Date: Wed, 28 Jun 2000 10:47:58 -0700 From: Alfred Perlstein <alfred@FreeBSD.org> To: Chris Costello <chris@calldei.com> Cc: cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/sys/miscfs/fdesc fdesc_vnops.c Message-ID: <20000628104758.Y275@fw.wintelcom.net> In-Reply-To: <20000627223423.H20702@holly.calldei.com>; from chris@calldei.com on Tue, Jun 27, 2000 at 10:34:24PM -0500 References: <200006272246.PAA76896@freefall.freebsd.org> <20000627223423.H20702@holly.calldei.com>
next in thread | previous in thread | raw e-mail | index | archive | help
* Chris Costello <chris@calldei.com> [000627 20:41] wrote:
> On Tuesday, June 27, 2000, Alfred Perlstein wrote:
> > alfred 2000/06/27 15:46:36 PDT
> >
> > Modified files:
> > sys/miscfs/fdesc fdesc_vnops.c
> > Log:
> > by changing the logic here we can support dynamic additions of new
> > filetypes.
>
> +
> + default:
> error = fo_stat(fp, &stb, ap->a_p);
> if (error == 0) {
> VATTR_NULL(vap);
> - if (fp->f_type == DTYPE_KQUEUE)
> + /* XXX Fake it! */
> + if (fp->f_type != DTYPE_PIPE && fp->f_type != DTYPE_SOCKET)
> vap->va_type = VFIFO;
>
> This has bad style (way past 80 columns) and it just does not
> make sense. This doesn't make it any easier to add new file
> types, and you're just hiding the DTYPE_KQUEUE, DTYPE_PIPE and
> DTYPE_SOCKET cases inside a default: handler. Are all new file
> types going to be FIFOs?
>
> What you should really do is ensure that any new file type
> returns proper stat information. I'm tempted to say that ones
> that don't should be treated as VNON, rather than VFIFO, from a
> design perspective.
so basically change:
vap->va_type = VFIFO;
to:
vap->va_type = VNON;
and fix the long line?
-Alfred
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000628104758.Y275>