mbox series

[v3,00/13] MT8195 SMI support

Message ID 20210810080859.29511-1-yong.wu@mediatek.com (mailing list archive)
Headers show
Series MT8195 SMI support | expand

Message

Yong Wu Aug. 10, 2021, 8:08 a.m. UTC
This patchset mainly adds SMI support for mt8195.

Comparing with the previous version, add two new functions:
a) add smi sub common
b) add initial setting for smi-common and smi-larb.

Change note:
v3:1) In the dt-binding:
       a. Change mediatek,smi type from phandle-array to phandle from Rob.
       b. Add a new bool property (mediatek,smi_sub_common)
          to indicate if this is smi-sub-common.
   2) Change the clock using bulk parting.
      keep the smi-common's has_gals flag. more strict.
   3) More comment about larb initial setting.
   4) Add a maintain patch
       
v2: https://lore.kernel.org/linux-mediatek/20210715121209.31024-1-yong.wu@mediatek.com/
    rebase on v5.14-rc1
    1) Adjust clk_bulk flow: use devm_clk_bulk_get for necessary clocks.
    2) Add two new little patches:
       a) use devm_platform_ioremap_resource
       b) Add error handle for smi_probe

v1: https://lore.kernel.org/linux-mediatek/20210616114346.18812-1-yong.wu@mediatek.com/

Yong Wu (13):
  dt-bindings: memory: mediatek: Add mt8195 smi binding
  dt-bindings: memory: mediatek: Add mt8195 smi sub common
  memory: mtk-smi: Use clk_bulk clock ops
  memory: mtk-smi: Rename smi_gen to smi_type
  memory: mtk-smi: Adjust some code position
  memory: mtk-smi: Add error handle for smi_probe
  memory: mtk-smi: Add device link for smi-sub-common
  memory: mtk-smi: Add clocks for smi-sub-common
  memory: mtk-smi: Use devm_platform_ioremap_resource
  memory: mtk-smi: mt8195: Add smi support
  memory: mtk-smi: mt8195: Add initial setting for smi-common
  memory: mtk-smi: mt8195: Add initial setting for smi-larb
  MAINTAINERS: Add entry for MediaTek SMI

 .../mediatek,smi-common.yaml                  |  36 +-
 .../memory-controllers/mediatek,smi-larb.yaml |   3 +
 MAINTAINERS                                   |   8 +
 drivers/memory/mtk-smi.c                      | 596 ++++++++++--------
 4 files changed, 395 insertions(+), 248 deletions(-)

Comments

Krzysztof Kozlowski Aug. 18, 2021, 8:41 p.m. UTC | #1
On 10/08/2021 10:08, Yong Wu wrote:
> This patchset mainly adds SMI support for mt8195.
> 
> Comparing with the previous version, add two new functions:
> a) add smi sub common
> b) add initial setting for smi-common and smi-larb.
> 
> Change note:
> v3:1) In the dt-binding:
>        a. Change mediatek,smi type from phandle-array to phandle from Rob.
>        b. Add a new bool property (mediatek,smi_sub_common)
>           to indicate if this is smi-sub-common.
>    2) Change the clock using bulk parting.
>       keep the smi-common's has_gals flag. more strict.
>    3) More comment about larb initial setting.
>    4) Add a maintain patch

The patchset looks good to me but I saw now comments from Rob, so I
expect a resend. Therefore there is also time for additional review -
maybe continued by Ikjoon Jang?

If you sent a v4 with fixes rather soon and get ack from Rob, I could
still try to get it into next merge window. After this weekend I won't
be taking patches for this cycle and it will wait for the merge window
to finish.


Best regards,
Krzysztof
Yong Wu Aug. 19, 2021, 5:34 a.m. UTC | #2
On Wed, 2021-08-18 at 22:41 +0200, Krzysztof Kozlowski wrote:
> On 10/08/2021 10:08, Yong Wu wrote:
> > This patchset mainly adds SMI support for mt8195.
> > 
> > Comparing with the previous version, add two new functions:
> > a) add smi sub common
> > b) add initial setting for smi-common and smi-larb.
> > 
> > Change note:
> > v3:1) In the dt-binding:
> >        a. Change mediatek,smi type from phandle-array to phandle
> > from Rob.
> >        b. Add a new bool property (mediatek,smi_sub_common)
> >           to indicate if this is smi-sub-common.
> >    2) Change the clock using bulk parting.
> >       keep the smi-common's has_gals flag. more strict.
> >    3) More comment about larb initial setting.
> >    4) Add a maintain patch
> 
> The patchset looks good to me but I saw now comments from Rob, so I
> expect a resend. Therefore there is also time for additional review -
> maybe continued by Ikjoon Jang?
> 
> If you sent a v4 with fixes rather soon and get ack from Rob, I could
> still try to get it into next merge window. After this weekend I
> won't
> be taking patches for this cycle and it will wait for the merge
> window
> to finish.

Hi Krzysztof,

Thanks very much for your information.

It looks the time is too short to get Rob's ack in this weekend. I will
wait for one or two weeks to see if there is other comments, then
resend.

> 
> 
> Best regards,
> Krzysztof