[GIT,PULL] soc: mediatek: updates for v5.6
mbox series

Message ID 9b365e76-e346-f813-d750-d7cfd0d16e4e@gmail.com
State New
Headers show
Series
  • [GIT,PULL] soc: mediatek: updates for v5.6
Related show

Pull-request

https://git.kernel.org/pub/scm/linux/kernel/git/matthias.bgg/linux.git/ tags/v5.5-next-soc

Message

Matthias Brugger Jan. 13, 2020, 10:17 a.m. UTC
Hi Olof,
Hi ARnds,

Please have a look on the following patches for mediatek soc drivers for v5.6

I includes the stable branch v5.5-next/cmdq-stable which should also be merged
in the DRM tree. I'm not very familiar with providing a stable tree, although I
double-checked that in the pull request to the DRM tree the commit IDs are the
same, I thought it is worth noting.

Regards,
Matthias

---

The following changes since commit e42617b825f8073569da76dc4510bfa019b1c35a:

  Linux 5.5-rc1 (2019-12-08 14:57:55 -0800)

are available in the Git repository at:

  https://git.kernel.org/pub/scm/linux/kernel/git/matthias.bgg/linux.git/
tags/v5.5-next-soc

for you to fetch changes up to 9c26abeb86ce6ab53b6397873bb3f556016c05ba:

  Merge branch 'v5.5-next/cmdq-stable' into v5.5-next/soc (2020-01-10 11:00:45
+0100)

----------------------------------------------------------------
cmdq:
- clean ups of unused code and debuggability
- add cmdq_instruction to make the function call interface more readable
- add functions for polling and providing info for the user of cmdq

scpsys:
- add bindings for MT6765

----------------------------------------------------------------
Bibby Hsieh (4):
      soc: mediatek: cmdq: remove OR opertaion from err return
      soc: mediatek: cmdq: define the instruction struct
      soc: mediatek: cmdq: add polling function
      soc: mediatek: cmdq: add cmdq_dev_get_client_reg function

Mars Cheng (1):
      dt-bindings: mediatek: add MT6765 power dt-bindings

Matthias Brugger (2):
      soc: mediatek: cmdq: delete not used define
      Merge branch 'v5.5-next/cmdq-stable' into v5.5-next/soc

 .../devicetree/bindings/soc/mediatek/scpsys.txt    |   6 +
 drivers/soc/mediatek/mtk-cmdq-helper.c             | 149 +++++++++++++++++----
 include/dt-bindings/power/mt6765-power.h           |  14 ++
 include/linux/mailbox/mtk-cmdq-mailbox.h           |  11 ++
 include/linux/soc/mediatek/mtk-cmdq.h              |  53 ++++++++
 5 files changed, 205 insertions(+), 28 deletions(-)
 create mode 100644 include/dt-bindings/power/mt6765-power.h

Comments

Olof Johansson Jan. 17, 2020, 7:22 p.m. UTC | #1
Hi,

On Mon, Jan 13, 2020 at 2:18 AM Matthias Brugger <matthias.bgg@gmail.com> wrote:
>
> Hi Olof,
> Hi ARnds,
>
> Please have a look on the following patches for mediatek soc drivers for v5.6
>
> I includes the stable branch v5.5-next/cmdq-stable which should also be merged
> in the DRM tree. I'm not very familiar with providing a stable tree, although I
> double-checked that in the pull request to the DRM tree the commit IDs are the
> same, I thought it is worth noting.

I'm a little confused as to how the dependencies are expected to line up here.

Does your non-DRM contents rely on the DRM pieces? If so, you need
them to be based on top of that branch to make sure they are
bisectable, not merge the DRM branch in on top (since a bisect might
land on a commit before the merge).

But... I also don't see how there's any actual dependency here? There
are a few cleanups, the DRM branch builds without them, and the branch
when checked out right before the DRM merge in your pull request, also
builds.


-Olof
Matthias Brugger Jan. 17, 2020, 9:52 p.m. UTC | #2
On 17/01/2020 20:22, Olof Johansson wrote:
> Hi,
> 
> On Mon, Jan 13, 2020 at 2:18 AM Matthias Brugger <matthias.bgg@gmail.com> wrote:
>>
>> Hi Olof,
>> Hi ARnds,
>>
>> Please have a look on the following patches for mediatek soc drivers for v5.6
>>
>> I includes the stable branch v5.5-next/cmdq-stable which should also be merged
>> in the DRM tree. I'm not very familiar with providing a stable tree, although I
>> double-checked that in the pull request to the DRM tree the commit IDs are the
>> same, I thought it is worth noting.
> 
> I'm a little confused as to how the dependencies are expected to line up here.
> 
> Does your non-DRM contents rely on the DRM pieces? If so, you need
> them to be based on top of that branch to make sure they are
> bisectable, not merge the DRM branch in on top (since a bisect might
> land on a commit before the merge).

No, it's the other way round. DRM relies on the four patches I took from Bibby
Hsieh. They are part of v5.5-next/cmdq-stable and this branch was merged into
v5.5-next/soc as well as in Ck Hu's DRM branch. My understanding is that Ck Hu
sends pull requests to the DRM branch.

> 
> But... I also don't see how there's any actual dependency here? There
> are a few cleanups, the DRM branch builds without them, and the branch
> when checked out right before the DRM merge in your pull request, also
> builds.
> 

The patches add some helper functions like cmdq_dev_get_client_reg on which the
DRM branch from Ck Hu depends [1].

Hope that clarifies your questions.
Regards,
Matthias

[1] https://github.com/ckhu-mediatek/linux.git-tags/commits/mediatek-drm-next-5.6
Olof Johansson Jan. 20, 2020, 7:25 p.m. UTC | #3
On Fri, Jan 17, 2020 at 1:52 PM Matthias Brugger <matthias.bgg@gmail.com> wrote:
>
>
>
> On 17/01/2020 20:22, Olof Johansson wrote:
> > Hi,
> >
> > On Mon, Jan 13, 2020 at 2:18 AM Matthias Brugger <matthias.bgg@gmail.com> wrote:
> >>
> >> Hi Olof,
> >> Hi ARnds,
> >>
> >> Please have a look on the following patches for mediatek soc drivers for v5.6
> >>
> >> I includes the stable branch v5.5-next/cmdq-stable which should also be merged
> >> in the DRM tree. I'm not very familiar with providing a stable tree, although I
> >> double-checked that in the pull request to the DRM tree the commit IDs are the
> >> same, I thought it is worth noting.
> >
> > I'm a little confused as to how the dependencies are expected to line up here.
> >
> > Does your non-DRM contents rely on the DRM pieces? If so, you need
> > them to be based on top of that branch to make sure they are
> > bisectable, not merge the DRM branch in on top (since a bisect might
> > land on a commit before the merge).
>
> No, it's the other way round. DRM relies on the four patches I took from Bibby
> Hsieh. They are part of v5.5-next/cmdq-stable and this branch was merged into
> v5.5-next/soc as well as in Ck Hu's DRM branch. My understanding is that Ck Hu
> sends pull requests to the DRM branch.

Ah, ok -- in that case there's no need to bring in the DRM branch to
your contents here, they'll merge them at their end.

It doesn't do any harm (besides adding dependencies that aren't
strictly needed), but we usually avoid doing it.

> > But... I also don't see how there's any actual dependency here? There
> > are a few cleanups, the DRM branch builds without them, and the branch
> > when checked out right before the DRM merge in your pull request, also
> > builds.
> >
>
> The patches add some helper functions like cmdq_dev_get_client_reg on which the
> DRM branch from Ck Hu depends [1].
>
> Hope that clarifies your questions.

Yeah, ok -- merging.


-Olof