Re: tlug: Re:MO & SCSI Logitec LMO-640E / Fujitsu + Mouse Problem

[Karl-Max Wagner <karlmax@example.com>]

> Question 1:
> 
> I used these options in make config:
> 
> NCR53C8XX SCSI support
> (CONFIG_SCSI_NCR53C8XX) [Y/m/n/?]
>  detect and read serial NVRAMs (CONFIG_SCSI_NCR53C8XX_NVRAM_DETECT) [Y/n/?]
> 
>   enable tagged command queueing (CONFIG_SCSI_NCR53C8XX_TAGGED_QUEUE)
> [Y/n/?]
>   use normal IO
> (CONFIG_SCSI_NCR53C8XX_IOMAPPED) [Y/n/?]
>   maximum number of queued commands
> (CONFIG_SCSI_NCR53C8XX_MAX_TAGS) [4]
> synchronous transfers frequency in MHz
> (CONFIG_SCSI_NCR53C8XX_SYNC) [5]
>  assume boards are SYMBIOS compatible
> (CONFIG_SCSI_NCR53C8XX_SYMBIOS_COMPAT) [Y/n/?]
OK.
> kernel:
> Serial driver version 4.13 with no serial options enabled
> tty00 at 0x03f8 (irq = 4) is a 16550A
>  tty01 at 0x02f8 (irq = 3) is a 16550A
> lp1 at 0x0378, (polling)
> PS/2 auxiliary pointing device detected -- driver installed.
> Floppy drive(s): fd0 is 1.44M
> FDC 0 is a National Semiconductor PC87306
> ncr53c8xx: at PCI bus 0, device 13, function 0
> ncr53c8xx: 53c875 detected with Symbios NVRAM
> ncr53c875-0: rev=0x03, base=0xfffbff00, io_port=0xfc00, irq=9
> ncr53c875-0: Symbios format NVRAM, ID 7, Fast-20, Parity Checking
> ncr53c875-0: on-board RAM at 0xfffbe000
> ncr53c875-0: restart (scsi reset). kernel: ncr53c875-0-<5,0>:
> FAST-10 SCSI 10.0 MB/s (100 ns, offset 15)
This was the detection of the adapter itself ! Detection of
attached devices comes afterwards.
> to mount /dev/sda1 :
> mount -t vfat /dev/sda1 /mo
Not good. Under DOS and Windows the partition name very often is
"/dev/sda4". Nobody knows why, but weirdness is the M$ way....

Better do a "fdisk /dev/sda" and when you are inside the program
a "p" for print. You get a partition list then. If not, this is
your great chance to install an ext2 partition now and write it
to disk.

You quit with "q" ( no writing ) or with "w" ( with writing ).

If fdisk barfs your MO wasn't recognized. You probably have to
do some kernel massaging then ( I think there's the issue of non
standard blocksizes or so. Not a big deal, anyway ).

There are some low level SCSI tools for Linux but for using them
you better know really well what you are up to or you run into
serious trouble. Low level manipulation of SCSI devices requires
in depth knowledge of the underlying hardware.

> These are the make config settings
> (My mouse is a three button PS/2 type (psaux) according to my working
> 2.0.0)
>  *
> * Character devices
> *
> Standard/generic serial support (CONFIG_SERIAL) [Y/m/n/?]
OK.
> Digiboard PC/Xx Support (CONFIG_DIGI) [N/y/?]
OK.
> Cyclades async mux support (CONFIG_CYCLADES) [N/y/m/?]
OK.
> Stallion multiport serial support (CONFIG_STALDRV) [N/y/?]
OK.
> SDL RISCom/8 card support (CONFIG_RISCOM8) [N/y/m/?]
OK.
> Parallel printer support (CONFIG_PRINTER) [Y/m/n/?]
OK.
> Specialix IO8+ card support (CONFIG_SPECIALIX) [N/y/m/?]
Not required.
> Mouse Support (not serial mice) (CONFIG_MOUSE) [Y/n/?]
> ATIXL busmouse support (CONFIG_ATIXL_BUSMOUSE) [Y/m/n/?]
Not required.
> Logitech busmouse support (CONFIG_BUSMOUSE) [Y/m/n/?]
> Microsoft busmouse support (CONFIG_MS_BUSMOUSE) [Y/m/n/?]
Not required.
> PS/2 mouse (aka "auxiliary device") support (CONFIG_PSMOUSE) [Y/m/n/?]
> C&T 82C710 mouse port support (as on TI Travelmate) (CONFIG_82C710_MOUSE)
> [Y/n/?]
Oh no ! Hands off that !!!!! Use it ONLY if really required or
you're in for trouble !!!!!

Better: compile all the above stuff as a module. The kernel will
load everything it needs automatically. I personally only
compile the stuff in required to get the machine up, everything
else is a module - including the serial and paralell port
drivers, all mouse drivers and all file systems not required for
booting ( i.e. all except ext2 ). This provides you always with
a clean setup without any unnecessary clutter in the kernel: it
is not a secret that some parts interfere with each other and
that the ensuing kernel may suffer from all sorts of weird
behaviour.

Another problem: With some installations with PS/2 mice X gets
into some weird sort of competition with gpm when starting up.
This may be your problem, too. Try killing gpm before starting
up X. You can also put a line doing that into the "startx"
script and restart it at the end of it ( it only gets there when
the X server terminates ).

I had this kind of problem with the Apricot Qi 300 and Qi 600
machines. These are MCA machines and MCA machines aren't exactly
brilliant.....

                                  Karl-Max Wagner
                                  karlmax@example.com
--------------------------------------------------------------
Next Meeting: 8 August, Tokyo Station Yaesu central gate 12:30
featuring Linux on multiple platforms:
i386, Sparc, PA-Risc, Amiga, SGI, Alpha, PalmPilot, ...
Next Nomikai: 18 September, 19:30 Tengu TokyoEkiMae 03-3275-3691
--------------------------------------------------------------
Sponsor: PHT, makers of TurboLinux http://www.pht.co.jp