[GIT,PULL] ARM: dts: aspeed: Changes for v4.15

Message ID CACPK8XdG=wjKuAJJ7yzu320UPxKAGA-A2DYtrL5=CwCN2DNpBg@mail.gmail.com
State New
Headers show
Series
  • [GIT,PULL] ARM: dts: aspeed: Changes for v4.15
Related show

Pull-request

git://git.kernel.org/pub/scm/linux/kernel/git/joel/aspeed.git

Message

Joel Stanley Oct. 6, 2017, 7:05 a.m.
Hello ARM SoC maintainers!

Here are some ASPEED device tree updates for v4.15.

Sadly our clk driver is still under review, so I have not been able to
include all of the patches I intended to. Alas, these improvements are
still useful.

Cheers,

Joel

The following changes since commit e19b205be43d11bff638cad4487008c48d21c103:

  Linux 4.14-rc2 (2017-09-24 16:38:56 -0700)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/joel/aspeed.git
tags/aspeed-4.15-devicetree

for you to fetch changes up to 27b5e338d2153978aa1d4a01189bbd4713dab406:

  ARM: dts: aspeed-romulus: Enable VUART (2017-10-05 16:30:13 +1030)

----------------------------------------------------------------
ASPEED devicetree updates for 4.15

 - Cleanups of the ASPEED device trees
 - Enable the i2c bus on all platforms
 - Turn VUART on for BMC platforms
 - Bind watchdog two for compatilbiy with shipping u-boot

----------------------------------------------------------------
Andrew Jeffery (1):
      ARM: dts: aspeed: Move pinctrl subnodes to improve readability

Joel Stanley (12):
      ARM: dts: aspeed: Reorder ADC node
      ARM: dts: aspeed: Add I2C buses
      ARM: dts: aspeed-romulus: Add I2C devices
      ARM: dts: aspeed-palmetto: Add I2C devices
      ARM: dts: aspeed-ast2500: Add I2C devices
      ARM: dts: aspeed: Add aliases for UARTs
      ARM: dts: aspeed: Correctly order UART nodes
      ARM: dts: aspeed: Clean up UART nodes
      ARM: dts: aspeed: Remove undocumented wdt properties
      ARM: dts: aspeed: Enable watchdog two
      ARM: dts: aspeed-palmetto: Enable VUART
      ARM: dts: aspeed-romulus: Enable VUART

 arch/arm/boot/dts/aspeed-ast2500-evb.dts      |   19 +
 arch/arm/boot/dts/aspeed-bmc-opp-palmetto.dts |   56 +-
 arch/arm/boot/dts/aspeed-bmc-opp-romulus.dts  |   58 +
 arch/arm/boot/dts/aspeed-g4.dtsi              | 1816 ++++++++++++++-----------
 arch/arm/boot/dts/aspeed-g5.dtsi              | 1806 +++++++++++++-----------
 5 files changed, 2199 insertions(+), 1556 deletions(-)

Comments

Arnd Bergmann Oct. 19, 2017, 9:49 p.m. | #1
On Fri, Oct 6, 2017 at 9:05 AM, Joel Stanley <joel@jms.id.au> wrote:
> Hello ARM SoC maintainers!
>
> Here are some ASPEED device tree updates for v4.15.
>
> Sadly our clk driver is still under review, so I have not been able to
> include all of the patches I intended to. Alas, these improvements are
> still useful.

Pulled into next/dt

>     ARM: dts: aspeed: Add aliases for UARTs

Can you explain this one? It looks like you now have aliases for
all uart and i2c entries on all machines, even when they don't
actually expose the ports. The purpose of the aliases is to
only list the ones that are actually usable, so they should normally
go into the board specific .dts files.

     Arnd
Joel Stanley Nov. 17, 2017, 2:57 a.m. | #2
On Fri, Oct 20, 2017 at 8:19 AM, Arnd Bergmann <arnd@arndb.de> wrote:
> On Fri, Oct 6, 2017 at 9:05 AM, Joel Stanley <joel@jms.id.au> wrote:
>> Hello ARM SoC maintainers!
>>
>> Here are some ASPEED device tree updates for v4.15.
>>
>> Sadly our clk driver is still under review, so I have not been able to
>> include all of the patches I intended to. Alas, these improvements are
>> still useful.
>
> Pulled into next/dt

I noticed this didn't make it into Linus tree for 4.15 :(

Was it because of your question below?

>
>>     ARM: dts: aspeed: Add aliases for UARTs
>
> Can you explain this one? It looks like you now have aliases for
> all uart and i2c entries on all machines, even when they don't
> actually expose the ports. The purpose of the aliases is to
> only list the ones that are actually usable, so they should normally
> go into the board specific .dts files.

Sorry for not responding to this at the time. I was on vacation.

The idea was to maintain expectations for userspace where they assume
the character devices match the numbering of the hardware. So
/dev/ttyS4 is always the fifth uart, even when the board only has one
or two wired up.

If we don't do that, our userspace developers have trouble telling
which UART is actually /dev/ttyS0, for example.

Cheers,

Joel
Joel Stanley Nov. 17, 2017, 3 a.m. | #3
On Fri, Nov 17, 2017 at 1:27 PM, Joel Stanley <joel@jms.id.au> wrote:
> On Fri, Oct 20, 2017 at 8:19 AM, Arnd Bergmann <arnd@arndb.de> wrote:
>> On Fri, Oct 6, 2017 at 9:05 AM, Joel Stanley <joel@jms.id.au> wrote:
>>> Hello ARM SoC maintainers!
>>>
>>> Here are some ASPEED device tree updates for v4.15.
>>>
>>> Sadly our clk driver is still under review, so I have not been able to
>>> include all of the patches I intended to. Alas, these improvements are
>>> still useful.
>>
>> Pulled into next/dt
>
> I noticed this didn't make it into Linus tree for 4.15 :(

Scratch that. My mistake, it is there (I was looking at the wrong part
of the diffstat). We should still clear up your query about the
aliases.

>
> Was it because of your question below?
>
>>
>>>     ARM: dts: aspeed: Add aliases for UARTs
>>
>> Can you explain this one? It looks like you now have aliases for
>> all uart and i2c entries on all machines, even when they don't
>> actually expose the ports. The purpose of the aliases is to
>> only list the ones that are actually usable, so they should normally
>> go into the board specific .dts files.
>
> Sorry for not responding to this at the time. I was on vacation.
>
> The idea was to maintain expectations for userspace where they assume
> the character devices match the numbering of the hardware. So
> /dev/ttyS4 is always the fifth uart, even when the board only has one
> or two wired up.
>
> If we don't do that, our userspace developers have trouble telling
> which UART is actually /dev/ttyS0, for example.
>
> Cheers,
>
> Joel