fanj%remb6489.wpd.sgi.com@sgi.com (Fan Jiao 9U-510 x1615) (10/19/90)
Should use /dev/at0 instead of /dev/fd0. Fan Jiao(Chiao) x 1615 | M/S 9U-510
postmaster@ssci.UUCP (Kevin Benton) (10/21/90)
In an article of <18 Oct 90 13:38:28 GMT>, traw@grad1.cis.upenn.edu writes: >From: traw@grad1.cis.upenn.edu >Date: 18 Oct 90 13:38:28 GMT >Message-ID: <31333@netnews.upenn.edu> >Newsgroups: comp.os.minix > I am trying to use dd to copy a 1.2M minix binaries diskette (one of >the ones supplied with minix). I get an error telling me that minix >can't read the disk (then 0 blocks in, 0 blocks out). I used the >following command line: > >$ dd /dev/at0 diskfile bs=7680 # several other blocksizes tried No problem.... Try this... dd if=/dev/at0 of=/usr/tmp/disk.image bs=1024 count=1200 if= and of= were missing. Also, make sure you have /dev/at0 there, and that it is the same as /dev/fd0. I have had similar problems on SCO, but realized that I was using the "wrong" device (major, minor number). Technically speaking, the bs and count aren't necessary... Kevin -- Kevin Benton - SysOp of... Project Thunder (ssci) 1:157/516.0 :-) once in a while... FidoNet 1:157/533.14 you might actually learn UseNet uunet!ssci!kevin to like it!
evans@syd.dit.CSIRO.AU (Bruce.Evans) (10/25/90)
In article <16.27214A99@ssci.UUCP> postmaster@ssci.UUCP (Kevin Benton) writes: >In an article of <18 Oct 90 13:38:28 GMT>, traw@grad1.cis.upenn.edu writes: > > I am trying to use dd to copy a 1.2M minix binaries diskette (one of > >the ones supplied with minix). I get an error telling me that minix > >can't read the disk (then 0 blocks in, 0 blocks out). I used the > >following command line: > > > >$ dd /dev/at0 diskfile bs=7680 # several other blocksizes tried > >No problem.... Try this... > >dd if=/dev/at0 of=/usr/tmp/disk.image bs=1024 count=1200 Better, us bs=15360 (both sides of one track). The original has half this size, but that may not work so well since it uses an odd number of sectors, depending on how well FS hides the mismatch. -- Bruce Evans evans@syd.dit.csiro.au
snewton@ux.acs.umn.edu (steven robert newton) (10/26/90)
It seems the early decision to have the driver assume a standard format for a block disk driver (aka: /Single_sidedDisk /sd0 /Dbl_sidedDisk /dd0) is very good for `tar cf /dev/dd0` but lousy for 85 tracks&11 sectors. I like the idea (posted earlier) to have /dev/fd? sense the dos disk table for #blocks and #sectors/track. (#tracks can be computed). The big deal is that ALL BIOS's could then continue to support 3.5 or at disks with a block-zero read that all drives should be able to do. Requirements: Just write boot block code with a jump +x1c ;around this important disk param. table .db 22 ;a int 41h and device driver Disk Parameter Table Reference ;can document all the needed fields. .text ;begin your code.(Stop using x1fe for #img-size for std xaa55 sig!!) When I had only a Tandy 2000(argh) I noticed it used the boot-param-tbl to decide to double step a 80 track(720k) drive to read/write a 360K disk. So, I moved the code +x27 within the bootblock and changed the block move source to point the the new starting byte spot. (I know it's obscure.. I could only work with debug ) I did all this--I still could handle patching all the hardware port changes. Next I bought a Sanyo 675 xt and installed a Seagate ST124 (616cly 20Mb HD) with a Seagate ST11(mfm 8 bit) card. I get fs panics--it fails to rw-block(0 even!) Should I get a ESDI SCSI or RLL conroller with my next computer?? Thanks for MINIX and all your work in improving it!!!