FreeBSD 9.1-RELEASE w/ SanDisk ImageMate S11202

[Joe Altman]

On Fri, Jan 18, 2013 at 10:19:25PM -0800, Waitman Gobble wrote:
> On Fri, Jan 18, 2013 at 12:10 AM, Waitman Gobble <gobble.wa at gmail.com>wrote:
> 
>  Hi,
> 
> I have a PCI Express card with VIA VL800 chipset which seems to work
> OK with a Seagate drive, so I presume the interface is working.

I think my question is relevant:

Does the working drive mean that you have the proper USB bits in your
kernel?

http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/usb-disks.html

19.5.1 Configuration
If you use a custom kernel, be sure that the following lines are present
in your kernel configuration file:

device scbus <==
device da    <==
device pass  <+=
device uhci
device ohci
device ehci
device usb
device umass <==

I don't see them mentioned below.

> > If I boot with a SanDisk ImageMate S11202 plugged into the USB 3.0 card,
> > the display shows messages about the SanDisk device, appears to properly
> > identify it, and with like 'querying ...' slots on the card reader, which
> > all fail - (there are no cards in the reader). Then the machine sails into
> > outerspace, kind of just sits there until (perhaps) the end of time, or
> > until I kill the power. It never gets the network interface up, so no ssh.

> > kernel built with:
> >
> > device        ahci        # AHCI-compatible SATA controllers
> > device        uhci        # UHCI PCI->USB interface
> > device        ohci        # OHCI PCI->USB interface
> > device        ehci        # EHCI PCI->USB interface (USB 2.0)
> > device        xhci        # XHCI PCI->USB interface (USB 3.0)

General question:

The man page is ambiguous to me; does it mean that one only needs xhci
for all three versions, or must one specify all? 

DESCRIPTION
     The xhci driver provides support for the USB eXtensible Host
     Controller Interface, which allows use of USB 1.0, 2.0 and 3.0
     devices on the same USB port.

Such that xhci can handle *all* such USB devices?

Also, ahci has a number of sysctl knobs for interrupts. Maybe that is a
path to resolving the issue?

Regards,

Joe