From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sudip Mukherjee Subject: Re: [PATCH 0/3] 8250: Split Fintek PCIE to UART to independent file Date: Fri, 22 Jan 2016 16:23:50 +0530 Message-ID: <20160122105350.GA7023@sudip-pc> References: <1453171266-15874-1-git-send-email-hpeter+linux_kernel@gmail.com> <20160119035649.GA1696@windriver.com> <569DF7C3.5050306@gmail.com> <20160119123327.3b519062@lxorguk.ukuu.org.uk> <1453209684.2521.115.camel@linux.intel.com> <569EF810.1000304@gmail.com> <20160120062214.GB3747@sudip-pc> <569F4444.3040104@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <569F4444.3040104@gmail.com> Sender: linux-kernel-owner@vger.kernel.org To: Peter Hung Cc: Rob Groner , Andy Shevchenko , One Thousand Gnomes , Paul Gortmaker , gregkh@linuxfoundation.org, jslaby@suse.com, heikki.krogerus@linux.intel.com, peter@hurleysoftware.com, soeren.grunewald@desy.de, udknight@gmail.com, adam.lee@canonical.com, arnd@arndb.de, yamada.masahiro@socionext.com, mans@mansr.com, scottwood@freescale.com, paul.burton@imgtec.com, matthias.bgg@gmail.com, manabian@gmail.com, peter.ujfalusi@ti.com, linux-kernel@vger.kernel.org, linux-serial@vger.kernel.org, peter_hong@fintek.com.tw, Peter Hung List-Id: linux-serial@vger.kernel.org On Wed, Jan 20, 2016 at 04:24:36PM +0800, Peter Hung wrote: > Hi Sudip, >=20 > Sudip Mukherjee =E6=96=BC 2016/1/20 =E4=B8=8B=E5=8D=88 02:22 =E5=AF=AB= =E9=81=93: > >On Wed, Jan 20, 2016 at 10:59:28AM +0800, Peter Hung wrote: >=20 > >But my personal opinion, if we move out the serial port related code > >into a new driver (a new Kconfig symbol) userspace of many system wi= ll > >break if this new symbol is not enabled by the distributions. But in= the > >way I have done the new symbol needs to be enabled only if the user > >wants to use the GPIO capability. If that is not enabled GPIO cannot= be > >used but it will never break the serial port related code for them. > >I think we should give a thought to that before splitting out the co= des > >from 8250_pci. >=20 > I agree with your opinion. I'm trying to implement GPIO with 2 ways, > One is like yours, add platform_device with in 8250_pci.c and impleme= nt > GPIOLIB platform driver with in 'driver/gpio", and the other is tryin= g > split out from 8250_pci.c to MFD. >=20 > In my personal opinion, the first method is less impact with compatib= le > old system. Looks like no one else is in support of our opinion. Fair enough, I wil= l split out the related code from 8250_pci and create the MFD driver this weekend for my hardware. regards sudip