From mboxrd@z Thu Jan 1 00:00:00 1970 From: NamJae Jeon Subject: Re: mmc ioctl for hw reset ,partitiing, enhanced area setting, boot partition update. Date: Tue, 5 Apr 2011 16:53:13 +0900 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset=EUC-KR Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mail-ww0-f44.google.com ([74.125.82.44]:49997 "EHLO mail-ww0-f44.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752470Ab1DEHxP convert rfc822-to-8bit (ORCPT ); Tue, 5 Apr 2011 03:53:15 -0400 Received: by wwa36 with SMTP id 36so136068wwa.1 for ; Tue, 05 Apr 2011 00:53:14 -0700 (PDT) In-Reply-To: Sender: linux-mmc-owner@vger.kernel.org List-Id: linux-mmc@vger.kernel.org To: Andrei Warkentin , linux-mmc@vger.kernel.org, Chris Ball Hi. Andrei. Can I know current status of progress accepting partition mangement pat= ch ? Thanks. 2011=B3=E2 4=BF=F9 4=C0=CF =BF=C0=C8=C4 9:13, Andrei Warkentin =B4=D4=C0=C7 =B8=BB: > When the boot partitioning patch is accepted, there will be a short > patch to add GP partitions in. > > Thanks, > A > > 2011/4/4 NamJae Jeon : >> Hi. Andrei >> >> I checked this patch [RFC] MMC: MMC boot partitions support). Can it >> support gerneal purpose partition ? Boot partition be only supported >> in this patch. >> >> Is there another patch for general purpose partitions ? >> >> Thanks. >> >> 2011=B3=E2 4=BF=F9 3=C0=CF =BF=C0=C8=C4 12:03, NamJae Jeon =B4=D4=C0=C7 =B8=BB: >>> Thanks for your reply. >>> >>> 2011. 4. 1. =BF=C0=C8=C4 4:58=BF=A1 "Andrei Warkentin" =B4=D4=C0=CC =C0=DB=BC=BA: >>> >>>> >>>> I highly suggest you read the relevant section in JEDEC eMMC spec,= which >>>> is free. >>>> >>>> Writeable attributes in EXT_CSD are manipulated through CMD6. >>>> >>>> Thanks, >>>> A >>>> >>>> On Apr 1, 2011 2:51 AM, "NamJae Jeon" wrote= : >>>> > Hi. Andrei. >>>> > >>>> > For partitioning, you suggest I try to use ioctl of ACMD. >>>> > >>>> > I don't understand your means. because EXT_CSD is changed for >>>> > partitioning not CMD. >>>> > >>>> > Would you plz explain more ? >>>> > >>>> > Thanks. >>>> > >>>> > 2011/3/31 NamJae Jeon : >>>> >> Hi, Andrei. >>>> >> >>>> >> Thanks. I will check your information. >>>> >> >>>> >> >>>> >> 2011/3/31 Andrei Warkentin : >>>> >>> Hi, >>>> >>> >>>> >>> On Wed, Mar 30, 2011 at 10:11 AM, NamJae Jeon >>>> >>> wrote: >>>> >>>> usally, bootloader is stored in boot partition. I suggested i= octl to >>>> >>>> update bootloader in boot partition not modifing boot partiti= on size. >>>> >>>> >>>> >>> >>>> >>> Maybe I didn't explain myself clearly enough, sorry. The patch= I have >>>> >>> exposes the boot partitions >>>> >>> as block devices and you can read and write to them without af= fecting >>>> >>> the rest of the MMC card. >>>> >>> >>>> >>>> and moviNAND(samsung) is needed by boot parittion size settin= g. and >>>> >>>> when boot parition size is changed, all user data will be res= et, it >>>> >>>> will be initial status. >>>> >>>> >>>> >>>> but iNAND(sandisk) is not needed it. boot partition size is >>>> >>>> fixed(1MB). >>>> >>>> >>>> >>> >>>> >>> I am unsure what you're talking about. The eMMC 4.41 spec (see= 7.2) >>>> >>> does not allow changing boot partition size. >>>> >>> It is fixed to a certain size, and that size is specified by t= he >>>> >>> BOOT_SIZE_MULTI field in EXT_CSD. For the Toshiba devices >>>> >>> I have around me (MMC08G, MMC16G and MMC32G), each boot partit= ion (two >>>> >>> total) is 2MB in size. >>>> >>> >>>> >>> I don't think ioctl for accessing device partitions is a good = idea, >>>> >>> because the device partitions are just block devices, so they = should >>>> >>> be exposed as such. >>>> >>> >>>> >> >>>> >>> For device partitioning, to support create General Purpose par= titions >>>> >>> (up to 4), an ioctl is probably needed. You could look at "[PA= TCH >>>> >>> resend] mmc: Added ioctl to let userspace apps send ACMDs" by = John >>>> >>> Calixto which was last send March 17th. That's generic enough = to >>>> >>> support partitioning (which is a one time programming procedur= e >>>> >>> anyway...so there may be no value of adding a specific ioctl f= or it, >>>> >>> considering how dangerous of an operation it is). >>>> >>> >>>> >>> A >>>> >>> >>>> >> >>> >> >