home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.linux
- Path: sparky!uunet!stanford.edu!agate!boulder!ophelia!drew
- From: drew@ophelia.cs.colorado.edu (Drew Eckhardt)
- Subject: Re: SCSI Controller
- Message-ID: <1992Dec14.041032.29121@colorado.edu>
- Keywords: scsi, st01
- Sender: news@colorado.edu (The Daily Planet)
- Nntp-Posting-Host: ophelia.cs.colorado.edu
- Organization: University of Colorado at Boulder
- References: <1ggs9uINNh2u@life.ai.mit.edu>
- Date: Mon, 14 Dec 1992 04:10:32 GMT
- Lines: 68
-
- In article <1ggs9uINNh2u@life.ai.mit.edu> jimc@hal.gnu.ai.mit.edu (James Carpenter) writes:
- >In article <1992Dec12.172120.23570@sun1.ruf.uni-freiburg.de> hartnegg@sun1.ruf.uni-freiburg.de (Klaus Hartnegg) writes:
- >
- >>FORGET THE ST01/02 !
- >>It just costs you a lot of time and money.
- >
- >>However:
- >>- the ST01 is not available any more in shops
- >>- sometimes it works, sometimes it doesn't
- >>- it depends on the bios version
- >>- it does not properly work with MY hard disk (Rodime 200 MB)
- >>- it is not (yet?) really supported by Linux (i.e. will be sloooow).
- >>- trying different controllers you will have to fdisk every time
- >> you switch controllers since they map the disk differently, even
- >> different ST01 controllers do it different (number of tracks and such).
- >>
- >>Of course you MIGHT be lucky and save a lot of money.
- >>But be warned that I WASTED a lot of money (and time!!) trying
- >>to find a cheaper solution. Linux apparently does not like cheap
- >>controllers. If there is a small problem, it does not work at all in Linux.
- >>Right now I have an St01 here that works most of the time under dos
- >>and it is being detected by Linux but then a few
- >>seconds later Linux CRASHES with a SCSI timeout message.
-
- The Seagate BIOS isn't interrupt driven - under DOS, with one
- process running at a time you can't do anything while you're
- waiting for disk I/O and it doesn't matter.
-
- The Linux seagate driver uses interrupts so that only the task waiting
- on the I/O has to sleep when the disk is seeking. For the driver
- to work, you *HAVE* to jumper the board for IRQ5 (as documented
- in the FAQ).
-
- Chances are that Klaus just isn't following directions.
-
- >Perfect timing. I was about ready to get a ST01 for myself. How many of you
- >are having problems with the ST01? I haven't seen too many complaints about
- >the ST01 but then again I just got back on the net not too long ago. But if
- >there is definitely serious problems with it PLEASE let me know. I would go
- >with the Adaptec but my budget is VERY tight.
-
- For most people, the Seagate boards work (albeit slowly). Data transfer
- should max out at 500K/sec with the current routines on a buffered
- drive, I get 120K/sec on an unbuffered drive formated with a 3:1 interleve,
- many people get 60K/sec (1 1K block per revolution on a 3600 RPM drive).
-
- Your mileage may vary, but chances are you'll be closest to the
- last figure.
-
- A few people have reported problems that I can't reproduce.
-
- >Any advice???
-
- If you can't afford an Adaptec 15/174x, but already have a disk and
- NEED a SCSI board, go for the seagate. The $20 investment isn't a lot
- to loose if you ever upgrade.
-
- If the drive is exclusively for Linux, you won't even have to repartition
- if you upgrade, as the Linux code treats SCSI devices as they should
- be treated, disks with blocks numbered from 0..N, and it only looks
- at the starting sector and length of partitions in the partition table.
-
-
-
- --
- Boycott AT&T for their absurd anti-BSDI lawsuit. | Drew Eckhardt
- Condemn Colorado for Amendment Two. | drew@cs.colorado.edu
- Use Linux, the fast, flexible, and free 386 unix |
-