From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751495AbcCADck (ORCPT ); Mon, 29 Feb 2016 22:32:40 -0500 Received: from kvm5.telegraphics.com.au ([98.124.60.144]:35369 "EHLO kvm5.telegraphics.com.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750888AbcCADci (ORCPT ); Mon, 29 Feb 2016 22:32:38 -0500 Date: Tue, 1 Mar 2016 14:32:23 +1100 (AEDT) From: Finn Thain To: "Martin K. Petersen" cc: James Bottomley , Michael Schmitz , linux-m68k@vger.kernel.org, linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 0/6] ncr5380: Exception handling fixes for v4.5 In-Reply-To: Message-ID: References: <20160222230703.994951661@telegraphics.com.au> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, 29 Feb 2016, Martin K. Petersen wrote: > >>>>> "Finn" == Finn Thain writes: > > Finn> These patches fix some exception handling and autosense bugs that > Finn> I accidentally introduced in v4.5-rc1. > > Finn> drivers/scsi/NCR5380.c | 133 +++++++++++++++++++------------------------ > Finn> drivers/scsi/atari_NCR5380.c | 133 +++++++++++++++++++------------------------ > Finn> 2 files changed, 118 insertions(+), 148 deletions(-) > > This is a pretty big lump of changes for a 4.5 bug fix! I'm sure this is a lot of rework when compared to the high standard set by well-funded corporate contributions. However, I don't have anyone paying me to write and execute thorough test plans for error paths on a wide variety of different hardware platforms. Nonetheless, these error paths have now been thoroughly tested on several platforms. Please keep in mind that these fixes are all rework of the changes I made in -rc1. So this submission is 2 files changed, 118 insertions(+), 148 deletions(-) which is all rework of the previous submission, which was 18 files changed, 2940 insertions(+), 3688 deletions(-) Better than 5% rate of rework. Bugs-per-line-of-code should drop quite quickly at that rate, presuming fixes get merged. > > Given the limited exposure based on the nature of the affected hardware > I have queued them for 4.6. > The affected hardware doesn't matter (though it is likely outside of any commercial support contract). It's the affected users that interest me. --