Re: pcmcia scsi

[Christos Zoulas]

In article <963911856.3119867.1501512841...@mail.yahoo.com>,
Björn Johannesson   wrote:
>Hello.
>
>I found an old Adaptec pcmcia scsi adapter in a box. It uses the aic(4)
>driver but
>it will not attach during boot/insertion it complains that it is "unable
>to detect chip!".
>
>I found this oneliner patch in the archives:
>http://mail-index.netbsd.org/netbsd-bugs/2005/12/04/0013.html
>This patch makes it attach and work.
>
>aic1 at pcmcia1 function 0: Version 0.01>
>scsibus0 at aic1: 8 targets, 8 luns per target
>scsibus0: waiting 2 seconds for devices to settle...
>sd0 at scsibus0 target 4 lun 0:  disk fixed
>
>Above is the external scsi-box for my DECstation connected to an i386 laptop.
>It mounts and works fine.
>
>Now, whose leg do I need to (figuratively) hump to get this committed?

I committed it but I still don't think it is right...

christos

Re: pcmcia scsi

[Michael van Elst]

rherdw...@yahoo.com (=?UTF-8?Q?Bj=C3=B6rn_Johannesson?=) writes:

>pcmcia1: function 0, config table entry 9: I/O card; irq mask 1e00; iomask a, 
>iospace 340-35f; io16 irqlevel
>pcmcia1: function 0, config table entry 8: I/O card; irq mask 1e00; iomask a, 
>iospace 140-15f; io16 irqlevel

Strange. The card requests an alignment of 0xA (== 1024) and the driver
cannot access the chip. But when you fake an alignment of only 32 (the size
of the iospace requested), then it works.

I could imagine that the address allocation fails because of the high
alignment requirement. But then you should get a different failure.


-- 
-- 
Michael van Elst
Internet: mlel...@serpens.de
"A potential Snark may lurk in every tree."

Re: pcmcia scsi

[Swift Griggs]

On Mon, 31 Jul 2017, Bj?rn Johannesson wrote:

pcmcia1: CIS version PCMCIA 2.0 or 2.1
pcmcia1: CIS info: Adaptec, Inc., APA-1460 SCSI Host Adapter, Version 0.01


I have one of these, too. I got it to use with my Amiga 1200. I wonder if 
NetBSD supports it on that platform.


If someone has experience, let me know. I'll probably just try and find 
out!


-Swift

Re: pcmcia scsi

[Björn Johannesson]

Hello.


On Mon, 7/31/17, J. Hannken-Illjes  wrote:

 >Looks like this should become a quirk.  Could you try an unmodified
 >kernel with option PCMCIAVERBOSE and post the dmesg?
 
Apparently it did not like me pasting the entire dmesg in the mail so I'll try 
again.
Snippet with that card follows:
pcmcia1: CIS version PCMCIA 2.0 or 2.1
pcmcia1: CIS info: Adaptec, Inc., APA-1460 SCSI Host Adapter, Version 0.01
pcmcia1: Manufacturer code 0x12f, product 0x1
pcmcia1: function 0: unspecified, ccr addr 2000 mask 1
pcmcia1: function 0, config table entry 9: I/O card; irq mask 1e00; iomask a, 
iospace 340-35f; io16 irqlevel
pcmcia1: function 0, config table entry 8: I/O card; irq mask 1e00; iomask a, 
iospace 140-15f; io16 irqlevel
aic1 at pcmcia1 function 0: 
aic1: unable to detect chip!

The entire dmesg can be found at https://pastebin.com/UEuTW7pb

/B

Re: pcmcia scsi

[J. Hannken-Illjes]

> On 31. Jul 2017, at 16:54, Björn Johannesson  wrote:
> 
> Hello.
> 
> I found an old Adaptec pcmcia scsi adapter in a box. It uses the aic(4) 
> driver but
> it will not attach during boot/insertion it complains that it is "unable to 
> detect chip!".
> 
> I found this oneliner patch in the archives:
> http://mail-index.netbsd.org/netbsd-bugs/2005/12/04/0013.html
> This patch makes it attach and work.
> 
> aic1 at pcmcia1 function 0:  Version 0.01>
> scsibus0 at aic1: 8 targets, 8 luns per target
> scsibus0: waiting 2 seconds for devices to settle...
> sd0 at scsibus0 target 4 lun 0:  disk fixed
> 
> Above is the external scsi-box for my DECstation connected to an i386 laptop.
> It mounts and works fine.
> 
> Now, whose leg do I need to (figuratively) hump to get this committed?

Looks like this should become a quirk.  Could you try an unmodified
kernel with option PCMCIAVERBOSE and post the dmesg?

--
J. Hannken-Illjes - hann...@eis.cs.tu-bs.de - TU Braunschweig (Germany)