Re: Endless CAM messages with recent CURRENT

After looking around I had been able to localise the cause for the 
described messages, see below:

On 14.01.2011 10:07 (UTC+1), Rainer Hurling wrote:
> Today I updated my 9.0-CURRENT system (amd64) to revision 199506: After
> rebooting I get the following messages two times per second in an
> endless run:
>
>
> ---------------------------------------------
> ...
> Jan 14 09:37:47 krabat kernel: (sg1:umass-sim0:0:0:0):
> cam_periph_release_locked: release 0xfffffe0009f41200 when refcount is zero
> Jan 14 09:37:47 krabat kernel:
> Jan 14 09:37:48 krabat kernel: (sg2:umass-sim0:0:0:1):
> cam_periph_release_locked: release 0xfffffe0009f41100 when refcount is zero
> Jan 14 09:37:48 krabat kernel:
> ...
> ---------------------------------------------

When turning off hald no more of these messages appear. hald is trying 
to poll the card reader, but it does not like this. So prohibiting the 
polling in /usr/local/share/hal/fdi/preprobe/20thirdparty/ solves my 
problem:

#cat 10-broken-usb-card-reader.fdi
<?xml version="1.0" encoding="UTF-8"?>

<deviceinfo version="0.2">
   <device>

     <!-- Generic STORAGE DEVICE 9744 -->
     <match key="storage.bus" string="usb">
       <!-- usb card reader Silverstone SST-FP35B doesn't like polling -->
       <match key="_at_storage.originating_device:usb.vendor_id" int="0x05e3">
         <match key="_at_storage.originating_device:usb.product_id" 
int="0x0716">
           <merge key="storage.media_check_enabled" 
type="bool">false</merge>
         </match>
       </match>
     </match>

   </device>
</deviceinfo>



> sg1 and sg2 are devices from my front panel card reader Silverstone
> SST-FP35B:
>
> #camcontrol devlist
> [..snip..]
> <Generic STORAGE DEVICE 9744> at scbus6 target 0 lun 0 (sg1,pass3,da0)
> <Generic STORAGE DEVICE 9744> at scbus6 target 0 lun 1 (sg2,pass4,da1)
>
>
> Is it possible that the last changes in usb code (xhci) or in cam code
> are responsible for this? Does anyone else observe this behaviour?

A more generic approach would be to integrate the polling info into 
/usr/local/share/hal/fdi/preprobe/10osvendor/20-broken-usb-sticks.fdi. 
Is anyone willing to integrate the code (.fdi file or some quirks) in 
the usb stuff?


In that case Linux is offering some more info about this device:

#lssub
Bus 002 Device 002: ID05e3:0716 Genesys Logic, Inc. USB2.0 Multislot 
Card Reader/Writer
idVendor 0x05e3 Genesys Logic, Inc.

#usb-devices
T: Bus=02 Lev=01 Prnt=01 Port=03 Cnt=01 Dev#= 2 Spd=480 MxCh= 0
D: Ver=2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs=  1
P: Vendor=05e3 ProdID=0176 Rev=97.44
S: Product=USB Storage
S: SerialNumber=000000009744
C: #Ifs= 1 Cfg#= 1 Atr=80 MxPwr=500mA
I: If#= 0 Alt= 0 #EPs= 2 Cls=08(stor.) Sub=06 Prot=50 Driver=usb-storage


Note, that the card reader from Benesys Logic is only one device of some 
more (firewire, eSATA, USB slots ...), which are provided from the 
'Silverstone SST-FP35B' front panel access unit.

> Please let me know if you need more info or if I can test something.

Thanks in advance,
Rainer Hurling