Bug 456989 - gs cores in gx_path_new at src/gxpath.c:441
Summary: gs cores in gx_path_new at src/gxpath.c:441
Keywords:
Status: CLOSED NEXTRELEASE
Alias: None
Product: Fedora
Classification: Fedora
Component: ghostscript
Version: 9
Hardware: All
OS: Linux
low
medium
Target Milestone: ---
Assignee: Tim Waugh
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2008-07-29 02:54 UTC by Tim Taiwanese Liim
Modified: 2008-09-25 00:06 UTC (History)
0 users

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2008-09-25 00:06:37 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
the ps file that caused the core dump (must go with proper args) (945.96 KB, application/postscript)
2008-07-29 02:54 UTC, Tim Taiwanese Liim
no flags Details

Description Tim Taiwanese Liim 2008-07-29 02:54:02 UTC
Description of problem:
    gs cores in gx_path_new at src/gxpath.c:441 for attached file,
    with ijsgutenprint.


Version-Release number of selected component (if applicable):
    ghostscript-8.62-4.fc9.i386

How reproducible:
    always, with said sample file and args.
    unable to reproduce otherwise.


Steps to Reproduce:
    1. in shell prompt,
          b=/usr/bin/gs 
          arg="-sDEVICE=ijs -sIjsServer=ijsgutenprint.5.0 -r360x360
080728.gs.cores.at.gxpath.c.441.ps"
          export b arg
    2. $b $arg
    3. gdb $b
       gdb> run $arg
  
Actual results:
    - in step 2, got "Segmentation fault" in 1 sec.
    - in step 3, got
        Program received signal SIGSEGV, Segmentation fault.
        0x05a68956 in gx_path_new (ppath=<value optimized out>) at src/gxpath.c:441
        441             rc_free_path_segments_local(psegs->rc.memory, psegs,
"gx_path_new");

Expected results:
    no core dump.

Additional info:
    The issue was first noticed when printing using CUPS, to a Epson 
    Stylus Color 880 printer.  The attached file would not print, and
    in /var/log/messages this line shows up:
       Jul 28 22:25:59 formosa kernel: gs[19194]: segfault at 4 
       ip 0447c956 sp bfd62380 error 4 in libgs.so.8.62[4108000+442000]

Comment 1 Tim Taiwanese Liim 2008-07-29 02:54:02 UTC
Created attachment 312839 [details]
the ps file that caused the core dump (must go with proper args)

Comment 2 Tim Taiwanese Liim 2008-07-29 03:23:01 UTC
Compiled gs with "-g"; running with gdb, got this:
    Program received signal SIGSEGV, Segmentation fault.
    0x0833cdfc in gx_path_new (ppath=0xbfa56200) at ./src/gxpath.c:441
    441             rc_free_path_segments_local(psegs->rc.memory, psegs,
"gx_path_new");
    (gdb) p psegs
    $1 = (gx_path_segments *) 0x0


Comment 3 Fedora Update System 2008-08-04 16:16:13 UTC
ghostscript-8.63-1.fc9 has been submitted as an update for Fedora 9

Comment 4 Tim Taiwanese Liim 2008-08-06 03:51:00 UTC
Verified with ghostscript-8.63-1.fc10.i386.  
Worked well with sample ps file.  (Haven't seen this fix
in fc9 updates-testing yet; couldn't wait, used rawhide.)

Comment 5 Fedora Update System 2008-08-07 23:52:03 UTC
ghostscript-8.63-1.fc9 has been pushed to the Fedora 9 testing repository.  If problems still persist, please make note of it in this bug report.
 If you want to test the update, you can install it with 
 su -c 'yum --enablerepo=updates-testing update ghostscript'.  You can provide feedback for this update here: http://admin.fedoraproject.org/updates/F9/FEDORA-2008-7035

Comment 6 Tim Taiwanese Liim 2008-08-09 03:17:13 UTC
Verified in ghostscript-8.63-1.fc9.  Works well.
Please close ticket.  Thanks for your effort.

Comment 7 Tim Taiwanese Liim 2008-09-11 02:45:28 UTC
Verified in ghostscript-8.63-1.fc9.  Works well.
Please close ticket.  Thanks for your effort.

Comment 8 Tim Taiwanese Liim 2008-09-24 03:31:16 UTC
Any plan to publish this fix?  It has been 1.5 months
since it was fixed.

Comment 9 Tim Waugh 2008-09-24 08:19:37 UTC
I pushed it yesterday so it will be available shortly.

Comment 10 Fedora Update System 2008-09-25 00:06:31 UTC
ghostscript-8.63-1.fc9 has been pushed to the Fedora 9 stable repository.  If problems still persist, please make note of it in this bug report.


Note You need to log in before you can comment on or make changes to this bug.