So this may be silly to point out but I noticed the value is multiplied by 2 later in txpower setup, which introduces the possibility that of overflow/wrapping. min(power_limit, int_min) would maybe run into trouble there. Since the behavior is to assume 2/5g tx power, I think it would be nice to set to those values (limits), then it would display sanely. But I don't have a good grasp on this stuff, I imagine regdom plays a role someplace too On Mon, Dec 16, 2019 at 11:58 PM Johannes Berg <johannes@xxxxxxxxxxxxxxxx> wrote: > > On Fri, 2019-12-13 at 15:03 -0800, greearb@xxxxxxxxxxxxxxx wrote: > > > > So, instead initialize the txpower to -1 in mac80211, and let drivers know that > > means the power has not been set and so should be ignored. > > Technically (or maybe just physically?), even -1 is a sort of valid TX > power. > > I know all of this is pretty messed up, but wouldn't it make more sense > to go with some kind of tx_power_valid bit, or perhaps something that > certainly will never make sense like MIN_INT instead of -1? > > johannes >
