First off, sorry for the delay, been swamped with non-usb stuff...
On Wed, Mar 05, 2003 at 12:35:26PM -0800, David Brownell wrote:
Greg, you seemed willing to merge the first version of this when it came around (2.5.50 or so), though you seemed to agree that it'd be worth trying to have that primitive outside of usbcore. At this point I'd rather merge it sooner, then maybe move it later.
I was willing when it moved the io_wait_completion() into the kernel core, and we modified the USB code to use it. As it stands, this patch doesn't buy us anything, nor change the way anything works, right?
It buys us a replacement for some code that I've never trusted, since it's playing rude games. It's the difference between code that I think is buggy *every* time I read it, and code that uses standard completion logic (which doesn't give me that negative reaction).
On the other hand, we know that at least some of those bulk/control diagnostics are likely not caused by that chunk of code, so maybe it's OK to just keep walking quietly around that logic.
Yes, I think io_wait_completion() looks like the right thing to have in the kernel core, just that I'm not the person to put it there :)
So sorry, I'm not going to apply this.
Well you shouldn't apply that version at any rate, I noticed that somewhere along the line the "free that urb" logic got mislaid!
Do I understand that if Linus (say) were to take the io_wait_completion() logic into kernel/sched.c (presumably), then you'd be willing to apply a USB patch to use it?
If so, I'll post that patch to LKML ... Linus isn't away for the winter holidays right now, so this kind of update might stand a better chance of getting in!
- Dave
thanks,
greg k-h
-------------------------------------------------------
This SF.net email is sponsored by:Crypto Challenge is now open! Get cracking and register here for some mind boggling fun and the chance of winning an Apple iPod:
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
