This is the mail archive of the gdb-prs@sources.redhat.com mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

java/1322: "internal-error: sect_index_text not initialized" after printing a java type


>Number:         1322
>Category:       java
>Synopsis:       "internal-error: sect_index_text not initialized" after printing a java type
>Confidential:   no
>Severity:       critical
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Wed Aug 06 20:18:00 UTC 2003
>Closed-Date:
>Last-Modified:
>Originator:     mec@shout.net
>Release:        gdb HEAD 2003-08-03 05:21:09 UTC
>Organization:
>Environment:
target=native
host=i686-pc-linux-gnu
osversion=red-hat-8.0
gdb=HEAD 2003-08-03 05:21:09 UTC
gcc=3.3
binutils=2.14
glibc=2.2.93-5-rh
gformat=dwarf-2
glevel=2
>Description:
gdb gets an internal error of "sect_index_text not initialized" in file dwarf2-frame.c function dwarf2_frame_find_fde line 1025.

The problem is that dwarf2_frame_find_fde does an ALL_OBJFILES loop and then looks at SECT_OFF_TEXT.  It hits the java "anonymous objfile" and borks out.

There is a nice stack trace here:

http://sources.redhat.com/ml/gdb/2003-08/msg00078.html

A similar problem happened with gdb 5.3 so this is not a regression against gdb 5.3.  However, gdb 2003-05-15 00:00:00 UTC works fine.  The problem probably came in with the new file dwarf2-frame.c on 2003-05-31.

Tom Tromey says this makes Java debugging unusable, hence the severity=critical marker.
>How-To-Repeat:
Run a Java program.
Set a breakpoint and take it.
Print a Java type with an array or a string in it
  (the arguments to "main" work fine).
Set a breakpoint in a shared library that has not had a breakpoint in it yet.  I like to use 'exit'.
Continue to that breakpoint.

This happens a lot more often if (1) you use more shared libraries, such as making the Java run-time shared, and (2) if you re-run the program instead of doing this in one run.

I am working on a patch to jmisc1.exp to make this easier to repeat.

>Fix:
Daniel Jacobowitz has a path to a fix:

http://sources.redhat.com/ml/gdb/2003-08/msg00080.html
>Release-Note:
>Audit-Trail:
>Unformatted:


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]