Konica Minolta PagePro 1300W USB Printing problem

When I turn on my printer, dmesg says:
[4295953.738000] usb 2-2: new full speed USB device using uhci_hcd and address 4
[4295953.871000] drivers/usb/class/usblp.c: usblp0: USB Bidirectional printer dev 4 if 0 alt 0 proto 2 vid 0x0686 pid 0x300C

When I try to print a test page, /var/log/cups/error_log says:
I [01/Apr/2006:09:25:49 +0300] Adding start banner page "none" to job 4.
I [01/Apr/2006:09:25:49 +0300] Adding end banner page "none" to job 4.
I [01/Apr/2006:09:25:49 +0300] Job 4 queued on "PagePro-1300W" by "vyruss".
I [01/Apr/2006:09:25:49 +0300] Started filter /usr/lib/cups/filter/pstops (PID 8711) for job 4.
I [01/Apr/2006:09:25:49 +0300] Started filter /usr/lib/cups/filter/foomatic-rip (PID 8712) for job 4.
I [01/Apr/2006:09:25:49 +0300] Started backend /usr/lib/cups/backend/usb (PID 8713) for job 4.
W [01/Apr/2006:09:25:49 +0300] [Job 4] Printer off-line.

and it continues saying "Printer off-line" forever.

I saw a few days ago where a similar problem was fixed in cups. Moving there since they may be able to better debug this.

please recheck, after a dist-upgrade to dapper 20060414, removal and resintallation of the printer

No, I'm afraid it still doesn't work. What's strange is that parallel printing with the same printer works! That's why I originally filed this as a kernel bug...

After upgrading from breezy to dapper I run into similar problems with my hp 1022 usb printer. I'm not sure whether I had tracked down this bug completely but finally I created a working patch for cups.

/var/log/cups/error.log says:

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
D [26/May/2006:20:56:35 +0200] Get-Jobs ipp://localhost.localdomain:631/printers/HP1022
D [26/May/2006:20:56:35 +0200] cupsdProcessIPPRequest: 5 status_code=0 (successful-ok)
D [26/May/2006:20:56:35 +0200] cupsdCloseClient: 5
D [26/May/2006:20:56:38 +0200] [Job 71] LPGETSTATUS returned a port status of 00...
W [26/May/2006:20:56:38 +0200] [Job 71] Printer fault!
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

this is the code snipped from backend/usb-unix.c that generates these logging output:

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>
  while (!ioctl(fd, LPGETSTATUS, &status))
  {
    fprintf(stderr, "DEBUG: LPGETSTATUS returned a port status of %02X...\n", status);

    if (status & LP_POUTPA)
    {
      fputs("WARNING: Media tray empty!\n", stderr);
      fputs("STATUS: +media-tray-empty-error\n", stderr);

      paperout = 1;
    }

    if (!(status & LP_PERRORP))
      fputs("WARNING: Printer fault!\n", stderr);
    else if (!(status & LP_PSELECD))
      fputs("WARNING: Printer off-line.\n", stderr);
    else
      break;

    sleep(5);
  }
<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

first !ioctl(fd, LPGETSTATUS, &status) is ever evaluated to true (even if status is 0). Maybe this is a side effect from a bug in the kernel. Second I really don't know why the break statement is never reached, so i applied following (ugly) patch that enabled printing for me:

--- cupsys-1.2.0/backend/usb-unix.c.orig 2006-05-26 22:08:53.000000000 +0200
+++ cupsys-1.2.0/backend/usb-unix.c 2006-05-26 22:09:43.000000000 +0200
@@ -158,7 +158,7 @@
   * Show the printer status before we send the file...
   */

- while (!ioctl(fd, LPGETSTATUS, &status))
+ while (!ioctl(fd, LPGETSTATUS, &status) && status != 0)
   {
     fprintf(stderr, "DEBUG: LPGETSTATUS returned a port status of %02X...\n", status);

Older CUPS versions didn't do a status check. That's why the printer did work previously. Unfortunately it seems Minolta printers themselves aren't 100% class compliant (-> general Minolta problem).

Oliver, thanks for your investigations. Indeed the break won't be reached, since !(0 && something) is always true.

However, from reading lp.h, it says

  #define LP_PERRORP 0x08 /* unchanged input, active low */

so if that bit is zero, it means 'yes', and thus the printer reports an error condition. I'll talk with upstream about this.

Forwarded upstream: http://www.cups.org/str.php?L1773

After updating to cupsys 1.2.2-0ubuntu0.6.06 printing works fine for me. Thanks.

Thanks for checking again!

The printer finally works without any workarounds in 9.04 (Jaunty).