Date: Thu, 29 Mar 2007 17:19:08 +0200 From: Markus Henschel <m.henschel@bally-wulff.de> To: Hans Petter Selasky <hselasky@c2i.net> Cc: freebsd-usb@freebsd.org Subject: Re: usb/110855: ugen: interrupt in msgs are truncated when buffer is full Message-ID: <460BD8EC.5060900@bally-wulff.de> In-Reply-To: <200703290838.57458.hselasky@c2i.net> References: <200703271940.l2RJe90x058558@freefall.freebsd.org> <200703281832.22527.hselasky@c2i.net> <460AB387.20304@bally-wulff.de> <200703290838.57458.hselasky@c2i.net>
next in thread | previous in thread | raw e-mail | index | archive | help
Hans Petter Selasky schrieb: > > Does it hang as long as before ? > > If you try to increase the delay again, does that help? > > Edit sys/dev/usb/umass.c > > Then lookup: > > usbd_delay_ms(uaa->device, 1000); > > Change it into: > > usbd_delay_ms(uaa->device, 5000); > >> Thank you for your time. Let me know if you want me to try other >> changes. May be it would be easier if you could create a branch in svn >> with the changes needed? > > Maybe. > > --HPS > Hi, changing usbd_delay_ms(uaa->device, 1000); to usbd_delay_ms(uaa->device, 5000); increases the overall delay be 4 seconds :-). The value given there doesn't seem to be related to the problem. The first returned CCW is always invalid for that stick. I did some experiments with the other change you suggested. When I change: sc->sc_last_xfer_index = UMASS_T_BBB_COMMAND; to sc->sc_last_xfer_index = UMASS_T_BBB_STATUS; the first CCW is still invalid but the hanging is gone (overall detection time is still much longer than with the old usb stuff). I also tried to use the original umass.c together with the new usb stack. It seems to work. There result is similar to changing sc->sc_last_xfer_index = UMASS_T_BBB_STATUS. -- Regards, Markus Henschel Development BALLY WULFF Automaten GmbH Maybachufer 48-51 12045 Berlin Phone: +49(30)62002 161 FAX: +49(30)62002 230 http://www.ballywulff.de
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?460BD8EC.5060900>