From what C=E9sar told me about a patch for the USB driver, I suspect
  that the problem is that the CUPS 1.2 USB driver expects to read from
  the printer.

I was confused - the mail about the patch was from Yorick Hardy, not
C=E9sar.

I've added a patch to disable the use of the back channel.  If you've
tried cups 1.2.x and failed to print on a USB printer, please update and
try again, and let me know if this helps.  I don't want to report this
upstream until we have more confidence that we understand what's going
on.



From: Greg Troxel <gdt@netbsd.org>
Subject: CVS commit: pkgsrc/print/cups
To: pkgsrc-changes@NetBSD.org
Date: Mon,  2 Apr 2007 12:11:01 +0000 (UTC)
Reply-To: gdt@netbsd.org


Module Name:	pkgsrc
Committed By:	gdt
Date:		Mon Apr  2 12:11:01 UTC 2007

Modified Files:
	pkgsrc/print/cups: Makefile distinfo
Added Files:
	pkgsrc/print/cups/patches: patch-au

Log Message:
Add patch to disable the use of the back channel for USB on NetBSD.
This is based on a suggestion by Yorick Hardy, who reports that it
improved behavior.  Without the patch, the cups usb driver tries to
read status from ulpt(4) (for most printers), and this results in no
output.


To generate a diff of this commit:
cvs rdiff -r1.120 -r1.121 pkgsrc/print/cups/Makefile
cvs rdiff -r1.48 -r1.49 pkgsrc/print/cups/distinfo
cvs rdiff -r0 -r1.6 pkgsrc/print/cups/patches/patch-au

Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.