Re: vfs.txt and i_ino

[Eli Carter <eli.carter@inet.com>]

Anton Altaparmakov wrote:
> 
> On Thu, 31 Jan 2002, Eli Carter wrote:
> > It appears that struct inode i_ino has a special value of 0.  I don't
> > see a mention of that in vfs.txt, and I haven't found anything obvious
> > in the fs code... Would it be possible to add some documentation of
> > that, along with an explaination of what i_ino==0 is supposed to
> > indicate?  (Bad/invalid inode?)
> 
> i_ino = 0 is perfectly valid and is in fact one of the system files in
> NTFS. And accessing inode 0 from user space works fine, too. The only
> thing which is odd is that a simple "ls" (or "ls -l") doesn't show the
> file with i_ino=0, while an explicit ls a-la "ls \$MFT" (or "ls -l \$MFT")
> does show the file. I believe this to be purely a userspace problem but
> when I looked at the /bin/ls source I got scared and ran away... A short
> investigation into /bin/ls source didn't make anything obvious appear but
> I do think it is /bin/ls at fault and not the kernel...
> 
> So I guess my point is that i_ino=0 is not special as far as the kernel is
> concerned.

Hmm... 'ls -al' doesn't show the file for me.  I was using i_ino=0 for
the root inode, and found that 'ls -al' did not display '.' or '..'.  It
very well may be a user-space error... do you know who I should ask
about it?

TIA,

Eli
--------------------.     Real Users find the one combination of bizarre
Eli Carter           \ input values that shuts down the system for days.
eli.carter(a)inet.com `-------------------------------------------------