sane-project-website/old-archive/1999-02/0079.html

<!-- received="Mon Feb 15 08:23:58 1999 PST" -->
<!-- sent="Mon, 15 Feb 1999 16:22:21 +0000 (GMT)" -->
<!-- name="H C Pumphrey" -->
<!-- email="hcp@met.ed.ac.uk" -->
<!-- subject="AT12 / DTC SCSI" -->
<!-- id="" -->
<!-- inreplyto="" -->
<title>sane-devel: AT12 / DTC SCSI</title>
<h1>AT12 / DTC SCSI</h1>
<b>H C Pumphrey</b> (<a href="mailto:hcp@met.ed.ac.uk"><i>hcp@met.ed.ac.uk</i></a>)<br>
<i>Mon, 15 Feb 1999 16:22:21 +0000 (GMT)</i>
<p>
<ul>
<li> <b>Messages sorted by:</b> <a href="date.html#79">[ date ]</a><a href="index.html#79">[ thread ]</a><a href="subject.html#79">[ subject ]</a><a href="author.html#79">[ author ]</a>
<!-- next="start" -->
<li> <b>Next message:</b> <a href="0080.html">Petter Reinholdtsen: "Re: xsane 0.11 available"</a>
<li> <b>Previous message:</b> <a href="0078.html">Robert J. Alexander: "Re: Please help me with unrecogn. scanner (Microtek 630)"</a>
<!-- nextthread="start" -->
<!-- reply="end" -->
</ul>
<!-- body="start" -->
Dear Chris / Sane:<br>
<p>
A while ago I asked the sane-devel mailing list if it were possible to<br>
get the Artec AT12 scanner working with Linux and SANE using the SCSI card<br>
supplied with the scanner. I had some helpful advice (for which many<br>
thanks) but I could never get the kernel to see the card. I have made<br>
some progress this weekend and am reporting it to you and sane-devel<br>
in case it is of use or interest.<br>
<p>
 <br>
The SCSI card in question is a nasty piece of work called a<br>
DTC3181X/3151X and the chip on it is called a DTCT436P. There are only<br>
two jumpers on it, J1 can be set to OFF=PNP or ON=PNR and is set to<br>
ON=PNR as the card is supplied. J2 can be set to ON=0WS and OFF=1WS<br>
(set to off as supplied). I have left these alone as I have not the<br>
faintest idea what they do.<br>
<p>
Up to now I have always used Linux kernel 2.0.34 because that is what<br>
you get with Debian 2.0. I found that you can either do without the<br>
SCSI support or have the kernel hang when you try to boot it. <br>
<p>
This weekend, just for a laugh, I tried out the brand, spanking new<br>
2.2.1 kernel. I compiled it with the generic NCR5380 scsi support and<br>
used the boot prompt argument dtc3181e=0x2c0,0 .  To my astonishment,<br>
on boot-up you see this:<br>
<p>
scsi0 : IRQ0 not free, interrupts disabled<br>
scsi0 : interrupts not enabled. for better interactive performance,<br>
scsi0 : please jumper the board for a free IRQ.<br>
scsi0 : at port 0x2c0 interrupts disabled options CAN_QUEUE=16<br>
CMD_PER_LUN=2 release=1 generic options AUTOPROBE_IRQ AUTOSENSE<br>
USLEEP, USLEEP_POLL=1 USLEEP_SLEEP=20 generic release=7<br>
scsi0 : Generic NCR5380/53C400 Driver<br>
scsi : 1 host.<br>
  Vendor: ULTIMA    Model: AT12              Rev: 2.10<br>
  Type:   Scanner                            ANSI SCSI revision: 02<br>
Detected scsi generic sga at scsi0, channel 0, id 5, lun 0<br>
<p>
Whoopee. Now we can see the scanner OK. (Well, sometimes. Sometimes it<br>
hangs at the line "scsi : 1 host".) The scanner refused grimly to work<br>
in sane 0.72 as supplied with Debian, so I pulled down the Sane 1.0<br>
source and built it, along with your brand-new artec driver. I can now<br>
boot with the scanner on and do exactly one faultless scan with<br>
xscanimage before the scanner hangs up and refuses to do anything<br>
more. The scan mechanism does return to the usual park position after<br>
the one good scan. This sounds very like the symptoms reporeted a few<br>
weeks ago by Francisco Rodrigo Escobedo Robles (<a href="mailto:frer@vnet.es">frer@vnet.es</a>) using a<br>
NCR53C810 card.<br>
<p>
I do <br>
<p>
export SANE_DEBUG_ARTEC=128<br>
export SANE_DEBUG_SANEI_SCSI=128<br>
<p>
before scanning and these were the last few lines of output in one<br>
case, although they don't seem repeatable.<br>
<p>
[sanei_scsi] scsi_req_enter: entered 0x808eaf8<br>
[sanei_scsi] sanei_scsi_req_wait: waiting for 0x808eaf8<br>
[sanei_scsi] sanei_scsi.issue: 0x808eaf8<br>
[sanei_scsi] sanei_scsi_req_wait: read 36 bytes<br>
[sanei_scsi] sanei_scsi_req_wait: SCSI command complained: Success<br>
[artec] sense:  8031 INT MEM R/W ERROR<br>
[artec] sense:  EEPROM R/W ERROR<br>
[artec] sense:  ASIC TEST ERROR<br>
[artec] sense:  LINE RAM R/W ERROR<br>
[artec] sense:  PSRAM R/W TEST ERROR<br>
[artec] sense:  POSITIONING ERROR<br>
[artec] sense:  TEST 6 ERROR<br>
[artec] sense:  TEST 7 ERROR<br>
[artec] sense:  TEST 8 ERROR<br>
[artec] sense:  TEST 9 ERROR<br>
[artec] sense:  TEST 10 ERROR<br>
[artec] sense:  TEST 11 ERROR<br>
[artec] sense:  TEST 12 ERROR<br>
[artec] sense:  TEST 13 ERROR<br>
[artec] sense:  TEST 14 ERROR<br>
[artec] sense:  TEST 15 ERROR<br>
[artec] wait_ready: 'Error during device I/O'<br>
[artec] wait for scanner ready failed: Error during device I/O<br>
[artec] sane_start<br>
[sanei_init_debug]: Setting debug level of sanei_scsi to 128.<br>
<p>
&lt;I try to start the second scan here, but nothing happens. Kill<br>
xscanimage and restart it and it won't scan, you just get this:&gt;<br>
<p>
orac:~# xscanimage artec:/dev/scanner<br>
[sanei_init_debug]: Setting debug level of artec to 128.<br>
[artec] Artec/Ultima backend version 0.4, last mod: 12/9/1998 01:11<br>
[artec] <a href="http://www4.infi.net/~cpinkham/sane-artec-doc.html">http://www4.infi.net/~cpinkham/sane-artec-doc.html</a><br>
[artec] artec.conf line: '/dev/scanner', len = 12<br>
[artec] attach_one<br>
[artec] attach: opening /dev/scanner<br>
[sanei_init_debug]: Setting debug level of sanei_scsi to 128.<br>
[sanei_scsi] sanei_scsi_open: sanei_scsi_max_request_size=32768 bytes<br>
[artec] attach: sending INQUIRY<br>
[sanei_scsi] scsi_req_enter: entered 0x808eaf8<br>
[sanei_scsi] sanei_scsi_req_wait: waiting for 0x808eaf8<br>
[sanei_scsi] sanei_scsi.issue: 0x808eaf8<br>
[sanei_scsi] sanei_scsi_req_wait: read 132 bytes<br>
[artec] attach: wait for scanner to come ready<br>
[artec] wait_ready<br>
[sanei_scsi] scsi_req_enter: entered 0x808eaf8<br>
[sanei_scsi] sanei_scsi_req_wait: waiting for 0x808eaf8<br>
[sanei_scsi] sanei_scsi.issue: 0x808eaf8<br>
<p>
** ERROR **: sigint caught<br>
<p>
Something similar happens if I use scanimage instead of xscanimage. If I do <br>
<p>
prompt&gt; scanimage -d artec:/dev/scanner &gt; smeg.ppm<br>
<p>
I get a full a4 (letter) size image. If I then do <br>
<p>
prompt&gt; scanimage -d artec:/dev/scanner &gt; smeg2.ppm<br>
<p>
.. I get nothing. <br>
<p>
Strange!<br>
<p>
<p>
I'll keep you posted if I have any further success.<br>
<p>
Many thanks for your efforts, <br>
 <br>
<p>
All the best<br>
<p>
Hugh <br>
<p>
<p>
==========================================================================<br>
Hugh C. Pumphrey, Dept. of -    | Tel. 0131-650-6026,Fax:0131-650-5780<br>
Meteorology, Univ. of Edinburgh | Replace 0131 with +44-131 if outside U.K<br>
EDINBURGH EH9 3JZ, Scotland     | Email <a href="mailto:hcp@met.ed.ac.uk">hcp@met.ed.ac.uk</a><br>
======P=l=e=a=s=e==N=o=t=e==t=h=e==N=e=w==F=A=X==N=u=m=b=e=r==============<br>
<p>
<p>
<p>
<p>
<pre>
--
Source code, list archive, and docs: <a href="http://www.mostang.com/sane/">http://www.mostang.com/sane/</a>
To unsubscribe: echo unsubscribe sane-devel | mail <a href="mailto:majordomo@mostang.com">majordomo@mostang.com</a>
</pre>
<!-- body="end" -->
<p>
<ul>
<!-- next="start" -->
<li> <b>Next message:</b> <a href="0080.html">Petter Reinholdtsen: "Re: xsane 0.11 available"</a>
<li> <b>Previous message:</b> <a href="0078.html">Robert J. Alexander: "Re: Please help me with unrecogn. scanner (Microtek 630)"</a>
<!-- nextthread="start" -->
<!-- reply="end" -->
</ul>