Date: Wed, 6 Oct 2004 04:28:34 -0700 From: Bruce M Simpson <bms@spc.org> To: freebsd-current@FreeBSD.org Subject: Re: Help wanted re formatting USB floppies. Message-ID: <20041006112834.GC1794@empiric.icir.org> In-Reply-To: <20041006092156.GA690@empiric.icir.org> References: <20041006092156.GA690@empiric.icir.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--L2Brqb15TUChFOBK
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
I realize this is a bit of a long post, but I believe I've found the answer.
Those of you who want to format a USB floppy on FreeBSD right at this
moment may have luck with the following uber sekret combination:
camcontrol cmd ${DEVNAME} \
-c "04 17 00 00 00 00 00 00 00 00 00 00" \
-o 12 "00 A0 00 08 00 00 0B 40 00 00 02 00"
This roughly means 'format everything on this disk with 2880 512-byte block=
s'
in USB-floppy-ese; where ${DEVNAME} is the device name of your USB floppy.
For those of you who were always told to show your working in school,
=66rom my notes:
%%%
- Format track 0 with USB-FDU defaults.
Looks like it's hanging. But there's additional data needed! It needs a=20
Format Parameter List
on the Bulk endpoint (!!).
Needs a parameter list as follows:
4-byte Defect List Header:
00 reserved
FOV | (0) | DCRT | (0) | (0) | (0) | (0) | (0) --> A0
A0 format options valid, disable certification, not immediate, all tracks.
00
08 defect list length (must always be 8 bytes, length of the format descrip=
tor
which follows)
8-byte Format Descriptor:
00 00 0B 40 -> 2880 blocks
00 reserved
00 02 00 -> 512 byte blocks
these values come from the format descriptor:
empiric# camcontrol cmd da0 -v -v -c "23 00 00 00 00 00 00 00 20 00 00 00" =
-i 0x
20 "{} *i3 {Len} i1 {Blocks} i4 {} *b6 {Code} b2 {Blocklen} i3"
32 2880 2 512
%%%
This is not an 'exact science' yet. A more user friendly USB/SCSI/ATAPI
removable disk formatting program should be forthcoming soon. A track by
track format is what I believe Windows does, though I haven't done any
USB reverse engineering to confirm this. newfs_msdos(8) seems to like it:
empiric# newfs_msdos da0 auto
/dev/da0: 2840 sectors in 355 FAT12 clusters (4096 bytes/cluster)
bps=3D512 spc=3D8 res=3D1 nft=3D2 rde=3D512 sec=3D2880 mid=3D0xf0 spf=3D2 s=
pt=3D32 hds=3D64 hid=3D0
Hex dumps using vim's xxd(1) of /dev/da0 on my system seem to confirm that
the device did indeed format properly.
Blimey, all that for something ordinary people do every day on Windows
machines without a second thought.
Have fun, I'm off to bed.
BMS
--L2Brqb15TUChFOBK
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Comment: ''
iD8DBQFBY9bhueUpAYYNtTsRAuGzAJ0apRznVxiPLa2RwMtp/cX0u2xPmACePoXN
oXDTut1C532y29X0PqvJPGs=
=LqYc
-----END PGP SIGNATURE-----
--L2Brqb15TUChFOBK--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041006112834.GC1794>