Unusual unusual device Sony-Ericsson W760i

Summary: This phone provides two luns which represent the internal memory and the other an optional memory-stick. While the memory stick suffers "only" from the capacity bug, the internal memory seems to return the "no sense" errors even for sectors farther away from the end.

I am currently a bit lacking ideas how to proceed with that. Any ideas would be welcome.

More detail:

Kernel versions: 2.6.28 but also confirmed to be the same on a recent mainline kernel (2.6.31)

All the information has been gathered on this bug report and it is confirmed with two different phones of the same model.
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/400652

T:  Bus=02 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 11 Spd=12  MxCh= 0
D:  Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P:  Vendor=0fce ProdID=e0c6 Rev= 0.00
S:  Manufacturer=Sony Ericsson
S:  Product=Memory Stick
S:  SerialNumber=3570670227111290
C:* #Ifs= 1 Cfg#= 2 Atr=80 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage
E:  Ad=05(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=85(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms

LUN#0 (internal memory) returns 0x1d51c(120093) when queried for capacity. The usbmon trace[1] however shows, that the read request of 8 sectors starting from 0x1d508(120072) fails with the known no-sense error and is repeated infinitely.

Preventing vol_id from checking for raid meta-data lets the device detection finish. Unfortunately calling fdisk causes the end of device area to be accessed and I forgot that those values are in /sys as well. But the size of 119992 sounds like there is some gap between the end of the defined partition and the end of the device.

This feels a bit like some sort of host protected area, if SCSI knows such a thing (I don't think so). In the end I am lacking a good clue how this device might be handled. SINGLE_LUN does not work as it is LUN#0 which behaves that way and people will want to access the internal storage anyways. All other quirk options seem to be not matching that case (beside of FIX_CAPACITY). So I would be thankful for any hint possible.

Please cc me on replies as I am not subscribed. Thanks

- Stefan

[1] Usbmon output of bus 2 while plugging and unplugging the phone on an umpatched kernel: http://launchpadlibrarian.net/29196150/normal-usbmon-2u.txt