From mboxrd@z Thu Jan 1 00:00:00 1970 From: Andy Grover Subject: Re: [PATCH] block: Export REQ_FLUSH status via sysfs Date: Wed, 14 May 2014 17:29:16 -0700 Message-ID: <53740A5C.1040103@redhat.com> References: <1400107908-16095-1-git-send-email-agrover@redhat.com> <1400112663.7127.64.camel@haakon3.risingtidesystems.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1400112663.7127.64.camel@haakon3.risingtidesystems.com> Sender: target-devel-owner@vger.kernel.org To: "Nicholas A. Bellinger" Cc: linux-scsi@vger.kernel.org, target-devel@vger.kernel.org List-Id: linux-scsi@vger.kernel.org On 05/14/2014 05:11 PM, Nicholas A. Bellinger wrote: > On Wed, 2014-05-14 at 15:51 -0700, Andy Grover wrote: >> Whether this is set or not is important for whether good performance is >> possible, when using the block device as a backstore for the kernel >> target. Exposing this will let kernel target configuration tools provide >> this information to the user when configuring LIO targets. >> > > The whole point is that IBLOCK detects if WCE is enabled on the > underlying block_device at creation time using these same flags, and > sets its own WCE bit automatically without any user interaction. > > The user should not be allowed to change the WCE bit to something > different from what the underlying block_device is reporting. > > That said, I don't see the point of this patch for target related logic. My understanding is that a LIO target backed by a block device that doesn't support REQ_FLUSH will have subpar performance. Userspace configuration tools could potentially warn the user if this is the case. This information doesn't appear to be currently available to userspace. Regards -- Andy