From mboxrd@z Thu Jan 1 00:00:00 1970 From: Florian Fainelli Subject: Re: [PATCH 1/4] MIPS: BMIPS: bcm7346: add nodes for NAND Date: Fri, 26 Jun 2015 09:32:37 -0700 Message-ID: <558D7EA5.9030202@gmail.com> References: <1544bf6110b43fbaa8dbb3b06a18e08ae87b386d.1435124524.git.jaedon.shin@gmail.com> <558B05B7.8010401@gmail.com> <558CC050.6040101@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Jaedon Shin Cc: Ralf Baechle , Kevin Cernekee , linux-mips-6z/3iImG2C8G8FEW9MqTrA@public.gmane.org, devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, computersforpeace-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org List-Id: devicetree@vger.kernel.org On 26/06/15 06:54, Jaedon Shin wrote: >=20 >> On Jun 26, 2015, at 12:00 PM, Florian Fainelli wrote: >> >> Le 06/25/15 01:52, Jaedon Shin a =C3=A9crit : >>> >>>> On Jun 25, 2015, at 4:32 AM, Florian Fainelli wrote: >>>> >>>> +Brian, >>>> >>>> On 23/06/15 23:08, Jaedon Shin wrote: >>>>> Add NAND device nodes to BMIPS based BCM7346 platform. >>>>> >>>>> Signed-off-by: Jaedon Shin >>>>> --- >>>> >>>> [snip] >>>> >>>>> + >>>>> +&nand0 { >>>>> + status =3D "okay"; >>>>> + >>>>> + nandcs@1 { >>>>> + compatible =3D "brcm,nandcs"; >>>>> + reg =3D <1>; >>>>> + nand-ecc-step-size =3D <512>; >>>>> + nand-ecc-strength =3D <8>; >>>>> + nand-on-flash-bbt; >>>>> + >>>>> + #size-cells =3D <2>; >>>>> + #address-cells =3D <2>; >>>>> + >>>>> + flash1.rootfs0@0 { >>>>> + reg =3D <0x0 0x0 0x0 0x80000000>; >>>>> + }; >>>>> + >>>>> + flash1.rootfs1@80000000 { >>>>> + reg =3D <0x0 0x80000000 0x0 0x80000000>; >>>>> + }; >>>>> + }; >>>>> +}; >>>> >>>> Should we create something like brcmnand-cs1-512-8 to reduce the a= mount >>>> of duplication between DTS files? >>>> --=20 >>>> Florian >>> >>> I Think that is not duplication. >>> >>> I have no reference boards, but this node is maybe explaining for h= ardware >>> description of the BCM97346DBSMB reference board. The nodes are cha= nged by >>> EBI CS and ECC capabilities of NAND flash. I used brcmnand-cs2-512-= 4 and >>> brcmnand-cs1-512-4 for others. >> >=20 > The *others* means BCM7346 based set-top box of manufacturer. they ar= e not > bcm973XX{dbsmb,svmb} boards. Understood. >=20 >> Then I am confused, your 4 patches add identical NAND flash chip >> properties for 7346, 7358, 7360 and 7362: CS#1, 512 bytes of ECC ste= p >> size and 8 bits of ECC strength, am I missing something? >> --=20 >> Florian >=20 > The mips based reference boards have postfix of DBSMB, SVMB, SFF and = others. > they have different properties of DDR, #CS, SPI-NOR, NOR, NAND. if we= write > the DT of bcm97346sff.dts, therefore it has different fields. >=20 > Then I expected that mips 40nm based SVMB type reference boards have = the same > properties. but, I don't have confidence. AISE I have no reference bo= ards > unfortunately. if you can contact with people who have reference boar= d, would > tell me information of #CS and NAND. I agree with you that outside of reference boards, these flash settings will vary, however, Broadcom reference boards tend to follow the same design from one chip/board to another, so instead of replicating 4 time= s the same DT snippet in Device Tree, we could come up with differerent sets of include files which abstract commonly found NAND flash types an= d properties. Code might talk better, so I will try to submit something that illustrates what I have in mind. Thanks! --=20 =46lorian -- To unsubscribe from this list: send the line "unsubscribe devicetree" i= n the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html