From mboxrd@z Thu Jan 1 00:00:00 1970 From: Daniel Lezcano Subject: Re: [PATCH v2] clocksource: exynos_mct: fix for sleeping in atomic ctx handling cpu hotplug notif. Date: Wed, 27 May 2015 14:43:46 +0200 Message-ID: <5565BC02.7010101@linaro.org> References: <1426151496-5509-1-git-send-email-d.eppel@samsung.com> <55509013.3040901@linaro.org> <1432567470.13951.2.camel@AMDC1552.digital.local> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <1432567470.13951.2.camel@AMDC1552.digital.local> Sender: stable-owner@vger.kernel.org To: d.eppel@samsung.com Cc: tglx@linutronix.de, kgene@kernel.org, linux-kernel@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-samsung-soc@vger.kernel.org, kyungmin.park@samsung.com, k.kozlowski@samsung.com, m.jabrzyk@samsung.com, stable@vger.kernel.org List-Id: linux-samsung-soc@vger.kernel.org On 05/25/2015 05:24 PM, Damian Eppel wrote: > On Mon, 2015-05-11 at 13:18 +0200, Daniel Lezcano wrote: [ ... ] >> The code sounds very complex for what it is supposed to do. >> >> Perhaps I am missing something but you have more or less the same >> functionality than the smp_twd timers and these ones don't look so c= omplex. >> >> Could you please look at the smp_twd.c implementation ? > > Hi Daniel, > > exynos_mct.c driver looks more complex as it supports two types of ti= mer > interrupts - private and shared peripheral interrupts (for exynos4412 > and exynos4210 accordingly). In smp_twd.c driver I can see only PPI t= ype > of irqs supported. SPI and PPI irqs differs slightly in setup - thus = two > different code paths appears in the driver in initialization and > handling of CPU notifications. The fix is addressing issue that appea= rs > only for hardware using SPI irqs so it is hard to compare it to > smp_twd.c. > BTW, If we remove support for SPI irqs in exynos_mct.c it would look > almost the same as smp_twd.c. Ok, thanks. I will have a deeper look this afternoon. -- Daniel --=20 Linaro.org =E2=94=82 Open source software fo= r ARM SoCs =46ollow Linaro: Facebook | Twitter | Blog From mboxrd@z Thu Jan 1 00:00:00 1970 From: daniel.lezcano@linaro.org (Daniel Lezcano) Date: Wed, 27 May 2015 14:43:46 +0200 Subject: [PATCH v2] clocksource: exynos_mct: fix for sleeping in atomic ctx handling cpu hotplug notif. In-Reply-To: <1432567470.13951.2.camel@AMDC1552.digital.local> References: <1426151496-5509-1-git-send-email-d.eppel@samsung.com> <55509013.3040901@linaro.org> <1432567470.13951.2.camel@AMDC1552.digital.local> Message-ID: <5565BC02.7010101@linaro.org> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 05/25/2015 05:24 PM, Damian Eppel wrote: > On Mon, 2015-05-11 at 13:18 +0200, Daniel Lezcano wrote: [ ... ] >> The code sounds very complex for what it is supposed to do. >> >> Perhaps I am missing something but you have more or less the same >> functionality than the smp_twd timers and these ones don't look so complex. >> >> Could you please look at the smp_twd.c implementation ? > > Hi Daniel, > > exynos_mct.c driver looks more complex as it supports two types of timer > interrupts - private and shared peripheral interrupts (for exynos4412 > and exynos4210 accordingly). In smp_twd.c driver I can see only PPI type > of irqs supported. SPI and PPI irqs differs slightly in setup - thus two > different code paths appears in the driver in initialization and > handling of CPU notifications. The fix is addressing issue that appears > only for hardware using SPI irqs so it is hard to compare it to > smp_twd.c. > BTW, If we remove support for SPI irqs in exynos_mct.c it would look > almost the same as smp_twd.c. Ok, thanks. I will have a deeper look this afternoon. -- Daniel -- Linaro.org ? Open source software for ARM SoCs Follow Linaro: Facebook | Twitter | Blog