[Devel] Re: [PATCH 4/4] keys: make procfiles per-user-namespace

[Serge E. Hallyn]

Quoting David Howells (dhowells at redhat.com):
> Serge E. Hallyn <serue at us.ibm.com> wrote:
> 
> > > Hmmm...  I wonder if we can do better by making the file position indicate
> > > the key ID rather than being a count of the number of keys read.  It might
> > > make this cleaner.
> > 
> > file position?  as in the result of lseek(fd, 0, SEEK_CUR)?
> > 
> > I don't understand what you're suggesting.
> 
> Currently the file position on /proc/keys indicates the number of keys that
> have been read.  It is incremented by 1 for each key read, irrespective of the
> length of the line that was read for that key.
> 
> We could, instead, map file positions to key IDs, and skip any file positions
> that don't actually map to an extant key.

So you want users to be able to mmap the file and lseek to a particular
spot?  Is that bc you have users with so many keys that
	grep keyid /proc/keys
becomes slow?

Or am I still misunderstanding?

Meanwhile, do you have any objections to these 4 patches?  :)

thanks,
-serge
_______________________________________________
Containers mailing list
Containers at lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers