From owner-freebsd-hackers Tue Jun 27 17:26:42 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id RAA10934 for hackers-outgoing; Tue, 27 Jun 1995 17:26:42 -0700 Received: from cs.weber.edu (cs.weber.edu [137.190.16.16]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id RAA10928 for ; Tue, 27 Jun 1995 17:26:41 -0700 Received: by cs.weber.edu (4.1/SMI-4.1.1) id AA04357; Tue, 27 Jun 95 18:19:37 MDT From: terry@cs.weber.edu (Terry Lambert) Message-Id: <9506280019.AA04357@cs.weber.edu> Subject: Re: 205R on Digital Hinote laptop, some experiences To: wilko@yedi.iaf.nl (Wilko Bulte) Date: Tue, 27 Jun 95 18:19:36 MDT Cc: FreeBSD-hackers@freebsd.org In-Reply-To: <199506272118.XAA00640@yedi.iaf.nl> from "Wilko Bulte" at Jun 27, 95 11:18:44 pm X-Mailer: ELM [version 2.4dev PL52] Sender: hackers-owner@freebsd.org Precedence: bulk > - the hinote has a US Robotics 28K8 PCMCIA modem. Of course I tried to make it > work. Things like Windoze, MSdos Diagnostics etc correctly identify a second > 16550A type comport (==the modem). FreeBSD does not, the probe fails. > > - the hinote also has a SCSI PCMCIA card. It was made by Adaptec and badge > engineered by Digital. I think it is a APA-1460, it uses ASPI2DOS (so > 1522 style) drivers with DOS. This aspi driver is for 6260/6360 type chips. > This made me attempt to use the aic driver. No luck though. The probe fails. > > The latter 2 try-outs might fail because I miss something on the PCMCIA > front. Do they need extra initialising to become active (???) or something > like that? > > Ideas and suggestions are welcome. The PCMCIA bridge chips are very similar to the PCI chips. Basically, they map the card IRQ/DMA/address ranges onto the (ISA in this case) bus. Once this is done, then the cards should function. Currently, the PCMCIA address range mapping mechanisms (generalized by an ENPIC type enabler) aren't generalized. The upshot of this is that only one of the five chipset types is supported at this time (the Intel -- by far the most popular), and the bridge enabling is done by specific drivers knowing about the address ranges instead of a general "card services" mechanism. The barrier to generalization is that the module would like to handle all of PCMCIA/PCI/EISA/MCA devices as bus attaches, and this code is pretty much not done (the PCI stuff is the closest thing FreeBSD has). I believe someone posted about having hacks for the USR modem to the list at one time. For all other cards, it's pretty much unsupported and not general. For now I would suggest sticking with only supported hardware. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.