Andy Green wrote: > But I'm not willing to cede all decisions about our product to > each upstream sort of mindlessly like you are there.
You may be surprised to find that I actually did ask myself to what extent we care about upstream in this case. I took it as a hint that, given that we have already have six interfaces that affect WLAN power in one way or another and still can't give a simple answer to what seems to be a simple question, "do it just because we can" doesn't seem to be an efficient policy when it comes to actually providing a solution in this case. For those who haven't kept count, here are the options: 1) SIOCSIWTXPOW 2) AR6000_IOCTL_EXTENDED/AR6000_XIOCTRL_WMI_SET_WLAN_STATE 3) sysfs power_on 4) global suspend/resume 5) module load/unload 6) bind/unbind 7) rfkill - Werner
