Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 29 Sep 2009 01:29:50 +0800
From:      Eugene Grosbein <eugen@kuzbass.ru>
To:        Hans Petter Selasky <hselasky@freebsd.org>
Cc:        usb@freebsd.org
Subject:   Re: 8.0-RC1: AMD CS5536 (Geode) USB 2.0 controller strange behavour
Message-ID:  <20090928172950.GA6865@svzserv.kemerovo.su>
In-Reply-To: <200909280849.22346.hselasky@freebsd.org>
References:  <20090928034208.GA64444@svzserv.kemerovo.su> <200909280849.22346.hselasky@freebsd.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Sep 28, 2009 at 08:49:21AM +0200, Hans Petter Selasky wrote:

> > > If the USB stack puts the new job into the schedule and the USB
> > > controller does not pick it up, it is not an USB stack problem ...
> >
> > If so, any workarounds possible?
> 
> None which I know about.

I've just found one. I noticed that USB controller here uses IRQ10
and shares it with NIC and FireWire. This system does not support IO-APIC.
I've disabled ACPI and NIC has moved to another IRQ,
USB and FireWire still share one IRQ but now there are no stalled transfers.
I get stable 16.2 megabyte/s writing speed to USB HDD.

Now I wonder if there is another way to assign distinct IRQ for USB,
BIOS Setup does not help.

Eugene Grosbein.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090928172950.GA6865>