lkml.org 
[lkml]   [2011]   [Feb]   [14]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
DateMon, 14 Feb 2011 15:09:02 -0800
From"Paul E. McKenney" <>
SubjectRe: [PATCH 0/2] jump label: 2.6.38 updates
On Mon, Feb 14, 2011 at 06:03:01PM -0500, Mathieu Desnoyers wrote:
> * Matt Fleming (matt@console-pimps.org) wrote:
> > On Mon, 14 Feb 2011 13:46:00 -0800 (PST)
> > David Miller <davem@davemloft.net> wrote:
> >
> > > From: Steven Rostedt <rostedt@goodmis.org>
> > > Date: Mon, 14 Feb 2011 16:39:36 -0500
> > >
> > > > Thus it is not about global, as global is updated by normal means
> > > > and will update the caches. atomic_t is updated via the ll/sc that
> > > > ignores the cache and causes all this to break down. IOW... broken
> > > > hardware ;)
> > >
> > > I don't see how cache coherency can possibly work if the hardware
> > > behaves this way.
> >
> > Cache coherency is still maintained provided writes/reads both go
> > through the cache ;-)
> >
> > The problem is that for read-modify-write operations the arbitration
> > logic that decides who "wins" and is allowed to actually perform the
> > write, assuming two or more CPUs are competing for a single memory
> > address, is not implemented in the cache controller, I think. I'm not a
> > hardware engineer and I never understood how the arbitration logic
> > worked but I'm guessing that's the reason that the ll/sc instructions
> > bypass the cache.
> >
> > Which is why the atomic_t functions worked out really well for that
> > arch, such that any accesses to an atomic_t * had to go through the
> > wrapper functions.

???

What CPU family are we talking about here? For cache coherent CPUs,
cache coherence really is supposed to work, even for mixed atomic and
non-atomic instructions to the same variable.

Thanx, Paul

> If this is true, then we have bugs in lots of xchg/cmpxchg users (which
> do not reside in atomic.h), e.g.:
>
> fs/fs_struct.c:
> int current_umask(void)
> {
> return current->fs->umask;
> }
> EXPORT_SYMBOL(current_umask);
>
> kernel/sys.c:
> SYSCALL_DEFINE1(umask, int, mask)
> {
> mask = xchg(&current->fs->umask, mask & S_IRWXUGO);
> return mask;
> }
>
> The solution to this would be to force all xchg/cmpxchg users to swap to
> atomic.h variables, which would force the ll semantic on read. But I'd
> really like to see where this is documented first -- or which PowerPC
> engineer we should talk to.
>
> Thanks,
>
> Mathieu
>
> --
> Mathieu Desnoyers
> Operating System Efficiency R&D Consultant
> EfficiOS Inc.
> http://www.efficios.com


\
 
 \ /
  Last update: 2011-02-15 00:11    [from the cache]
©2003-2010