My results are similiar to Brian's. Here is what I see:
I have 2.2.15 + usb-2.3.99-pre7-1-for-2.2.14.diff + Vojtech Pavlik
printer-unidir.diff version .4, and use usb-uhci. When I try to print, I
get a single line of only garbage from the printer before the printer ejects
the page. When this occurs the three LED's on the printer are rapidly
blinking about 5-10 times a second. Here is what I get in the log when this
happens:
May 16 20:41:49 timon kernel: usb.c: USB new device connect, assigned device
number 2
May 16 20:41:49 timon kernel: Manufacturer: Hewlett-Packard
May 16 20:41:49 timon kernel: Product: DeskJet 810C
May 16 20:41:49 timon kernel: SerialNumber: MY95C19366FY
May 16 20:41:49 timon kernel: printer.c: usblp0: USB Bidirectional printer
dev 2 if 0 alt 1
May 16 20:43:01 timon kernel: usb-uhci.c: interrupt, status 2, frame# 1436
May 16 20:43:01 timon kernel: usb-uhci-debug.h: TD @ c2ccd860/02CCD860,
MaxLen=3f DT0 EP=2 Dev=2 PID=( IN ) buf=02cca000
May 16 20:43:01 timon kernel: usb-uhci-debug.h: Len=7ff e0 SPD Stalled
CRC/Timeo
May 16 20:43:01 timon kernel: usb-uhci-debug.h: Link points to TD @
02ccda60, Breadth first
May 16 20:43:01 timon kernel: printer.c: nonzero read/write bulk status
received: -84
May 16 20:43:01 timon kernel: usb-uhci-debug.h: TD @ c2d27300/02D27300,
MaxLen=3f DT1 EP=1 Dev=2 PID=( OUT ) buf=02cc81c0
May 16 20:43:01 timon kernel: usb-uhci-debug.h: Len=3f e0 SPD Stalled
CRC/Timeo
May 16 20:43:01 timon kernel: usb-uhci-debug.h: Link points to TD @
02d27340, Breadth first
May 16 20:43:01 timon kernel: printer.c: nonzero read/write bulk status
received: -110
May 16 20:43:01 timon kernel: usb-uhci.c: interrupt, status 2, frame# 1447
May 16 20:43:01 timon kernel: usb-uhci-debug.h: TD @ c2d27f40/02D27F40,
MaxLen=07 DT0 EP=0 Dev=2 PID=(SETUP) buf=02d48ae0
May 16 20:43:01 timon kernel: usb-uhci-debug.h: Len=07 e0 SPD Stalled
CRC/Timeo
May 16 20:43:01 timon kernel: usb-uhci-debug.h: Link points to TD @
02d27f00, Breadth first
May 16 20:43:01 timon kernel: printer.c: failed reading usblp status
When I comment out setting bidir = 1 in the _probe function, the printer
behaves fine and prints correctly.
-Matt
----- Original Message -----
From: "Brian F. Hensch" <[EMAIL PROTECTED]>
To: "Vojtech Pavlik" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Tuesday, May 16, 2000 5:07 PM
Subject: Re: [linux-usb] patch printer.c
> On Mon, 15 May 2000, Vojtech Pavlik wrote:
>
> > On Sun, May 14, 2000 at 10:11:20PM -0400, Brian F. Hensch wrote:
> >
> > > Trying to get the printer driver to work I ran across a small problem
when
> > > I forced it to probe for the uni-directional interface. The probe
> > > function assumed that the out endpoint was the first end point. This
> > > should be sensible but my HP doesn't make sense and has an in endpoint
> > > first. The attached patch fixes this assumption.
> >
> > Well, in unidirectional mode the device should have the out endpoint
> > *only*. This is in the specification. Ok, I'll Extend the probng as you
> > suggested.
>
> I'm sorry, what document and what section are you reading the *only*
> requirement in? I'm not finding it. I have the Printer Device Class
> Definition document version 1.082 draft, Jan 2000. I'm only finding that
> a uni-directonal must support a BULK out endpoint and a bidi must support
> a BULK in and a BULK out endpoint.
>
> > > I still get extra random characters with the bidi interface unless I
> > > comment out the submit read urb "usb_submit_urb(&usblp->readurb);" in
> > > usblp_open(). It also works if I force the uni-directional interface.
> >
> > You mean extra random characters printed? This is really odd. Please try
> > with the attached patch.
>
> I guess they don't seem like extra chars. but rather random chars will be
> modified into hearts or accented a's. Stuff I can't represent in text
> e-mail.
>
> I'll try an example:
> A file containing several lines of the text:
> This is a test of the HP Deskjet 812C.
> Some characters will get messed up.
>
> Some lines will come out:
> This is a test of the HP Deskjet 812C.
> Some characters will get messW_@#@.
>
> Unless I set the alts = 0 or bidi = 0 in the probe or I comment out the
> submit read URB in the open. Then it works fine. It also works well with
> 2.2.14 with a very old backport patch, before the printer driver was
> URBified.
>
> I forgot to let you know I am working with kernel version 2.3.99-pre6. I
> can't get pre7 or pre8 to boot.
>
> > > It looks to me that the read is just totally broken. Not being very
> > > familiar with this I'm not sure how to fix it.
> >
> > What seems broken to you? If you have any comments, tell me, I could
> > have made some mistake there.
>
> I Shouldn't have said that. Reading is at least broken with my printer.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [EMAIL PROTECTED]
> For additional commands, e-mail: [EMAIL PROTECTED]
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
