Messages in this thread | | | Date | Mon, 8 Aug 2011 19:07:35 +0100 | From | Ben Hutchings <> | Subject | Re: [Stable-review] [07/55] SUNRPC: Fix a race between work-queue and rpc_killall_tasks |
| |
On Mon, Aug 08, 2011 at 10:03:00AM -0700, Greg KH wrote: > On Sun, Aug 07, 2011 at 06:38:01PM +0100, Ben Hutchings wrote: > > On Fri, 2011-08-05 at 17:01 -0700, Greg KH wrote: > > > 2.6.32-longterm review patch. If anyone has any objections, please let us know. > > > > > > ------------------ > > > > > > From: Trond Myklebust <Trond.Myklebust@netapp.com> > > > > > > commit b55c59892e1f3b6c7d4b9ccffb4263e1486fb990 upstream. > > > > > > Since rpc_killall_tasks may modify the rpc_task's tk_action field > > > without any locking, we need to be careful when dereferencing it. > > [...] > > > > This isn't nearly careful enough to avoid races. You must at least use > > the ACCESS_ONCE macro, otherwise the compiler can just optimise away the > > local variable this introduces. > > Well, that's an upstream issue as well, right? Not much I can do here > in the stable updates :)
Right, but if it doesn't fix a bug then it is not suitable for stable.
Ben.
-- Ben Hutchings We get into the habit of living before acquiring the habit of thinking. - Albert Camus
| |