[Bug ada/43769] New: Ada Bug Box when -O1 or higher

[Bug ada/43769] New: Ada Bug Box when -O1 or higher

[ve3wwg at gmail dot com]

Using -O1 (or higher) in the compile:

gnatmake -O1 -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr z9.adb

provokes a bug box:

gnatmake -O1 -v -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9.adb

GNATMAKE  4.3.4 20090804 (release) 1
Copyright (C) 1995-2007, Free Software Foundation, Inc.
  "z9.ali" being checked ...
  -> "z9.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9.adb
  "z9-command_input.ali" being checked ...
  -> "z9-command_input.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9-command_input.adb
  "z9-command_options.ali" being checked ...
  -> "z9-command_options.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9-command_options.adb
  "z9-commands.ali" being checked ...
  -> "z9-commands.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9-commands.adb
  "z9-gc_styles.ali" being checked ...
  -> "z9-gc_styles.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9-gc_styles.adb
  "z9-global.ali" being checked ...
  -> "z9-global.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9-global.adb
  "z9-lex_streams.ali" being checked ...
  -> "z9-lex_streams.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9-lex_streams.adb
  "z9-lexers.ali" being checked ...
  -> "z9-lexers.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9-lexers.adb
  "z9-line_buffers.ali" being checked ...
  -> "z9-line_buffers.ali" missing.
gcc -c -O1 -save-temps -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
z9-line_buffers.adb
+===========================GNAT BUG DETECTED==============================+
| 4.3.4 20090804 (release) 1 (i686-pc-cygwin) GCC error:                   |
| in bitfield_overlaps_p, at tree-sra.c:2937                               |
| Error detected around z9-line_buffers.adb:1479                           |
| Please submit a bug report; see http://gcc.gnu.org/bugs.html.            |
| Use a subject line meaningful to you and us to track the bug.            |
| Include the entire contents of this bug box in the report.               |
| Include the exact gcc or gnatmake command that you entered.              |
| Also include sources listed below in gnatchop format                     |
| (concatenated together with no headers between files).                   |
+==========================================================================+

Please include these source files with error report
Note that list may not be accurate in some cases,
so please double check that the problem can still
be reproduced with the set of files listed.

z9-line_buffers.adb
z9-line_buffers.ads
z9.ads
z9-global.ads
z9-string_containers.ads
simpmap.ads
z9-tokens.ads
z9-terminal.ads
z9-funcs.ads
z9-keywords.ads
z9-functions.ads
z9-error_codes.ads
z9-command_history.ads
rings.ads
z9-variables.ads
subscripts.ads
z9-lexers.ads
z9-lex_streams.ads
z9-gc_styles.ads
z9-vstrings.ads
z9-line_buffers-load_unnumbered.adb
simpmap.adb


raised TYPES.UNRECOVERABLE_ERROR : comperr.adb:398
End of compilation
gnatmake: "z9-line_buffers.adb" compilation error
make: *** [all] Error 4
----
$ cat Makefile

BUG = -v -save-temps
#DEBUG  = -g -O0
DEBUG  = -O1
AFLAGS = $(DEBUG) $(BUG) -gnat05 -Wall -gnatwl -gnata -gnatVa -gnatf -gnatwr
CFLAGS = $(DEBUG) -I.. $(BUG)
CXFLAGS = $(DEBUG) -I.. $(BUG)

all:    z9.adb
        gnatmake $(AFLAGS) z9.adb

z9.adb: z9.ada
        gnatchop z9.ada

clean:
        rm -f *.ali *.o *.exe *.exe.stackdump core *.a
        rm -f b~*.a* *.tmp *.s
        rm -f errs.t

clobber: clean
        rm *.adb *.ads
----
Where do I attach my 8,223 line z9.ada file in this Bug-zilla?  Surely
not in this message? First time using gcc bugzilla-- hopefully it lets
me attach it in the next step(s). Otherwise email me, and I'll gladly
forward the z9.ada (gnatchop-able) source code.

Warren


-- 
           Summary: Ada Bug Box when -O1 or higher
           Product: gcc
           Version: 4.3.4
            Status: UNCONFIRMED
          Severity: critical
          Priority: P3
         Component: ada
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: ve3wwg at gmail dot com
 GCC build triplet: i686-pc-cygwin
  GCC host triplet: i686-pc-cygwin
GCC target triplet: i686-pc-cygwin


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug ada/43769] Ada Bug Box when -O1 or higher

[ve3wwg at gmail dot com]

------- Comment #1 from ve3wwg at gmail dot com  2010-04-16 14:34 -------
Created an attachment (id=20396)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20396&action=view)
gnatchop-able source code

You will need this source code to reproduce the bug box.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug ada/43769] Ada Bug Box when -O1 or higher

[ve3wwg at gmail dot com]

------- Comment #2 from ve3wwg at gmail dot com  2010-04-16 14:35 -------
Created an attachment (id=20397)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20397&action=view)
Makefile

This is a simple Makefile for your convenience:

make  # will gnatchop and reproduce the bug box

make  # clobber cleans out everything but the Makefile and z9.ada file.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[rguenth at gcc dot gnu dot org]

------- Comment #3 from rguenth at gcc dot gnu dot org  2010-04-16 14:44 -------
Use -fno-tree-sra as a workaround.


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|ada                         |tree-optimization
            Summary|Ada Bug Box when -O1 or     |ICE in bitfield_overlaps_p,
                   |higher                      |at tree-sra.c:2937


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[ebotcazou at gcc dot gnu dot org]

------- Comment #4 from ebotcazou at gcc dot gnu dot org  2010-04-16 15:15 -------
Known problem in the SRA pass.


-- 

ebotcazou at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ebotcazou at gcc dot gnu dot
                   |                            |org
           Severity|critical                    |normal
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2010-04-16 15:15:05
               date|                            |
            Summary|ICE in bitfield_overlaps_p, |[4.3 regression] ICE in
                   |at tree-sra.c:2937          |bitfield_overlaps_p, at
                   |                            |tree-sra.c:2937


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[ebotcazou at gcc dot gnu dot org]

------- Comment #5 from ebotcazou at gcc dot gnu dot org  2010-04-16 15:17 -------
Created an attachment (id=20398)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=20398&action=view)
Potential, untested fix.

        * tree-sra.c (bitfield_overlaps_p): If the length of the element is
        self-referential, try to compute an upper bound.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[ebotcazou at gcc dot gnu dot org]

------- Comment #6 from ebotcazou at gcc dot gnu dot org  2010-04-16 15:20 -------
Richard, do you think this kind of patches is worth installing on the branch at
this point?  If no, we should mark the PR as WONTFIX, the workaround is easy.


-- 

ebotcazou at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[rguenther at suse dot de]

------- Comment #7 from rguenther at suse dot de  2010-04-16 15:32 -------
Subject: Re:  [4.3 regression] ICE in
 bitfield_overlaps_p, at tree-sra.c:2937

On Fri, 16 Apr 2010, ebotcazou at gcc dot gnu dot org wrote:

> ------- Comment #6 from ebotcazou at gcc dot gnu dot org  2010-04-16 15:20 -------
> Richard, do you think this kind of patches is worth installing on the branch at
> this point?  If no, we should mark the PR as WONTFIX, the workaround is easy.

Yes, the patch looks like it can't make things worse and so is
certainly fine (4.4 looks also affected?  if not, how was it
fixed there - maybe that fix should be backported instead)

Thanks,
Richard.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[ebotcazou at gcc dot gnu dot org]

------- Comment #8 from ebotcazou at gcc dot gnu dot org  2010-04-16 15:35 -------
> Yes, the patch looks like it can't make things worse and so is
> certainly fine (4.4 looks also affected?  if not, how was it
> fixed there - maybe that fix should be backported instead)

Yes, 4.4 is very likely affected as well, I'll test.


-- 

ebotcazou at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |ebotcazou at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2010-04-16 15:15:05         |2010-04-16 15:35:24
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3/4.4 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[ebotcazou at gcc dot gnu dot org]

------- Comment #9 from ebotcazou at gcc dot gnu dot org  2010-04-18 07:15 -------
Present on x86 and x86-64/Linux, as well as on the 4.4 branch.


-- 

ebotcazou at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  GCC build triplet|i686-pc-cygwin              |
   GCC host triplet|i686-pc-cygwin              |
 GCC target triplet|i686-pc-cygwin              |
            Summary|[4.3 regression] ICE in     |[4.3/4.4 regression] ICE in
                   |bitfield_overlaps_p, at     |bitfield_overlaps_p, at
                   |tree-sra.c:2937             |tree-sra.c:2937


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3/4.4 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[ebotcazou at gcc dot gnu dot org]

------- Comment #10 from ebotcazou at gcc dot gnu dot org  2010-04-18 15:56 -------
Subject: Bug 43769

Author: ebotcazou
Date: Sun Apr 18 15:56:32 2010
New Revision: 158490

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=158490
Log:
        PR tree-optimization/43769
        * tree-sra.c (bitfield_overlaps_p): If the length of the element is
        self-referential, try to compute an upper bound.

Added:
    branches/gcc-4_4-branch/gcc/testsuite/gnat.dg/rep_clause5.adb
      - copied unchanged from r158489,
trunk/gcc/testsuite/gnat.dg/rep_clause5.adb
    branches/gcc-4_4-branch/gcc/testsuite/gnat.dg/rep_clause5.ads
      - copied unchanged from r158489,
trunk/gcc/testsuite/gnat.dg/rep_clause5.ads
    branches/gcc-4_4-branch/gcc/testsuite/gnat.dg/rep_clause5_pkg.ads
      - copied unchanged from r158489,
trunk/gcc/testsuite/gnat.dg/rep_clause5_pkg.ads
Modified:
    branches/gcc-4_4-branch/gcc/ChangeLog
    branches/gcc-4_4-branch/gcc/testsuite/ChangeLog
    branches/gcc-4_4-branch/gcc/tree-sra.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3/4.4 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[ebotcazou at gcc dot gnu dot org]

------- Comment #11 from ebotcazou at gcc dot gnu dot org  2010-04-18 15:57 -------
Subject: Bug 43769

Author: ebotcazou
Date: Sun Apr 18 15:56:56 2010
New Revision: 158491

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=158491
Log:
        PR tree-optimization/43769
        * tree-sra.c (bitfield_overlaps_p): If the length of the element is
        self-referential, try to compute an upper bound.

Added:
    branches/gcc-4_3-branch/gcc/testsuite/gnat.dg/rep_clause5.adb
      - copied unchanged from r158489,
trunk/gcc/testsuite/gnat.dg/rep_clause5.adb
    branches/gcc-4_3-branch/gcc/testsuite/gnat.dg/rep_clause5.ads
      - copied unchanged from r158489,
trunk/gcc/testsuite/gnat.dg/rep_clause5.ads
    branches/gcc-4_3-branch/gcc/testsuite/gnat.dg/rep_clause5_pkg.ads
      - copied unchanged from r158489,
trunk/gcc/testsuite/gnat.dg/rep_clause5_pkg.ads
Modified:
    branches/gcc-4_3-branch/gcc/ChangeLog
    branches/gcc-4_3-branch/gcc/testsuite/ChangeLog
    branches/gcc-4_3-branch/gcc/tree-sra.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769

[Bug tree-optimization/43769] [4.3/4.4 regression] ICE in bitfield_overlaps_p, at tree-sra.c:2937

[ebotcazou at gcc dot gnu dot org]

------- Comment #12 from ebotcazou at gcc dot gnu dot org  2010-04-18 16:00 -------
On all branches.


-- 

ebotcazou at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://gcc.gnu.org/ml/gcc-
                   |                            |patches/2010-
                   |                            |04/msg01135.html
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |4.3.5


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43769