Re: [PATCH] base/class.c: prevent ooops due to insert/remove race (v3)

On Thu, 29 Nov 2007, Mark Lord wrote:

> > So again, the problem is in the higher up scsi layer, and that is where
> > the problem should already be fixed.
> ..
> 
> Ahhh.. so you figure the Oops should also have been fixed
> as part of the 2.6.24 SCSI fixes ?  That's what I was missing here.

Yes indeed.  I wish I could point you to the exact patch containing the 
fix, but the git software seems to have lost track of it (it's combined
in with a large number of other patches with no obvious way to separate 
it out).  It's also available in the various mailing list archives, but 
I don't have a pointer to it and there's no reasonable way to search 
for it.

The patch in question was written by Matthew Wilcox; it added code to 
the SCSI async-scanning routines to utilize the scan_mutex.  IMO it 
should have been applied to 2.6.23 but it wasn't.

Alan Stern


-------------------------------------------------------------------------
SF.Net email is sponsored by: The Future of Linux Business White Paper
from Novell.  From the desktop to the data center, Linux is going
mainstream.  Let it simplify your IT future.
http://altfarm.mediaplex.com/ad/ck/8857-50307-18918-4
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel