From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tomasz Figa Subject: Re: [PATCH v2 00/12] Samsung PM consolidation part 2 (multiplatform) Date: Tue, 11 Feb 2014 18:52:21 +0100 Message-ID: <52FA6355.5070806@samsung.com> References: <1391713977-22300-1-git-send-email-t.figa@samsung.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: Received: from mailout2.w1.samsung.com ([210.118.77.12]:43214 "EHLO mailout2.w1.samsung.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751460AbaBKRw0 (ORCPT ); Tue, 11 Feb 2014 12:52:26 -0500 Received: from eucpsbgm1.samsung.com (unknown [203.254.199.244]) by mailout2.w1.samsung.com (Oracle Communications Messaging Server 7u4-24.01(7.0.4.24.0) 64bit (built Nov 17 2011)) with ESMTP id <0N0U000B7EZ7EFB0@mailout2.w1.samsung.com> for linux-samsung-soc@vger.kernel.org; Tue, 11 Feb 2014 17:52:19 +0000 (GMT) In-reply-to: <1391713977-22300-1-git-send-email-t.figa@samsung.com> Sender: linux-samsung-soc-owner@vger.kernel.org List-Id: linux-samsung-soc@vger.kernel.org To: linux-samsung-soc@vger.kernel.org Cc: linux-arm-kernel@lists.infradead.org, Kukjin Kim , Arnd Bergmann , Doug Anderson , Olof Johansson , Kyungmin Park , Marek Szyprowski , Tomasz Figa On 06.02.2014 20:12, Tomasz Figa wrote: > Current Samsung PM code is heavily unprepared for multiplatform systems. > The design implies accessing functions and global variables defined in > particular mach- subdirectory from common code in plat-, which is not > allowed when building ARCH_MULTIPLATFORM. In addition there is a lot of > forced code unification, which makes common function handle any possible > quirks of all supported SoCs. In the end this design turned out to not > work too well, ending with a lot of empty functions exported from mach-, > just because code in common pm.c calls them. Moreover, recent trend of > moving lower level suspend/resume code to proper drivers, like pinctrl > or clk, made a lot of code there redundant, especially on DT-only platforms > like Exynos. > > This patch series attempts to untie Exynos PM support from the legacy > Samsung PM core and make it multiplatform-aware, by isolating truly > generic parts of the latter, making them multiplatform-friendly and then > reimplementing Exynos PM support in a multiplatform-capable way by using > those generic parts. The result is that now PM initialization is started > from mach-exynos*-dt, which calls Exynos-specific initialization code that > registers platform_suspend_ops, so control flow is basically reversed > ending with mach- code calling more generic plat- code if needed. > > This is limited to Exynos right now, but remaining SoCs could follow > in further series. > > Depends on Samsung PM consolidation part 1 (clocks) series: > - http://thread.gmane.org/gmane.linux.kernel.samsung-soc/26816 > > On Exynos4210-based Trats, Exynos4412-based Trats2 and Exynos5250-based > Arndale boards (except suspend/resume, which is broken because of > unrelated reasons): > > Tested-by: Tomasz Figa > > Changes since v1 (RFC): > - fixed l2x0 resume, > - fixed checkpatch complaints (about issues in existing code being moved), > - rebased on top of current linux-next, > - slightly reordered patches to make the order more logical. > > Tomasz Figa (12): > ARM: EXYNOS: Do not resume l2x0 if not enabled before suspend > ARM: SAMSUNG: Add soc_is_s3c2410() helper > ARM: SAMSUNG: pm: Save UART DIVSLOT register based on SoC type > ARM: SAMSUNG: pm: Use debug_ll_addr() to get UART base address > ARM: SAMSUNG: pm: Consolidate PM debug functions > ARM: SAMSUNG: pm: Move Samsung PM debug code into separate file > ARM: SAMSUNG: Move common save/restore helpers to separate file > ARM: SAMSUNG: pm: Move s3c_pm_check_* prototypes to plat/pm-common.h > ARM: EXYNOS: Kconfig: Fix abuse of CONFIG_PM > ARM: EXYNOS: Remove PM initcalls and useless indirection > ARM: EXYNOS: Stop using legacy Samsung PM code > ARM: exynos: Allow wake-up using GIC interrupts > > arch/arm/mach-exynos/Kconfig | 16 +-- > arch/arm/mach-exynos/Makefile | 2 +- > arch/arm/mach-exynos/common.c | 1 + > arch/arm/mach-exynos/common.h | 14 ++ > arch/arm/mach-exynos/include/mach/pm-core.h | 75 ----------- > arch/arm/mach-exynos/pm.c | 172 +++++++++++++++++++------ > arch/arm/mach-exynos/regs-pmu.h | 2 + > arch/arm/mach-exynos/sleep.S | 85 ++++++++++++ > arch/arm/mach-s3c64xx/pm.c | 1 - > arch/arm/mach-s5p64x0/pm.c | 1 - > arch/arm/plat-samsung/Makefile | 2 + > arch/arm/plat-samsung/include/plat/cpu.h | 6 + > arch/arm/plat-samsung/include/plat/pm-common.h | 110 ++++++++++++++++ > arch/arm/plat-samsung/include/plat/pm.h | 80 +----------- > arch/arm/plat-samsung/pm-check.c | 2 +- > arch/arm/plat-samsung/pm-common.c | 75 +++++++++++ > arch/arm/plat-samsung/pm-debug.c | 98 ++++++++++++++ > arch/arm/plat-samsung/pm.c | 146 --------------------- > arch/arm/plat-samsung/s5p-sleep.S | 43 ------- > 19 files changed, 531 insertions(+), 400 deletions(-) > delete mode 100644 arch/arm/mach-exynos/include/mach/pm-core.h > create mode 100644 arch/arm/mach-exynos/sleep.S > create mode 100644 arch/arm/plat-samsung/include/plat/pm-common.h > create mode 100644 arch/arm/plat-samsung/pm-common.c > create mode 100644 arch/arm/plat-samsung/pm-debug.c > With patch [PATCH] ARM: dts: exynos5250-arndale: Keep G3D regulator always on on Exynos5250-based Arndale board, including suspend to RAM: Tested-by: Tomasz Figa Best regards, Tomasz