: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
$
-----------------------------------------------

It also segaults when the "-O1 -finline-functions" is changed for
"-O2 -finline-functions" or for "-O3".
I also tried to replace the -O1 with the following options "-fdefer-pop
-fmerge-constants -fthread-jumps -floop-optimize -fif-conversion
-fif-conversion2 -fdelayed-branch -fguess-branch-probability -fcprop-registers",
which is supposed to be equivalent to the -O1 according to the gcc.info file,
but surprisingly enough it passes without the SEGFAULT (so my guess is -O1 is
not equivalent to those options). It also passes, when either of following
conditions occurs:

1) Line 21 in the source is changed from "f(ac, 6, 3, -1, 0)" to "f(ac, 0, 3,
-1, 0)".
2) Line 3 and 21 in the source is commented out.
3) All "short int" types in the source are changed to "int".
4) You compile without the "-finline-functions".
5) You replace the "-O1" with "-O0".
6) You compile with the GCC compiled from the same sources as the x86_64 one
mentioned above, but when you compile it as a native compiler on the
i686-pc-linux-gnu architecture, i.e. the following (with exactly the same
compilation options, of course):

-----------------------------------------------
$ gcc -v
Using built-in specs.
Configured with: ../../../gcc-CVS-20050107/gcc-CVS-20050107/configure
--host=i686-pc-linux-gnu --prefix=/usr/local/opt/gcc-4.0
--exec-prefix=/usr/local/opt/gcc-4.0 --sysconfdir=/etc
--libdir=/usr/local/opt/gcc-4.0/lib --libexecdir=/usr/local/opt/gcc-4.0/libexec
--sharedstatedir=/var --localstatedir=/var --program-suffix=-4.0
--with-x-includes=/usr/X11R6/include --with-x-libraries=/usr/X11R6/lib
--enable-shared --enable-static --with-gnu-as --with-gnu-ld --with-stabs
--enable-threads=posix --enable-version-specific-runtime-libs --disable-coverage
--enable-gather-detailed-mem-stats --disable-libgcj --disable-checking
--enable-multilib --with-x --enable-cmath --enable-libstdcxx-debug
--enable-fast-character --enable-hash-synchronization --with-system-zlib
--with-libbanshee --with-demangler-in-ld --with-arch=athlon-xp --enable-libada
--enable-languages=c,c++,f95,objc,ada
Thread model: posix
gcc version 4.0.0 20050107 (experimental)
$
-----------------------------------------------

BTW. I also tried to compile with the same native x86_64 for different targets,
but the SEGFAULT is still there.

-- 
           Summary: Compilation SEGFAULTs with -O1 -finline-functions on the
                    x86_64 architecture.
           Product: gcc
           Version: 4.0.0
            Status: UNCONFIRMED
          Severity: critical
          Priority: P1
         Component: target
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: drab at kepler dot fjfi dot cvut dot cz
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: x86_64-pc-linux-gnu
  GCC host triplet: x86_64-pc-linux-gnu
GCC target triplet: x86_64-pc-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-01-08 19:42 -------
Confirmed, here is the backtrace:
#0  0x0014c5f4 in maybe_fold_offset_to_array_ref (base=0x416dd15c, offset=0x4164caa0, 
orig_type=0x416a1570) at ../../gcc/tree-ssa-ccp.c:1351
#1  0x0014dcfc in maybe_fold_stmt_indirect (expr=0x416ffc60, base=0x416dd15c, 
offset=0x4164caa0) at ../../gcc/tree-ssa-ccp.c:1559
#2  0x0014f0dc in fold_stmt_r (expr_p=0x41702188, walk_subtrees=0xbfffe744, data=0xbfffe97c) at 
../../gcc/tree-ssa-ccp.c:1753
#3  0x0074389c in walk_tree (tp=0x41702188, func=0x14ee20 <fold_stmt_r>, data=0xbfffe97c, 
pset=0x0) at ../../gcc/tree-inline.c:2017
#4  0x00744574 in walk_tree (tp=0x416ffc5c, func=0x14ee20 <fold_stmt_r>, data=0xbfffe97c, 
pset=0x0) at ../../gcc/tree-inline.c:2114
#5  0x0014f1e8 in fold_stmt_r (expr_p=0x41703140, walk_subtrees=0xbfffe8e4, data=0xbfffe97c) at 
../../gcc/tree-ssa-ccp.c:1763
#6  0x0074389c in walk_tree (tp=0x41703140, func=0x14ee20 <fold_stmt_r>, data=0xbfffe97c, 
pset=0x0) at ../../gcc/tree-inline.c:2017
#7  0x00150dfc in fold_stmt (stmt_p=0x416fc6b0) at ../../gcc/tree-ssa-ccp.c:2046
#8  0x0019ee98 in optimize_stmt (walk_data=0xbfffee00, bb=0x416e2ef4, si={tsi = {ptr = 
0x416fc6a8, container = 0x416dbd68}, bb = 0x416e2ef4}) at ../../gcc/tree-ssa-dom.c:2922


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pinskia at gcc dot gnu dot
                   |                            |org
             Status|UNCONFIRMED                 |NEW
          Component|target                      |tree-optimization
     Ever Confirmed|                            |1
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|0000-00-00 00:00:00         |2005-01-08 19:42:52
               date|                            |
            Summary|Compilation SEGFAULTs with -|[4.0 Regression] Compilation
                   |O1 -finline-functions on the|SEGFAULTs with -O1 -finline-
                   |x86_64 architecture.        |functions on the x86_64
                   |                            |architecture.
   Target Milestone|---                         |4.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-01-08 19:43 -------
Also happens on ppc64.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
  GCC build triplet|x86_64-pc-linux-gnu         |
   GCC host triplet|x86_64-pc-linux-gnu         |
 GCC target triplet|x86_64-pc-linux-gnu         |any LP64 target


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-01-08 19:46 -------
Hmm, IV-OPTS looks like it is creating invalid gimple:
&(*(short int[4] *) &aa)[0](0)

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |rakdver at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-01-08 19:50 -------
But that is the difference which is causing it.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[rakdver at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |rakdver at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[rakdver at gcc dot gnu dot org]

------- Additional Comments From rakdver at gcc dot gnu dot org  2005-01-10 08:51 -------
http://gcc.gnu.org/ml/gcc-patches/2005-01/msg00512.html

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |patch


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[drab at kepler dot fjfi dot cvut dot cz]

------- Additional Comments From drab at kepler dot fjfi dot cvut dot cz  2005-01-11 14:15 -------
(In reply to comment #5)
> http://gcc.gnu.org/ml/gcc-patches/2005-01/msg00512.html

Quite thouroughly tested, the patch works fine. Thanks.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-01-12 05:12 -------
Hmm, I think this is a bad gcc extension:
extern short int aa[][];
Note with -pedantic we warning about it.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[drab at kepler dot fjfi dot cvut dot cz]

------- Additional Comments From drab at kepler dot fjfi dot cvut dot cz  2005-01-12 11:25 -------
(In reply to comment #7)
> Hmm, I think this is a bad gcc extension:
> extern short int aa[][];
> Note with -pedantic we warning about it.

I agree, I would personally never use it. However, this came out during
compilation of XviD (if I remember correctly).

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[rakdver at gcc dot gnu dot org]

------- Additional Comments From rakdver at gcc dot gnu dot org  2005-01-14 23:47 -------
The conclusion of the discussion on the patches list seems to be that

extern short int aa[][];

is invalid and should be caught by frontend in the first place.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|ice-on-valid-code, patch    |ice-on-invalid-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2005-01-27 16:05 -------
In that case, isn't the following patch enough (modulo testing, doc 
fixes, and test cases)? 
 
 
 
Index: c-decl.c 
=================================================================== 
RCS file: /cvs/gcc/gcc/gcc/c-decl.c,v 
retrieving revision 1.627 
diff -u -3 -p -r1.627 c-decl.c 
--- c-decl.c    27 Jan 2005 12:38:25 -0000      1.627 
+++ c-decl.c    27 Jan 2005 16:04:20 -0000 
@@ -4142,9 +4142,9 @@ grokdeclarator (const struct c_declarato 
                itype = build_range_type (sizetype, size_zero_node, 
NULL_TREE); 
              } 
 
-           /* If pedantic, complain about arrays of incomplete types.  */ 
-           if (pedantic && !COMPLETE_TYPE_P (type)) 
-             pedwarn ("array type has incomplete element type"); 
+           /* Complain about arrays of incomplete types.  */ 
+           if (!COMPLETE_TYPE_P (type)) 
+             error ("array type has incomplete element type"); 
 
            /* Build the array type itself, then merge any constancy 
               or volatility into the target type.  We must do it in 
 

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jsm28 at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2005-01-27 16:32 -------
Given http://gcc.gnu.org/ml/gcc-patches/2005-01/msg00948.html the 
patch looks like a good start. 

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2005-01-28 09:25 -------
With the patch from comment #10 I get the following extra failures: 
 
gcc.c-torture/compile/20011130-1.c  -O0  (test for excess errors) 
gcc.c-torture/compile/20011130-1.c  -O1  (test for excess errors) 
gcc.c-torture/compile/20011130-1.c  -O2  (test for excess errors) 
gcc.c-torture/compile/20011130-1.c  -O3 -fomit-frame-pointer  (test for excess 
errors) 
gcc.c-torture/compile/20011130-1.c  -O3 -g  (test for excess errors) 
gcc.c-torture/compile/20011130-1.c  -Os  (test for excess errors) 
gcc.dg/20030815-1.c (test for excess errors) 
gcc.dg/array-7.c  (test for errors, line 14) 
gcc.dg/array-7.c (test for excess errors) 
gcc.dg/noncompile/20000901-1.c  -O0  (test for excess errors) 
gcc.dg/noncompile/20000901-1.c  -O1  (test for excess errors) 
gcc.dg/noncompile/20000901-1.c  -O2  (test for excess errors) 
gcc.dg/noncompile/20000901-1.c  -O3 -fomit-frame-pointer  (test for excess 
errors) 
gcc.dg/noncompile/20000901-1.c  -O3 -g  (test for excess errors) 
gcc.dg/noncompile/20000901-1.c  -Os  (test for excess errors) 
gcc.dg/noncompile/init-2.c  -O0  (test for excess errors) 
gcc.dg/noncompile/init-2.c  -O1  (test for excess errors) 
gcc.dg/noncompile/init-2.c  -O2  (test for excess errors) 
gcc.dg/noncompile/init-2.c  -O3 -fomit-frame-pointer  (test for excess errors) 
gcc.dg/noncompile/init-2.c  -O3 -g  (test for excess errors) 
gcc.dg/noncompile/init-2.c  -Os  (test for excess errors) 
 
Much appreciated if a C language guru could see if these are not false 
positives. 
 

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] Compilation SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

[joseph at codesourcery dot com]

------- Additional Comments From joseph at codesourcery dot com  2005-01-28 11:09 -------
Subject: Re:  [4.0 Regression] Compilation
 SEGFAULTs with -O1 -finline-functions on the x86_64 architecture.

On Fri, 28 Jan 2005, steven at gcc dot gnu dot org wrote:

> gcc.c-torture/compile/20011130-1.c  -O0  (test for excess errors) 
> gcc.c-torture/compile/20011130-1.c  -O1  (test for excess errors) 
> gcc.c-torture/compile/20011130-1.c  -O2  (test for excess errors) 
> gcc.c-torture/compile/20011130-1.c  -O3 -fomit-frame-pointer  (test for excess 
> errors) 
> gcc.c-torture/compile/20011130-1.c  -O3 -g  (test for excess errors) 
> gcc.c-torture/compile/20011130-1.c  -Os  (test for excess errors) 
> gcc.dg/20030815-1.c (test for excess errors) 

Correct, I don't know whether it's useful to reorder the testcases 
minimally to avoid arrays of incomplete types.

> gcc.dg/array-7.c  (test for errors, line 14) 
> gcc.dg/array-7.c (test for excess errors) 
> gcc.dg/noncompile/20000901-1.c  -O0  (test for excess errors) 
> gcc.dg/noncompile/20000901-1.c  -O1  (test for excess errors) 
> gcc.dg/noncompile/20000901-1.c  -O2  (test for excess errors) 
> gcc.dg/noncompile/20000901-1.c  -O3 -fomit-frame-pointer  (test for excess 
> errors) 
> gcc.dg/noncompile/20000901-1.c  -O3 -g  (test for excess errors) 
> gcc.dg/noncompile/20000901-1.c  -Os  (test for excess errors) 
> gcc.dg/noncompile/init-2.c  -O0  (test for excess errors) 
> gcc.dg/noncompile/init-2.c  -O1  (test for excess errors) 
> gcc.dg/noncompile/init-2.c  -O2  (test for excess errors) 
> gcc.dg/noncompile/init-2.c  -O3 -fomit-frame-pointer  (test for excess errors) 
> gcc.dg/noncompile/init-2.c  -O3 -g  (test for excess errors) 
> gcc.dg/noncompile/init-2.c  -Os  (test for excess errors) 

For these I think the testcases should be updated to expect the additional 
or changed diagnostics.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] C front end accepts arrays of incomplete types

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2005-01-29 13:20 -------
Lemme take this one... 

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|rakdver at gcc dot gnu dot  |steven at gcc dot gnu dot
                   |org                         |org
           Severity|critical                    |normal
            Summary|[4.0 Regression] Compilation|[4.0 Regression] C front end
                   |SEGFAULTs with -O1 -finline-|accepts arrays of incomplete
                   |functions on the x86_64     |types
                   |architecture.               |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug tree-optimization/19333] [4.0 Regression] C front end accepts arrays of incomplete types

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-01-29 13:23 -------
*** Bug 19390 has been marked as a duplicate of this bug. ***

-- 
Bug 19333 depends on bug 19390, which changed state.

Bug 19390 Summary: Accepts invalid code (well I don't know if it is invalid or not but I think it should be)
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19390

           What    |Old Value                   |New Value
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |DUPLICATE

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug c/19333] [4.0 Regression] C front end accepts arrays of incomplete types

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-01-29 13:25 -------
Simple testcase now:
extern short int aa[][];

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|tree-optimization           |c
           Keywords|                            |accepts-invalid


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug c/19333] [4.0 Regression] C front end accepts arrays of incomplete types

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2005-01-31 09:54 -------
Proposed patch here: http://gcc.gnu.org/ml/gcc-patches/2005-01/ 


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |patch


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug c/19333] [4.0 Regression] C front end accepts arrays of incomplete types

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-02-01 00:09 -------
Subject: Bug 19333

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	steven@gcc.gnu.org	2005-02-01 00:09:40

Modified files:
	gcc            : ChangeLog c-decl.c c-typeck.c 
	gcc/testsuite  : ChangeLog 
	gcc/testsuite/gcc.c-torture/compile: 20011130-1.c 
	gcc/testsuite/gcc.dg: 20030815-1.c array-7.c pr18596-3.c 
	gcc/testsuite/gcc.dg/noncompile: 20000901-1.c init-2.c init-4.c 

Log message:
	gcc/
	PR c/19333
	* c-decl.c (start_decl): Do not warn about arrays of elements with
	an incomplete type here.
	(grokdeclarator): Do it here by making a pedwarn an error.
	* c-typeck.c (push_init_level): If there were previous errors with
	the constructor type, do not warn about braces for initializers.
	(process_init_element): Likewise for excess initializer elements.
	
	testsuite/
	PR c/19333
	* testsuite/gcc.c-torture/compile/20011130-1.c: Reorder to make
	the test case valid.
	* testsuite/gcc.dg/20030815-1.c: Remove invalid tests.
	* testsuite/gcc.dg/array-7.c: Adjust expected result.
	* testsuite/gcc.dg/pr18596-3.c: Likewise.
	* testsuite/gcc.dg/noncompile/20000901-1.c: Likewise.
	* testsuite/gcc.dg/noncompile/init-2.c: Likewise.
	* testsuite/gcc.dg/noncompile/init-4.c: Likewise.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.7349&r2=2.7350
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/c-decl.c.diff?cvsroot=gcc&r1=1.628&r2=1.629
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/c-typeck.c.diff?cvsroot=gcc&r1=1.415&r2=1.416
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.4964&r2=1.4965
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.c-torture/compile/20011130-1.c.diff?cvsroot=gcc&r1=1.1&r2=1.2
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/20030815-1.c.diff?cvsroot=gcc&r1=1.1&r2=1.2
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/array-7.c.diff?cvsroot=gcc&r1=1.1&r2=1.2
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/pr18596-3.c.diff?cvsroot=gcc&r1=1.1&r2=1.2
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/noncompile/20000901-1.c.diff?cvsroot=gcc&r1=1.1&r2=1.2
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/noncompile/init-2.c.diff?cvsroot=gcc&r1=1.2&r2=1.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/noncompile/init-4.c.diff?cvsroot=gcc&r1=1.2&r2=1.3



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug c/19333] [4.0 Regression] C front end accepts arrays of incomplete types

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2005-02-01 00:15 -------
http://gcc.gnu.org/ml/gcc-patches/2005-01/msg02245.html 

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug c/19333] [4.0 Regression] C front end accepts arrays of incomplete types

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2005-02-01 00:19 -------
I'm on there twice :) 

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333

[Bug c/19333] [4.0 Regression] C front end accepts arrays of incomplete types

[debian-gcc at lists dot debian dot org]

------- Additional Comments From debian-gcc at lists dot debian dot org  2005-02-01 11:55 -------
at least that breaks the bootstrap on a Debian x86_64 system (glibc-2.3.2 based):

  Matthias

./xgcc -B./ -B/usr/lib/gcc-snapshot/x86_64-linux-gnu/bin/ -isystem
/usr/lib/gcc-snapshot/x86_64-linux-gnu/include -isystem
/usr/lib/gcc-snapshot/x86_64-linux-gnu/sys-include
-L/home/packages/gcc/snap/gcc-snapshot-20050201/build/gcc/../ld -O2  -DIN_GCC  
 -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes
-Wold-style-definition  -isystem ./include  -fPIC -g -DHAVE_GTHR_DEFAULT
-DIN_LIBGCC2 -D__GCC_FLOAT_NOT_NEEDED  -I. -I. -I../../src/gcc -I../../src/gcc/.
-I../../src/gcc/../include -I../../src/gcc/../libcpp/include 
-fvisibility=hidden -DHIDE_EXPORTS -fexceptions -c ../../src/gcc/unwind-dw2.c -o
libgcc/./unwind-dw2.o
In file included from ../../src/gcc/gthr-posix.h:43,
                 from ./gthr-default.h:1,
                 from ../../src/gcc/gthr.h:114,
                 from ../../src/gcc/unwind-dw2.c:42:
/usr/include/pthread.h:655: error: array type has incomplete element type
make[5]: *** [libgcc/./unwind-dw2.o] Error 1
make[5]: Leaving directory `/home/packages/gcc/snap/gcc-snapshot-20050201/build/gcc'
make[4]: *** [stmp-multilib] Error 2
make[4]: Leaving directory `/home/packages/gcc/snap/gcc-snapshot-20050201/build/gcc'
make[3]: *** [stage1_build] Error 2


pthread.h:654: is
struct __jmp_buf_tag;
extern int __sigsetjmp (struct __jmp_buf_tag __env[1], int __savemask) __THROW;

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |doko at debian dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19333