Re: EXT2_UNRM_FL
Richard Gooch (rgooch@atnf.csiro.au)
Fri, 5 Mar 1999 14:45:49 +1100
david parsons writes:
> In article <linux.kernel.199903032337.SAA00994@dcl>,
> Theodore Y. Ts'o <tytso@MIT.EDU> wrote:
> >
> > (d) why not do it in userspace anyway? I did that years ago, although
> > I "moved" files to /tmp, but it would be easy enough to move to a
> > garbage/$LOGNAME directory on the same FS.
> >
> >Absolutely, agreed. As I said earlier, there are plenty of user-space
> >"rm" replacements, and doing it in the kernel is almost certainly not
> >worth the pain.
>
> Unless you want to do it the right way.
>
> Userspace bolt-ons are spiffy iff you stick within the environment
> the designer the bolt-on set up for you. If you walk outside that
> environment, all of a sudden things don't work anymore, so (for
> instance) a LD_PRELOAD will mysteriously fail when you run a
> statically linked rm command, or a hacked ELF library will fail
> when you have the temerity to run an a.out executable.
But why does it matter? Safe-delete is hardly a critical facility:
it's just there to protect you from doing silly things like:
% rm fred *
I'm not even convinced it belongs anywhere except in a replacement
utility for rm(1). It doesn't seem critical for other programmes. And
it's definately not critical for *all* programmes.
The only other place outside of rm(1) I can see it being useful is
inside a GUI like a file manager. In which case it's new code and I
don't think we have to worry about static binaries (statically linked
GUIs are bloatware, especially if they come as part of a desktop
environment, where you get lots of copies of the same code).
Regards,
Richard....
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/