mbox series

[v7,0/7] Add basic SoC support for mt6765

Message ID 1581067250-12744-1-git-send-email-macpaul.lin@mediatek.com (mailing list archive)
Headers show
Series Add basic SoC support for mt6765 | expand

Message

Macpaul Lin Feb. 7, 2020, 9:20 a.m. UTC
This patch adds basic SoC support for Mediatek's new 8-core SoC,
MT6765, which is mainly for smartphone application.

Changes in V7:
1. Adapt V6's patchset to latest kernel tree 5.5-rc1.
   Origin V6 patchset:
   https://patchwork.kernel.org/cover/11041963/
2. Correct 2 clock-controller type in documentation:
   mipi0 and venc_gcon.
   [v7 1/7] dt-bindings: clock: mediatek: document clk bindings
3. Remove V6's patch 03 because it has been taken into 5.5-next-soc
   [v6, 03/08] dt-bindings: mediatek: add MT6765 power dt-bindings
3. Update Reviewed-by: Rob Herring <robh@kernel.org> for
   [v6, 04/08] clk: mediatek: add mt6765 clock IDs
   --> [v7, 03/07] clk: mediatek: add mt6765 clock IDs
4. Update SPDX tag for
   [v6, 05/08] clk: mediatek: Add MT6765 clock support
   --> [v7, 04/07] clk: mediatek: Add MT6765 clock support

Changes in V6:
1. Adapt V5's patchset to latest kernel tree.
   Origin V5 patchset.
   https://lore.kernel.org/patchwork/cover/963612/
2. Due to clk's common code has been submit by other platform,
   this patch set will have dependencies with the following patchsets
   as the following orders.
   2.a. [v8,00/21] MT8183 IOMMU SUPPORT
        https://patchwork.kernel.org/cover/11023585/
   2.b. [v11,0/6] Add basic node support for Mediatek MT8183 SoC
        https://patchwork.kernel.org/cover/10962385/
   2.c. [v6,00/14] Mediatek MT8183 scpsys support
        https://patchwork.kernel.org/cover/11005751/
3. Correct power related patches into dt-binding patches.
4. Re-order V5's 4/11, 6/11, and 7/11 due clk common code change
   and make dependencies in order.
5. Update some commit message in clk related patches.

Changes in V5:
1. add clk support

Changes in V4:
1. add gic's settings in reg properties
2. remove some patches about dt-bindings since GKH already took them

Changes in V3:
1. split dt-binding document patchs
2. fix mt6765.dtsi warnings with W=12
3. remove uncessary PPI affinity for timer
4. add gicc base for gic dt node

Changes in V2:
1. fix clk properties in uart dts node
2. fix typo in submit title
3. add simple-bus in mt6765.dtsi
4. use correct SPDX license format

Mars Cheng (5):
  dt-bindings: clock: mediatek: document clk bindings for Mediatek
    MT6765 SoC
  dt-bindings: mediatek: Add smi dts binding for Mediatek MT6765 SoC
  clk: mediatek: add mt6765 clock IDs
  soc: mediatek: add MT6765 scpsys and subdomain support
  arm64: dts: mediatek: add mt6765 support

Owen Chen (2):
  clk: mediatek: Add MT6765 clock support
  arm64: defconfig: add CONFIG_COMMON_CLK_MT6765_XXX clocks

 .../arm/mediatek/mediatek,apmixedsys.txt      |   1 +
 .../bindings/arm/mediatek/mediatek,audsys.txt |   1 +
 .../bindings/arm/mediatek/mediatek,camsys.txt |   1 +
 .../bindings/arm/mediatek/mediatek,imgsys.txt |   1 +
 .../arm/mediatek/mediatek,infracfg.txt        |   1 +
 .../bindings/arm/mediatek/mediatek,mipi0a.txt |  28 +
 .../bindings/arm/mediatek/mediatek,mmsys.txt  |   1 +
 .../arm/mediatek/mediatek,pericfg.txt         |   1 +
 .../arm/mediatek/mediatek,topckgen.txt        |   1 +
 .../arm/mediatek/mediatek,vcodecsys.txt       |  27 +
 .../mediatek,smi-common.txt                   |   1 +
 arch/arm64/boot/dts/mediatek/Makefile         |   1 +
 arch/arm64/boot/dts/mediatek/mt6765-evb.dts   |  33 +
 arch/arm64/boot/dts/mediatek/mt6765.dtsi      | 253 +++++
 arch/arm64/configs/defconfig                  |   6 +
 drivers/clk/mediatek/Kconfig                  |  86 ++
 drivers/clk/mediatek/Makefile                 |   7 +
 drivers/clk/mediatek/clk-mt6765-audio.c       | 100 ++
 drivers/clk/mediatek/clk-mt6765-cam.c         |  74 ++
 drivers/clk/mediatek/clk-mt6765-img.c         |  70 ++
 drivers/clk/mediatek/clk-mt6765-mipi0a.c      |  68 ++
 drivers/clk/mediatek/clk-mt6765-mm.c          |  96 ++
 drivers/clk/mediatek/clk-mt6765-vcodec.c      |  70 ++
 drivers/clk/mediatek/clk-mt6765.c             | 952 ++++++++++++++++++
 drivers/soc/mediatek/mtk-scpsys.c             | 130 +++
 include/dt-bindings/clock/mt6765-clk.h        | 313 ++++++
 26 files changed, 2323 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/arm/mediatek/mediatek,mipi0a.txt
 create mode 100644 Documentation/devicetree/bindings/arm/mediatek/mediatek,vcodecsys.txt
 create mode 100644 arch/arm64/boot/dts/mediatek/mt6765-evb.dts
 create mode 100644 arch/arm64/boot/dts/mediatek/mt6765.dtsi
 create mode 100644 drivers/clk/mediatek/clk-mt6765-audio.c
 create mode 100644 drivers/clk/mediatek/clk-mt6765-cam.c
 create mode 100644 drivers/clk/mediatek/clk-mt6765-img.c
 create mode 100644 drivers/clk/mediatek/clk-mt6765-mipi0a.c
 create mode 100644 drivers/clk/mediatek/clk-mt6765-mm.c
 create mode 100644 drivers/clk/mediatek/clk-mt6765-vcodec.c
 create mode 100644 drivers/clk/mediatek/clk-mt6765.c
 create mode 100644 include/dt-bindings/clock/mt6765-clk.h

Comments

Stephen Boyd Feb. 12, 2020, 11:44 p.m. UTC | #1
Quoting Macpaul Lin (2020-02-07 01:20:43)
> This patch adds basic SoC support for Mediatek's new 8-core SoC,
> MT6765, which is mainly for smartphone application.

Clock patches look OK to me. Can you resend them without the defconfig
and dts patches and address Matthias' question?
Matthias Brugger Feb. 15, 2020, 1:47 a.m. UTC | #2
Hi Stephen,

On 13/02/2020 00:44, Stephen Boyd wrote:
> Quoting Macpaul Lin (2020-02-07 01:20:43)
>> This patch adds basic SoC support for Mediatek's new 8-core SoC,
>> MT6765, which is mainly for smartphone application.
> 
> Clock patches look OK to me. Can you resend them without the defconfig
> and dts patches and address Matthias' question?
> 

I'm not sure if I understand you. Do you prefer to have just the clock parts
send as an independent version so that you can easier apply the patches to your
tree?

Patch 2, 5, 6 and 7 should go through my tree.
So do you want a series with patches 1, 3 and 4?

Regards,
Matthias
Macpaul Lin Feb. 18, 2020, 4:12 a.m. UTC | #3
On Sat, 2020-02-15 at 02:47 +0100, Matthias Brugger wrote:

Hi Stephen,

> Hi Stephen,
> 
> On 13/02/2020 00:44, Stephen Boyd wrote:
> > Quoting Macpaul Lin (2020-02-07 01:20:43)
> >> This patch adds basic SoC support for Mediatek's new 8-core SoC,
> >> MT6765, which is mainly for smartphone application.
> > 
> > Clock patches look OK to me. Can you resend them without the defconfig
> > and dts patches and address Matthias' question?
> > 
> 
> I'm not sure if I understand you. Do you prefer to have just the clock parts
> send as an independent version so that you can easier apply the patches to your
> tree?
> 
> Patch 2, 5, 6 and 7 should go through my tree.
> So do you want a series with patches 1, 3 and 4?
> 
> Regards,
> Matthias

Yup, I've got a little bit confused, too.
Should I separate and resend these patches into 2 patch sets?
The 1st patch set includes #1, #3, and #4?
And the other includes #2, #5, #6, and #7?

Regards,
Macpaul Lin
Matthias Brugger Feb. 18, 2020, 4:45 p.m. UTC | #4
On 18/02/2020 05:12, Macpaul Lin wrote:
> On Sat, 2020-02-15 at 02:47 +0100, Matthias Brugger wrote:
> 
> Hi Stephen,
> 
>> Hi Stephen,
>>
>> On 13/02/2020 00:44, Stephen Boyd wrote:
>>> Quoting Macpaul Lin (2020-02-07 01:20:43)
>>>> This patch adds basic SoC support for Mediatek's new 8-core SoC,
>>>> MT6765, which is mainly for smartphone application.
>>>
>>> Clock patches look OK to me. Can you resend them without the defconfig
>>> and dts patches and address Matthias' question?
>>>
>>
>> I'm not sure if I understand you. Do you prefer to have just the clock parts
>> send as an independent version so that you can easier apply the patches to your
>> tree?
>>
>> Patch 2, 5, 6 and 7 should go through my tree.
>> So do you want a series with patches 1, 3 and 4?
>>
>> Regards,
>> Matthias
> 
> Yup, I've got a little bit confused, too.
> Should I separate and resend these patches into 2 patch sets?
> The 1st patch set includes #1, #3, and #4?
> And the other includes #2, #5, #6, and #7?
> 

Yes please do so. I think that's what Stephen referred to.

Regards,
Matthias
Stephen Boyd Feb. 19, 2020, 6:41 p.m. UTC | #5
Quoting Matthias Brugger (2020-02-18 08:45:42)
> 
> 
> On 18/02/2020 05:12, Macpaul Lin wrote:
> > On Sat, 2020-02-15 at 02:47 +0100, Matthias Brugger wrote:
> > 
> > Hi Stephen,
> > 
> >> Hi Stephen,
> >>
> >> On 13/02/2020 00:44, Stephen Boyd wrote:
> >>> Quoting Macpaul Lin (2020-02-07 01:20:43)
> >>>> This patch adds basic SoC support for Mediatek's new 8-core SoC,
> >>>> MT6765, which is mainly for smartphone application.
> >>>
> >>> Clock patches look OK to me. Can you resend them without the defconfig
> >>> and dts patches and address Matthias' question?
> >>>
> >>
> >> I'm not sure if I understand you. Do you prefer to have just the clock parts
> >> send as an independent version so that you can easier apply the patches to your
> >> tree?
> >>
> >> Patch 2, 5, 6 and 7 should go through my tree.
> >> So do you want a series with patches 1, 3 and 4?
> >>
> >> Regards,
> >> Matthias
> > 
> > Yup, I've got a little bit confused, too.
> > Should I separate and resend these patches into 2 patch sets?
> > The 1st patch set includes #1, #3, and #4?
> > And the other includes #2, #5, #6, and #7?
> > 
> 
> Yes please do so. I think that's what Stephen referred to.
> 

If those are the ones that aren't dts or defconfig patches sounds good
to me.
Macpaul Lin Feb. 21, 2020, 10:23 a.m. UTC | #6
Quoting Matthias Brugger and Stephen Boyd wrote:
> Quoting Matthias Brugger (2020-02-18 08:45:42)
> > 
> > On 18/02/2020 05:12, Macpaul Lin wrote:
> > > On Sat, 2020-02-15 at 02:47 +0100, Matthias Brugger wrote:
> > > 
> > > Hi Stephen,
> > > 
> > >> Hi Stephen,
> > >>
> > >> On 13/02/2020 00:44, Stephen Boyd wrote:
> > >>> Quoting Macpaul Lin (2020-02-07 01:20:43)
> > >>>> This patch adds basic SoC support for Mediatek's new 8-core SoC,
> > >>>> MT6765, which is mainly for smartphone application.
> > >>>
> > >>> Clock patches look OK to me. Can you resend them without the defconfig
> > >>> and dts patches and address Matthias' question?
> > >>>
> > >>
> > >> I'm not sure if I understand you. Do you prefer to have just the clock parts
> > >> send as an independent version so that you can easier apply the patches to your
> > >> tree?
> > >>
> > >> Patch 2, 5, 6 and 7 should go through my tree.
> > >> So do you want a series with patches 1, 3 and 4?
> > >>
> > >> Regards,
> > >> Matthias
> > > 
> > > Yup, I've got a little bit confused, too.
> > > Should I separate and resend these patches into 2 patch sets?
> > > The 1st patch set includes #1, #3, and #4?
> > > And the other includes #2, #5, #6, and #7?
> > > 
> > 
> > Yes please do so. I think that's what Stephen referred to.
> > 
> 
> If those are the ones that aren't dts or defconfig patches sounds good
> to me.

Here comes the spilt patch sets.
1. [New] Add basic clock support for mt6765.
  https://patchwork.kernel.org/cover/11395997/
2. [PATCH v8] Add basic SoC support for mt6765
  https://patchwork.kernel.org/patch/11396019/
  But it's a little bit strange I cannot find patch v8's cover-letter
  in patchwork. Only records which patches has been taken from v7.
  If resend cover-letter is required please let me know.

Regards,
Macpaul Lin