[fix] Too async libata breakage

[Arjan van de Ven <arjan@infradead.org>]

> > Additional async_synchronize_full() helps and box boot to the end.

Hi Linus,

can you add the patch below to your tree? This makes the
async_synchronize_full() a real absolute async barrier, rather than
just synchronizing the work scheduled upto the starting point.
It's a bit too much for some cases (so later on I might introduce the
softer synchronization again) but it solves some real problems
for people so this should go into -rc1 please...

>From ca1827b142e41b54c612026b3c0b5d239f21a715 Mon Sep 17 00:00:00 2001
From: Arjan van de Ven <arjan@linux.intel.com>
Date: Thu, 8 Jan 2009 12:35:11 -0800
Subject: [PATCH] async: make async_synchronize_full() more serializing

turns out that there are real problems with allowing async
tasks that are scheduled from async tasks to run after
the async_synchronize_full() returns.

This patch makes the _full more strict and a complete
synchronization. Later I might need to add back a lighter
form of synchronization for other uses.. but not right now.

Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
---
 kernel/async.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/kernel/async.c b/kernel/async.c
index 9737338..64cc916 100644
--- a/kernel/async.c
+++ b/kernel/async.c
@@ -206,7 +206,9 @@ EXPORT_SYMBOL_GPL(async_schedule_special);
 
 void async_synchronize_full(void)
 {
-	async_synchronize_cookie(next_cookie);
+	do {
+		async_synchronize_cookie(next_cookie);
+	} while (!list_empty(&async_running) || !list_empty(&async_pending));
 }
 EXPORT_SYMBOL_GPL(async_synchronize_full);
 
-- 
1.5.5.1



-- 
Arjan van de Ven 	Intel Open Source Technology Centre
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org