Patchwork [GIT,PULL] Highbank fixes for 3.12

login
register
mail settings
Submitter Rob Herring
Date Aug. 14, 2013, 3:01 a.m.
Message ID <520AF2ED.9030900@gmail.com>
Download mbox
Permalink /patch/266945/
State New
Headers show

Pull-request

git://sources.calxeda.com/kernel/linux.git

Comments

Rob Herring - Aug. 14, 2013, 3:01 a.m.
Kevin, Olof,

Please pull highbank updates for 3.12.

There's one trivial core ARM code change in the series to allow a
DMA_ZONE size of 4GB. I believe this should not conflict with Russell's
DMA mask series.

Rob

The following changes since commit d4e4ab86bcba5a72779c43dc1459f71fea3d89c8:

  Linux 3.11-rc5 (2013-08-11 18:04:20 -0700)

are available in the git repository at:

  git://sources.calxeda.com/kernel/linux.git
tags/highbank-lpae-fixes-for-3.12

for you to fetch changes up to ce3ec1d61c5ed6af3ceaace6f57acd588c41b4d6:

  ARM: highbank: fix debug uart virtual address for LPAE (2013-08-13
20:50:16 -0500)

----------------------------------------------------------------
A set of fixes for highbank to enable LPAE and selecting of necesssary
ARM errata.

----------------------------------------------------------------
Rob Herring (6):
      ARM: use phys_addr_t for DMA zone sizes
      ARM: highbank: enable DMA zone for LPAE
      ARM: highbank: select ARCH_HAS_HOLES_MEMORYMODEL
      ARM: highbank: select required errata work-arounds
      ARM: highbank: select ARCH_DMA_ADDR_T_64BIT for LPAE
      ARM: highbank: fix debug uart virtual address for LPAE

 arch/arm/include/asm/mach/arch.h  | 2 +-
 arch/arm/include/debug/highbank.S | 2 +-
 arch/arm/mach-highbank/Kconfig    | 6 ++++++
 arch/arm/mach-highbank/highbank.c | 3 +++
 arch/arm/mm/init.c                | 2 +-
 5 files changed, 12 insertions(+), 3 deletions(-)
Kevin Hilman - Aug. 14, 2013, 4:04 p.m.
Hi Rob,

Rob Herring <robherring2@gmail.com> writes:

> Kevin, Olof,
>
> Please pull highbank updates for 3.12.
>
> There's one trivial core ARM code change in the series to allow a
> DMA_ZONE size of 4GB. I believe this should not conflict with Russell's
> DMA mask series.
>
> Rob
>
> The following changes since commit d4e4ab86bcba5a72779c43dc1459f71fea3d89c8:
>
>   Linux 3.11-rc5 (2013-08-11 18:04:20 -0700)
>
> are available in the git repository at:
>
>   git://sources.calxeda.com/kernel/linux.git
> tags/highbank-lpae-fixes-for-3.12
>
> for you to fetch changes up to ce3ec1d61c5ed6af3ceaace6f57acd588c41b4d6:
>
>   ARM: highbank: fix debug uart virtual address for LPAE (2013-08-13
> 20:50:16 -0500)
>
> ----------------------------------------------------------------
> A set of fixes for highbank to enable LPAE and selecting of necesssary
> ARM errata.
>
> ----------------------------------------------------------------
> Rob Herring (6):
>       ARM: use phys_addr_t for DMA zone sizes
>       ARM: highbank: enable DMA zone for LPAE
>       ARM: highbank: select ARCH_HAS_HOLES_MEMORYMODEL
>       ARM: highbank: select required errata work-arounds
>       ARM: highbank: select ARCH_DMA_ADDR_T_64BIT for LPAE

This one seems to have broken the AMBA PL08x DMA driver, and is still
objected to by Russell, so I'll drop this one for now.

>       ARM: highbank: fix debug uart virtual address for LPAE

The rest look OK and I can pull them if the rest are OK without the
objectionable one.  Let me know if that's OK, or you prefer to wait for
resolution.

Kevin


>  arch/arm/include/asm/mach/arch.h  | 2 +-
>  arch/arm/include/debug/highbank.S | 2 +-
>  arch/arm/mach-highbank/Kconfig    | 6 ++++++
>  arch/arm/mach-highbank/highbank.c | 3 +++
>  arch/arm/mm/init.c                | 2 +-
>  5 files changed, 12 insertions(+), 3 deletions(-)
Rob Herring - Aug. 14, 2013, 6:50 p.m.
On Wed, Aug 14, 2013 at 11:04 AM, Kevin Hilman <khilman@linaro.org> wrote:
> Hi Rob,
>
> Rob Herring <robherring2@gmail.com> writes:
>
>> Kevin, Olof,
>>
>> Please pull highbank updates for 3.12.
>>
>> There's one trivial core ARM code change in the series to allow a
>> DMA_ZONE size of 4GB. I believe this should not conflict with Russell's
>> DMA mask series.
>>
>> Rob
>>
>> The following changes since commit d4e4ab86bcba5a72779c43dc1459f71fea3d89c8:
>>
>>   Linux 3.11-rc5 (2013-08-11 18:04:20 -0700)
>>
>> are available in the git repository at:
>>
>>   git://sources.calxeda.com/kernel/linux.git
>> tags/highbank-lpae-fixes-for-3.12
>>
>> for you to fetch changes up to ce3ec1d61c5ed6af3ceaace6f57acd588c41b4d6:
>>
>>   ARM: highbank: fix debug uart virtual address for LPAE (2013-08-13
>> 20:50:16 -0500)
>>
>> ----------------------------------------------------------------
>> A set of fixes for highbank to enable LPAE and selecting of necesssary
>> ARM errata.
>>
>> ----------------------------------------------------------------
>> Rob Herring (6):
>>       ARM: use phys_addr_t for DMA zone sizes
>>       ARM: highbank: enable DMA zone for LPAE
>>       ARM: highbank: select ARCH_HAS_HOLES_MEMORYMODEL
>>       ARM: highbank: select required errata work-arounds
>>       ARM: highbank: select ARCH_DMA_ADDR_T_64BIT for LPAE
>
> This one seems to have broken the AMBA PL08x DMA driver, and is still
> objected to by Russell, so I'll drop this one for now.

Exynos does the same selection, but I guess it is not multi-platform
though and doesn't select ARM_AMBA.

>>       ARM: highbank: fix debug uart virtual address for LPAE
>
> The rest look OK and I can pull them if the rest are OK without the
> objectionable one.  Let me know if that's OK, or you prefer to wait for
> resolution.

Yes, I'm fine with dropping this one for now.

Rob
Kevin Hilman - Aug. 14, 2013, 7:55 p.m.
Rob Herring <robherring2@gmail.com> writes:

> On Wed, Aug 14, 2013 at 11:04 AM, Kevin Hilman <khilman@linaro.org> wrote:
>> Hi Rob,
>>
>> Rob Herring <robherring2@gmail.com> writes:
>>
>>> Kevin, Olof,
>>>
>>> Please pull highbank updates for 3.12.
>>>
>>> There's one trivial core ARM code change in the series to allow a
>>> DMA_ZONE size of 4GB. I believe this should not conflict with Russell's
>>> DMA mask series.
>>>
>>> Rob
>>>
>>> The following changes since commit d4e4ab86bcba5a72779c43dc1459f71fea3d89c8:
>>>
>>>   Linux 3.11-rc5 (2013-08-11 18:04:20 -0700)
>>>
>>> are available in the git repository at:
>>>
>>>   git://sources.calxeda.com/kernel/linux.git
>>> tags/highbank-lpae-fixes-for-3.12
>>>
>>> for you to fetch changes up to ce3ec1d61c5ed6af3ceaace6f57acd588c41b4d6:
>>>
>>>   ARM: highbank: fix debug uart virtual address for LPAE (2013-08-13
>>> 20:50:16 -0500)
>>>
>>> ----------------------------------------------------------------
>>> A set of fixes for highbank to enable LPAE and selecting of necesssary
>>> ARM errata.
>>>
>>> ----------------------------------------------------------------
>>> Rob Herring (6):
>>>       ARM: use phys_addr_t for DMA zone sizes
>>>       ARM: highbank: enable DMA zone for LPAE
>>>       ARM: highbank: select ARCH_HAS_HOLES_MEMORYMODEL
>>>       ARM: highbank: select required errata work-arounds
>>>       ARM: highbank: select ARCH_DMA_ADDR_T_64BIT for LPAE
>>
>> This one seems to have broken the AMBA PL08x DMA driver, and is still
>> objected to by Russell, so I'll drop this one for now.
>
> Exynos does the same selection, but I guess it is not multi-platform
> though and doesn't select ARM_AMBA.
>
>>>       ARM: highbank: fix debug uart virtual address for LPAE
>>
>> The rest look OK and I can pull them if the rest are OK without the
>> objectionable one.  Let me know if that's OK, or you prefer to wait for
>> resolution.
>
> Yes, I'm fine with dropping this one for now.

OK, dropped that one and applied the rest.

Kevin
Kevin Hilman - Aug. 15, 2013, 3:01 p.m.
Kevin Hilman <khilman@linaro.org> writes:

> Rob Herring <robherring2@gmail.com> writes:
>
>> On Wed, Aug 14, 2013 at 11:04 AM, Kevin Hilman <khilman@linaro.org> wrote:
>>> Hi Rob,
>>>
>>> Rob Herring <robherring2@gmail.com> writes:
>>>
>>>> Kevin, Olof,
>>>>
>>>> Please pull highbank updates for 3.12.
>>>>
>>>> There's one trivial core ARM code change in the series to allow a
>>>> DMA_ZONE size of 4GB. I believe this should not conflict with Russell's
>>>> DMA mask series.
>>>>
>>>> Rob
>>>>
>>>> The following changes since commit d4e4ab86bcba5a72779c43dc1459f71fea3d89c8:
>>>>
>>>>   Linux 3.11-rc5 (2013-08-11 18:04:20 -0700)
>>>>
>>>> are available in the git repository at:
>>>>
>>>>   git://sources.calxeda.com/kernel/linux.git
>>>> tags/highbank-lpae-fixes-for-3.12
>>>>
>>>> for you to fetch changes up to ce3ec1d61c5ed6af3ceaace6f57acd588c41b4d6:
>>>>
>>>>   ARM: highbank: fix debug uart virtual address for LPAE (2013-08-13
>>>> 20:50:16 -0500)
>>>>
>>>> ----------------------------------------------------------------
>>>> A set of fixes for highbank to enable LPAE and selecting of necesssary
>>>> ARM errata.
>>>>
>>>> ----------------------------------------------------------------
>>>> Rob Herring (6):
>>>>       ARM: use phys_addr_t for DMA zone sizes
>>>>       ARM: highbank: enable DMA zone for LPAE
>>>>       ARM: highbank: select ARCH_HAS_HOLES_MEMORYMODEL
>>>>       ARM: highbank: select required errata work-arounds
>>>>       ARM: highbank: select ARCH_DMA_ADDR_T_64BIT for LPAE
>>>
>>> This one seems to have broken the AMBA PL08x DMA driver, and is still
>>> objected to by Russell, so I'll drop this one for now.
>>
>> Exynos does the same selection, but I guess it is not multi-platform
>> though and doesn't select ARM_AMBA.
>>
>>>>       ARM: highbank: fix debug uart virtual address for LPAE
>>>
>>> The rest look OK and I can pull them if the rest are OK without the
>>> objectionable one.  Let me know if that's OK, or you prefer to wait for
>>> resolution.
>>
>> Yes, I'm fine with dropping this one for now.
>
> OK, dropped that one and applied the rest.

OK, with 2 of the 5 patches still having problems/objections, and the
one causing a new build warning, I've backed out this branch
from for-next.

Could you just respin a branch with the first 3 patches and the fix
folded in?  I know I requested the opposite on IRC, but having a new
pull request allows me to include your signed tag on the whole bit
instead of me having to manually massage the branch and merge without a
signed tag.

Thanks,

Kevin
Rob Herring - Aug. 15, 2013, 8:51 p.m.
On 08/15/2013 10:01 AM, Kevin Hilman wrote:
> Kevin Hilman <khilman@linaro.org> writes:
> 
>> Rob Herring <robherring2@gmail.com> writes:
>>
>>> On Wed, Aug 14, 2013 at 11:04 AM, Kevin Hilman <khilman@linaro.org> wrote:
>>>> Hi Rob,
>>>>
>>>> Rob Herring <robherring2@gmail.com> writes:
>>>>
>>>>> Kevin, Olof,
>>>>>
>>>>> Please pull highbank updates for 3.12.
>>>>>
>>>>> There's one trivial core ARM code change in the series to allow a
>>>>> DMA_ZONE size of 4GB. I believe this should not conflict with Russell's
>>>>> DMA mask series.
>>>>>
>>>>> Rob
>>>>>
>>>>> The following changes since commit d4e4ab86bcba5a72779c43dc1459f71fea3d89c8:
>>>>>
>>>>>   Linux 3.11-rc5 (2013-08-11 18:04:20 -0700)
>>>>>
>>>>> are available in the git repository at:
>>>>>
>>>>>   git://sources.calxeda.com/kernel/linux.git
>>>>> tags/highbank-lpae-fixes-for-3.12
>>>>>
>>>>> for you to fetch changes up to ce3ec1d61c5ed6af3ceaace6f57acd588c41b4d6:
>>>>>
>>>>>   ARM: highbank: fix debug uart virtual address for LPAE (2013-08-13
>>>>> 20:50:16 -0500)
>>>>>
>>>>> ----------------------------------------------------------------
>>>>> A set of fixes for highbank to enable LPAE and selecting of necesssary
>>>>> ARM errata.
>>>>>
>>>>> ----------------------------------------------------------------
>>>>> Rob Herring (6):
>>>>>       ARM: use phys_addr_t for DMA zone sizes
>>>>>       ARM: highbank: enable DMA zone for LPAE
>>>>>       ARM: highbank: select ARCH_HAS_HOLES_MEMORYMODEL
>>>>>       ARM: highbank: select required errata work-arounds
>>>>>       ARM: highbank: select ARCH_DMA_ADDR_T_64BIT for LPAE
>>>>
>>>> This one seems to have broken the AMBA PL08x DMA driver, and is still
>>>> objected to by Russell, so I'll drop this one for now.
>>>
>>> Exynos does the same selection, but I guess it is not multi-platform
>>> though and doesn't select ARM_AMBA.
>>>
>>>>>       ARM: highbank: fix debug uart virtual address for LPAE
>>>>
>>>> The rest look OK and I can pull them if the rest are OK without the
>>>> objectionable one.  Let me know if that's OK, or you prefer to wait for
>>>> resolution.
>>>
>>> Yes, I'm fine with dropping this one for now.
>>
>> OK, dropped that one and applied the rest.
> 
> OK, with 2 of the 5 patches still having problems/objections, and the
> one causing a new build warning, I've backed out this branch
> from for-next.

Wow, that went well. Sorry about that.

> Could you just respin a branch with the first 3 patches and the fix
> folded in?  I know I requested the opposite on IRC, but having a new
> pull request allows me to include your signed tag on the whole bit
> instead of me having to manually massage the branch and merge without a
> signed tag.

Will do. I've got another patch to add and will see if I can get pl081
issue sorted out in the next few days.

Rob