| 22:08 | cbxbiker61 | i just release linux kernel 3.5.7.15 with F2FS support and CuBox graphics support |
| 22:10 | cbxbiker61 | _rmk_ , with your recent kernel patches is the watchdog supposed to work on CuBox? |
| 22:22 | _rmk_ | cbxbiker61: not with DT, but it works here with my 3.9 + rabeeh based kernel |
| 22:24 | cbxbiker61 | _rmk_ still something missing for device tree then, i saw the discussion about a missing irq |
| 22:26 | _rmk_ | irqs aren't necessary for the watchdog - what the watchdog driver does is clear any pending interrupt before enabling the ability for the watchdog to reset the system |
| 22:26 | _rmk_ | presumably it does that to prevent a pending IRQ from resetting the system when the reset path is enabled |
| 22:34 | cbxbiker61 | _rmk_ i applied your two watchdog patches and added the def for rtc@ back into the dtsi, I though plugging the rtc back into the dtsi might make it work |
| 22:34 | cbxbiker61 | sorry, i meant watchdog |
| 22:35 | cbxbiker61 | orion_wdt f1020300.wdt: Orion Watchdog missing clock |
| 22:37 | cbxbiker61 | finally I'm at a point where I can dig back into xbmc/cubox, they've restructed EGL code, so I need to fix that |
| 22:45 | _rmk_ | you probably need to add to the driver a struct of_device_id declaration similar to the one in drivers/rtc/rtc-mv.c, along with the .of_match_table initializer in the platform_driver structure |
| 22:48 | _rmk_ | sorry for the slow responses; I'm nackered this evening |
| 22:48 | _rmk_ | and I might also be knackered too |
| 22:48 | cbxbiker61 | yeah, i'm kinda bouncin around between projects right now |