> On 06/02/2024 21:41:10+0000, Chris Packham wrote: > > > > On 7/02/24 10:12, Alexandre Belloni wrote: > > > On 06/02/2024 20:19:20+0000, Chris Packham wrote: > > >> That is an incredibly good point. The max31335 binding covers one > specific chip. This binding covers more and with that there are a few more > properties that the max31335 on it's own doesn't have (e.g. the clock > consumer, the ability to have different i2c addresses). Binding wise I could > probably roll all of the max31335 into this max313xx binding. > > >> > > >> Driver wise things are a bit trickier. I've only got access to one of > > >> the variants so I am hoping to leverage the work Ibrahim had already > > >> done. I could attempt to incorporate max31335 support into the > > >> max313xx driver but I wouldn't really be able to test it properly and > > >> there is a reasonably high chance of regressing something. > > > But I won't take a separate driver. Everything would be better if Analog > > > was sharing the datasheets... > > > > The datasheets are pretty accessible so I'd give Analog a pass on that > > (they're certainly better than some vendors). I'll include some links on > > the next update. > > > > The max31335 is not available > Indeed, the max31355 datasheet is not available yet. This is also stated in the cover letter of the patch series for max31335. It was an urgent request from a customer to have it mainline as soon as possible. If there are questions about the part functionality, I can definitely help. > > I'll attempt to roll the max31335 into the max313xx driver. > > I guess this would be the opposite. Renaming a driver breaks existing > kernel configurations. > > -- > Alexandre Belloni, co-owner and COO, Bootlin > Embedded Linux and Kernel engineering > https://urldefense.com/v3/__https://bootlin.com__;!!A3Ni8CS0y2Y!7A0VM > Nj1SLsNWZ8SnPrFwG6KX2TqjeLQi38EYIV164_5MWlQ7yCFvp8yOjaswIRNLC > 0EK-_bhHEfy6xAXEJimVEW8BZXdn_r$
