Summary: | crash when reading EXIF from USB PtP camera or newly imported album | ||
---|---|---|---|
Product: | [Applications] digikam | Reporter: | Bruce M. Simpson <bms> |
Component: | Import-Gphoto2 | Assignee: | Digikam Developers <digikam-bugs-null> |
Status: | RESOLVED WORKSFORME | ||
Severity: | crash | CC: | caulier.gilles |
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | FreeBSD Ports | ||
OS: | FreeBSD | ||
Latest Commit: | Version Fixed In: | 8.2.0 |
Description
Bruce M. Simpson
2007-07-02 16:29:27 UTC
You must compile digiKam using .configure --enable-debug=full to have a suitable backtrace... Gilles Caulier I have recompiled digikam from ports, modifying the Makefile to pass --enable-debug=full to the configure script, and I still get the same KCrash message about the backtrace being corrupt. FYI: As of this morning I forced another recompile of digikam from ports. I also forced a rebuild of libexif, libkexif and libexif-gtk. I still see the same failure, this time on digikam startup. A build with the configure option --enable-debug=full still produces a corrupt backtrace. since digiKam 0.9.0, we don't use libexif, which is obsolete library since a long time, but libExiv2 to handle Exif, makernote, and iptc... Gilles Caulier I forced a portupgrade of libkexivm libexiv2 and exiv2 ports. digikam still crashes. it appears to link against libexif somewhere down the line: %%% empiric:~ % ldd `which digikam` | grep ex libkexiv2.so.2 => /usr/local/lib/libkexiv2.so.2 (0x28c3f000) libexiv2.so.0 => /usr/local/lib/libexiv2.so.0 (0x28c63000) libXext.so.6 => /usr/local/lib/libXext.so.6 (0x29e5c000) libexif.so.12 => /usr/local/lib/libexif.so.12 (0x2a01a000) libexpat.so.6 => /usr/local/lib/libexpat.so.6 (0x2a4ba000) %%% I will now try rebuilding digikam against the rebuilt exiv libraries. digikam is linked indirectly to libexif because it use libgphoto2 which is linked to libexif so, i recommend to recompile libgphoto2 too... Gilles I forced a rebuild of gphoto2, libgphoto2, and digikam with the same result -- a crash on startup. However I noticed that gdb is in fact dumping core also when KCrash invokes it... #0 0x0819f4e9 in cplus_demangle_mangled_name () #1 0x0819f637 in cplus_demangle_mangled_name () #2 0x0819f13a in cplus_demangle_fill_dtor () #3 0x0819f1c9 in cplus_demangle_fill_dtor () #4 0x081a2426 in cplus_demangle_init_info () #5 0x081a2571 in cplus_demangle_v3 () #6 0x0819af57 in cplus_demangle () #7 0x080e0ec3 in symbol_init_language_specific () #8 0x080e102d in symbol_set_names () #9 0x080ac275 in prim_record_minimal_symbol_and_info () #10 0x081312df in f_print_type () #11 0x081315e7 in f_print_type () #12 0x0813193c in f_print_type () #13 0x080e7911 in syms_from_objfile () #14 0x080e8ced in new_symfile_objfile () #15 0x080aeec5 in free_so () #16 0x080c73d0 in throw_exception () #17 0x080c751a in catch_errors () #18 0x080af1b1 in solib_add () #19 0x0807488e in attach_command () #20 0x080c753d in catch_errors () #21 0x080c73d0 in throw_exception () #22 0x080c751a in catch_errors () #23 0x080c757e in catch_command_errors () #24 0x08071918 in main () I have also informed the FreeBSD port maintainer about this bug... thanks for all your help to date... After a complete restart of X and KDE, everything seems to be working. Sorry for the confusion and thanks for all your help so far. |