ToDo, Bugs and Problems
- Resume problems — I think I'll have to only implement BAT[01]_LOW handling.
- wm97xx isn't detected sometimes. Maybe it's related to the fact that I kexec the developing kernel, so it doesn't get fully reset. It's high priority because batteries and touchscreen are routed through the wm9712 chip. — mostly workarounded by using patch from RP patchseries to reset codec in a loop.
Split tosa-power to use class_power & company.Debug acceleration reenable problems of tmio fb driver.- wm97xx can't resume properly — again workarounded by patch from RP patchset.
- The tmio_ohci needs to store full SRAM in a buffer during suspend/resume. Check if this is also necessary for the tmiofb (code was there but got removed)
- Denoise touchscreen samples by synchronising with VGA sync signal
The bootloader can charge the battery till 4.2V, but I get "full" signal near 4.0V. Check how to enable full charging.- Dyntics breaks resume at 2.6.24.
pxa2xx_udc isn't working with recent (2.6.24-rc6) kernels.— patch proposed, but not yet in mainline- Check linux-wlan-ng. Maybe support for USB Prism2 cards can be merged with hostap and/or ported to some newer API.
- Check whether we can use deferred io fb framework for tmiofb to remove hack from xserver.
Last modified at: Friday, February 8, 2008 1:52:57 AM.