Messages in this thread | | | Subject | Re: tulip driver in 2.1.11* - 2.1.21 is broken - new driver | Date | Sun, 20 Sep 1998 02:23:38 -0500 (CDT) | From | kwrohrer@ce ... |
| |
And lo, Justin A. Kolodziej saith unto me: > David S. Miller wrote: > > From: Linus Torvalds <torvalds@transmeta.com> > > > > It's way too easy to just add new broken members. We need to clean cruft > > out too. > > > > But please give me one answer, how do we deal officially with the > > driver maintenance issue, ie. the Becker problems? I don't want to > > step on toes, at the same time I want to fix all his broken code so > > 2.2.x can be a reality. > > I thought that the point of open code was for anyone that wants to fix > problems to be able to without a hassle. I thought the point of common courtesy was that when you fix problems in a piece of software which has a maintainer, you funnel those fixes through the maintainer and avoid forking distributions if at all possible. On one hand, I don't know how obstinate Becker is being about the whole issue--I haven't noticed a single message from him on this thread--but on the other hand, over 6 billion humans could care less whether or not this issue gets "fixed" and will go with whatever they (or their distribution vendor) happen to get...or with Becker's version, if it adds a feature they need or fixes a bug they're actually experiencing.
> My answer would be to go right > ahead, fix all the obviously broken code that you find, and not worry > about what Becker thinks. The only approval that I would consider > getting is Linus's. Of course, make sure that it works the same way > first so that 50 million other lines of code do not need to be > rewritten. Or is that how broken the code is? :) Well, everywhere "tbusy" appears in a network driver, someone has to figure out *how* it was being used, and either change the name (if it was being used "right") or *fix* it. That's not going to be quick, especially for someone who's not familiar with a given driver, since all drivers seem to use it the "right" way in addition to some/most drivers "misusing" it.
This whole controversy seems to stem from inadequate or outdated kernel internals documentation. Unfortunately, the right solution (i.e. thorough documentation of the kernel that gets updated when things get changed) isn't really worth anybody's time or money to do...or, if it is, only for the reason Hawking's _A Brief History of Time_ sold...
Keith (least-read bestseller of the decade, I heard...)
-- "The avalanche has already started; |Linux: http://www.linuxhq.com |"Zooty, it is too late for the pebbles to |KDE: http://www.kde.org | zoot vote." Kosh, "Believers", Babylon 5 |Keith: kwrohrer@enteract.com | zoot!" www.midwinter.com/lurk/lurker.html |http://www.enteract.com/~kwrohrer | --Rebo
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |