From owner-freebsd-current Sun Sep 5 0:15:51 1999 Delivered-To: freebsd-current@freebsd.org Received: from gndrsh.dnsmgr.net (GndRsh.dnsmgr.net [198.145.92.4]) by hub.freebsd.org (Postfix) with ESMTP id 2179D14C93 for ; Sun, 5 Sep 1999 00:15:48 -0700 (PDT) (envelope-from freebsd@gndrsh.dnsmgr.net) Received: (from freebsd@localhost) by gndrsh.dnsmgr.net (8.9.3/8.9.3) id AAA95613; Sun, 5 Sep 1999 00:14:02 -0700 (PDT) (envelope-from freebsd) From: "Rodney W. Grimes" Message-Id: <199909050714.AAA95613@gndrsh.dnsmgr.net> Subject: Re: PNP ids missing in sio.c In-Reply-To: <199909050447.AAA26573@khavrinen.lcs.mit.edu> from Garrett Wollman at "Sep 5, 1999 00:47:29 am" To: wollman@khavrinen.lcs.mit.edu (Garrett Wollman) Date: Sun, 5 Sep 1999 00:14:01 -0700 (PDT) Cc: mike@smith.net.au (Mike Smith), gurney_j@resnet.uoregon.edu (John-Mark Gurney), freebsd-current@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL54 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG > < said: > > > The enumerator should assign these resources to a placeholder; I was > > thinking the nexus was as good an owner as any. If there's an > > "unknown" device that's probably even better. > > Some of them should be claimed by real devices -- for example, the > pseudo-i8237 ISA DMA controller should be claimed by the ISA bus > (which I don't think it does now, unless someone added the code to do > it while I wasn't looking). Similarly the PIC, the PIT, the RTC, and > other random bits of ``Industry Standard'' hardware. It might even be > worth having these be their own unique devices, just to help diagnosis > if they ever go away... Yes please please please. Those of us doing strange stuff with strange machines would be eternally greatful if there were device nodes for PIC's, PIT's, RTC's and ``other random bits''. Note that though the ISA standard only has 1 of each of these in them, some really crazy hardware guys like me know all to well how to add anyone of them to an existing design. I have had x86 boxen with 4 PIC's in them and had to hand glue the interrupt code togeather to deal with it. I don't really care if the interrupt code is ever tought about this stuff, but at least having the device code know about these devices and report there existence goes a long way to dealing with these strange systems. Multi PIT's are extreamly useful, and I think there have already been some hooks added by either bde or phk to deal with a real hardware stat clock, having a real PIT driver would make a $10.00 add in PIT card possible for FreeBSD. Also note that some of these devices are already special cased and actually have multiple devices inside of them, infact the PIC and DMA are really 2 physical devices in the original design, just the cascade line is common. rtc0 at 0x040 on isa rtc1 at 0x048 on isa pic0 at 0x020 on isa pic1 at 0x0a0 on isa dmapg0 at 0x080 on isa dma0 at 0x000 on isa dma1 at 0x0c0 on isa -- Rod Grimes - KD7CAX - (RWG25) rgrimes@gndrsh.dnsmgr.net To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message