Ingo Bormuth wrote:

>Hi Edward,
>
>I just played around with reiser4's cryptcompress plugin
>an realized an odd behaviour.
>
>It seems, every file on any partition made with '-o ccreg40'
>has its SATTY_BITS set:
>  
>

Wow, it seems Ingo has found the culprit that confused Gentoo
manager.

I was inaccurate when introducing some dummy methods
to make file operations supplied for VFS invariant:
ioctl_cryptcompress should return -ENOSYS, otherwise userspace
application can make a wrong decision about file's nature.

Dushan, would you please check, if the attached patch really
fixes the python-prompt-problem?

Thanks to everyone!
Edward.

>$mkfs.reiser4 /dev/hdaXX
>$mount /dev/hdaXX /mnt/test
>$touch /mnt/test/x; exec 3<>/mnt/test/x; test -t 3 && echo T || echo F; exec 3>&-
>F
>$umount /mnt/test
>$mkfs.reiser4 -o create=ccreg40 /dev/hdaXX
>$mount /dev/hdaXX /mnt/test
>$touch /mnt/test/x; exec 3<>/mnt/test/x; test -t 3 && echo T || echo F; exec 3>&-
>T
>
>$mkfs.reiser4 -V
>mkfs.reiser4 1.0.6
>Copyright (C) 2001-2005 by Hans Reiser, licensing governed by reiser4progs/COPYING.
>
>$cat /proc/version
>Linux version 2.6.23.9-reiser4-r2 (gcc version 4.1.2) #1 PREEMPT Fri Dec 14 19:34:05 CET 2007
>
>
>Great performance andspace efficiency, no further problems, great job, btw.
>
>
>
>
>  
>