On Wed, 2007-05-09 at 11:14 -0400, Geoffrey Tam wrote: > Here is the patch created as per instructions in > Documentation/SubmittingPatches. > > Does anyone know if there is a MontaVista USB developer mailing list?
The latest code for this MUSB driver is maintained as part of the OMAP git tree. The major rework of the Mentor code was done primarily by David Brownell, and more recently with several patches from the OMAP git developers. I suggest you send your patch to the OMAP mailing list[1]. Once it is approved/merged into the OMAP tree, it will make it into the DaVinci tree. Note however, that the DaVinci tree is not actively maintained. Right now we are focusing efforts on getting the DaVinci codebase into mainline instead of maintaining a separate tree. Kevin [1] http://linux.omap.com/mailman/listinfo/linux-omap-open-source > > Hi, > > > > On Wed, May 09, 2007 at 12:52:46AM -0700, ext Greg KH wrote: > > > On Thu, May 03, 2007 at 04:02:59PM -0400, Geoffrey Tam wrote: > > > > I am new to the Linux open source community. I am not > > sure if I am posting > > > > this to the correct mailing list. Please tell me where > > this should be posted > > > > if it is in the wrong spot. I have subscribed to this > > mailing list. > > > > > > > > I am using Linux 2.6.20. It was retrieved from the > > MontaVista git tree > > > > (source.mvista.com/git/linux-davinci-2.6.gitCVS:) in > > February 2007. I am > > > > using the TI Davinci processor. > > > > > > > > My USB gadget supports 2 endpoint 0 control transfers: > > "get variable" and > > > > "set variable". > > > > > > > > The "get" function must stall endpoint 0 if the variable > > id retrieved from > > > > the endpoint 0 setup stage packet is invalid. The current > > implementation of > > > > the USB gadget supports this. > > > > > > > > The "set" function receives the variable id in the > > endpoint 0 setup stage > > > > packet. The value to set the variable to is received in > > the data stage > > > > packet. If this value is outside of the variable's valid > > range, it must > > > > stall endpoint 0. The current implementation of the USB > > gadget does not > > > > support this. > > > > > > > > This is a legacy protocol, so it cannot be changed as > > this would break other > > > > applications that already use it. > > > > > > > > Through my debugging, I believe that the souce of my problems are: > > > > - The state machine is going to the setup stage > > (MGC_END0_STAGE_SETUP) too > > > > early. > > > > - The USB peripheral hardware is being notified with "end > > of data stage" too > > > > early (MGC_M_CSR0_P_DATAEND bit in MGC_O_HDRC_CSR0 being > > written too early). > > > > > > > > I noticed that on 22 Mar 2007, an update to the git tree > > was made to correct > > > > problems on endpoint 0 > > > > > > (source.mvista.com/git/gitweb.cgi?p=linux-davinci-2.6.git;a=lo > > g). I scanned > > > > the changes that were made, and do not believe that my > > changes are included. > > > > > > > > I made changes to the code, and this has fixed the > > problem for me. I would > > > > appreciate if someone could review them for correctness, > > and if possible, > > > > add them to the git tree. > > > > > > > > Only 1 file had to be modified: drivers/usb/musb/g_ep0.c > > > > > > > > There are 4 files included in this e-mail: > > > > - g_ep0-2_6_20.c is the original Linux 2.6.20 g_ep0.c > > from the git tree. > > > > - g_ep0.c is my modified file. > > > > - PATCH contains the differences between the 2 files. > > > > - README contains information on how PATCH was generated, > > and why the > > > > changes were made. > > > > > > Can you take a look at the file, Documentation/SubmittingPatches and > > > resend this based on the directions written there for how > > to create a > > > patch in a manner so that we can apply it? > > > > > > Also, if you are having mvista specific issues, I suggest contacting > > > them, there is nothing we can do about their kernels, and > > we do not know > > > what they add or remove from them, based on the kernel.org releases. > > > > And also, musb driver is not yet pushed into linux-2.6.git. Maybe you > > would like to sign to linux-omap mailing list. Here's the link: > > http://linux.omap.com/mailman/listinfo/linux-omap-open-source > > > > Also, you would like to send this email to Kevin Hilman > > <[EMAIL PROTECTED]>. He's the maintainer > > of the linux-davinci gittree. > > > > > > > > > > thanks, > > > > > > greg k-h > > > > > > > > -------------------------------------------------------------- > > ----------- > > > This SF.net email is sponsored by DB2 Express > > > Download DB2 Express C - the FREE version of DB2 express and take > > > control of your XML. No limits. Just data. Click to get it now. > > > http://sourceforge.net/powerbar/db2/ > > > _______________________________________________ > > > [email protected] > > > To unsubscribe, use the last form field at: > > > https://lists.sourceforge.net/lists/listinfo/linux-usb-devel > > > > -- > > Best Regards, > > > > Felipe Balbi > > [EMAIL PROTECTED] > > +55 92 2126-1003 > > > > Kernel Developers Team > > > ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ [email protected] To unsubscribe, use the last form field at: https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
