Cc: Paul Mackerras <paulus@...>, <heiko.carstens@...>, <horms@...>, Stefan Richter <stefanr@...>, Satyam Sharma <satyam@...>, Linux Kernel Mailing List <linux-kernel@...>, David Miller <davem@...>, Paul E. McKenney <paulmck@...>, <ilpo.jarvinen@...>, <ak@...>, <cfriesen@...>, <rpjday@...>, Netdev <netdev@...>, <jesper.juhl@...>, <linux-arch@...>, Andrew Morton <akpm@...>, <zlynx@...>, <schwidefsky@...>, Chris Snook <csnook@...>, Herbert Xu <herbert@...>, Linus Torvalds <torvalds@...>, <wensong@...>, <wjiang@...>
>> "volatile" has nothing to do with reordering. atomic_dec() writes
Read again: I said the C "volatile" construct has nothing to do
with CPU memory access reordering.
The _compiler_ isn't allowed to reorder things here. Yes, of course
you do need stronger barriers for many purposes, volatile isn't all
that useful you know.
Segher
-