[Bug java/17779] New: ICE: Seg fault

[Bug java/17779] New: ICE: Seg fault

[ericw at evcohs dot com]

Using:
gcj (GCC) 3.4.1 (cygming special)

Building a project called Avrora found here:
<http://compilers.cs.ucla.edu/avrora/release.html>
Version 1.2.0-Beta JAR

With command line:
gcj --main=Main -o avrora.exe avrora-beta-1.2.0.jar

I get an ICE:
Arith.java:276: 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.

I am not familiar at all with building Java projects, so I'm not sure how to
provide a smaller test case. There looks to be a an Arith.java file in the 
src\avrora\core\isdl\ast
subdirectory of the source code for this project. I don't know how to build just
that module without a bunch of other errors due to my incomplete command-line.
The web page on how to report a bug <http://gcc.gnu.org/bugs.html>, was not very
helpful when it comes to java projects.

Thanks
Eric Weddington

-- 
           Summary: ICE: Seg fault
           Product: gcc
           Version: 3.4.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: java
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: ericw at evcohs dot com
                CC: gcc-bugs at gcc dot gnu dot org,java-prs at gcc dot gnu
                    dot org
  GCC host triplet: cygwin


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[tromey at gcc dot gnu dot org]

------- Additional Comments From tromey at gcc dot gnu dot org  2004-10-01 17:20 -------
This also fails with current cvs head.

BTW, Eric, you probably want "--main=avrora.Main".
"--main" takes the fully qualified class name,
(just like the "java" command...)


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
      Known to fail|                            |4.0.0
   Last reconfirmed|0000-00-00 00:00:00         |2004-10-01 17:20:40
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-01 17:32 -------
Here is the backtrace for a day old 4.0.0:
#0  make_class_data (type=0xf6e16244) at /home/gates/pinskia/src/gnu/gcc/src/gcc/java/class.c:
1563
#1  0x0809ee15 in finish_class () at /home/gates/pinskia/src/gnu/gcc/src/gcc/java/class.c:1844
#2  0x080c2b1c in parse_class_file () at /home/gates/pinskia/src/gnu/gcc/src/gcc/java/jcf-parse.c:
800
#3  0x080c6497 in java_parse_file (set_yydebug=0) at /home/gates/pinskia/src/gnu/gcc/src/gcc/
java/jcf-parse.c:1248
#4  0x083e61b5 in toplev_main (argc=0, argv=0xfef831e4) at /home/gates/pinskia/src/gnu/gcc/src/
gcc/toplev.c:985
#5  0x00314ad4 in __libc_start_main () from /lib/tls/libc.so.6
#6  0x08049cb1 in _start ()


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-valid-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[ericw at evcohs dot com]

------- Additional Comments From ericw at evcohs dot com  2004-10-01 17:52 -------
Subject: Re:  ICE: Seg fault

tromey at gcc dot gnu dot org wrote:

>------- Additional Comments From tromey at gcc dot gnu dot org  2004-10-01 17:20 -------
>This also fails with current cvs head.
>
>BTW, Eric, you probably want "--main=avrora.Main".
>"--main" takes the fully qualified class name,
>(just like the "java" command...)
>
>
>  
>
Thanks for the tip! As I mentioned, I don't know anything about building 
Java projects.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[tromey at gcc dot gnu dot org]

------- Additional Comments From tromey at gcc dot gnu dot org  2004-10-01 17:58 -------
I looked into this a little.

The bug is that we never call push_super_field() for the class
in question.  In layout_class, push_super_field can be called,
but we fail this test and skip the call:

  if (super_class && !CLASS_BEING_LAIDOUT (super_class))

In our case, CLASS_BEING_LAIDOUT (super_class) is true.

layout_class is only called once for the class in question,
so we never get another chance to call push_super_field.
Later we die because TYPE_FIELDS is NULL_TREE.

Offhand I'm not sure what to do.
Andrew has been working in this area lately, perhaps he
has an idea.  I will ping him.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[aph at gcc dot gnu dot org]

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-05 14:47 -------
Patch here: <http://gcc.gnu.org/ml/gcc-patches/2004-10/msg00380.html>.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-10-05 14:55 -------
Subject: Bug 17779

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	aph@gcc.gnu.org	2004-10-05 14:55:40

Modified files:
	gcc/java       : ChangeLog jcf-parse.c 

Log message:
	2004-10-05  Andrew Haley  <aph@redhat.com>
	
	PR java/17779
	* jcf-parse.c (parse_zip_file_entries): If a class has a
	superclass and a TYPE_SIZE of zero, lay it out.
	
	2004-09-30  Andrew Haley  <aph@redhat.com>
	
	PR java/17733
	* jcf-parse.c (compute_class_name): Rewrite.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/java/ChangeLog.diff?cvsroot=gcc&r1=1.1478&r2=1.1479
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/java/jcf-parse.c.diff?cvsroot=gcc&r1=1.174&r2=1.175



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-05 15:50 -------
Fixed.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |4.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[ericw at evcohs dot com]

------- Additional Comments From ericw at evcohs dot com  2004-10-05 16:04 -------
Can this be backported at all to the 3.4.x series?

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779

[Bug java/17779] ICE: Seg fault

[aph at redhat dot com]

------- Additional Comments From aph at redhat dot com  2004-10-05 16:08 -------
Subject:  ICE: Seg fault

In principle, yes.  I'm not strongly inclined to do so myself, but
someone else could certainly apply the patch and do the tests.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17779