On Sat, Jan 03, 2009 at 11:03:57PM +0100, michael wrote: > Hi, > > Balaji Rao wrote: >> On Sat, Jan 03, 2009 at 10:56:09PM +0100, michael wrote: >>> Hi, >>> >>> Balaji Rao wrote: >>>> On Sat, Jan 03, 2009 at 09:04:22PM +0000, Andy Green wrote: >>>>> -----BEGIN PGP SIGNED MESSAGE----- >>>>> Hash: SHA1 >>>>> >>>>> Somebody in the thread at some point said: >>>>> >>>>> | but the main usage of it says >>>>> | >>>>> | ~ /* register power off handler with core power management */ >>>>> | ~ /* FIXME : pm_power_off = &pcf50633_go_standby; */ >>>>> | >>>>> | Balaji does this ring any bells from the regulator changes? >>>>> | >>>>> | -Andy >>>>> |> >>>>> | Like this? >>>>> >>>>> | I must test... >>>>> >>>>> Yes it's the right lines but I wonder why Balaji pulled it from the >>>>> pcf50633-core that went upstream, he may have some thinking on it. But >>>>> it would seem any system that had a pcf50633 in it would want to point >>>>> pm_power_off at it so it's at least arguable it belongs in there. >>>>> >>>> Ah yes, yes! It's my mistake. I thought it was part of the APM emulation >>>> code and didn't bother about it at all :( Why didn't I ever think of >>>> this! :( >>>> >>>> Thinking about it, I feel it's better to put it in mach-gta02.c and >>>> strictly not in our driver. It's easily doable in mach-gta02.c no ? >>>> >>> Ok, but like this: >>> >>> pm_power_off = pcf_get_poweroff_func(); >> >> not exactly. Simpler, we can directly write into the appropriate register >> from mach_gta02.c. >>> This function return NULL in case that the pcf50633 core is failed to >>> register. One >>> question: >>> >>> - what happen if compile pcf50633 as a module? >> >> For GTA02, we can _not_ use it as a module. >> >> - Balaji >> > Ok, but i think that you need some i2c related stuff.
We can simply use pcf50633_reg_write to write into that register. I'll try it out tomorrow. It should work. Thanks Balaji
