From mboxrd@z Thu Jan 1 00:00:00 1970 From: Pat LaVarre Subject: KiB/cdb limits Date: 14 Oct 2003 17:45:33 -0600 Sender: linux-scsi-owner@vger.kernel.org Message-ID: <1066175133.3399.12.camel@patehci2> Mime-Version: 1.0 Content-Type: text/plain Content-Transfer-Encoding: 7bit Return-path: Received: from email-out1.iomega.com ([147.178.1.82]:56194 "EHLO email.iomega.com") by vger.kernel.org with ESMTP id S261824AbTJNXpr (ORCPT ); Tue, 14 Oct 2003 19:45:47 -0400 List-Id: linux-scsi@vger.kernel.org To: dougg@torque.net Cc: linux-scsi@vger.kernel.org > I find some bugs "interesting". I fell into that possibly fatal 'sg_dd bpt= count=' by mistyping a command line designed to help me nail down what the KiB/cdb limit is. Looked to me like cdb's didn't match command line for such command lines as: sudo sg_dd of=/dev/sg0 if=/dev/zero bs=2k bpt=$n count=$n I thought I saw trouble arose somewhere down below something small, like below 8 MiB = 0x1000 blocks per cdb. (As you know, the t10 claim of cdb max for ops like x2A Write 10 is xFFFF.) If noone pops up here to beat me to the conclusion, I will myself try this again in a few days and post to confirm/ deny this tentative first impression of mine. Pat LaVarre