mbox

[GIT,PULL,1/4] omap soc changes for v4.5 merge window

Message ID pull-1450731519-787330
State New
Headers show

Pull-request

git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap tags/omap-for-v4.5/soc-initcall

Message

Tony Lindgren Dec. 21, 2015, 9:06 p.m. UTC
The following changes since commit 1ec218373b8ebda821aec00bb156a9c94fad9cd4:

  Linux 4.4-rc2 (2015-11-22 16:45:59 -0800)

are available in the git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap tags/omap-for-v4.5/soc-initcall

for you to fetch changes up to f2e6a0a913b53f5af87c7e9a783ceae5bb4ae2c4:

  ARM: OMAP2+: Remove device creation for omap-pcm-audio (2015-12-17 10:34:41 -0800)

----------------------------------------------------------------
SoC changes for omaps for v4.5 merge window. The main change here is to
change the omap initcall levels a bit to initialize things later to allow
early device drivers at core_initcall level. This makes things easier
for us as most clocks can be made into regular device drivers except for
a few early clocks needed to initialize system timers. I wanted to have
these changes sit in Linux next for a few weeks before sending out a pull
request, and so far now issues have showed up.

The other changes in this series are timer changes for making use of the
new PWM driver, and timer changes to support more high security SoCs.
Also few minor improvments for module autoidle settings for ti81xx spinbox
and dra7 debug on uart4 in hwmod code. The rest is pretty much just removal
of platform data for SoCs that are all device tree only nowadays.

----------------------------------------------------------------
Aaro Koskinen (1):
      ARM: OMAP: RX-51: fix a typo in log writing

Franklin S Cooper Jr (1):
      ARM: omap4: hwmod: Remove elm address space from hwmod data

J.D. Schroeder (1):
      ARM: DRA7: hwmod: Enable DEBUG_LL for UART4

Neil Armstrong (3):
      ARM: OMAP: dmtimer: Add clock source from DT
      ARM: OMAP: add DT support for ti,dm816-timer
      ARM: OMAP2+: Add hwmod spinbox support for dm816x

Peter Ujfalusi (2):
      ARM: OMAP1: Remove device creation for omap-pcm-audio
      ARM: OMAP2+: Remove device creation for omap-pcm-audio

Suman Anna (6):
      ARM: OMAP2+: Remove legacy device instantiation of IOMMUs
      ARM: OMAP3: hwmod data: Remove legacy IOMMU data
      ARM: OMAP4: hwmod data: Remove legacy IOMMU attr and addrs
      ARM: OMAP2+: Remove omap_mmu_dev_attr structure
      ARM: OMAP: dmtimer: check for fixed timers during config
      ARM: OMAP2+: timer: Remove secure timer for DRA7xx HS devices

Tony Lindgren (2):
      ARM: OMAP2+: Change core_initcall levels to postcore_initcall
      Merge branch 'omap-for-v4.5/initcall' into omap-for-v4.5/soc-v2

 arch/arm/mach-omap1/devices.c                | 19 --------
 arch/arm/mach-omap2/Makefile                 |  2 -
 arch/arm/mach-omap2/board-rx51-peripherals.c |  2 +-
 arch/arm/mach-omap2/devices.c                | 25 ++---------
 arch/arm/mach-omap2/omap-iommu.c             | 66 ----------------------------
 arch/arm/mach-omap2/omap2-restart.c          |  2 +-
 arch/arm/mach-omap2/omap_device.c            |  2 +-
 arch/arm/mach-omap2/omap_hwmod.c             |  2 +-
 arch/arm/mach-omap2/omap_hwmod_3xxx_data.c   | 42 ------------------
 arch/arm/mach-omap2/omap_hwmod_44xx_data.c   | 41 -----------------
 arch/arm/mach-omap2/omap_hwmod_7xx_data.c    |  2 +-
 arch/arm/mach-omap2/omap_hwmod_81xx_data.c   | 35 +++++++++++++++
 arch/arm/mach-omap2/pdata-quirks.c           |  2 +-
 arch/arm/mach-omap2/serial.c                 |  2 +-
 arch/arm/mach-omap2/timer.c                  |  6 +--
 arch/arm/plat-omap/dmtimer.c                 | 42 +++++++++++++++++-
 include/linux/platform_data/iommu-omap.h     |  9 ----
 17 files changed, 90 insertions(+), 211 deletions(-)
 delete mode 100644 arch/arm/mach-omap2/omap-iommu.c

Comments

Olof Johansson Dec. 22, 2015, 10:53 p.m. UTC | #1
Hi Tony,

On Mon, Dec 21, 2015 at 01:06:35PM -0800, Tony Lindgren wrote:
> The following changes since commit d893656e61040f3ff7b5f72a986052a348f3c94e:
> 
>   ARM: OMAP2+: Remove useless check for legacy booting for dm814x (2015-12-09 16:53:46 -0800)
> 
> are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap tags/omap-for-v4.5/81xx-devices
> 
> for you to fetch changes up to 8f2cf92a912ca7173d29c045d225588c9630ed94:
> 
>   ARM: dts: Add usb support for j5-eco evm (2015-12-21 10:43:29 -0800)
> 
> ----------------------------------------------------------------
> Now with dm814x support fixed up enough where we can make use of some clocks,
> we can adds some more basic devices for dm814x. And we can also add minimal
> support for dra62x j5eco-evm that is another ti81xx variant related to dm814x.
> 
> I would have rather merged this series toghether with the ADPLL clock driver,
> and was waiting for that to get merged before sending out these patches. But
> the ADPLL clock driver can be merged separately, and at least dm814x-evm,
> dra62x-j5eco-evm, and t410 all boot with this series alone. So let's merge
> these patches separately as that allows people to do development on these
> platforms for other device drivers using the mainline kernel.
> 
> Note that this series depends on the earlier 81xx-fixes and omap initcall
> changes. It also depends on the DMA commit ae0add740cd0 that's already in
> mainline. As this series is mostly DTS changes with a few related hwmod
> changes, it's probably best merged along with the other DTS changes.

I'm a little puzzled by how you chose to organize this branch. We usually
ask for SoC enablement and new DT files separately. Is there a reason you
couldn't split out the SoC pieces and include those in the soc branch, and
leave the DT contents for this one?


-Olof
Olof Johansson Dec. 22, 2015, 10:54 p.m. UTC | #2
On Mon, Dec 21, 2015 at 01:06:36PM -0800, Tony Lindgren wrote:
> The following changes since commit 8005c49d9aea74d382f474ce11afbbc7d7130bec:
> 
>   Linux 4.4-rc1 (2015-11-15 17:00:27 -0800)
> 
> are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap tags/omap-for-v4.5/wakeup-m3
> 
> for you to fetch changes up to cdd5de500b2c90d5181ebc963826019a0a4234ba:
> 
>   soc: ti: Add wkup_m3_ipc driver (2015-12-03 07:24:29 -0800)
> 
> ----------------------------------------------------------------
> TI wakeup M3 IPC device driver for v4.5 merge window. This driver will
> eventually allow am33xx and am437x to support PM with their Cortex-M3
> power management processor.
> 
> This driver has been waiting to get merged for quite a while but has
> had dependencies to the remoteproc that are now out of the way.

Merged, thanks.


-Olof
Olof Johansson Dec. 22, 2015, 10:55 p.m. UTC | #3
On Mon, Dec 21, 2015 at 01:06:37PM -0800, Tony Lindgren wrote:
> The following changes since commit 89c6f2e5ab7a0499c6bcb25d281977ada205765d:
> 
>   ARM: dts: omap3-igep0030: Use MMC pwrseq to init SDIO WiFi (2015-12-07 16:11:43 -0800)
> 
> are available in the git repository at:
> 
>   git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap tags/omap-for-v4.5/dt-pt2
> 
> for you to fetch changes up to 6c95771f8c71cd4dd6ed7ab84a3aea7d509d7704:
> 
>   ARM: OMAP2+: LogicPD Torpedo: Add Touchscreen Support (2015-12-21 11:10:40 -0800)
> 
> ----------------------------------------------------------------
> Second set of omap device tree changes for v4.5 merge window. This series
> updates the EDMA bindings based on the various fixes to split EDMA into
> separate independent devices. It also adds support for more devices on the
> CompuLab cm-t335 and LogicPD Torpedo boards, and enables the new bindings
> for qspi for am437x and dra7. There are also few dra7 regulator fixes, and
> change of gpoi-key,wakeup to wakeup-source.
> 
> These depend on commit ae0add740cd0 ("dmaengine: edma: DT: Change reserved
> slot array from 16bit to 32bit type") already merged into mainline.

Thanks, merged.


-Olof
Tony Lindgren Dec. 22, 2015, 11:33 p.m. UTC | #4
* Olof Johansson <olof@lixom.net> [151222 14:58]:
> Hi Tony,
> 
> On Mon, Dec 21, 2015 at 01:06:35PM -0800, Tony Lindgren wrote:
> > The following changes since commit d893656e61040f3ff7b5f72a986052a348f3c94e:
> > 
> >   ARM: OMAP2+: Remove useless check for legacy booting for dm814x (2015-12-09 16:53:46 -0800)
> > 
> > are available in the git repository at:
> > 
> >   git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap tags/omap-for-v4.5/81xx-devices
> > 
> > for you to fetch changes up to 8f2cf92a912ca7173d29c045d225588c9630ed94:
> > 
> >   ARM: dts: Add usb support for j5-eco evm (2015-12-21 10:43:29 -0800)
> > 
> > ----------------------------------------------------------------
> > Now with dm814x support fixed up enough where we can make use of some clocks,
> > we can adds some more basic devices for dm814x. And we can also add minimal
> > support for dra62x j5eco-evm that is another ti81xx variant related to dm814x.
> > 
> > I would have rather merged this series toghether with the ADPLL clock driver,
> > and was waiting for that to get merged before sending out these patches. But
> > the ADPLL clock driver can be merged separately, and at least dm814x-evm,
> > dra62x-j5eco-evm, and t410 all boot with this series alone. So let's merge
> > these patches separately as that allows people to do development on these
> > platforms for other device drivers using the mainline kernel.
> > 
> > Note that this series depends on the earlier 81xx-fixes and omap initcall
> > changes. It also depends on the DMA commit ae0add740cd0 that's already in
> > mainline. As this series is mostly DTS changes with a few related hwmod
> > changes, it's probably best merged along with the other DTS changes.
> 
> I'm a little puzzled by how you chose to organize this branch. We usually
> ask for SoC enablement and new DT files separately. Is there a reason you
> couldn't split out the SoC pieces and include those in the soc branch, and
> leave the DT contents for this one?

Hmm yeah I guess I kept all the 81xx stuff in a single branch because of the
earlier largeish set of 81xx fixes that had interlaced dts and SoC changes
that had to be in a specific order to keep system booting because of bugs in
the dts defined clocks.

If you want to, I can respin these patches into two branches, one for dts and
one for SoC changes. Probably both branches have to be based on the earlier
branch "omap-for-v4.5/81xx-fixes-signed" to avoid merge conflicts though.

Regards,

Tony
Olof Johansson Dec. 22, 2015, 11:41 p.m. UTC | #5
On Tue, Dec 22, 2015 at 3:33 PM, Tony Lindgren <tony@atomide.com> wrote:
> * Olof Johansson <olof@lixom.net> [151222 14:58]:
>> Hi Tony,
>>
>> On Mon, Dec 21, 2015 at 01:06:35PM -0800, Tony Lindgren wrote:
>> > The following changes since commit d893656e61040f3ff7b5f72a986052a348f3c94e:
>> >
>> >   ARM: OMAP2+: Remove useless check for legacy booting for dm814x (2015-12-09 16:53:46 -0800)
>> >
>> > are available in the git repository at:
>> >
>> >   git://git.kernel.org/pub/scm/linux/kernel/git/tmlind/linux-omap tags/omap-for-v4.5/81xx-devices
>> >
>> > for you to fetch changes up to 8f2cf92a912ca7173d29c045d225588c9630ed94:
>> >
>> >   ARM: dts: Add usb support for j5-eco evm (2015-12-21 10:43:29 -0800)
>> >
>> > ----------------------------------------------------------------
>> > Now with dm814x support fixed up enough where we can make use of some clocks,
>> > we can adds some more basic devices for dm814x. And we can also add minimal
>> > support for dra62x j5eco-evm that is another ti81xx variant related to dm814x.
>> >
>> > I would have rather merged this series toghether with the ADPLL clock driver,
>> > and was waiting for that to get merged before sending out these patches. But
>> > the ADPLL clock driver can be merged separately, and at least dm814x-evm,
>> > dra62x-j5eco-evm, and t410 all boot with this series alone. So let's merge
>> > these patches separately as that allows people to do development on these
>> > platforms for other device drivers using the mainline kernel.
>> >
>> > Note that this series depends on the earlier 81xx-fixes and omap initcall
>> > changes. It also depends on the DMA commit ae0add740cd0 that's already in
>> > mainline. As this series is mostly DTS changes with a few related hwmod
>> > changes, it's probably best merged along with the other DTS changes.
>>
>> I'm a little puzzled by how you chose to organize this branch. We usually
>> ask for SoC enablement and new DT files separately. Is there a reason you
>> couldn't split out the SoC pieces and include those in the soc branch, and
>> leave the DT contents for this one?
>
> Hmm yeah I guess I kept all the 81xx stuff in a single branch because of the
> earlier largeish set of 81xx fixes that had interlaced dts and SoC changes
> that had to be in a specific order to keep system booting because of bugs in
> the dts defined clocks.
>
> If you want to, I can respin these patches into two branches, one for dts and
> one for SoC changes. Probably both branches have to be based on the earlier
> branch "omap-for-v4.5/81xx-fixes-signed" to avoid merge conflicts though.

Yeah, that sounds good. Please do.


-Olof
Tony Lindgren Dec. 23, 2015, 1:47 a.m. UTC | #6
Hi,

* Olof Johansson <olof@lixom.net> [151222 15:41]:
> On Tue, Dec 22, 2015 at 3:33 PM, Tony Lindgren <tony@atomide.com> wrote:
> > * Olof Johansson <olof@lixom.net> [151222 14:58]:
> >>
> >> I'm a little puzzled by how you chose to organize this branch. We usually
> >> ask for SoC enablement and new DT files separately. Is there a reason you
> >> couldn't split out the SoC pieces and include those in the soc branch, and
> >> leave the DT contents for this one?
> >
> > Hmm yeah I guess I kept all the 81xx stuff in a single branch because of the
> > earlier largeish set of 81xx fixes that had interlaced dts and SoC changes
> > that had to be in a specific order to keep system booting because of bugs in
> > the dts defined clocks.
> >
> > If you want to, I can respin these patches into two branches, one for dts and
> > one for SoC changes. Probably both branches have to be based on the earlier
> > branch "omap-for-v4.5/81xx-fixes-signed" to avoid merge conflicts though.
> 
> Yeah, that sounds good. Please do.

OK updated pull requests sent. It's now three separate pull requests starting
with "[GIT PULL 1/3] reworked fix for earlier ti81xx changes for v4.5 merge
window".

Regards,

Tony