Justus Winter, le Fri 29 Aug 2014 11:33:44 +0200, a écrit : > Previously, peropen objects were created with a reference count of > zero. Therefore, if diskfs_create_protid fails, passing such an > object to diskfs_release_peropen would lead to a reference count > underflow.
I see, it is indeed more convenient this way. Please also update the doc/hurd.texi documentation. It seems that libnetfs doesn't even care about calling netfs_release_peropen on diskfs_make_protid errors, thus leaking the peropen memory. Are you having a look already? Samuel
