Date: Mon, 13 Jan 1997 14:14:41 -0700 (MST) From: Nate Williams <nate@mt.sri.com> To: Terry Lambert <terry@lambert.org> Cc: nate@mt.sri.com (Nate Williams), joerg_wunsch@uriah.heep.sax.de, freebsd-hackers@freebsd.org Subject: Re: mount -o async on a news servre Message-ID: <199701132114.OAA26992@rocky.mt.sri.com> In-Reply-To: <199701132058.NAA28205@phaeton.artisoft.com> References: <199701130539.WAA23833@rocky.mt.sri.com> <199701132058.NAA28205@phaeton.artisoft.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Terry Lambert writes: > > > <grumble....> > > > > > > Maybe it could if it were hooked to a PCMCIA card. 8-). > > > > If it were hooked to a PCMCIA card, the kernel would realize it *was* > > ejected *AFTER* you physically removed the media which means it would be > > tool late to update the dirty buffers. The only thing the PCIC > > controllers buys you is the ability to see insertion/removal request > > plus the ability to setup resource allocation a bit differently. > > It can also scream "Put the card back! Put the card back!". Not without an incredible amount of kludges to *every* device driver. Currently, the PCCARD code knows *nothing* about the individual device drivers, and shouldn't. Adding intrinsic device driver knowledge to the generic support code is something that I'll fight against. Nate
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199701132114.OAA26992>