From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stephen Neuendorffer Subject: RE: Xilinx SYSACE driver and 8-bit attachment Date: Wed, 7 Oct 2009 09:18:40 -0700 Message-ID: <51ada07b-493e-4552-b79b-7a4044ae1b4c@VA3EHSMHS028.ehs.local> References: <1d3f23370910070013u6307bc42hf7ab243b52d007df@mail.gmail.com> <192797ad-5dd4-4ebe-8db3-d381bb949e1e@VA3EHSMHS029.ehs.local> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Class: urn:content-classes:message In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Grant Likely Cc: David DeBonis , devicetree-discuss , Linux Kernel list , Michal Simek , John Linn , John Williams List-Id: devicetree@vger.kernel.org > -----Original Message----- > From: devicetree-discuss-bounces+stephen.neuendorffer=3Dxilinx.com@li= sts.ozlabs.org [mailto:devicetree- > discuss-bounces+stephen.neuendorffer=3Dxilinx.com@lists.ozlabs.org] O= n Behalf Of Grant Likely > Sent: Wednesday, October 07, 2009 7:53 AM > To: John Linn > Cc: David DeBonis; Stephen Neuendorffer; devicetree-discuss; Linux Ke= rnel list; Michal Simek; John > Linn; John Williams > Subject: Re: Xilinx SYSACE driver and 8-bit attachment >=20 > On Wed, Oct 7, 2009 at 8:42 AM, John Linn wrot= e: > >> -----Original Message----- > >> From: John Linn > >> Sent: Wednesday, October 07, 2009 8:24 AM > >> To: 'Grant Likely'; John Williams > >> Cc: Linux Kernel list; Michal Simek; David DeBonis; Stephen Neuend= orffer; devicetree-discuss > >> Subject: RE: Xilinx SYSACE driver and 8-bit attachment > >> > >> I'll look to see if something has changed. =A0It's not clear to me= if it has ever been right. > >> > >> It look like we normally use 16 bit mode as that's how BSB generat= es the system by default. > >> > >> Thanks, > >> John > >> > >> > -----Original Message----- > >> > From: glikely@secretlab.ca [mailto:glikely@secretlab.ca] On Beha= lf Of Grant Likely > >> > Sent: Wednesday, October 07, 2009 7:35 AM > >> > To: John Williams > >> > Cc: Linux Kernel list; Michal Simek; John Linn; David DeBonis; S= tephen Neuendorffer; devicetree- > >> > discuss > >> > Subject: Re: Xilinx SYSACE driver and 8-bit attachment > >> > > >> > [Added devicetree-discuss to cc: list] > >> > > >> > On Wed, Oct 7, 2009 at 1:13 AM, John Williams > >> > wrote: > >> > > Hi Grant (it's your driver :) and David D ( and your DTS gener= ator :) > >> > [...] > >> > > if (of_find_property(op->node, "8-bit", NULL)) > >> > [...] > >> > > doesn't match the properties generated by Xilinx's device tree= generator: > >> > [...] > >> > > =A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0=A0= =A0=A0 xlnx,mem-width =3D <0x8>; > > > > Do you know if the 8 bit mode was ever tested with the "8-bit" prop= erty in the tree? > > > > I don't see anything in the device tree generator history to say we= ever did that "8-bit", but > maybe I'm missing something. > > > > We are just generating the parameters from the h/w and the memory w= idth is it. =A0We can always put > something in there special, but it seems silly if it was never used a= nyway. >=20 > Don't forget that Virtex and Microblaze are not the only users of thi= s > driver. There is a AMCC 440 board which uses the sysace as a CF > adapter. 8-bit was definitely tested and is in use. In this case, I'm pretty sure this tag was never automatically generate= d by the DTS generator. Personally, I think it's bad practice to always assume that the device = tree generator should be modified to match what a Linux driver is expec= ting, however given the precedence in this case, I agree that it might be reasonable to put a special case in to generat= e this. Steve =20 This email and any attachments are intended for the sole use of the nam= ed recipient(s) and contain(s) confidential information that may be pro= prietary, privileged or copyrighted under applicable law. If you are no= t the intended recipient, do not read, copy, or forward this email mess= age or any attachments. Delete this email message and any attachments i= mmediately.