Re: [PATCH 2/3] integrity: Linux Integrity Module(LIM)

["Serge E. Hallyn" <serue@us.ibm.com>]

Quoting Christoph Hellwig (hch@infradead.org):
> >  /*
> > diff --git a/include/linux/fs.h b/include/linux/fs.h
> > index 32477e8..349d548 100644
> > --- a/include/linux/fs.h
> > +++ b/include/linux/fs.h
> > @@ -683,6 +683,9 @@ struct inode {
> >  #ifdef CONFIG_SECURITY
> >  	void			*i_security;
> >  #endif
> > +#ifdef CONFIG_INTEGRITY
> > +	void			*i_integrity;
> > +#endif
> 
> Sorry, but as said before bloating the inode for this is not an option.
> Please use something like the MRU approach I suggested in the last
> review round.

Hi Christoph, Mimi is looking into uinsg a tree but is still trying to
get the locking right.  So in the meantime I just have to ask again -
for something which will always be either compiled out, or filled in for
every inode, why have the overhead of having an external cache?  Either
the i_integrity won't be compiled in, or you'll have to go through the
external cache for every inode operation anyway.  It doesn't make sense
to me...  Is there a case I'm missing that risks getting hard-hit here
(performance-wise), or is this mainly a style thing?

thanks,
-serge