[git pull lucid] mvl-dove: sync to LSP 5.2.1 (LP591249)

[Eric Miao]

On Wed, Jun 9, 2010 at 9:45 PM, Stefan Bader <stefan.bader at canonical.com> wrote:
> On 06/09/2010 03:26 PM, Eric Miao wrote:
>> On Wed, Jun 9, 2010 at 9:15 PM, Stefan Bader <stefan.bader at canonical.com> wrote:
>>> On 06/09/2010 02:51 PM, Eric Miao wrote:
>>>> On Wed, Jun 9, 2010 at 5:40 PM, Eric Miao <eric.miao at canonical.com> wrote:
>>>>> On Wed, Jun 9, 2010 at 5:13 PM, Stefan Bader <stefan.bader at canonical.com> wrote:
>>>>>> On 06/09/2010 03:02 AM, Eric Miao wrote:
>>>>>>> On Wed, Jun 9, 2010 at 4:52 AM, Chase Douglas
>>>>>>> <chase.douglas at canonical.com> wrote:
>>>>>>>> On Tue, 2010-06-08 at 22:28 +0800, Eric Miao wrote:
>>>>>>>>> Stefan,
>>>>>>>>>
>>>>>>>>> Below are the patches needed to sync our mvl-dove branch with Marvell LSP 5.2.1
>>>>>>>>> which Chase is looking for. All 10 code fixes. Please help pull.
>>>>>>>>>
>>>>>>>>> The following changes since commit 191d3ad543c66bbe69c61877637decdf3653bfa2:
>>>>>>>>>   Eric Miao (1):
>>>>>>>>>         UBUNTU: [Config]: sync with Dove LSP 5.1.0
>>>>>>>>>
>>>>>>>>> are available in the git repository at:
>>>>>>>>>
>>>>>>>>>   git://kernel.ubuntu.com/ycmiao/ubuntu-lucid.git lp591249
>>>>>>>>>
>>>>>>>>> Chao Deng (1):
>>>>>>>>>       Fix wrong stride issue for RGB565 on overlay
>>>>>>>>>
>>>>>>>>> Ethan Ku (1):
>>>>>>>>>       fix spi flash type to mx2513205d for AVD1 rev2
>>>>>>>>>
>>>>>>>>> Green Wan (1):
>>>>>>>>>       Refine alogrithm of filling yuv pitch length and do correction
>>>>>>>>> when input pitch value <= 0.
>>>>>>>>>
>>>>>>>>> Joseph Lo (2):
>>>>>>>>>       MRVL BMM module update
>>>>>>>>>       BMM module: remove ioremap_xxx for kernel space VA mapping
>>>>>>>>>
>>>>>>>>> Saeed Bishara (3):
>>>>>>>>>       dove pm: restore cp15 Marvell Auxiliary Function and PMC register
>>>>>>>>>       dove: fix build issues when MV_HAL_DRIVERS_SUPPORT disabled
>>>>>>>>>       dove: register the usb as host mode when HAL drivers disabled
>>>>>>>>>
>>>>>>>>> Tawfik Bayouk (1):
>>>>>>>>>       Nand: fix PIO support for 8bit devices.
>>>>>>>>>
>>>>>>>>> stephenkou (1):
>>>>>>>>>       Fix colorkey issue with PM
>>>>>>>>>
>>>>>>>>>  arch/arm/mach-dove/common.c                        |    5 +-
>>>>>>>>>  arch/arm/mach-dove/dove-db-setup.c                 |    4 +-
>>>>>>>>>  arch/arm/mach-dove/dove-rd-avng-setup.c            |    5 +-
>>>>>>>>>  arch/arm/mach-dove/mv_hal_support/Makefile         |    2 +-
>>>>>>>>>  .../mv_hal_drivers/mv_hal/pmu/mvPmu_asm.S          |   15 +++-
>>>>>>>>>  drivers/misc/mrvl-bmm/bmm_drv.c                    |   88 ++++++++++++----
>>>>>>>>>  drivers/misc/mrvl-bmm/bmm_drv.h                    |   48 +++++-----
>>>>>>>>>  drivers/mtd/nand/dove_nand_hal.c                   |   50 ++++++---
>>>>>>>>>  drivers/video/marvell/dovefb_base.c                |   34 +++++-
>>>>>>>>>  drivers/video/marvell/dovefb_ovly.c                |  105 +++++++++++++-------
>>>>>>>>>  10 files changed, 242 insertions(+), 114 deletions(-)
>>>>>>>>
>>>>>>>> A lot of these changes are cryptic and do not contain meaningful commit
>>>>>>>> messages, but after dealing with LSP patches for the Hedley project I
>>>>>>>> can't fault Eric :). That said, it is hard to figure out what is going
>>>>>>>> on in some of them, and I really don't know what many of the changes
>>>>>>>> really do. However, they do come from Marvell, and they only touch
>>>>>>>> Marvell specific code, so I feel I can trust them. Most of the changes
>>>>>>>> are really bug fixes anyways and not huge changes to the source code.
>>>>>>>>
>>>>>>>> Acked-by: Chase Douglas <chase.douglas at canonical.com>
>>>>>>>>
>>>>>>>
>>>>>>> Stefan,
>>>>>>>
>>>>>>> These are actually rebased on top of your last pull (the one again rebased
>>>>>>> on top of the security patches). If you haven't yet committed all the crap,
>>>>>>> you may have noticed the title of commit 191d3ad:
>>>>>>>
>>>>>>>>> Eric Miao (1):
>>>>>>>>>         UBUNTU: [Config]: sync with Dove LSP 5.1.0
>>>>>>>
>>>>>>> is wrong, which should be "sync with Dove LSP 5.1.1". It will make things
>>>>>>> a bit clearer if you can 'reword' that commit message?
>>>>>>>
>>>>>>> Sorry for the mistake and let me know you prefer me to have a change
>>>>>>> here at my side and send a pull again?
>>>>>>>
>>>>>>> Thanks
>>>>>>> - eric
>>>>>>
>>>>>> Hi Eric,
>>>>>>
>>>>>> thanks for the hint, I will keep that in mind when we try to merge everything
>>>>>> back. I will need to see how to proceed there anyway. From the last time
>>>>>> experience somehow it felt like your base has somehow diverged from the one I am
>>>>>> using. Like this patch which changes a file which has DOS newlines in the tree I
>>>>>> have, but your patch is formed like the file is cleaned. I will have to see how
>>>>>> the most recent patches go.
>>>>>
>>>>> Stefan,
>>>>>
>>>>> Hold on for a while, I do think the DOS newlines things caused some mess.
>>>>> There is a build failure with GC600. Let me try to fix that first.
>>>>>
>>>>
>>>> Hi Stefan,
>>>>
>>>> It looks like you made a commit to address part of the DOS newline issue,
>>>>
>>> Right, because without that one of the patches would not apply and I tried to
>>> preserve that patch as much as possible.
>>>
>>>> f28114ee UBUNTU: SAUCE: Convert file to propper newline
>>>>
>>>> But this doesn't convert all the files with DOS newline issue. Do you want
>>>> me to file a patch to address all those files, or maybe we can just leave
>>>> it as is to avoid further Marvell patches from being failed to apply?
>>>
>>> Right, as long as patches are based on that the file layout it would break
>>> future patches. So those should remain the same. I only touched that other file
>>> because it seemed (from the failing patch) that somewhere that file had been
>>> converted.
>>>
>>
>> OK, I'll make a merge of these two pulls into one and based that on
>> top of below commits:
>>
>> 88b50ff UBUNTU: Start new release
>> 1ae5fbb UBUNTU: Ubuntu-2.6.32-205.18
>>
>> Let me know if it's OK.
>
> Wait a little more. Andy is likely still updating the master branch. When that
> is done I would prepare a proposed upload for master and then rebase mvl-dove
> (and the other trees) on top of that. For doing a new effort I would delay until
> that is done. I will give you a ping.
>

OK.

Chase,

So for your rebase I have just pushed a branch 'LSP_5_2_1' to

git://kernel.ubuntu.com/ycmiao/ubuntu-lucid.git LSP_5_2_1

Which you might use as a reference temporarily. Changes on top of
that should be quite easy to rebase once again.