[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20081201100350.34af705e@lxorguk.ukuu.org.uk>
Date: Mon, 1 Dec 2008 10:03:50 +0000
From: Alan Cox <alan@...rguk.ukuu.org.uk>
To: weigelt@...ux.de
Cc: linux kernel list <linux-kernel@...r.kernel.org>
Subject: Re: scanner interface proposal was: [TALPA] Intro to a linux
interface for on access scanning (fwd)
On Mon, 1 Dec 2008 01:51:42 +0100
Enrico Weigelt <weigelt@...ux.de> wrote:
> * Alan Cox <alan@...rguk.ukuu.org.uk> wrote:
> > > Fine. Why not just putting this into a userland filesystem ?
> >
> > 1. Performance
>
> Does it really hurt so bad, compared with all actual AV stuff ?
> It has to go to userland anyway, in case you don't intend to
> put the scanner into the kernel ;-o
That depends on the scanning and what you are using it for - at least
part of the interest is in asynchronous indexing systems in which case
the answer appears to be no, the scanner doesn't trash your performance.
>
> > 2. Networked file systems
>
> What's the problem ?
> (btw: 9P already *IS* an network filesystem ;-P)
Try 9P with an NFS backing store. Try a cluster file system.
> > 3. Ioctls
>
> Ah, just forgot a while that this crap still exists ;-o
>
> BUT: do the affected dirs have to contain devices ?
> Is there any point for pulling /dev through the AV scanner ?
Lots of ioctls apply to files. Please think about what is involved rather
than just assuming.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists