From: Richard Maine (maine@altair.dfrf.nasa.gov)
Date: 05/21/92


From: maine@altair.dfrf.nasa.gov (Richard Maine)
Subject: Re: Probleems with Linux-0.96 and 5.25" 360k diskette
Date: Thu, 21 May 1992 16:16:17 GMT

On 21 May 92 08:30:34 GMT, almesber@nessie.cs.id.ethz.ch (Werner Almesberger) said:
Werner> In article <1992May21.032746.8899@athena.mit.edu> Epstein@DOCKMASTER.NCSC.MIL writes:
> Prior to 0.96 could use 5.25" 360 diskette as B: with new auto detect
> fd0 and fd1 mknod /dev/fd0 b 2 0 mknod /dev/fd1 b 2 1 mknod /dev/at0 b 2
> 8 mknod /dev/fd0-360 b 2 20 mknod /dev/fd1-360 b 2 5

Werner> The fd1-360 device shouldn't depend on the CMOS setup, however. I
Werner> don't know what might cause it to fail with the new driver. Does
Werner> it also happen if you don't use mtools to access the drive ? Is
Werner> anybody experiencing similar problems ?

Yep. Me too. I hadn't mentioned it because I was too busy to track
it down in detail and I almost never use the 360k under linux anyway.
But now that you mention it...

I have a 1.44mb drive A and a 360k drive B. The floppy driver reports
it correctly on boot. Drive A works fine (if a bit slow - it takes
forever it seems to write a couple hundred k file to the drive with mtools).
By the way, I love the new autodetect for drive A.

Drive B doesn't seem to work at all. I've tried both autodetect with
  mknod /dev/fd1 b 2 1
and with the old specific 360k density
  mknod /dev/fd1 b 2 5 (I think it was 5; my notes are at home).
Same results. My mtools line is
  B /dev/fd1 12 40 2 9

I'm fairly sure (not 100%, though) that I tried reading
from the 360k with tar and got the same error. I can try again tonight
if you really want to know for sure.

Werner> Hmm ... is this new with 0.96 or did it do that with earlier versions
Werner> too ?

This is new to 0.96. Worked fine with 0.12 and one of the 0.95 reeases.