From mboxrd@z Thu Jan 1 00:00:00 1970 From: Richard Zhao Subject: Re: [PATCH V3 4/7] cpufreq: add generic cpufreq driver Date: Wed, 21 Dec 2011 22:19:11 +0800 Message-ID: <20111221141907.GA1777@richard-laptop> References: <1324264903-15395-1-git-send-email-richard.zhao@linaro.org> <20111221012046.GE15863@b20223-02.ap.freescale.net> <20111221092739.GI15863@b20223-02.ap.freescale.net> <201112210943.34310.arnd@arndb.de> <20111221121246.GB15309@opensource.wolfsonmicro.com> Mime-Version: 1.0 Return-path: Content-Disposition: inline In-Reply-To: Sender: cpufreq-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Kay Sievers Cc: Mark Brown , Arnd Bergmann , Richard Zhao , linux@arm.linux.org.uk, mark.langsdorf@calxeda.com, patches@linaro.org, marc.zyngier@arm.com, catalin.marinas@arm.com, devicetree-discuss@lists.ozlabs.org, bryanh@codeaurora.org, cpufreq@vger.kernel.org, grant.likely@secretlab.ca, jamie@jamieiles.com, rdunlap@xenotime.net, eric.miao@linaro.org, kernel@pengutronix.de, davej@redhat.com, linaro-dev@lists.linaro.org, davidb@codeaurora.org, shawn.guo@linaro.org, rob.herring@calxeda.com, linux-arm-kernel@lists.infradead.org, Greg KH On Wed, Dec 21, 2011 at 01:49:07PM +0100, Kay Sievers wrote: > On Wed, Dec 21, 2011 at 13:12, Mark Brown > wrote: > > On Wed, Dec 21, 2011 at 12:44:57PM +0100, Kay Sievers wrote: > > > >> We will convert all classes to buses over time time, and have a single > >> type of device and a single type of subsystem. > > > > Are there any conversions that have been done already that I can look at > > for reference? > > The first step is the conversion from 'sys_device' to 'device', which is here: > http://git.kernel.org/?p=linux/kernel/git/kay/patches.git;a=tree > > That should hit the tree soon, if all works according to plan. All > sys_devices and sysdev classes will be gone for forever. Even cpu node is device, I still need to find a way to get it. I think it's better have another patch to fix the regulator dt binding in cpu node. I'll not include it in this patch series. Richard > > The 'class' to 'bus' work is simpler, because the logic in both of > them is very similar and both use the same 'struct device' already. > > We'll need to add some convenience APIs to bus, and add code to make > sure the converted stuff has compat symlinks in /sys/class when > needed. Then we can convert-over one 'struct class' to 'struct > bus_type' after the other until 'struct class' can be deleted. > > This work has not yet started, because we are busy with the sys_device > stuff at the moment. > > No new stuff should use 'struct class' or 'struct sys_device', they > should all start right away with 'struct bus_type'. > > Kay From mboxrd@z Thu Jan 1 00:00:00 1970 From: richard.zhao@linaro.org (Richard Zhao) Date: Wed, 21 Dec 2011 22:19:11 +0800 Subject: [PATCH V3 4/7] cpufreq: add generic cpufreq driver In-Reply-To: References: <1324264903-15395-1-git-send-email-richard.zhao@linaro.org> <20111221012046.GE15863@b20223-02.ap.freescale.net> <20111221092739.GI15863@b20223-02.ap.freescale.net> <201112210943.34310.arnd@arndb.de> <20111221121246.GB15309@opensource.wolfsonmicro.com> Message-ID: <20111221141907.GA1777@richard-laptop> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Wed, Dec 21, 2011 at 01:49:07PM +0100, Kay Sievers wrote: > On Wed, Dec 21, 2011 at 13:12, Mark Brown > wrote: > > On Wed, Dec 21, 2011 at 12:44:57PM +0100, Kay Sievers wrote: > > > >> We will convert all classes to buses over time time, and have a single > >> type of device and a single type of subsystem. > > > > Are there any conversions that have been done already that I can look at > > for reference? > > The first step is the conversion from 'sys_device' to 'device', which is here: > http://git.kernel.org/?p=linux/kernel/git/kay/patches.git;a=tree > > That should hit the tree soon, if all works according to plan. All > sys_devices and sysdev classes will be gone for forever. Even cpu node is device, I still need to find a way to get it. I think it's better have another patch to fix the regulator dt binding in cpu node. I'll not include it in this patch series. Richard > > The 'class' to 'bus' work is simpler, because the logic in both of > them is very similar and both use the same 'struct device' already. > > We'll need to add some convenience APIs to bus, and add code to make > sure the converted stuff has compat symlinks in /sys/class when > needed. Then we can convert-over one 'struct class' to 'struct > bus_type' after the other until 'struct class' can be deleted. > > This work has not yet started, because we are busy with the sys_device > stuff at the moment. > > No new stuff should use 'struct class' or 'struct sys_device', they > should all start right away with 'struct bus_type'. > > Kay