See downstream bug report: http://bugs.gentoo.org/270664 Running "valgrind --tool=memcheck --leak-check=full ristretto" And then opening the image, gives this output: ==31404== LEAK SUMMARY: ==31404== definitely lost: 3,234 bytes in 13 blocks. ==31404== indirectly lost: 10,068 bytes in 505 blocks. ==31404== possibly lost: 197,247 bytes in 236 blocks. ==31404== still reachable: 1,076,691,176 bytes in 29,200 blocks. ==31404== suppressed: 0 bytes in 0 blocks. ==31404== Reachable blocks (those to which a pointer was found) are not shown. ==31404== To see them, rerun with: --leak-check=full --show-reachable=yes There is an example image that causes this on that bug report, feel free to ask if you need anymore info. Thanks.
This seems to be a bug in the image + gtk, it can not be opened in the gimp here either. I'll have to do some more investigation though.
(In reply to comment #1) > This seems to be a bug in the image + gtk, it can not be opened in the gimp > here either. I'll have to do some more investigation though. Probably not a big deal then. Thanks for the comment and sorry for my late response. I'm closing the downstream gentoo report
Well, could you try to reproduce with the latest gtk+ (and both ristretto and the gimp), if it is a Gtk+ bug it should still be fixed ofcourse, but at a different upstream. ;-)
This happens with GTK 2.16.5 (2.16 is the latest stable). In some programs it displays fine; in others not. The programs in which it doesn't work: * display, ImageMagick. * F-Spot * Gimp * Ristretto * Evince Programs in which it does work: * Whatever makes the Nautilus thumbnails * gthumb * Eye of Gnome ImageMagick and GTK both use libtiff, so maybe it is libtiff that is to blame here. but that doesn't explain why in some programs it does work.
Gonna close this bug then :)