Hi Rakesh, Yes, this is plain wrong.
The logic of this function is bad. The code for creating a new record is a few lines below, and activated through an irrelevant condition. I have to rework that... Regards Jean-Pierre Rakesh Pandit wrote: > Hi Jean-Pierre, > > May you check that while playing redo operations in play_redos if we > don't need old record (or ignore it), we allocate a buffer without > initializing but then use it later. At least in redo_create_file this > was done for one of the NTFS storage devices I have. > > Attached is possible way to avoid it. Note that this might be an issue > in other places but I have just skimmed through some actions but seems > worth fix even if trivial. > > Best regards, > > > ------------------------------------------------------------------------------ > > > > _______________________________________________ > ntfs-3g-devel mailing list > ntfs-3g-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel > ------------------------------------------------------------------------------ _______________________________________________ ntfs-3g-devel mailing list ntfs-3g-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ntfs-3g-devel
