Brandon,
I think the DaVinci SDK is based on 2.6.10 kernel.
The MUSB driver on 2.6.26 might have issues in the gadget mode. I
will be submitting patches in this regards shortly (probably next week) on the
latest GIT MUSB driver.
Regards
Swami
PSP downloads at :
http://software.ti.com/swcoe/intranet/reports/pds/PSP_releases.php
Office : +91-80-25048629
________________________________
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Brandon
Hotmail.com
Sent: Wednesday, September 17, 2008 8:05 PM
To: [email protected]
Cc: Mike; Danny Marsh; Hieu
Subject: musb gagetfs usb testing.
I am performing some test to check the abilities of the USB on the davinci sdk.
I am using the Inventra High speed dual role controller that is provided on
the board. On top of this driver, I am applying gadgetfs bring the USB
controls into user space. Finally, I am using the freely available USB.c
testing program that I widely available on the net to test if gadgetfs is
working.
My first test was to run the musb driver and gadgetfs built into the kernel. I
ran the usb.c programming and everything appeared fine. There were endpoints
in /dev/gadget/ and I was able to send data out of the host with no errors
using a bulk out endpoint.. After 12,000 bytes are sent, the usb.c was no
longer responding to usb bulk transfers. More specifically, I get the
following error ->
read: Bad address
fd 5, umclaimed = 0
and it crashes the my usb.c program in user space.
*The endpoints are still shown on gadget fs.
Next, I rerun the usb.c program and I try to send out packets from the host and
I get the same error right away.
I assume that it is a lock up of either musb or gadgetfs so I rebuild them as
modules and conducts the tests again.
I send 12,00 bytes with musb, gadgetfs, and usb.c all running and it gives me
the same error as before. It also crashes my program as before. Next, I rmmod
the gadgetfs and insmod it back into the kernel. I try usb.c again but the
gadget is still locked up.
Naturally I assumed that the musb driver is froze and I removed musb and
gagdtfs from the kernel. Next, I put them back in and run usb.c and finally
the thing is working again.
I am running the 2.6.26 kernel. Has anyone seen this before? It seems like
there is some sort of buffer full problem or something alike. The end result
is the musb driver malfunctioning but I am still trying to figure out the
cause. For further analysis, I am going to put the musb driver into full debug
mode and see what that gives me.
Can Anyone help me?
Thanks,
Brandon Levander
_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
