[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <bb99742f40762d18baa342e7f5ee697fe5a86c09.camel@mediatek.com>
Date: Tue, 3 May 2022 10:54:09 +0800
From: Rex-BC Chen <rex-bc.chen@...iatek.com>
To: AngeloGioacchino Del Regno
<angelogioacchino.delregno@...labora.com>,
Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
"mturquette@...libre.com" <mturquette@...libre.com>,
"sboyd@...nel.org" <sboyd@...nel.org>,
"matthias.bgg@...il.com" <matthias.bgg@...il.com>,
"robh+dt@...nel.org" <robh+dt@...nel.org>,
"krzysztof.kozlowski+dt@...aro.org"
<krzysztof.kozlowski+dt@...aro.org>
CC: "p.zabel@...gutronix.de" <p.zabel@...gutronix.de>,
Chun-Jie Chen (陳浚桀)
<Chun-Jie.Chen@...iatek.com>,
"wenst@...omium.org" <wenst@...omium.org>,
Runyang Chen (陈润洋)
<Runyang.Chen@...iatek.com>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
"linux-clk@...r.kernel.org" <linux-clk@...r.kernel.org>,
"linux-arm-kernel@...ts.infradead.org"
<linux-arm-kernel@...ts.infradead.org>,
"linux-mediatek@...ts.infradead.org"
<linux-mediatek@...ts.infradead.org>,
Project_Global_Chrome_Upstream_Group
<Project_Global_Chrome_Upstream_Group@...iatek.com>
Subject: Re: [PATCH V5 12/16] dt-bindings: reset: mediatek: Add infra_ao
reset bit for MT8195
On Mon, 2022-05-02 at 16:54 +0800, AngeloGioacchino Del Regno wrote:
> Il 29/04/22 23:13, Krzysztof Kozlowski ha scritto:
> > On 28/04/2022 13:56, Rex-BC Chen wrote:
> > > To support reset of infra_ao, add the bit definitions for MT8195.
> > > The infra_ao reset includes 5 banks and 32 bits for each bank.
> > >
> > > Signed-off-by: Rex-BC Chen <rex-bc.chen@...iatek.com>
> > > ---
> > > include/dt-bindings/reset/mt8195-resets.h | 170
> > > ++++++++++++++++++++++
> > > 1 file changed, 170 insertions(+)
> > >
> > > diff --git a/include/dt-bindings/reset/mt8195-resets.h
> > > b/include/dt-bindings/reset/mt8195-resets.h
> > > index a26bccc8b957..463114014483 100644
> > > --- a/include/dt-bindings/reset/mt8195-resets.h
> > > +++ b/include/dt-bindings/reset/mt8195-resets.h
> > > @@ -7,6 +7,7 @@
> > > #ifndef _DT_BINDINGS_RESET_CONTROLLER_MT8195
> > > #define _DT_BINDINGS_RESET_CONTROLLER_MT8195
> > >
> > > +/* TOPRGU resets */
> > > #define MT8195_TOPRGU_CONN_MCU_SW_RST 0
> > > #define MT8195_TOPRGU_INFRA_GRST_SW_RST 1
> > > #define MT8195_TOPRGU_APU_SW_RST 2
> > > @@ -26,4 +27,173 @@
> > >
> > > #define MT8195_TOPRGU_SW_RST_NUM 16
> > >
> > > +/* INFRA RST0 */
> > > +#define MT8195_INFRA_RST0_THERM_CTRL_SWRST 0
> > > +#define MT8195_INFRA_RST0_RSV0 1
> > > +#define MT8195_INFRA_RST0_DISP_PWM1_SWRST 2
> > > +#define MT8195_INFRA_RST0_RSV1 3
> > > +#define MT8195_INFRA_RST0_MSDC3_SWRST 4
> > > +#define MT8195_INFRA_RST0_MSDC2_SWRST 5
> > > +#define MT8195_INFRA_RST0_MSDC1_SWRST 6
> > > +#define MT8195_INFRA_RST0_MSDC0_SWRST 7
> > > +#define MT8195_INFRA_RST0_RSV2 8
> > > +#define MT8195_INFRA_RST0_AP_DMA_SWRST 9
> > > +#define MT8195_INFRA_RST0_MIPI_D_SWRST 10
> > > +#define MT8195_INFRA_RST0_RSV3 11
> > > +#define MT8195_INFRA_RST0_RSV4 12
> > > +#define MT8195_INFRA_RST0_SSUSB_TOP_SWRST 13
> > > +#define MT8195_INFRA_RST0_DISP_PWM_SWRST 14
> > > +#define MT8195_INFRA_RST0_AUXADC_SWRST 15
> > > +#define MT8195_INFRA_RST0_RSV5 16
> > > +#define MT8195_INFRA_RST0_RSV6 17
> > > +#define MT8195_INFRA_RST0_RSV7 18
> > > +#define MT8195_INFRA_RST0_RSV8 19
> > > +#define MT8195_INFRA_RST0_RSV9 20
> > > +#define MT8195_INFRA_RST0_RSV10 21
> > > +#define MT8195_INFRA_RST0_RSV11 22
> > > +#define MT8195_INFRA_RST0_RSV12 23
> > > +#define MT8195_INFRA_RST0_RSV13 24
> > > +#define MT8195_INFRA_RST0_RSV14 25
> > > +#define MT8195_INFRA_RST0_RSV15 26
> > > +#define MT8195_INFRA_RST0_RSV16 27
> > > +#define MT8195_INFRA_RST0_RSV17 28
> > > +#define MT8195_INFRA_RST0_RSV18 29
> > > +#define MT8195_INFRA_RST0_RSV19 30
> > > +#define MT8195_INFRA_RST0_RSV20 31
> >
> > These are not proper IDs... don't work-around usage of bits with
> > fake
> > reserved IDs...
>
> Hello Krzysztof,
>
> Actually, I get that it may seem that Rex is trying to cheat with
> fake
> reserved numbers... but it's really how the registers are laid out:
> there
> really are reserved bits in between used reset bits.
>
> I don't think that the reserved bits are doing anything though, so
> the
> best way to proceed is to just remove them and map the dt-bindings
> IDs to
> the HW register's bits in the driver instead.
> Even though the current approach is very simplistic, I agree that
> this is
> not how it's supposed to be done (and I'm sort-of sad about that).
>
> Rex, please map these values in the reset driver so that, in this
> header,
> you'll get something like:
>
> #define MT8195_INFRA_RST0_THERM_CTRL_SWRST 0
> #define MT8195_INFRA_RST0_DISP_PWM1_SWRST 1
> #define MT8195_INFRA_RST0_MSDC3_SWRST 2
> #define .... (etc)
>
> Cheers,
> Angelo
>
> >
> > Best regards,
> > Krzysztof
>
>
Hello Krzysztof and Angelo,
Thanks for your advice and review.
I will modify my driver using index and I will just add some reset we
curreently use.
reset.h will list like this:
#define MT8195_INFRA_RST0_THERM_CTRL_SWRST 0
#define MT8195_INFRA_RST3_THERM_CTRL_PTP_SWRST 1
#define MT8195_INFRA_RST4_THERM_CTRL_MCU_SWRST 2
For this, I will add a new mode for input argument because we alos need
to be compatible with previous reset drivers.
For input argument with different mode:
enum MTK_RST_CTRL_MODE {
MTK_RST_CTRL_BIT_MODE = 0,
MTK_RST_CTRL_INDEX_MODE,
};
If register MTK_RST_CTRL_INDEX_MODE for reset controller, I will
implent new xlate function to transfer the index to offsets.
BRs,
Rex
Powered by blists - more mailing lists