Re: umass(4) regression in 9.0-RC1.

On Friday 28 October 2011 21:09:47 Pawel Jakub Dawidek wrote:
> On Fri, Oct 28, 2011 at 09:11:42AM +0200, Hans Petter Selasky wrote:
> > On Thursday 27 October 2011 20:51:15 Pawel Jakub Dawidek wrote:
> > > On Thu, Oct 27, 2011 at 08:42:09PM +0200, Hans Petter Selasky wrote:
> > > > This is the root HUB. Can you also show the actual device?
> > > 
> > > Sorry, it wasn't connected, here it goes:
> > > 
> > > ugen0.2: <USB2.0-CRW Generic> at usbus0, cfg=255 md=HOST spd=HIGH
> > > (480Mbps) pwr=ON
> > > 
> > >   bLength = 0x0012
> > >   bDescriptorType = 0x0001
> > >   bcdUSB = 0x0200
> > >   bDeviceClass = 0x0000
> > >   bDeviceSubClass = 0x0000
> > >   bDeviceProtocol = 0x0000
> > >   bMaxPacketSize0 = 0x0008
> > >   idVendor = 0x0bda
> > >   idProduct = 0x0119
> > >   bcdDevice = 0x1981
> > >   iManufacturer = 0x0001  <retrieving string failed>
> > >   iProduct = 0x0002  <retrieving string failed>
> > >   iSerialNumber = 0x0003  <retrieving string failed>
> > >   bNumConfigurations = 0x0001
> > 
> > Hi,
> > 
> > The control request in question is mandatory according to the UMASS
> > specification, and I wonder why it times out and all other control
> > requests aswell.
> > 
> > Could you try setting the no-synchronize cache quirk instead, and then
> > plug your device.
> > 
> > I'm sorry, but this problem needs further investigation before we can
> > make a patch.
> 
> It wasn't immediately obvious for me how to set the no-synchronize cache
> quirk, but I think I found it:
> 
> 	# usbconfig add_quirk UQ_MSC_NO_SYNC_CACHE
> 
> And it seems to work:
> 
> umass0: <Generic USB2.0-CRW, class 0/0, rev 2.00/19.81, addr 2> on usbus0
> (probe0:umass-sim0:0:0:0): TEST UNIT READY. CDB: 0 0 0 0 0 0
> (probe0:umass-sim0:0:0:0): CAM status: SCSI Status Error
> (probe0:umass-sim0:0:0:0): SCSI status: Check Condition
> (probe0:umass-sim0:0:0:0): SCSI sense: UNIT ATTENTION asc:28,0 (Not ready
> to ready change, medium may have changed) da0 at umass-sim0 bus 0 scbus13
> target 0 lun 0
> da0: <Generic- SD/MMC 1.00> Removable Direct Access SCSI-0 device
> da0: 40.000MB/s transfers
> da0: 30799MB (63076352 512 byte sectors: 255H 63S/T 3926C)

Hi Pawel,

REALTEK, which is the manufacturer of your device has already been quirked:

http://svnweb.freebsd.org/base?view=revision&revision=225777

I think the patch is just MFC'ed yet.

--HPS