From mboxrd@z Thu Jan 1 00:00:00 1970 From: Kukjin Kim Subject: Re: [PATCH v12 2/3] ARM: EXYNOS: Change System MMU platform device definitions Date: Fri, 16 Mar 2012 13:47:50 -0700 Message-ID: <4F63A6F6.9050506@samsung.com> References: <000201cd0286$2e6fbb40$8b4f31c0$%cho@samsung.com> <20120315111227.GA2384@amd.com> <4F61FF1D.5000502@samsung.com> <4F630841.2070307@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <4F630841.2070307@linaro.org> Sender: linux-samsung-soc-owner@vger.kernel.org To: Subash Patel Cc: Kukjin Kim , Joerg Roedel , Cho KyongHo , linux-arm-kernel@lists.infradead.org, linux-samsung-soc@vger.kernel.org, iommu@lists.linux-foundation.org, linux-kernel@vger.kernel.org, 'Younglak Kim' , 'Kyungmin Park' , 'Sanghyun Lee' List-Id: iommu@lists.linux-foundation.org Subash Patel wrote: > Hi KyongHo, Kukjin, > > On 03/15/2012 08:09 PM, Kukjin Kim wrote: >> On 03/15/12 04:12, Joerg Roedel wrote: >>> On Thu, Mar 15, 2012 at 05:32:39PM +0900, Cho KyongHo wrote: >> >> KyongHo, I looked at 'Cho KyongHo ' and 'KyongHo Cho >> ' in this series, which one do you want to use? >> >>>> Handling System MMUs with an identifier is not flexible to manage >>>> System MMU platform devices because of the following reasons: >>>> 1. A device driver which needs to handle System MMU must know the ID. >>>> 2. A System MMU may not present in some implementations of Exynos >>>> family. >>>> 3. Handling System MMU with IOMMU API does not require an ID. >>>> >>>> This patch is the result of removing ID of System MMUs. >>>> Instead, a device driver that needs to handle its System MMU must >>>> use IOMMU API while its descriptor of platform device is given. >>>> >>>> This patch also includes the following enhancements: >>>> - A System MMU device becomes a child if its power domain device. >>>> - clkdev >>>> >>>> Cc: Kukjin Kim >>>> Signed-off-by: KyongHo Cho >>>> --- >>>> arch/arm/mach-exynos/Kconfig | 10 +- >>>> arch/arm/mach-exynos/Makefile | 2 +- >>>> arch/arm/mach-exynos/clock-exynos4.c | 79 ++-- >>>> arch/arm/mach-exynos/clock-exynos4.h | 2 + >>>> arch/arm/mach-exynos/clock-exynos4210.c | 11 + >>>> arch/arm/mach-exynos/clock-exynos4212.c | 28 ++- >>>> arch/arm/mach-exynos/clock-exynos5.c | 90 +++++ >>>> arch/arm/mach-exynos/dev-sysmmu.c | 457 ++++++++++++---------- >>>> arch/arm/mach-exynos/include/mach/irqs.h | 25 +- >>>> arch/arm/mach-exynos/include/mach/map.h | 38 ++ >>>> arch/arm/mach-exynos/include/mach/regs-clock.h | 5 + >>>> arch/arm/mach-exynos/include/mach/regs-sysmmu.h | 28 -- >>>> arch/arm/mach-exynos/include/mach/sysmmu.h | 88 +++-- >>>> arch/arm/mach-exynos/mach-armlex4210.c | 1 - >>>> arch/arm/mach-exynos/mach-smdkv310.c | 1 - >>>> 15 files changed, 529 insertions(+), 336 deletions(-) >>>> delete mode 100644 arch/arm/mach-exynos/include/mach/regs-sysmmu.h >>> >>> This patch doesn't apply cleanly against upstream Linux (v3.3-rcX). >>> Please rebase to upstream and resend. The other 2 patches apply fine. >>> >> >> Yes, as KyoungHo said, this has been created on top of Samsung tree >> for-next because this touches many exynos stuff which has been changed >> from Samsung tree for upcoming merge window. So if you're ok on this, >> 1st patch and this should be sent to upstream via samsung tree even >> though I can't find any topic branch for this yet. And if you want >> topic branch for your tree, let me know, I can provide it. >> >> Thanks. > Wouldn't it be good to send exynos4 SYSMMU patches now, and send only > the exynos5 delta when your for-next(exynos5) changes Hmm, it can be a solution for this, but I'm not sure exynos4 and exynos5 can be splitted. > get merged to mainline? Manually editing the patches is really pain-some YES, absolutely. Nobody like it. > and sometimes error-some :( Can this patch in particular > be split? Hmm... Now we don't have much time for upcoming merge window, and this should be in linux-next before open merge window. Anyway Joerg, how do you want to handle this? Do you want to pick up 1/2/3 all of them in your tree? If so, 2nd patch should be on top of some samsung topic stuff. Thanks. Best regards, Kgene. -- Kukjin Kim , Senior Engineer, SW Solution Development Team, Samsung Electronics Co., Ltd. From mboxrd@z Thu Jan 1 00:00:00 1970 From: kgene.kim@samsung.com (Kukjin Kim) Date: Fri, 16 Mar 2012 13:47:50 -0700 Subject: [PATCH v12 2/3] ARM: EXYNOS: Change System MMU platform device definitions In-Reply-To: <4F630841.2070307@linaro.org> References: <000201cd0286$2e6fbb40$8b4f31c0$%cho@samsung.com> <20120315111227.GA2384@amd.com> <4F61FF1D.5000502@samsung.com> <4F630841.2070307@linaro.org> Message-ID: <4F63A6F6.9050506@samsung.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Subash Patel wrote: > Hi KyongHo, Kukjin, > > On 03/15/2012 08:09 PM, Kukjin Kim wrote: >> On 03/15/12 04:12, Joerg Roedel wrote: >>> On Thu, Mar 15, 2012 at 05:32:39PM +0900, Cho KyongHo wrote: >> >> KyongHo, I looked at 'Cho KyongHo ' and 'KyongHo Cho >> ' in this series, which one do you want to use? >> >>>> Handling System MMUs with an identifier is not flexible to manage >>>> System MMU platform devices because of the following reasons: >>>> 1. A device driver which needs to handle System MMU must know the ID. >>>> 2. A System MMU may not present in some implementations of Exynos >>>> family. >>>> 3. Handling System MMU with IOMMU API does not require an ID. >>>> >>>> This patch is the result of removing ID of System MMUs. >>>> Instead, a device driver that needs to handle its System MMU must >>>> use IOMMU API while its descriptor of platform device is given. >>>> >>>> This patch also includes the following enhancements: >>>> - A System MMU device becomes a child if its power domain device. >>>> - clkdev >>>> >>>> Cc: Kukjin Kim >>>> Signed-off-by: KyongHo Cho >>>> --- >>>> arch/arm/mach-exynos/Kconfig | 10 +- >>>> arch/arm/mach-exynos/Makefile | 2 +- >>>> arch/arm/mach-exynos/clock-exynos4.c | 79 ++-- >>>> arch/arm/mach-exynos/clock-exynos4.h | 2 + >>>> arch/arm/mach-exynos/clock-exynos4210.c | 11 + >>>> arch/arm/mach-exynos/clock-exynos4212.c | 28 ++- >>>> arch/arm/mach-exynos/clock-exynos5.c | 90 +++++ >>>> arch/arm/mach-exynos/dev-sysmmu.c | 457 ++++++++++++---------- >>>> arch/arm/mach-exynos/include/mach/irqs.h | 25 +- >>>> arch/arm/mach-exynos/include/mach/map.h | 38 ++ >>>> arch/arm/mach-exynos/include/mach/regs-clock.h | 5 + >>>> arch/arm/mach-exynos/include/mach/regs-sysmmu.h | 28 -- >>>> arch/arm/mach-exynos/include/mach/sysmmu.h | 88 +++-- >>>> arch/arm/mach-exynos/mach-armlex4210.c | 1 - >>>> arch/arm/mach-exynos/mach-smdkv310.c | 1 - >>>> 15 files changed, 529 insertions(+), 336 deletions(-) >>>> delete mode 100644 arch/arm/mach-exynos/include/mach/regs-sysmmu.h >>> >>> This patch doesn't apply cleanly against upstream Linux (v3.3-rcX). >>> Please rebase to upstream and resend. The other 2 patches apply fine. >>> >> >> Yes, as KyoungHo said, this has been created on top of Samsung tree >> for-next because this touches many exynos stuff which has been changed >> from Samsung tree for upcoming merge window. So if you're ok on this, >> 1st patch and this should be sent to upstream via samsung tree even >> though I can't find any topic branch for this yet. And if you want >> topic branch for your tree, let me know, I can provide it. >> >> Thanks. > Wouldn't it be good to send exynos4 SYSMMU patches now, and send only > the exynos5 delta when your for-next(exynos5) changes Hmm, it can be a solution for this, but I'm not sure exynos4 and exynos5 can be splitted. > get merged to mainline? Manually editing the patches is really pain-some YES, absolutely. Nobody like it. > and sometimes error-some :( Can this patch in particular > be split? Hmm... Now we don't have much time for upcoming merge window, and this should be in linux-next before open merge window. Anyway Joerg, how do you want to handle this? Do you want to pick up 1/2/3 all of them in your tree? If so, 2nd patch should be on top of some samsung topic stuff. Thanks. Best regards, Kgene. -- Kukjin Kim , Senior Engineer, SW Solution Development Team, Samsung Electronics Co., Ltd.