From mboxrd@z Thu Jan 1 00:00:00 1970 Date: Sun, 10 Dec 1995 21:44:19 -0500 From: Luther Huffman lutherh@mail.infinet.com Subject: Adding pcdist to CD-ROM file system? Topicbox-Message-UUID: 382bedec-eac8-11e9-9e20-41e7f4b1d025 Message-ID: <19951211024419.cKbUnRc7qAksoNIHXShQ1bE0lnzA1ivI9ry3dItF0Tk@z> Aha! Thanks. I'd been wondering about why it was named atapi2. I was expecting atapi1 or even atapi0. > From: "jim mckie " > To: 9fans@cse.psu.edu > Date: Sun, 10 Dec 1995 10:49:57 -0500 > Subject: re: Adding pcdist to CD-ROM file system? > Reply-to: 9fans@cse.psu.edu > The slave/master jumper may have been a factor. Early ATA 'standards' (for > want of a better word) forbade a slave without a master. The later ones say > it's possible but do not sanction it. The way the ATA driver currently looks > for drives is to use the 'execute device diagnostics' command then look at the > status for the master and then the slave. Because of problems with some > controllers i found it not advisable to try look at the status of the slave if > the master doesn't exist. > > The mapping between ATA controller number, port and drive number is fixed: > primary port 0x3F0 drives 0, 1 > secondary port 0x1F0 drives 2, 3 > tertiary port 0x1E8 drives 4, 5 > quaternary port 0x168 drives 6, 7 > This means there can be 'holes' in the drive name space. > Although it is part of the ATA driver, ATAPI drives are named differently > to not confuse them with hard drives. For example, with one hard drive on > the primary and a CD-ROM on the secondary the drives are named > /dev/hd0disk > /dev/atapi2disk > > >