[9fans] bitsy questions

[9fans] bitsy questions

[Fco.J.Ballesteros]

I started to play w/ the bitsy, in part to adapt the sdata thing,
which seems that no longer works in the 4th edition. So my first
question is, anyone already doing this?

The second question is that I'm getting what I think are
ramfs crashes. They are quite reproducible, if I open in acme
/ or /bin I usually get a series of "Bseek: unknown state" messages
and then the fs no longer works. Anyone saw the same thing?

thanks

PS: I saw the console putchar disabled, enabled it, and it seems to work
fine. Ditto for suspend (which resets the thing by default). Bug or
feature?

Re: [9fans] Bitsy questions

[lucio]

> Well I've got one in the cupboard behind me.  A 5 1/4" Full Height beast,
> 70 or 80 MB if I recall.  Now where is that controller card...
> 
How many controller cards do you need?  I can give you a few
choices...

++L

Re: [9fans] Bitsy questions

[Derek Fawcus]

On Mon, Mar 15, 2004 at 01:46:46AM -0800, Geoff Collyer wrote:
> Not too surprising; the PC has ~20 years of legacy cruft to be
> compatible with.  Our ata drivers would probably drive ST506 disks if
> any could be found.

Well I've got one in the cupboard behind me.  A 5 1/4" Full Height beast,
70 or 80 MB if I recall.  Now where is that controller card...

DF

Re: [9fans] Bitsy questions

[Geoff Collyer]

[-- Attachment #1: Type: text/plain, Size: 335 bytes --]

Not too surprising; the PC has ~20 years of legacy cruft to be
compatible with.  Our ata drivers would probably drive ST506 disks if
any could be found.

 From a fortune file:
Backward compatibility means never being able to say 'that was a mistake'.
If you build for compatibility with a donkey cart, you don't get a Concorde.

[-- Attachment #2: Type: message/rfc822, Size: 1918 bytes --]

From: Fco.J.Ballesteros <nemo@lsub.org>
To: 9fans@cse.psu.edu
Subject: Re: [9fans] Bitsy questions
Date: Mon, 15 Mar 2004 10:38:20 +0100
Message-ID: <cacf87ec40a9bf33f6d6a29246d7f501@plan9.escet.urjc.es>

> Is life easier on the bitsy wrt automatic choosing irq's?

I'm not responsible for any of the sdata drivers (I think jmk did
both), but I found the bitsy a lot cleaner than the PC.

Re: [9fans] Bitsy questions

[Fco.J.Ballesteros]

> Is life easier on the bitsy wrt automatic choosing irq's?

I'm not responsible for any of the sdata drivers (I think jmk did
both), but I found the bitsy a lot cleaner than the PC.

Re: [9fans] Bitsy questions

[Axel Belinfante]

[-- Attachment #1: Type: text/plain, Size: 1503 bytes --]

I asked:
> > Is the support for pcmcia disks only in the
> > ipaq (bitsy) kernel, or also in the pc one?

and jmk replied:

> there is some disabled code in pc/sdata.c which i put in just to see
> if it would work:

thanks! I have it working now.
took me (quite) some time to find working  numbers for port and irq.
I tried guessing them using info from aux/pcmcia, /dev/irqalloc and
/dev/ioalloc, I suppose I'm not smart at this.
Is there a better way?

I choose the first default configuration from aux/pcmcia --
naive question alert :-)   just curious:
Are there in general reasons to favor one over the other?
(I attach the output of aux/pcmcia at the end)

Somehow this worked more automatic on the bitsy, I recall.
Is life easier on the bitsy wrt automatic choosing irq's?
Or has more effort (default irq's) been put in for the bitsy?
(I know/think more effort has been put in there wrt the possibility
 of sending a configure command to - what was it? - pcm?ctl ?)

Rephrased/Related: just curious: are there intrincic
differences between the bitsy and the pc that make things
easier in one place and harder in another?

For sake of completeness, for my laptop the following
(inserted at the right place in sdata.c) made it work.
I really have no idea how generally useful the irq is.

	if(channel == -1) {
		isa.port = 0x200;		/* toshiba tecra 8000 */
		isa.irq = 10;			/* toshiba tecra 8000 */
		channel = pcmspecial("MK5002MPL", &isa);
	}


Axel.



[-- Attachment #2: mk5002mpl --]
[-- Type: text/plain , Size: 1316 bytes --]

device 2048 bytes of 100ns IO+MEM
version 4.1
	TOSHIBA
	MK5002MPL
configuration registers at (0)0x200 (1)0x202 (2)0x204 (3)0x206
configuration 0 (default)
	memory device,  Ready/Busy active Memory Wait required
	Vcc:  Vnominal 5000000uV Iavg 400mA Ipeak 500mA Ipowerdown 60mA
	memory address range 0x0 - 0x800
configuration 0
	Vcc:  Vnominal 3300000uV Ipeak 500mA
configuration 1 (default)
	I/O device,  Ready/Busy active
	Vcc:  Vnominal 5000000uV Iavg 400mA Ipeak 500mA Ipowerdown 60mA
	IO space 4 address lines selectable 8bit or 8&16bit access
	interrupts:level:pulse:shared, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
configuration 1
	Vcc:  Vnominal 3300000uV Ipeak 500mA
configuration 2 (default)
	I/O device,  Ready/Busy active
	Vcc:  Vnominal 5000000uV Iavg 400mA Ipeak 500mA Ipowerdown 60mA
	IO space 10 address lines selectable 8bit or 8&16bit access
		1f0 - 1f7
		3f6 - 3f7
	interrupts:level:pulse:shared, 14
configuration 2
	Vcc:  Vnominal 3300000uV Ipeak 500mA
configuration 3 (default)
	I/O device,  Ready/Busy active
	Vcc:  Vnominal 5000000uV Iavg 400mA Ipeak 500mA Ipowerdown 60mA
	IO space 10 address lines selectable 8bit or 8&16bit access
		170 - 177
		376 - 377
	interrupts:level:pulse:shared, 14
configuration 3
	Vcc:  Vnominal 3300000uV Ipeak 500mA