Date: Mon, 27 Feb 2006 12:34:31 +0300 From: Gleb Smirnoff <glebius@FreeBSD.org> To: Andre Oppermann <andre@FreeBSD.org>, Yar Tikhiy <yar@comp.chem.msu.su> Cc: arch@FreeBSD.org, yar@FreeBSD.org, jlemon@FreeBSD.org Subject: Re: changing EINVAL for SIOCSIFCAP to something else Message-ID: <20060227093431.GX55275@cell.sick.ru> In-Reply-To: <20060227091417.GF6435@comp.chem.msu.su> <4402C09C.C3FB0064@freebsd.org> References: <20060227083815.GW55275@cell.sick.ru> <20060227091417.GF6435@comp.chem.msu.su> <20060227083815.GW55275@cell.sick.ru> <4402C09C.C3FB0064@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Andre, Yar, On Mon, Feb 27, 2006 at 10:04:28AM +0100, Andre Oppermann wrote: A> > I prefer this variant: A> > A> > if (ifp->if_ioctl == NULL) A> > return (ENOTTY); A> > if (ifr->ifr_reqcap & ~ifp->if_capabilities) A> > return (ENODEV); A> > A> > Any objections? A> A> I don't think ENOTTY is appropriate here even though the comment to this A> error code would fit. But the define still says no TTY which is totally A> unrelated and confusing. It contains a confusing word "tty", but it means "Inappropriate ioctl for device". This error code is used in many places throughout the kernel. We already have some ENOTTY returns in src/sys/net. Y> I'm afraid that this is a case when EINVAL is used properly: an Y> argument to ioctl doesn't make sense to a particular device. It's Y> true that EINVAL may be abused in other places though. I wish each Y> EINVAL being returned to the userland were accompanied by log(). I don't agree. EINVAL can logically fit to almost any error condition. We should fine error codes fitting better. If "ioctl doesn't make sense to a particular device", then we should say "Operation not supported by device", which is ENODEV. -- Totus tuus, Glebius. GLEBIUS-RIPN GLEB-RIPE
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060227093431.GX55275>