kopia lustrzana https://gitlab.com/sane-project/website
71 wiersze
3.4 KiB
HTML
71 wiersze
3.4 KiB
HTML
|
<!-- received="Thu Feb 26 22:56:52 1998 PST" -->
|
||
|
<!-- sent="Fri, 27 Feb 1998 00:27:26 +0100 (MET)" -->
|
||
|
<!-- name="Rogier Wolff" -->
|
||
|
<!-- email="R.E.Wolff@BitWizard.nl" -->
|
||
|
<!-- subject="Re: Mustek scanner: success!" -->
|
||
|
<!-- id="199802262327.AAA20281@cave.BitWizard.nl" -->
|
||
|
<!-- inreplyto="Pine.LNX.3.96.980226150215.1430A-100000@info04.uni-trier.de" -->
|
||
|
<title>sane-devel: Re: Mustek scanner: success!</title>
|
||
|
<h1>Re: Mustek scanner: success!</h1>
|
||
|
<b>Rogier Wolff</b> (<a href="mailto:R.E.Wolff@BitWizard.nl"><i>R.E.Wolff@BitWizard.nl</i></a>)<br>
|
||
|
<i>Fri, 27 Feb 1998 00:27:26 +0100 (MET)</i>
|
||
|
<p>
|
||
|
<ul>
|
||
|
<li> <b>Messages sorted by:</b> <a href="date.html#184">[ date ]</a><a href="index.html#184">[ thread ]</a><a href="subject.html#184">[ subject ]</a><a href="author.html#184">[ author ]</a>
|
||
|
<!-- next="start" -->
|
||
|
<li> <b>Next message:</b> <a href="0185.html">REINHARD.ARZBERGER@quelle.de: "Microtek 630"</a>
|
||
|
<li> <b>Previous message:</b> <a href="0183.html">Rogan Dawes: "Dynamic loading libraries for AIX"</a>
|
||
|
<!-- nextthread="start" -->
|
||
|
<!-- reply="end" -->
|
||
|
</ul>
|
||
|
<!-- body="start" -->
|
||
|
Norbert Mueller wrote:<br>
|
||
|
<i>> Just an idea: Does anybody know whether there is a possibility to change </i><br>
|
||
|
<i>> the fixed sized buffer in sg.c to a circular list of e.g. 20 buffers of </i><br>
|
||
|
<i>> size such that the scanner may write to the list while at the same time </i><br>
|
||
|
<i>> the application reads from the list? As usually most CPU's are capable to </i><br>
|
||
|
<i>> read the data faster than the scanner can deliver, the circular list would </i><br>
|
||
|
<i>> never be full and backtracking would no longer happen.</i><br>
|
||
|
<p>
|
||
|
Sounds "neat", but it isn't as simple as that.<br>
|
||
|
<p>
|
||
|
The SCSI low level drivers would then need to be able to do<br>
|
||
|
"scatter-gather". I think that the API isn't that smart. Also, you<br>
|
||
|
could get into trouble that way: You'd have to tell the scanner: "Ok,<br>
|
||
|
I want 1Mb of data". and then you have to trust the user-level<br>
|
||
|
application to actually gather and store that data before your buffer<br>
|
||
|
overflows. What do you do if the application suddenly goes comatose?<br>
|
||
|
<p>
|
||
|
Another problem is that you'd be telling the scanner to send a whole<br>
|
||
|
lot of data, where you don't know how long that's going to take. The<br>
|
||
|
lowlevel SCSI driver (e.g. the NCR810 one) might time out after say 10<br>
|
||
|
seconds (It assumes a command like a "read" will finish within about<br>
|
||
|
10 seconds: This is valid for most disks.... ;-).<br>
|
||
|
<p>
|
||
|
Regards,<br>
|
||
|
<p>
|
||
|
Roger. <br>
|
||
|
<p>
|
||
|
<p>
|
||
|
<pre>
|
||
|
--
|
||
|
If it's there and you can see it, it's REAL <a href="mailto:|___R.E.Wolff@BitWizard.nl">|___R.E.Wolff@BitWizard.nl</a> |
|
||
|
If it's there and you can't see it, it's TRANSPARENT | Tel: +31-15-2137555 |
|
||
|
If it's not there and you can see it, it's VIRTUAL |__FAX:_+31-15-2138217 |
|
||
|
If it's not there and you can't see it, it's GONE! -- Roy Wilks, 1983 |_____|
|
||
|
<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="0185.html">REINHARD.ARZBERGER@quelle.de: "Microtek 630"</a>
|
||
|
<li> <b>Previous message:</b> <a href="0183.html">Rogan Dawes: "Dynamic loading libraries for AIX"</a>
|
||
|
<!-- nextthread="start" -->
|
||
|
<!-- reply="end" -->
|
||
|
</ul>
|