Date: Sun, 11 May 2003 14:30:17 +0200 From: Bernd Walter <ticso@cicely9.cicely.de> To: Marcin Dalecki <mdcki@gmx.net> Cc: "Yevmenkin, Maksim" <Maksim.Yevmenkin@cw.com> Subject: Re: USB link cable? Message-ID: <20030511123016.GG1922@cicely9.cicely.de> In-Reply-To: <3EBE3CB4.1000401@gmx.net> References: <2E7E8A35375D1449A6F28D5E022E67310AC4D2@USSC8MS04.Global.Cwintra.Com> <3EBD2F95.9090807@gmx.net> <20030511005127.GD1922@cicely9.cicely.de> <3EBE3CB4.1000401@gmx.net>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, May 11, 2003 at 02:06:12PM +0200, Marcin Dalecki wrote: > Bernd Walter wrote: > >I really doubt that the udbp driver is guilty of this. > >An USB driver has no chance to break a device permanently by accident > >unless it has a very broken design. > > > >What does FreeBSD tell you now on probing the device? > > > > The working end results in: > > ugen0: Prolific Technology Inc. PL2301 Host-Host interface, rev 1.00/0.00, > addr 2 > > The kaputt end results in: > > ugen0: at uhub0 port 1 (addr 2) disconnected > ugen0: detached > uhub0: device problem, disabling port 1 > > Linux is a bit more elaborative on error reporting: FreeBSD requires USB_DEBUG for this kind of message. > hub.c: new USB device 00:02.3-1.1, assigned address 5 > usb.c: USB device not accepting new address=5 (error=-32) > hub.c: new USB device 00:02.3-1.1, assigned address 6 > usb.c: USB device not accepting new address=6 (error=-32) > > It looks really as if the IO drivers on this end are burned. You may want to check with FreeBSD and USB_DEBUG again, as setting a new adress is not the first communication with the device (at least with FreeBSD). > Unfortunately the device is one of those melded in soft PVC kind of, so > I can't open it for further investigation. Hmm I may try anyway with a knife > out of couriosity... It certainly started during the experimentation > on the BSD side. But I did feed all data through udbp0: and never used > ugen0 devices. > > It's really just a bit curious now. Basicly the device is powered from the USB bus - of which there are two. It's quite possible that the vendor of your device failed to handle that situation correctly. Basicly it's possible to get an enourmous overvoltage if the +5V line gets connected before GND and your devices have different GND bases. Unless the vendor put the required overvoltage protection circuits into that device and they tried to take power from both sides it's strictly not hot-plug-able. -- B.Walter BWCT http://www.bwct.de ticso@bwct.de info@bwct.de
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030511123016.GG1922>