I am getting Windows version of segfault with the latest builds. This is with glib-2.39.1. Should I file a bugreport or it will sort itself out?

Here is what I see during the crash:

0x000000006862110F (0x0000000101ED4790 0x0000000101ED4780 0x0000000101ED4790 0x0000000002229D95), g_slice_free1() + 0x17F bytes(s) 0x0000000002248B1D (0x3F847AE147AE147B 0x0000000000000036 0x00000000841EBF40 0x00000000841EBF40), g_type_free_instance() + 0x18D bytes(s)
0x000000006207F64A (0x0000000000000002 0x00000000136F3500 0x00000000136F3500 0x0000000013534B00), gegl_tile_cache_destroy() + 0x2EA bytes(s)
0x000000000222A345 (0x0000000000000000 0x00000000685D86BD 0x0000000000000000 0x0000000000000000), g_object_unref() + 0x115 bytes(s) 0x000000006207E138 (0x0000000000000000 0x00000000137A0670 0x00000000620956F0 0x00000000841EB5B0), gegl_tile_backend_swap_cleanup() + 0x468 bytes(s)
0x000000000222A345 (0x000000002181A740 0x000000002181A740 0x000000003C8BD210 0x000000008420FDE0), g_object_unref() + 0x115 bytes(s) 0x000000000224D027 (0x00000000817AD9E0 0x000000002181A740 0x00000000841EB5B0 0x0000000000000000), g_value_unset() + 0x27 bytes(s) 0x0000000062089C22 (0x000000000023F8C0 0x0000000010FD1340 0x00000000899005D0 0x000000000023F8B0), gegl_operation_context_purge() + 0x32 bytes(s)
0x0000000062095CC2 (0x0000000000000002 0x0000000002249ABE 0x0000000000000001 0x00000000007F3932), gegl_graph_process() + 0x82 bytes(s) 0x00000000620953A1 (0x0000000000000100 0x00000000007F3A6E 0x0000000100000000 0x00000000137C26D0), gegl_eval_manager_apply() + 0x111 bytes(s)
0x000000006208CC75 (0x0000000000000001 0x00000000686212CC 0x0000000000000000 0x00000000685F5AE0), gegl_node_blit_buffer() + 0x45 bytes(s)
0x0000000000753840 (0x0000000000000000 0x000000007FFFFFFF 0x000000000023FC00 0x00000000033DB890) 0x00000000685FB665 (0x0000000000000000 0x000000006862BE90 0x0000000003431420 0x00000000033DB890), g_main_context_dispatch() + 0x135 bytes(s)
0x00000000685FB998 (0x0000000000000000 0x00000000006B003D 0x0000000000000000 0x0000000000000000), g_main_context_dispatch() + 0x468 bytes(s)
0x00000000685FBE27 (0x00000000034763A0 0x0000000000000000 0x0000000000A40830 0x0000000000409F80), g_main_loop_run() + 0x67 bytes(s) 0x000000000040182B (0x0000000000000001 0x000000000000002B 0x0000000000A40830 0x0000000000A4C960) 0x000000000080CF12 (0x0000000000000001 0x0000000000A4C960 0x0000000000000000 0x0000000000000000) 0x00000000004013D5 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000) 0x00000000004014D8 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000) 0x00007FFF9FBC15CD (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), BaseThreadInitThunk() + 0xD bytes(s) 0x00007FFFA00343D1 (0x0000000000000000 0x0000000000000000 0x0000000000000000 0x0000000000000000), RtlUserThreadStart() + 0x1D bytes(s)

Thanks. I will.

I will also rebuild glib and see if the crash goes away.

In response to your questions.

1. The last version that worked is the one I uploaded to my website. However, one caveat, I was using it on Windows 7. 2. I was editing a 16-bit tiff and then opening another 16-bit tiff.

On Sun, Dec 15, 2013 at 1:25 PM, Daniel Sabo wrote:

This appears to be something new, I would file a bug.

Standard questions: What was the last version that worked? What were you doing when it crashed?

I rebuilt glib-2.0 and I am still experiencing crashes. However, it's no longer the above message. Usually after opening 2 or 3 large 16-bit tiffs and running unsharp mask and some curves, I get the following with gdb:

warning: HEAP[gimp-2.9.exe]: warning: HEAP: Free Heap block 0000000082DCA470 modified at 0000000082DE3850 after it was freed

Program received signal SIGTRAP, Trace/breakpoint trap. 0x00007fffa00b55f6 in ?? ()
(gdb) bt
#0 0x00007fffa00b55f6 in ?? ()
#1 0x00007fffa000503c in ?? ()
#2 0x00007fffa0061dc0 in ?? ()
#3 0x0000000082dca470 in ?? ()
#4 0x0000000082de3850 in ?? ()
#5 0x00000000003902c0 in ?? ()
#6 0x000000000000007f in ?? ()
#7 0x000000008a0085c0 in ?? ()
#8 0x0000000050000100 in ?? ()
#9 0x0000000000000000 in ?? ()

So, can't reproduce the prior error, but it's something else now.

Thanks, Partha

On Sun, Dec 15, 2013 at 4:34 PM, Partha Bagchi wrote:

Thanks. I will.

I will also rebuild glib and see if the crash goes away.

In response to your questions.

1. The last version that worked is the one I uploaded to my website. However, one caveat, I was using it on Windows 7. 2. I was editing a 16-bit tiff and then opening another 16-bit tiff.

On Sun, Dec 15, 2013 at 1:25 PM, Daniel Sabo wrote:

This appears to be something new, I would file a bug.

Standard questions: What was the last version that worked? What were you doing when it crashed?

36MP.

On Sun, Dec 15, 2013 at 7:47 PM, Liam R E Quin wrote:

On Sun, 2013-12-15 at 18:01 -0500, Partha Bagchi wrote:

I rebuilt glib-2.0 and I am still experiencing crashes. However, it's no longer the above message. Usually after opening 2 or 3 large 16-bit tiffs and running unsharp mask and some curves, I get the following with gdb:

How large exactly, in pixels? 64-bit gimp?

I was having problems recently feathering the selection that related to an integer overflow in the size of a gegl buffer. Not saying you have the same problem of course, but that the dimensions of the image may matter.

Liam

--
Liam Quin - XML Activity Lead, W3C, http://www.w3.org/People/Quin/ Pictures from old books: http://fromoldbooks.org/ Ankh: irc.sorcery.net irc.gnome.org freenode/#xml