On Wed, Mar 27, 2002 at 03:33:20PM +0100, Nemosoft Unv. wrote:
>
> Then why does nobody do it? It's all I'm asking....
It's a pain in the butt at times, and much faster to fix something that
you just broke, by patching everyplace in the kernel you broke.
Look, someone changed the v4l core code. This broke your driver (and a
lot of other drivers.) So they then went and fixed them all up. This
is usually one of the requirements for someone changing a kernel api.
Be _glad_ for this! When me and Oliver changed some of the USB core api
calls, we went through and fixed up all of the kernel drivers that used
them. If we were to have to go through and send patches through all of
the individual maintainers of the drivers, it would have taken a lot
longer for that development to be done. It also would have left a lot
of people with broken drivers, while we insured that no one had a broken
driver (much nicer IMHO.)
I will be glad to accept a patch from you that states at the top of the
pwc files something to this effect:
DO NOT CHANGE THIS FILE! SEND ALL PATCHES THROUGH...
But you will find yourself doing more work than before :)
Might I suggest a good work environment that allows you to merge kernel
updates easily and allow you to see when changes happen to your code?
Yes, I'm not saying that it was not a "nice" thing to not notify you
when someone touched your driver, but again take a look at the changelog
showing the people who touched your code. I don't want to tell those
people to not change your code :)
greg k-h
_______________________________________________
[EMAIL PROTECTED]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
