From mboxrd@z Thu Jan 1 00:00:00 1970 From: Fam Zheng Subject: Re: sg driver, sg_io and sg tablesize Date: Thu, 2 Mar 2017 19:41:39 +0800 Message-ID: <20170302114139.GA22977@lemon.lan> References: <791707934.40662785.1488383072891.JavaMail.zimbra@redhat.com> <50601133.40845997.1488406291147.JavaMail.zimbra@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from mx1.redhat.com ([209.132.183.28]:55574 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752068AbdCBLuk (ORCPT ); Thu, 2 Mar 2017 06:50:40 -0500 Content-Disposition: inline In-Reply-To: <50601133.40845997.1488406291147.JavaMail.zimbra@redhat.com> Sender: linux-scsi-owner@vger.kernel.org List-Id: linux-scsi@vger.kernel.org To: Laurence Oberman Cc: dgilbert@interlog.com, Linux SCSI List , Ewan Milne On Wed, 03/01 17:11, Laurence Oberman wrote: > This was suggested by Ewan and is the best way to know what the actual max I/O > size for sg_io would be for the LPFC driver. This sounds silly, why is there a "lame" max I/O size as in max_sectors_kb, that is unusable because of a more limiting max_segments, and an "actual max I/O size" as derived from the latter? What is the reason of the contradiction here? Fam