Messages in this thread | | | Date | Tue, 12 Aug 1997 00:01:08 -0600 (CST) | From | Adam McKee <> | Subject | Re: [patch] QNX-style scheduling v1.06 |
| |
Hi there.
It's against 2.0 because a) I run 2.0, and b) most people run 2.0. I don't think there's a great likelihood of the patch being adopted into the offical 2.1/2.2 kernel -- maybe I'm wrong about that. However, I have been toying with the idea of switching to 2.1 for some time now, and as soon as I do I'll forward-port the patch. I'll probably do this sometime in 2.1.5x. It doesn't look like a big deal -- the main difference in the affected code is that cli()/sti() pairs have been replaced with a much better locking scheme.
-- Adam
On Mon, 11 Aug 1997, Aaron Tiensivu wrote:
> > I was going to wait for 2.0.31, but I thought "oh Hell", and I am > > releasing this now to see how it works for people. With any luck it'll > > patch cleanly against '31 when it is released. > > It might be preferred to make this patch against 2.1.49 instead of 2.0.x since > I believe something like this will probably not be added into 2.0 and more > likely to be thrashed about in 2.1.x. > > If it were against 2.1.x, I'd test it to death. ;-) > Or if it wasn't too much trouble, make two versions. > > I haven't tried to apply it to 2.1.x yet, so I'm just making rash assumptions > that it wouldn't go in cleanly. > > -- > Which is worse: ignorance or apathy? Who knows? Who cares? > OK, so you're a Ph.D. Just don't touch anything. > Software is like sex; it's better when it's free. - Linus Torvalds > IBM: It may be slow, but at least it's expensive. > Microsoft: What do you want to own today? >
| |