[Prism54-devel] USB devel Fw: prism USB driver

[Feyd]

This is a multi-part message in MIME format.

--Multipart=_Tue__6_Apr_2004_16_13_16_+0200_FizOmIK0uVTDyfu3
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit



Begin forwarded message:

Date: Mon, 29 Mar 2004 13:41:05 +0200
From: Johannes Steingraeber <Jo_Stein@web.de>
To: Feyd <feyd@seznam.cz>
Subject: Re: prism USB driver


Hello Feyd,

mail with updated driver arrived, I will look into it later...

My posting did not made it to prism-devel, so I send you a copy of it.

Johannes
-- 
Not using mail encoding is like always writing on postcards.
Key ID = 1024D/8014B0E9
Key fingerprint = 71FB 4193 DFC5 9827 7D06  B9F4 A6C2 D414 8014 B0E9

--Multipart=_Tue__6_Apr_2004_16_13_16_+0200_FizOmIK0uVTDyfu3
Content-Type: message/rfc822
Content-Disposition: inline
Content-Transfer-Encoding: 7bit

Return-Path: <Jo_Stein>
Received: (from joey@localhost)
	by brodlin.org (8.9.3/8.9.3/SuSE Linux 8.9.3-0.1) id SAA01476;
	Sat, 27 Mar 2004 18:15:56 +0100
X-Authentication-Warning: brodlin.org: joey set sender to Jo_Stein using -f
Date: Sat, 27 Mar 2004 18:15:56 +0100
From: Johannes Steingraeber <Jo_Stein@web.de>
To: prism54-devel@prism54.org
Subject: USB packet observations
Message-Id: <20040327181556.A1212@brodlin.uni-kiel.de>
Mail-Followup-To: prism54-devel@prism54.org
References: <20040324152612.A606@brodlin.uni-kiel.de> <20040325135425.4115906c.feyd@seznam.cz> <20040326163747.B692@brodlin.uni-kiel.de> <20040327145654.32f30a82.feyd@seznam.cz> <20040327165428.696bce20.feyd@seznam.cz>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 1.0pre3i
In-Reply-To: <20040327165428.696bce20.feyd@seznam.cz>

On Sat, Mar 27, 2004 at 04:54:28PM +0100, Feyd wrote:
> On Sat, 27 Mar 2004 14:56:54 +0100
> Feyd <feyd@seznam.cz> wrote:
> 
> > 
> > 0x0d write reg
> > 0x0e write reg
> > 0x8d read reg
> > 0x8e read reg
> 
> After looking on the output with timestamps, I think I understand it a bit
> more:
> 
> 0x0d/0x8d: accesses the usb<->pci bridge (may be)
> 0x0e/0x8e: accesses the prism chip

Some additional remarks:

Endpoints 0x82 and 0x02 are used to read/write 0x2000 bytes after
firmware upload has finished. The written data (ep 0x02) is all
0x00. But look at four bytes at offset 0x0C..0x0F for offset/length
information. The same scheme is used for incoming data (ep 0x82) at
offset 0x1C..0x1F.

joey@brodlin:~ > grep -A 1 "^.2 .. 00000000: 0" filtered.log 
82 <- 00000000: 0c 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00
   <- 00000010: 00 80 00 04 70 be 1e 84 0c 00 00 00 00 00 fc 03
--
02 -> 00000000: 00 80 00 04 70 be 1e 84 0c 00 00 00 00 00 fc 03
   -> 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
--
82 <- 00000000: 0c 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00
   <- 00000010: 00 80 00 04 50 7f 39 81 0c 00 00 00 fc 03 fc 03
--
02 -> 00000000: 00 80 00 04 50 7f 39 81 0c 00 00 00 fc 03 fc 03
   -> 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[...]
82 <- 00000000: 0c 04 00 00 00 00 00 00 00 00 00 00 00 00 00 00
   <- 00000010: 00 80 00 04 50 7f 39 81 0c 00 00 00 e4 1b fc 03
--
02 -> 00000000: 00 80 00 04 50 7f 39 81 0c 00 00 00 e4 1b fc 03
   -> 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
--
02 -> 00000000: 00 80 24 00 70 be 1e 84 0c 00 00 00 e0 1f 20 00
   -> 00000010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

So while for incoming packets the first 16 bytes only contain the
USB packet length information the length of the "bacon" is also
contained at offset 0x1E. The outgoing packet do not have the 10 byte
header with USB packet length but only the "bacon" part with length
information at offset 0x0E.

Johannes
-- 
Not using mail encoding is like always writing on postcards.
Key ID = 1024D/8014B0E9
Key fingerprint = 71FB 4193 DFC5 9827 7D06  B9F4 A6C2 D414 8014 B0E9


--Multipart=_Tue__6_Apr_2004_16_13_16_+0200_FizOmIK0uVTDyfu3--