On 01/31/2011 02:49 PM, Juha Kuikka wrote:
On Mon, Jan 31, 2011 at 11:47 AM, Steve Chen <[email protected] <mailto:[email protected]>> wrote:On Mon, Jan 31, 2011 at 1:36 PM, Juha Kuikka <[email protected] <mailto:[email protected]>> wrote: > On Mon, Jan 31, 2011 at 11:04 AM, Gary Thomas <[email protected] <mailto:[email protected]>> wrote: > > > > I have a board which is uses the OMAP-L138. It's similar in > > many respects to the EVM/DA850 modules, but there are differences. > > In particular, my board uses MMC/SD1 since it also uses NOR FLASH. > > > > I was able to modify U-Boot to handle this without much trouble. > > However, it doesn't run in Linux. I added the driver bits for > > this, but what I see is that the very first command issued sets > > up for a completion interrupt (which the registers tell me happens), > > but the driver IRQ handler is never called. > > > > Am I alone in this? as I didn't find any support for MMC/SD1 > > in the Linux kernel (I checked the most recent public 2.6.37) > > > > Does anyone have experience using MMC/SD1 on the OMAP-L138 > > with the Linux driver? > > > I think I ran into the same issue. > Check this conversation on the subject, especially the third > mail: http://davinci-linux-open-source.1494791.n2.nabble.com/DA850-and-MMCSD1-EDMA-td5077613.html > I guess I should have pushed that patch but for some reason I never got > Sudhakar's comment on it. :( > If this is the solution to your problem maybe you can refresh it against the > latest kernel and submit it? I don't have access to the HW any more so I > cannot verify it. > - Juha Hello Juha, Followed the link you provided, it seems that Sudhakar did respond. He suggested you to submit your patch to the davinci mailing list. Sorry, my comment was unclear, I see now that he did respond but I never got that email when the issue was originally discussed so I dropped the ball. Anyway, we cannot just change the default event queues on all the devices, see this chain. http://www.mail-archive.com/[email protected]/msg18038.html Then I got busy and didn't have time to implement and test the proper solution.
Do you have suggestions for what a proper solution is? I just merged these changes into [mainline] 2.6.37. I had to make a few tiny adjustments, e.g. part of the changes are already in place, just not all, but it seems to be working for me. I'd like to see the correct code figured out and pushed. Thanks -- ------------------------------------------------------------ Gary Thomas | Consulting for the MLB Associates | Embedded world ------------------------------------------------------------ _______________________________________________ Davinci-linux-open-source mailing list [email protected] http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source
