atom feed24 messages in org.freebsd.freebsd-archkldunload DIAGNOSTIC idea...
FromSent OnAttachments
Poul-Henning KampJul 20, 2004 11:21 am 
Brian Fundakowski FeldmanJul 20, 2004 11:32 am 
Poul-Henning KampJul 20, 2004 11:40 am 
Brian Fundakowski FeldmanJul 20, 2004 11:53 am 
Willem Jan WithagenJul 20, 2004 12:08 pm 
Scott LongJul 20, 2004 12:09 pm 
Scott LongJul 20, 2004 12:11 pm 
David SchultzJul 20, 2004 1:40 pm 
Andrea CampiJul 20, 2004 11:18 pm 
Poul-Henning KampJul 20, 2004 11:47 pm 
Julian ElischerJul 21, 2004 12:09 am 
Doug RabsonJul 21, 2004 2:10 am 
Poul-Henning KampJul 21, 2004 2:22 am 
Doug RabsonJul 21, 2004 3:37 am 
Poul-Henning KampJul 21, 2004 4:02 am 
Doug RabsonJul 21, 2004 4:20 am 
Poul-Henning KampJul 21, 2004 4:37 am 
Doug RabsonJul 21, 2004 5:20 am 
Poul-Henning KampJul 21, 2004 5:30 am 
M. Warner LoshJul 21, 2004 7:51 am 
Brian Fundakowski FeldmanJul 21, 2004 8:08 am 
Poul-Henning KampJul 21, 2004 9:38 am 
M. Warner LoshJul 21, 2004 10:14 am 
Poul-Henning KampJul 21, 2004 10:16 am 
Subject:kldunload DIAGNOSTIC idea...
From:Poul-Henning Kamp (ph@phk.freebsd.dk)
Date:Jul 21, 2004 2:22:06 am
List:org.freebsd.freebsd-arch

In message <2004@nlsystems.com>, Doug Rabson writes:

The original intention was that drivers use the device_busy()/device_unbusy() counter to handle these things. In some cases, just calling device_busy() from fooopen() and device_unbusy() from fooclose() is sufficient.

That is not enough. All methods in cdevsw, and things not in cdevsw (clone handlers, call backs, etc etc) needs to refcount.

I have a lot of this working in a tree here, and will commit it once I have gone over it a few more times.