At Wed, 04 Apr 2012 17:49:29 +0800, Liu Yuan wrote: > > On 04/04/2012 01:19 AM, MORITA Kazutaka wrote: > > > Hmm, does this work well when multiple write requests arrive at the > > same time? I cannot come up with a better approach, though. > > Hi Kazum, > > I give a second try, finding that Guest(RHEL 6) will wait after issue > a write request till its completion at very eary bootup before 'login'. > I didn't see concurrent IO from the log at that stage, so I recommend > not do over-design without a practical real case that cause inconsistency. We can use VDIs other than root disks. > > Here is the V2 path > -V2 > correct object_cache_lookup() usage > > From a167146c55f0354a6f44fa7b28cd9a3a4e4ee089 Mon Sep 17 00:00:00 2001 > From: Liu Yuan <tailai.ly at taobao.com> > Date: Wed, 4 Apr 2012 17:11:45 +0800 > Subject: [PATCH v2] object cache: add flush_and_delete operation > > If 1) VDI is opened without cache enabled and 2) we unfortunately have > a cache for it previously, we should flush the cache then delete it. > > Signed-off-by: Liu Yuan <tailai.ly at taobao.com> > --- > sheep/object_cache.c | 42 ++++++++++++++++++++++++++++++++++++++++++ > sheep/sheep_priv.h | 1 + > sheep/store.c | 30 +++++++++++++++++++++--------- > 3 files changed, 64 insertions(+), 9 deletions(-) Applied, thanks. But let's fix the concurrent write problem in future patch. Kazutaka |