java/1376: Core dump in GC_find_limit() in os_dep.c

java/1376: Core dump in GC_find_limit() in os_dep.c

[jthwaites]

>Number:         1376
>Category:       java
>Synopsis:       Core dump in GC_find_limit() in os_dep.c
>Confidential:   no
>Severity:       serious
>Priority:       low
>Responsible:    tromey
>State:          feedback
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Dec 20 12:19:25 PST 2000
>Closed-Date:    
>Last-Modified:  Thu Oct 12 12:16:00 PDT 2000
>Originator:     jthwaites@michaelhouse.com
>Release:        2.95
>Organization:
>Environment:
LinuxPPC 2000 (2.2.15 kernel) on Mac Powerbok
>Description:
Compiled hello.java using gcj -ohello --main=hello hello.java
with CLASSPATH=/usr/local/lib/libgcj.zip

Resulting executable core dumps immediately as described in synopsis (using GDB).
>How-To-Repeat:

>Fix:

>Release-Note:

>Audit-Trail:

Formerly PR gcj/349


From: Tom Tromey <tromey@cygnus.com>
To: jthwaites@michaelhouse.com
Cc: java-gnats@sourceware.cygnus.com
Subject: Re: gcj/349: Core dump in GC_find_limit() in os_dep.c
Date: 15 Sep 2000 16:03:45 -0600

 >> Synopsis:       Core dump in GC_find_limit() in os_dep.c
 
 This "crash" is normal -- the GC catches it and handles it.
 Please try debugging again and this time use "cont" in the debugger to
 go past this segv.
 
 Tom
Responsible-Changed-From-To: apbianco->tromey
Responsible-Changed-By: tromey
Responsible-Changed-When: Thu Oct 12 12:04:55 2000
Responsible-Changed-Why:
    Mine
State-Changed-From-To: open->feedback
State-Changed-By: tromey
State-Changed-When: Thu Oct 12 12:04:55 2000
State-Changed-Why:
    I'm putting this in `feedback' because we need more 
    information before we can address the real problem.
    The segv you saw is normal, and you should be able
    to "continue" past it in gdb to find the real segv.

From: tromey@cygnus.com
To: apbianco@cygnus.com, java-gnats@sourceware.cygnus.com,
  jthwaites@michaelhouse.com, tromey@cygnus.com
Cc:  
Subject: Re: gcj/349
Date: 12 Oct 2000 19:04:55 -0000

 Synopsis: Core dump in GC_find_limit() in os_dep.c
 
 Responsible-Changed-From-To: apbianco->tromey
 Responsible-Changed-By: tromey
 Responsible-Changed-When: Thu Oct 12 12:04:55 2000
 Responsible-Changed-Why:
     Mine
 State-Changed-From-To: open->feedback
 State-Changed-By: tromey
 State-Changed-When: Thu Oct 12 12:04:55 2000
 State-Changed-Why:
     I'm putting this in `feedback' because we need more 
     information before we can address the real problem.
     The segv you saw is normal, and you should be able
     to "continue" past it in gdb to find the real segv.
 
 http://sources.redhat.com/cgi-bin/gnatsweb.pl?cmd=view&pr=349&database=java
>Unformatted:
----gnatsweb-attachment----
Content-Type: application/octet-stream; name="hello.java"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="hello.java"
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