[RFC 1/4] ARM: tegra: Move SoC drivers to drivers/soc/tegra

Sat Jun 28 10:15:09 PDT 2014

On Friday 27 June 2014 07:27 PM, Thierry Reding wrote:
> On Fri, Jun 27, 2014 at 01:30:04PM -0400, Santosh Shilimkar wrote:
>> +Arnd, Greg, Catalin and Kumar,
>>
>> On Friday 27 June 2014 12:58 PM, Thierry Reding wrote:
>>> From: Thierry Reding <treding at nvidia.com>
>>>
>>> These drivers are closely coupled and need to be moved as a whole. One
>>> reason for moving them out of arch/arm/mach-tegra is to allow them to be
>>> shared with 64-bit ARM.
>>>
>>> Signed-off-by: Thierry Reding <treding at nvidia.com>
>>> ---
>>>  arch/arm/mach-tegra/Makefile                       | 32 --------
>>>  arch/arm/mach-tegra/common.h                       |  4 -
>>>  arch/arm/mach-tegra/io.c                           | 27 +++++-
>>>  arch/arm/mach-tegra/pmc.h                          | 62 --------------
>>>  arch/arm/mach-tegra/tegra.c                        |  9 --
>>>  drivers/soc/Makefile                               |  1 +
>>>  drivers/soc/tegra/Makefile                         | 34 ++++++++
>>>  .../soc/tegra}/cpuidle-tegra114.c                  |  0
>>>  .../soc/tegra}/cpuidle-tegra20.c                   |  6 +-
>>>  .../soc/tegra}/cpuidle-tegra30.c                   |  0
>>>  .../arm/mach-tegra => drivers/soc/tegra}/cpuidle.c |  0
>>>  .../arm/mach-tegra => drivers/soc/tegra}/cpuidle.h |  3 -
>>>  .../mach-tegra => drivers/soc/tegra}/flowctrl.c    |  0
>>>  .../mach-tegra => drivers/soc/tegra}/flowctrl.h    |  2 -
>>>  .../arm/mach-tegra => drivers/soc/tegra}/headsmp.S |  0
>>>  .../arm/mach-tegra => drivers/soc/tegra}/hotplug.c |  0
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/iomap.h |  0
>>>  .../arm/mach-tegra => drivers/soc/tegra}/irammap.h |  0
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/irq.c   |  1 -
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/irq.h   |  6 --
>>>  .../arm/mach-tegra => drivers/soc/tegra}/platsmp.c |  5 --
>>>  .../mach-tegra => drivers/soc/tegra}/pm-tegra20.c  |  0
>>>  .../mach-tegra => drivers/soc/tegra}/pm-tegra30.c  |  0
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/pm.c    |  0
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/pm.h    |  4 +-
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/pmc.c   |  0
>>>  drivers/soc/tegra/pmc.h                            | 35 ++++++++
>>>  .../mach-tegra => drivers/soc/tegra}/powergate.c   |  0
>>>  .../soc/tegra}/reset-handler.S                     |  0
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/reset.c |  0
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/reset.h |  2 -
>>>  .../soc/tegra}/sleep-tegra20.S                     |  0
>>>  .../soc/tegra}/sleep-tegra30.S                     |  0
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/sleep.S |  0
>>>  {arch/arm/mach-tegra => drivers/soc/tegra}/sleep.h |  2 -
>>>  include/linux/tegra-soc.h                          | 95 ++++++++++++++++++++++
>>>  36 files changed, 195 insertions(+), 135 deletions(-)
>>>  delete mode 100644 arch/arm/mach-tegra/common.h
>>>  delete mode 100644 arch/arm/mach-tegra/pmc.h
>>>  create mode 100644 drivers/soc/tegra/Makefile
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle-tegra114.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle-tegra20.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle-tegra30.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle.h (91%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/flowctrl.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/flowctrl.h (98%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/headsmp.S (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/hotplug.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/iomap.h (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/irammap.h (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/irq.c (99%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/irq.h (83%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/platsmp.c (98%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/pm-tegra20.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/pm-tegra30.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/pm.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/pm.h (94%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/pmc.c (100%)
>>>  create mode 100644 drivers/soc/tegra/pmc.h
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/powergate.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/reset-handler.S (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/reset.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/reset.h (97%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/sleep-tegra20.S (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/sleep-tegra30.S (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/sleep.S (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/sleep.h (98%)
>>>
>> NAK for this patch.
>>
>> You are using drivers/soc/* as a dump yard for your SOC code which
>> is not the intention we created drivers/soc/.
> 
> That was not my intention. What remains in arch/arm/mach-tegra at this
> point doesn't actually have a corresponding subsystem (well, except
> maybe the cpuidle code).
> 
>> Its really for subsystem drivers which doesn't have appropriate
>> home in Linux kernel today. From your above patch ...
>>
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle-tegra114.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle-tegra20.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle-tegra30.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/cpuidle.h (91%)
>> This should go into drivers/idle/*. if you have dependencies, please sort
>> them out.
> 
> What exactly is the difference between drivers/idle and drivers/cpuidle?
> There's an intel_idle driver in drivers/idle that includes cpuidle.h and
> registers with that subsystem. But there's also an i7300_idle driver
> that doesn't.
> 
> drivers/cpuidle seems like a better fit. I'll look into moving the code
> there.
> 
I meant drivers/cpuidle.

>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/reset-handler.S (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/reset.c (100%)
>>>  rename {arch/arm/mach-tegra => drivers/soc/tegra}/reset.h (97%)
>> subsystem: drivers/power/reset/
> 
> drivers/power/reset seems to be for drivers that register functions to
> reset a board. The above code for Tegra doesn't do that. Rather it sets
> up the reset handlers for secondary CPUs and for suspend/resume.
>
So this is the suspend code which should go along with rest of your
PM code.
 
>> For tegra/*pm*/, you can use drivers/power or drivers/base/power/
> 
> drivers/power seems to be exclusively battery charger drivers. The pm.c,
> pm-*.c and sleep-*.S set up suspend/resume. That doesn't seem to belong
> in drivers/base/power either.
>
It does have AVS PM code as well and also home for power controller
code. Similar efforts are happening on OMAP [1] to move the PM code
under drivers/power/* 
 
> pmc.c implements various routines to access the power management
> controller, some of which is needed by suspend/resume, some of it is
> needed by SMP. powergate.c implements a subset of the PMC that needs to
> be exported to drivers to enable power partitions on the SoC. I'm not
> aware of subsystems that deal with this kind of driver.
>
Please see above.
 
>> For SMP boot, ARMv8 expecting to have either PSCI based implementation or
>> device tree based boot scheme. you can move towards that model if possible.
> 
> But we also have the code for SMP on 32-bit ARM. Should that remain in
> arch/arm/mach-tegra or can it move to drivers/soc/tegra?
> 
You answered yourself. I don't see any point moving such a code to drivers
which is really SMP bring up code for ARM 32 bit. For ARM 64 bit, as
outline by maintainers (Catalin, Arnd), you should use PSCI or DT based
boot scheme.

> So the only thing in the above that I think could be moved somewhere
> else is the cpuidle drivers. Or do you have any other suggestions for
> the remaining code?
> 
Infact all of your code already has a potential home and I
suggest you discuss it with those subsystem maintainers.

[1] "[PATCH 00/55]: ARM: OMAP2+: PRCM move to drivers"
http://www.spinics.net/lists/arm-kernel/msg319360.html