On Fri, Jun 10, 2005 at 12:25:39AM -0400, Mark M. Hoffman wrote: > Hi Jean: > > * Mark M. Hoffman <mhoffman at lightlink.com> [2005-06-04 23:41:37 -0400]: > > It would be easy enough to do hwmon1, hwmon2, etc. just like the usb_host > > class does. Everyone OK with that? > > * Jean Delvare <khali at linux-fr.org> [2005-06-04 18:24:28 +0200]: > > > Last, it would be nice if the IDs were reused on driver cycling, just > > > like the i2c bus IDs are. You should be able to pick the code in i2c-dev > > > and reuse it in the hwmon class. > > * Mark M. Hoffman <mhoffman at lightlink.com> [2005-06-04 23:41:37 -0400]: > > It would be nice... I'll look into it. > > Actually, the bus IDs are assigned in i2c-core. The i2c-dev ID is just a > copy of that. The mechanism for the ID recycling is from lib/idr.c. > > AFAICT, the ID recycling is a mere side-effect of that (idr.c) modules's > real purpose. The way it's used in i2c-core.c feels like bloat to me. > I'm not certain I want to duplicate that in hwmon.c. Opinions? Hm, idr.c is used to generate unique ids, that's what it is there for. > Well, i2c_get_adapter() could actually use i2c_adapter_idr properly: > (untested, not signed off, may eet ur branes, /me needs sleep) Yeah, this looks probably right. Anyone want to test it out? :) thanks, greg k-h
