Hi... Once I met similar situation, it's just I am not that great like you do :) OK, let's jump to your questions.. On Tue, Aug 24, 2010 at 20:40, Lutz <linsenhof@xxxxxxxxxxxxxx> wrote: > Two questions: > > 1. i'm wondering if there is a way to avoid situations like this. Can't > expect every new implementation efforts to be announced on the kernel > mailing list, can i? IMHO, send your patch earlier....even if it's still premature. Thus, it's like sending RFC (Request For Comment). But this doesn't mean you can talk bull shit :) I am sure you know what I mean.. > 2. my driver utilizes a sysfs interface which works fine and in my > opinion offers some advantages. Thus it doesn't conform to > Documentation/watchdog/watchdog-api.txt (which i considered to be sort > of a guideline) Do you think this were acceptable even if quite every > watchdog driver i had a look at implements watchdog-api? The way I see it: it's classic engineering problem. Sometimes, to create something more efficient, you have to break the old "terms" and make new one. Even Greg KH said "there's no stable kernel API". We just can only expect stable user space API or user-to-kernel API (read: syscalls) Then again, how do we know if your approach is acceptable? simple...send it earlier :D And if somebody ask "why the f**k do you do that? use existing APIs sir..".... then there you can defend your work and (positive) debate start. the best example, IMHO, on how it could go...is when Con Kolivas defend Rotating Staircase Deadline Scheduler (RSDL) and it's still out of the ring :D :D :D -- regards, Mulyadi Santosa Freelance Linux trainer and consultant blog: the-hydra.blogspot.com training: mulyaditraining.blogspot.com -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
