On Sat, Nov 20, 2010 at 1:42 PM, Wolfram Sang <w.s...@pengutronix.de> wrote: > Hi, > >> As far as I could tell, using compatible = <24c64>; didn't load the right >> module (module name is at24) and using at24 caused a device id mismatch >> because at24 is not a known device ID. I could be wrong here and if so, I'd >> very much like a source code hint as to why... > > Have you tried "<vendor>, 24c64"? All I can say is that it works for me. Plain > at24 worked for years with pcm030.dts and pcm032.dts, so I don't know which > issue you are facing. This patch is just about extending the support.
According to of_modalias_node, the prefix is stripped anyway, so that wouldn't help. The code probably worked using the buggy probing... > >> My patch worked by changing drivers/of/of_i2c.c to allow a generic >> kind = " " statement that was filled in as i2c_board_info.type, to allow >> the module name and the device id to be different. >> This makes the at24 driver no longer rely on probing (which it claims >> is buggy anyway) and also benefits other drivers that support multiple >> devices, but where probing is difficult (e.g. lm90 driver). >> >> I'm in the process of getting employer approval to get these patches >> upstream. > It'll get approved ;) It's the first time I'm taking these steps, but the procedures are there, so at most it'll take some extra time. > I hope you will get it approved, it is a lot easier to talk about code :) > > Best regards, > > Wolfram > > -- > Pengutronix e.K. | Wolfram Sang | > Industrial Linux Solutions | http://www.pengutronix.de/ | > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.9 (GNU/Linux) > > iEYEARECAAYFAkznwjIACgkQD27XaX1/VRssnwCgg55UCwZFLcMI8kJI3mhCJQxL > N7kAoJHpLn5BJpNjET+ngaQFrGxUBQm1 > =tyTb > -----END PGP SIGNATURE----- > > _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
