You know guys, this is getting really old, really fast. The fact of the matter is, the original issue was raised so long ago that it's fallen off of my 1000+ local message archive. And, to be quite honest, I've _forgotten_ what the issue is. All you two are doing is yelling at each other that your not addressing each other's points, and not listening to each other, and being vaguely insulting the whole time.
It's not like either of you is really talking about whatever the hell it is that is bothering you. I, for one, am very tired of it. So, here's the deal. Explain it to _me_. What the hell are you arguing about, anyway? Start at ground zero (tho I have also written refernce-counting garbage collectors for other programming systems), and explain to me what the big fuss is about. Because I doubt it's really that big a fuss. What is the problem that needs solving, what solution is proposed, and what do people think is wrong with it? And, frankly, my new monitor just arrived. So my development machine will get set up tonight, and I'll be climing back into the driver's seat for usb-storage, so I should probably figure out what all this crap is about anyway. Either that, or you can drop your pants and we'll use a ruler to see who's is bigger. Because that's all that this 'discussion' feels like anymore. Matt On Thu, May 09, 2002 at 04:51:35PM -0700, David Brownell wrote: > > > Will you ever constructively respond to the points I made > > > about uhci_free_dev()? OR are you going to continue to > > > ignore the fact that when it returns, "uhci" is still holding > > > on to device resources it's just said it has freed? > > > > > > (And no, changing the definition of that routine is not > > > in the category of "constructive".) > > > > I have really no idea how else to explain this to you. I've tried a > > couple of different ways but you still don't understand. Everyone else > > I've asked seems to understand me, so I can only guess that the problem > > isn't with my explanation. > > Then you're omitting some parts of the problem. Such as the > one you're still omitting in this "response" ... > > Feels to me like I'm describing "elephant" and you're focusing > on "grey", then you're blaming me for saying that "grey" isn't > the whole story. > > > > The fact you're focusing on uhci_free_dev() and that it doesn't do anything > > reflects that you don't understand how reference counting works. > > And the fact that you're STILL not responding to a point-blank question > reflects ... well, I'll avoid going there, since I don't want to come across > to you like you're coming across to me. > > (Hey, that's the bogus dynamic: I'm expecting a response to a question, > you're giving a response to a different question!) > > > > Like I've said, go read up on reference counting theory and then take a > > look how it's implemented elsewhere in the kernel. > > Let's just say I've worked with quite a few reference counting systems, > implemented them too, and don't need a freshman tutorial on them. > You could look at the 2.5.current "ehci-mem.c", and you'd see proof > of at least that much. > > So assuming that's a problem comes across as condescending. > (And so it is counter-productive.) > > - Dave > > > > > > _______________________________________________________________ > > Have big pipes? SourceForge.net is looking for download mirrors. We supply > the hardware. You get the recognition. Email Us: [EMAIL PROTECTED] > _______________________________________________ > [EMAIL PROTECTED] > To unsubscribe, use the last form field at: > https://lists.sourceforge.net/lists/listinfo/linux-usb-devel -- Matthew Dharm Home: [EMAIL PROTECTED] Maintainer, Linux USB Mass Storage Driver Umm, these aren't the droids you're looking for. -- Bill Gates User Friendly, 11/14/1998
msg06351/pgp00000.pgp
Description: PGP signature
