From mboxrd@z Thu Jan  1 00:00:00 1970
From: hch@infradead.org (Christoph Hellwig)
Date: Mon, 15 May 2017 07:59:52 -0700
Subject: Unexpected issues with 2 NVME initiators using the same target
In-Reply-To: <20170515143632.GH3616@mtr-leonro.local>
References: <1948057100.362217696.1491824459274.JavaMail.zimbra@kalray.eu>
 <33e2cc35-f147-d4a4-9a42-8f1245e35842@mellanox.com>
 <517fb6d2-a010-1926-b4cb-0e6041722c28@grimberg.me>
 <af9044c2-fb7c-e8b3-d8fc-4874cfd1bb67@mellanox.com>
 <c2a1436f-48a6-bdea-c910-168fc6031364@grimberg.me>
 <DE927C68B458BE418D582EC97927A92855088C6F@ORSMSX113.amr.corp.intel.com>
 <82dd5b24-5657-ae5e-8a33-646fddd8b75b@grimberg.me>
 <20170515133122.GG3616@mtr-leonro.local>
 <9465cd0c-83db-b058-7615-5626ef60dbb0@grimberg.me>
 <20170515143632.GH3616@mtr-leonro.local>
Message-ID: <20170515145952.GA7871@infradead.org>

On Mon, May 15, 2017@05:36:32PM +0300, Leon Romanovsky wrote:
> I understand you and both Max and me are feeling the same as you. For more
> than 2 months, we constantly (almost on daily basis) asked for a solution from
> architecture group, but received different answers. The proposals were
> extremely broad from need for strong fence for all cards to no need for
> strong fence at all.

So let's get the patch to do a strong fence everywhere now, and relax
it later where possible.

Correntness before speed..