[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <20160526141922.163198062@infradead.org>
Date: Thu, 26 May 2016 16:19:22 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: linux-kernel@...r.kernel.org, torvalds@...ux-foundation.org,
manfred@...orfullife.com, dave@...olabs.net,
paulmck@...ux.vnet.ibm.com, will.deacon@....com
Cc: boqun.feng@...il.com, Waiman.Long@....com, tj@...nel.org,
pablo@...filter.org, kaber@...sh.net, davem@...emloft.net,
oleg@...hat.com, netfilter-devel@...r.kernel.org,
sasha.levin@...cle.com, hofrat@...dl.org,
"Peter Zijlstra (Intel)" <peterz@...radead.org>
Subject: [PATCH -v2 0/6] spin_unlock_wait borkage
This version rewrites all spin_unlock_wait() implementations to provide the
acquire order against the spin_unlock() release we wait on, ensuring we can
fully observe the critical section we waited on.
It pulls in the smp_cond_acquire() rewrite because it introduces a lot more
users of it. All simple spin_unlock_wait implementations end up being an
smp_cond_load_acquire().
And while this still introduces smp_acquire__after_ctrl_dep() it keeps its
usage contained to a few sites.
Please consider.. carefully.
Powered by blists - more mailing lists