There are instances of IO_ADDRESS() being used for system module
(sysmod) register access. Eliminate this in favor of a ioremap()
based access. ioremap() the entire sysmod address space once during
boot-up and provide a helper macro to access specific register
offsets within the address space.

With this, also eliminate ioremap() of specific sysmodule registers
related to VPIF happening in DM646x EVM code.

While at it, also eliminate some duplicate sysmod register offset macros
defined in code and place offset definitions at one place in davinci.h
Signed-off-by: NManjunath Hadli <manjunath.hadli@ti.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
[nsekhar@ti.com: removed the addition of ifndef __ASSEMBLER__
in davinci.h, eliminate IO_ADDRESS() usage left out in dm646x.c,
cleanup VPIF sysmodule register access as part of this patch and
keep all sysmod offsets in davinci.h Also, convert the WARN_ON()
on failure to setup sysmod base to BUG_ON()]
Signed-off-by: NSekhar Nori <nsekhar@ti.com>

Remove individual platform header files for dm365, dm355, dm644x
and dm646x and consolidate it into a single and common
header file davinci.h placed in arch/arm/mach-davinci.

This reduces the pollution in the include/mach and is consistent
with Russell's suggestions as part of his "pet peaves" mail.
(See #4 in: http://lists.infradead.org/pipermail/linux-arm-kernel/2011-November/071516.html)

While at it, fix the forward declaration of spi_board_info,
and include the right header file instead.

The further patches in the series take  advantage of this consolidation
for easy implementation of IO_ADDRESS elimination.
Signed-off-by: NManjunath Hadli <manjunath.hadli@ti.com>
[nsekhar@ti.com: make davinci.h the first local include file,
fix forward declaration of spi_board_info and add back Deep Root
Systems, LLC copyright]
Signed-off-by: NSekhar Nori <nsekhar@ti.com>

Move register base addresses and offsets used only in dm646x.c
from arch/arm/mach-davinci/include/mach/dm646x.h in to the C
file as these definitions are used only in C file.

This helps reduce code in arch/arm/mach-davinci/include/mach/
which is not really needed by rest of the kernel.
Signed-off-by: NManjunath Hadli <manjunath.hadli@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>

Rather than using DaVinci specific davinci_soc_info based
restart hook, use the restart hook available in the machine
descriptor instead.

Tested on DM365 and AM18x EVMs.

v2:
Changed to use restart hook in machine descriptor
per Russell's comment.
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Fix the incorrect classification of DSP clock into a
seperate DSP domain on DM646x.

Per the reference guide (http://www.ti.com/lit/ug/spruep9e/spruep9e.pdf)
there is only one "AlwaysON" power domain on DM6467.
Signed-off-by: NSekhar Nori <nsekhar@ti.com>

Davinci platforms may define a default queue for each channel
controller. If one is not defined, the default queue is set to EVENTQ_1.
However, there's no way to distinguish between an unset default queue to
one that is set to EVENTQ_0, as EVENTQ_0 = 0.

Explicitly specify the default queue for all channel controllers on all
Davinci platforms to EVENTQ_1, and don't overwrite it in the EDMA probe
function.

One exception is the DA850 board, for which EVENTQ_1 is not a valid
option for its second channel controller. Use EVENTQ_0 instead for that
channel controller.
Signed-off-by: NIdo Yariv <ido@wizery.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>

The <mach/gpio.h> file is included from upper directories
and deal with generic GPIO and gpiolib stuff. Break out the
platform and driver specific defines and functions into its own
header file.

Cc: Sekhar Nori <nsekhar@ti.com>
Cc: Kevin Hilman <khilman@ti.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

The DM6467 and DM6467T EVMs use different reference clock
frequencies. This difference is currently supported by having
the SoC code call a public board routine which sets up the reference
clock frequency. This does not scale as more boards are added.

Instead, use the clk_set_rate() API to setup the reference clock
frequency to a different value from the board file.
Suggested-by: NKevin Hilman <khilman@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Acked-by: NKevin Hilman <khilman@ti.com>

Remove linux/mm.h inclusion from netdevice.h -- it's unused (I've checked manually).

To prevent mm.h inclusion via other channels also extract "enum dma_data_direction"
definition into separate header. This tiny piece is what gluing netdevice.h with mm.h
via "netdevice.h => dmaengine.h => dma-mapping.h => scatterlist.h => mm.h".
Removal of mm.h from scatterlist.h was tried and was found not feasible
on most archs, so the link was cutoff earlier.

Hope people are OK with tiny include file.

Note, that mm_types.h is still dragged in, but it is a separate story.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On Davinci SRAM is mapped as MT_DEVICE becasue of the section
mapping pre-requisite instead of intended MT_MEMORY_NONCACHED

Since the section mapping limitation gets fixed with first
patch in this series, the MT_MEMORY_NONCACHED can be used now.
Signed-off-by: NSantosh Shilimkar <santosh.shilimkar@ti.com>
Acked-by: NKevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

This patch removes davinci architecture code that has now been rendered
useless by the previous patches in the MDIO separation series.

In addition, the earlier phy_mask definitions have been replaced with
corresponding phy_id definitions.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Tested-by: NMichael Williamson <michael.williamson@criticallink.com>
Tested-by: NCaglar Akyuz <caglarakyuz@gmail.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch adds mdio platform devices on SoCs that have the necessary
hardware.  Clock lookup entries (aliases) have also been added, so that the
MDIO and EMAC drivers can independently enable/disable a shared underlying
clock.  Further, the EMAC MMR region has been split down into separate MDIO
and EMAC regions.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Acked-by: NDavid S. Miller <davem@davemloft.net>
Tested-by: NMichael Williamson <michael.williamson@criticallink.com>
Tested-by: NCaglar Akyuz <caglarakyuz@gmail.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

Not all the channels and slots available on the DM646x EVM
are used by the devices on the EVM. These resources can be
used by the DSP to speed up codec operations.

This patch reserves these channels for the DSP.
Signed-off-by: NSudhakar Rajashekhara <sudhakar.raj@ti.com>
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch modifies the EDMA driver to expect the channel
controller (CC) infomation passed on by the platform as a fixed
size (EDMA_MAX_CC) array of pointers to structures.

Doing so helps catch errors of the sort where the resource
structure has information for more channel controllers than
the number channel controller info structures defined.

Such insufficient platform data would lead to illegal memory
accesses.
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch modifies the pinmux implementation so as to ioremap() the pinmux
register area on first use.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch implements the following:

 - interrupt initialization uses ioremap() instead of passing a virtual address
   via davinci_soc_info.

 - machine definitions directly point to cp_intc_init() or davinci_irq_init()

 - davinci_intc_type and davinci_intc_base now get initialized in controller
   specific init functions instead of davinci_common_init()

 - minor fix in davinci_irq_init() to use intc_irq_num instead of
   DAVINCI_N_AINTC_IRQ
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch modifies the psc and clock control code to use ioremap()ed
registers.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch replaces the jtag id base info in davinci_soc_info with a physical
address which is then ioremap()ed within common code.

This patch (in combination with a similar change for PSC) will allow us to
eliminate the SYSCFG nastiness in DA8xx code.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch modifies the gpio_base definition in davinci_soc_info to be a
physical address, which is then ioremap()ed by the gpio initialization
function.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

The earlier watchdog reset mechanism had a couple of limitations.  First, it
embedded a reference to "davinci_wdt_device" inside common code.  This
forced all derived platforms (da8xx and tnetv107x) to define such a device.
This also would have caused problems in including multiple socs in a single
build due to symbol redefinition.

With this patch, davinci_watchdog_reset() now takes the platform device as an
argument.  The davinci_soc_info struct has been extended to include a reset
function and a watchdog platform_device.  arch_reset() then uses these
elements to reset the system in a SoC specific fashion.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Tested-by: NSandeep Paulraj <s-paulraj@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

Pinmux registers are sequential, and do not need to be enumerated out as they
currently are.  This reduces code volume and keeps things simple.

If some future SoC comes up with a discontiguous register map, PINMUX() can
then be expanded with local token pasting.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch allows for gpio controllers that deviate from those found on
traditional davinci socs.  davinci_soc_info has an added field to indicate the
soc-specific gpio controller type.  The gpio initialization code then bails
out if necessary.

More elements (tnetv107x) to be added later into enum davinci_gpio_type.
Signed-off-by: NCyril Chemparathy <cyril@ti.com>
Tested-by: NSandeep Paulraj <s-paulraj@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

The IDE platform device is registered in three different places (2 board files
for DM644x and in dm646x.c for DM646x) while both the IDE base address and the
IDE IRQ are the same for both SoCs -- therefore,  the proper place for the IDE
platform seems to be in devices.c. Merge the IDE platform data and registration
code and create davinci_init_ide() in place of dm646x_init_ide()...
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

- Convert data/functions to static
- include headers for missing declarations
- pointer cleanups:  struct foo *__iomem f --> struct foo __iomem *f;
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

Remove unneeded 'struct davinci_clk' wrapper around 'struct clk_lookup'
and use clkdev_add_table() to add the list of clocks in one go.
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

Currently, the edma_noevent list is passed from platform data.
But on some architectures, there will be many EDMA channels
which will not be used at all. This patch scans all the
platform devices and then builds a list of events which are
not being used. The unused event list will be used to allocate
EDMA channels in case of EDMA_CHANNEL_ANY usage instead of the
edma_noevent being used earlier for this purpose.

This patch is based on David Brownells's suggestion at
http://article.gmane.org/gmane.linux.davinci/15176.
Signed-off-by: NSudhakar Rajashekhara <sudhakar.raj@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

DM6467T (T for Turbo) is a newer and faster DM6467
part from TI. The new part supports 1080p video and
has the ARM running at 495MHz. More SoC information:

http://focus.ti.com/docs/prod/folders/print/tms320dm6467t.html

Spectrum Digital, Inc has a new EVM for this part.
It is _mostly_ same as the older DM6467 EVM except
for a 33MHz crystal input and THS8200 video encoder
for 1080p support.

The meat of this patch is dedicated to initializing
the crystal frequency from EVM board file.

Additional notes:
I did consider some alternative ways to make the crystal
input board specific including - (1) having board code
initialize the crystal frequency using the first member
of soc_info->cpu_clks array (2) introducing a new ref_clk_rate
member in soc_info structure.

But, the current way seems to be the simplest and least
intruding considering that both the clock array and SoC
info structure are actually private to the SoC file. Also
the fact that davinci_common_init() initializes both the
soc_info and clocks in one go.
Signed-off-by: NSekhar Nori <nsekhar@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

There have accumulated quite a lot of them after the code reorganizations...

In several cases I had to replace #include <linux/dma-mapping.h> which wasn't
needed directly but happened to #include <linux/err.h> which was needed.
Signed-off-by: NSergei Shtylyov <sshtylyov@ru.mvista.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

DM6467 silicon revisions 3.x have variant field in JTAGID register as '1'.
This path adds entry for the same in dm646x_ids to be able to boot on boards
with 3.x revision chips.

Also modifies name for 'variant=0' (revisions 1.0, 1.1).
Signed-off-by: NHemant Pedanekar <hemantp@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

VPIF display changes (Chaithrika)

Add platform device and resource structures. Also define a platform specific
clock setup function that can be accessed by the driver to configure the clock
and CPLD.

VPIF caputure changes (Murali)

1) Modify vpif_subdev_info to add board_info, routing information and
   vpif interface configuration. Remove addr since it is part of
   board_info

2) Add code to setup channel mode and input decoder path for vpif
   capture driver

Also incorporated comments against version v0 of the patch series and
added a spinlock to protect writes to common registers

Tested on DM6467 on channel 0 using TVP514x. Following bootargs used
for drivers:

   vpif_capture.ch0_bufsize=829440 vpif_display.ch2_bufsize=829440
Signed-off-by: NManjunath Hadli <mrh@ti.com>
Signed-off-by: NBrijesh Jadav <brijesh.j@ti.com>
Signed-off-by: NChaithrika U S <chaithrika@ti.com>
Reviewed-by: NHans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: NMuralidharan Karicheri <m-karicheri2@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

There is no need to pass clock name strings in platform_data.
Instead, setup clkdev nodes to have correct ASoC device names.
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch adds platform data and init function for IDE which could be called
from board specific file to register IDE device.

Note that for 594MHz device the transfer mode is limited to UDMA4 since ideclk
rate is less than 100 MHz, which forces udma_mask in palm_bk3710.c to UDMA4,
while for 729MHz device, it is UDMA5.
Signed-off-by: NHemant Pedanekar <hemantp@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

This patch adds clock data for IDE and also updates pin mux mask for ATA so as
to disable PCI when ATA is selected.
Signed-off-by: NHemant Pedanekar <hemantp@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

1) Registers the platform devices for ASP on dm355, dm644x and dm646x
   so that the machine driver can probe to get ASP related platform
   data.
2) Move towards definition of the asp clocks using physical name(for
   dm355 and dm644x)
3) Add platform data to board specific files.
Signed-off-by: NNaresh Medisetty <naresh@ti.com>
Signed-off-by: NChaithrika U S <chaithrika@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

watchdog info is not needed in soc_info, platform_device can
be used directly in core code.
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

Adds McASP clock support for the two instances of mcasp (mcasp0,mcasp1). This
patch is part of the audio support for dm646x series.
Signed-off-by: NNaresh Medisetty <naresh@ti.com>
Signed-off-by: NChaithrika U S <chaithrika@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

Enables module clock for DM646x EDMA channel controller and transfer
controller.
Signed-off-by: NNaresh Medisetty <naresh@ti.com>
Signed-off-by: NSudhakar Rajashekhara <sudhakar.raj@ti.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

- restructure to support multiple channel controllers by using
  additional struct resources for each CC

- interface changes visible to EDMA clients

  Introduce macros to build IDs from controller and channel number,
  and to extract them. Modify the edma_alloc_slot function to take an
  extra argument for the controller.

  Also update ASoC drivers to use API.  ASoC changes
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

- Move queue related mappings to dm<soc>.c

  EDMA in DM355 and DM644x has two transfer controllers while DM646x
  has four transfer controllers. Moving the queue to tc mapping and
  queue priority mapping to dm<soc>.c will be helpful to probe these
  mappings from platform device so that the machine_is_* testing will
  be avoided.

- add channel mapping logic

  Channel mapping logic is introduced in dm646x EDMA. This implies
  that there is no fixed association for a channel number to a
  parameter entry number. In other words, using the DMA channel
  mapping registers (DCHMAPn), a PaRAM entry can be mapped to any
  channel. While in the case of dm644x and dm355 there is a fixed
  mapping between the EDMA channel and Param entry number.
Signed-off-by: NNaresh Medisetty <naresh@ti.com>
Signed-off-by: NSudhakar Rajashekhara <sudhakar.raj@ti.com>
Reviewed-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

Package on-chip SRAM.  It's always accessible from the ARM, so
set up a standardized virtual address mapping into a 128 KiB
area that's reserved for platform use.

In some cases (dm6467) the physical addresses used for EDMA are
not the same as the ones used by the ARM ... so record that info
separately in the SOC data, for chips (unlike the OMAP-L137)
where SRAM may be used with EDMA.

Other blocks of SRAM, such as the ETB buffer or DSP L1/L2 RAM,
may be unused/available on some system.  They are ignored here.
Signed-off-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>

Different SoC have different numbers of pinmux registers and other
resources that overlap with each other.  To clean up the code and
eliminate defines that overlap with each other, move the PINMUX
defines to the SoC specific files.
Signed-off-by: NMark A. Greer <mgreer@mvista.com>
Signed-off-by: NKevin Hilman <khilman@deeprootsystems.com>