From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: MIME-Version: 1.0 In-Reply-To: <20130718183845.8FF0379@viridian.itc.virginia.edu> References: <20130718183333.GA19558@kroah.com> <20130718183845.8FF0379@viridian.itc.virginia.edu> Date: Thu, 18 Jul 2013 16:16:38 -0400 Message-ID: Subject: Re: [PATCH] Adds Digi driver to staging. From: Lidza Louina Content-Type: multipart/alternative; boundary=20cf307c9ef83b6e1004e1cee3f8 List-ID: To: Bill Pemberton Cc: Greg KH , Mark Hounschell , driverdev-devel@linuxdriverproject.org --20cf307c9ef83b6e1004e1cee3f8 Content-Type: text/plain; charset=ISO-8859-1 I've got all of your feedback. I'll send my patch up again with * SVN directory removed * Makefile modified so it builds into the kernel * Better description of the driver * Attribute the original author * .h files in the main folder If there's anything else I need to add/modify, please let me know. Thanks, Lidza On Thu, Jul 18, 2013 at 2:38 PM, Bill Pemberton < wfp5p@viridian.itc.virginia.edu> wrote: > Greg KH writes: > > > > On Thu, Jul 18, 2013 at 02:25:11PM -0400, Mark Hounschell wrote: > > > On 07/18/2013 01:40 PM, Greg KH wrote: > > > > On Thu, Jul 18, 2013 at 08:26:55AM -0400, Mark Hounschell wrote: > > > >> On 07/17/2013 05:49 PM, Greg KH wrote: > > > >>> On Wed, Jul 17, 2013 at 04:18:01PM -0400, Lidza Louina wrote: > > > >>>> This patch adds the Digi driver to staging. > > > >>> > > > >>> Which digi driver? For what hardware? What type of driver is it? > > > >>> Please provide some more details here. > > > >>> > > > >> > > > >> From the readme: > > > >> > > > >> This Digi software package includes device drivers for the > PCI models > > > >> of the Digi Neo and ClassicBoard series of products. > > > >> It is currently supported on the following hardware > platforms: > > > >> > > > >> o Standard i386/i486 and Pentium PC (x86 32bit) > > > >> o x86 64bit > > > > > > > > I was "hinting" that this needed to be explained in the changelog > entry. > > > > The part about the architectures is not needed, as, unless something > > > > really strange is going on in the driver, it should work on all cpus > > > > that support PCI, right? > > > > > > > > > > That I cannot attest to. The readme also says: > > > > > > o Standard i386/i486 and Pentium PC (x86 32bit) > > > o x86 64bit > > > > > > I can only personally vouch for Intel/Amd CPUs running in 32 bit mode. > > > > I'll look at the driver code, but odds are, unless you are assuming a > > specific endian format of some data, all should be fine on other cpu > > types. > > > > If it's anything like dgrp, it will have some places that it assumes a > specific endian format. > > -- > Bill > --20cf307c9ef83b6e1004e1cee3f8 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I've got all of your feedback. I'll send= my patch up again with=A0

* SVN directory removed

* Makefile modified so it builds into the kernel

* Better description of the driver

* Attribute the original author

* .h files in the main folder

If there's anything else I need to add/modify, please let me know.

Thanks,

Lidza

On Thu, Jul 18, 2013 at 2:38 PM, Bill Pemberton <= wfp5p@viridian.itc.virginia.edu> wrote:

Greg= KH writes:
>
> On Thu, Jul 18, 2013 at 02:25:11PM -0400, Mark Hounschell wrote:
> > On 07/18/2013 01:40 PM, Greg KH wrote:
> > > On Thu, Jul 18, 2013 at 08:26:55AM -0400, Mark Hounschell wr= ote:
> > >> On 07/17/2013 05:49 PM, Greg KH wrote:
> > >>> On Wed, Jul 17, 2013 at 04:18:01PM -0400, Lidza Loui= na wrote:
> > >>>> This patch adds the Digi driver to staging.
> > >>>
> > >>> Which digi driver? =A0For what hardware? =A0What typ= e of driver is it?
> > >>> Please provide some more details here.
> > >>>
> > >>
> > >> =A0 From the readme:
> > >>
> > >> =A0 =A0 =A0 =A0This Digi software package includes devic= e drivers for the PCI models
> > >> =A0 =A0 =A0 =A0of the Digi Neo and ClassicBoard series o= f products.
> > >> =A0 =A0 =A0 =A0It is currently supported on the followin= g hardware platforms:
> > >>
> > >> =A0 =A0 =A0 =A0o Standard i386/i486 and Pentium PC (x86 = 32bit)
> > >> =A0 =A0 =A0 =A0o x86 64bit
> > >
> > > I was "hinting" that this needed to be explained i= n the changelog entry.
> > > The part about the architectures is not needed, as, unless s= omething
> > > really strange is going on in the driver, it should work on = all cpus
> > > that support PCI, right?
> > >
> >
> > That I cannot attest to. The readme also says:
> >
> > =A0 =A0 =A0o Standard i386/i486 and Pentium PC (x86 32bit)
> > =A0 =A0 =A0 o x86 64bit
> >
> > I can only personally vouch for Intel/Amd CPUs running in 32 bit = mode.
>
> I'll look at the driver code, but odds are, unless you are assumin= g a
> specific endian format of some data, all should be fine on other cpu > types.
>

If it's anything like dgrp, it will have some places that i= t assumes a
specific endian format.

--
Bill

--20cf307c9ef83b6e1004e1cee3f8--