Hi Guenter, On Sun, 3 Apr 2011 14:21:13 -0700, Guenter Roeck wrote: > On Sun, Apr 03, 2011 at 11:37:48AM -0400, Jean Delvare wrote: > > Hi Guenter, > > > > On Sat, 2 Apr 2011 09:06:19 -0700, Guenter Roeck wrote: > > > When writing hardware monitoring drivers, there are some common pitfalls which > > > keep coming up in code reviews. This patch provides a document describing all > > > those pitfalls and how to avoid them. > > > > FWIW, for the i2c subsystem I decided to publish my recommendations on > > a wiki, rather than in the kernel tree: > > https://i2c.wiki.kernel.org/index.php/Linux_2.6_I2C_development_FAQ > > > > We could have a similar wiki for the hwmon subsystem if you think this > > would be useful. It would also be possible to create a page in the > > lm-sensors.org wiki, as both projects are tightly related anyway. > > > Yes, I think that would be useful. I think we should have the document > in the kernel as well, though, since many people won't read the wiki, > and because having it in the kernel makes it official (or at least more so). Meaning we'd have to maintain the document in two different places? Maybe... but think twice before you go that route. > > > (...) > > > +* Avoid function macros. While it may save a line or so in the source, it > > > + obfuscates the code and makes code review more difficult. It may also result > > > + in code which is more complicated than necessary. > > > > See my review of your max16065 driver ;) > > > Guess you mean to avoid complex maros as well. Ok, no problem. Maybe I misunderstood "function macros" in your sentence. Did you mean functions with parameters, or macro-generated functions? I thought the first, but maybe you meant the second. Well, this point needs clarification apparently ;) > (...) > Overall this was mostly to initiate a discussion. Maybe it would make sense to add it > to the wiki first, and copy it into the kernel after it stabilizes. This way it would > not just be my document, but give the community a chance to participate. > What do you think ? We two are the maintainers of the hwmon subsystem, so this documentation must be ours. Don't ask the community at large how they want to contribute, you won't like the answer ;) In fact, the reason you felt the need to write this document is, in part, because you want to set the rules contributors should follow to make you (and me) happy. I'm fine with the document (both its existence and its content. In fact I find it curious that I wrote a guidance document to i2c subsystem contributors and never felt the need to do the same for hwmon subsystem contributors. I can't really explain it. -- Jean Delvare _______________________________________________ lm-sensors mailing list lm-sensors@xxxxxxxxxxxxxx http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
