From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: virtio-scsi spec (was Re: [PATCH] Add virtio-scsi to the virtio spec) Date: Thu, 01 Dec 2011 09:49:20 +0100 Message-ID: <4ED73F90.5050603@redhat.com> References: <1322661042-28191-1-git-send-email-pbonzini@redhat.com> <1322661042-28191-2-git-send-email-pbonzini@redhat.com> <4ED63AE1.8090105@suse.de> <4ED65BA4.3000003@redhat.com> <4ED74E54.3070303@suse.de> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4ED74E54.3070303@suse.de> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: virtualization-bounces@lists.linux-foundation.org Errors-To: virtualization-bounces@lists.linux-foundation.org To: Hannes Reinecke Cc: "Michael S. Tsirkin" , LKML , Stefan Hajnoczi , virtualization List-Id: virtualization@lists.linuxfoundation.org On 12/01/2011 10:52 AM, Hannes Reinecke wrote: >>>> >>> I would like to have the other request_queue limitations exposed >>> here, too. >>> Most notably we're missing the maximum size of an individual segment >>> and the maximum size of the overall I/O request. >> >> The virtio transport does not put any limit, as far as I know. >> > Virtio doesn't, but the underlying device/driver might. > And if we don't expose these values we cannot format the request correctly. These limits should be per target/LUN, so it seems like material for another controlq command when the need arises. For now, I'd really prefer to have the spec match the implementation (plus a few SAM bogosities). Paolo