[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <1555034293.349.6.camel@mtkswgap22>
Date: Fri, 12 Apr 2019 09:58:13 +0800
From: Light Hsieh <light.hsieh@...iatek.com>
To: Sean Wang <sean.wang@...nel.org>
CC: Linus Walleij <linus.walleij@...aro.org>,
<linux-mediatek@...ts.infradead.org>, <linux-gpio@...r.kernel.org>,
<linux-kernel@...r.kernel.org>, <wsd_upstream@...iatek.com>
Subject: Re: [PATCH] pinctrl: Add kernel config PINCTRL_MTK_V2
Dear reviewer,
The points of Sean are right. Please forget this patch proposal.
On Thu, 2019-04-11 at 15:04 -0700, Sean Wang wrote:
> Hi, Light
>
> On Thu, Apr 11, 2019 at 2:32 AM Light Hsieh <light.hsieh@...iatek.com> wrote:
> >
> > Since no single Mediatek chip use code for PINCTRL_MTK and code for
> > PINCTRL_MTK_MOORE/PINCTRL_MTK_PARIS simultaneously, it is better to use
> > different config to determine if related code will be built or not on
> > building non-generic kernel.
> >
> > Add kernel config PINCTRL_MTK_V2 selected by either PINCTRL_MTK_MOORE
> > or PINCTRL_MTK_PARIS.
> > Use PINCTRL_MTK and PINCTRL_MTK_V2 to control building of
> > drivers/pinctrl/medaitek/.
> > Remove selection of EINT_MTK from PINCTRL_MTK since code for EINT_MTK is
> > only related to PINCTRL_MTK_MOORE/PINCTRL_MTK_PARIS, i.e. PINCTL_MTK_V2.
> >
>
> PINCTRL_MTK also depends on EINT_MTK such as the symbol
> mtk_eint_do_init, it is a commonlibrary for the two kinds of the
> pinctrl core.
>
Yes, you are right.
It is my fault that I don't see some mtk_eint_* functions originally in
pinctrl-mtk-common.c had been moved to mtk-eint.c since kernel-4.18 and
now pinctrl-mtk-common.c depends on mtk-eint.c.
> > ---
> > drivers/pinctrl/Makefile | 3 ++-
> > drivers/pinctrl/mediatek/Kconfig | 15 ++++++++++++---
> > 2 files changed, 14 insertions(+), 4 deletions(-)
> >
> > diff --git a/drivers/pinctrl/Makefile b/drivers/pinctrl/Makefile
> > index 712184b..fcee0e0 100644
> > --- a/drivers/pinctrl/Makefile
> > +++ b/drivers/pinctrl/Makefile
> > @@ -65,6 +65,7 @@ obj-$(CONFIG_PINCTRL_SUNXI) += sunxi/
> > obj-y += ti/
> > obj-$(CONFIG_PINCTRL_UNIPHIER) += uniphier/
> > obj-$(CONFIG_ARCH_VT8500) += vt8500/
> > -obj-y += mediatek/
> > +obj-$(CONFIG_PINCTRL_MTK) += mediatek/
> > +obj-$(CONFIG_PINCTRL_MTK_V2) += mediatek/
>
> I would think it is good if deciding V1 or not should be done inside
> the vendor directory and the change also would cause COMPILE_TEST not
> be applied to
Agree.
> > obj-$(CONFIG_PINCTRL_ZX) += zte/
> > obj-y += cirrus/
> > diff --git a/drivers/pinctrl/mediatek/Kconfig b/drivers/pinctrl/mediatek/Kconfig
> > index a005cbc..5e26462 100644
> > --- a/drivers/pinctrl/mediatek/Kconfig
> > +++ b/drivers/pinctrl/mediatek/Kconfig
> > @@ -2,10 +2,15 @@ menu "MediaTek pinctrl drivers"
> > depends on ARCH_MEDIATEK || COMPILE_TEST
> >
> > config EINT_MTK
> > - bool "MediaTek External Interrupt Support"
> > - depends on PINCTRL_MTK || PINCTRL_MTK_MOORE || PINCTRL_MTK_PARIS || COMPILE_TEST
> > + bool "MediaTek External Interrupt driver that is based on PINCTRL_MTK_V2"
> > + depends on PINCTRL_MTK_MOORE || PINCTRL_MTK_PARIS || COMPILE_TEST
> > select GPIOLIB
> > select IRQ_DOMAIN
> > + help
> > + Say yes here to enable support for MediaTek External Interrupt
> > + (EINT) driver based on PINCTRL_MTK version 2.
> > + This driver is combined with MediaTek Pinctrl driver version 2
> > + so PINCTRL_MTK_V2 shall be set first.
> >
> > config PINCTRL_MTK
> > bool
> > @@ -13,9 +18,11 @@ config PINCTRL_MTK
> > select PINMUX
> > select GENERIC_PINCONF
> > select GPIOLIB
> > - select EINT_MTK
> > select OF_GPIO
> >
> > +config PINCTRL_MTK_V2
> > + bool "MediaTek Pinctrl Support V2"
> > +
> > config PINCTRL_MTK_MOORE
> > bool
> > depends on OF
> > @@ -24,6 +31,7 @@ config PINCTRL_MTK_MOORE
> > select GENERIC_PINMUX_FUNCTIONS
> > select GPIOLIB
> > select OF_GPIO
> > + select PINCTRL_MTK_V2
> >
> > config PINCTRL_MTK_PARIS
> > bool
> > @@ -33,6 +41,7 @@ config PINCTRL_MTK_PARIS
> > select GPIOLIB
> > select EINT_MTK
> > select OF_GPIO
> > + select PINCTRL_MTK_V2
> >
> > # For ARMv7 SoCs
> > config PINCTRL_MT2701
> > --
> > 1.8.1.1.dirty
> >
Powered by blists - more mailing lists