The code that reprograms the SDRC memory controller during CORE DVFS,
mach-omap2/sram34xx.S:omap3_sram_configure_core_dpll(), does not
ensure that all L3 initiators are prevented from accessing the SDRAM
before modifying the SDRC AC timing and MR registers.  This can cause
memory to be corrupted or cause the SDRC to enter an unpredictable
state.  This patch places that code behind a Kconfig option,
CONFIG_OMAP3_SDRC_AC_TIMING for now, and adds a note explaining what
is going on.  Ideally the code can be added back in once supporting
code is present to ensure that other initiators aren't touching the
SDRAM.  At the very least, these registers should be reprogrammable
during kernel init to deal with buggy bootloaders.  Users who know
that all other system initiators will not be touching the SDRAM can
also re-enable this Kconfig option.

This is a modification of a patch originally written by Rajendra Nayak
<rnayak@ti.com> (the original is at http://patchwork.kernel.org/patch/51927/).
Rather than removing the code completely, this patch just comments it out.

Thanks to Benoît Cousson <b-cousson@ti.com> and Christophe Sucur
<c-sucur@ti.com> for explaining the technical basis for this and for
explaining what can be done to make this path work in future code.
Thanks to Richard Woodruff <r-woodruff2@ti.com>, Nishanth Menon
<nm@ti.com>, and Olof Johansson <olof@lixom.net> for their comments.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>
Cc: Rajendra Nayak <rnayak@ti.com>
Cc: Christophe Sucur <c-sucur@ti.com>
Cc: Benoît Cousson <b-cousson@ti.com>
Cc: Richard Woodruff <r-woodruff2@ti.com>
Cc: Nishanth Menon <nm@ti.com>
Cc: Olof Johansson <olof@lixom.net>

IS_ERR returns only 1 or 0, and the functions return a negative error
in other cases anyways.
Signed-off-by: NRoel Kluin <roel.kluin@gmail.com>
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

This patch rearranges the order of structure members in struct powerdomain
to avoid wasting memory due to alignment restrictions.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

Replace some bare constants with power states.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>
Cc: Kevin Hilman <khilman@deeprootsystems.com>

OMAP24xx chips don't support software-configurable sleep dependencies.
Test early for this so the compiler can redact the entire function body
on OMAP24xx.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

Avoid cluttering the Kconfig space with debug options that are rarely
used.  These can now be enabled and disabled by patching the "#undef DEBUG"
in the source files with "#define DEBUG", conforming to the practice for
the rest of the linux-omap code.

Also, while we're here, some lines in plat-omap/Kconfig use sets of
leading spaces when those lines should start with tabs.  Convert most
of them to use tabs.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>
Cc: Tony Lindgren <tony@atomide.com>

Device drivers and loadable modules should not be calling these
prm_* and cm_* functions, so stop exporting them.  Only core code
and device driver integration code (in arch/arm/*omap*) should
call these functions.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

The OMAP1 clock code currently #includes a large .h file full of static
data structures.  Instead, define the data in a .c file.

Russell King <linux@arm.linux.org.uk> proposed this new arrangement:

    http://marc.info/?l=linux-omap&m=125967425908895&w=2

This patch also deals with most of the flagrant checkpatch violations.

While here, separate the mpu_rate data structures out into their own
files, opp.h and opp_data.c.  In the long run, these mpu_rate tables
should be replaced with OPP code.

Also includes a patch from Felipe Balbi <felipe.balbi@nokia.com> to
mark omap1_clk_functions as __initdata to avoid a section warning:

    http://patchwork.kernel.org/patch/64366/Signed-off-by: NPaul Walmsley <paul@pwsan.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Felipe Balbi <felipe.balbi@nokia.com>
Cc: Nishanth Menon <nm@ti.com>

mach-omap1/clock.c:omap1_clk_disable_unused() contains a test that
assumes that the clock structures are available in the file's
namespace.  After a following patch, this will no longer be the case.
So we need to reimplement that test.  It turns out that we already
have a facility in the clock framework to handle this case - the
ENABLE_ON_INIT flag - used on OMAP2/3.  Remove the offending test and
mark the clocks that it was intended to catch as ENABLE_ON_INIT.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

The OMAP2 clock code currently #includes a large .h file full of static
data structures.  Instead, define the data in a .c file.

Russell King <linux@arm.linux.org.uk> proposed this new arrangement:

    http://marc.info/?l=linux-omap&m=125967425908895&w=2

This patch also deals with most of the flagrant checkpatch violations.

While here, separate the prcm_config data structures out into their own
files, opp2xxx.h and opp24{2,3}0_data.c, and only build in the OPP tables
for the target device.  This should save some memory.  In the long run,
these prcm_config tables should be replaced with OPP code.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Richard Woodruff <r-woodruff2@ti.com>
Cc: Nishanth Menon <nm@ti.com>

The OMAP3 clock code currently #includes a large .h file full of static
data structures.  Instead, define the data in a .c file.

Russell King <linux@arm.linux.org.uk> proposed this new arrangement:

    http://marc.info/?l=linux-omap&m=125967425908895&w=2Signed-off-by: NPaul Walmsley <paul@pwsan.com>
Cc: Russell King <linux@arm.linux.org.uk>

cpu_mask is reused in the OMAP2xxx clock code to match against both the
CPU-specific rate flags (e.g., RATE_IN_2420) and the OMAP clkdev integration
code CPU flags (e.g., CK_242X).  This means that any patch that renumbers the
CK_* macros, as the next patch does, will probably break.  This patch
separates the clkdev_omap and clksel_rate CPU type detection flags so
the CK_* macros can be renumbered freely.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

clock34xx.c contains some macros which probably belong in mach-omap2/sdrc.h.
Move those macros to mach-omap2/sdrc.h.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

Similar to the previous patch, the APLL code relied on the presence of the
static struct clks in its own namespace.  The APLL code didn't use them for
validation, however - it adjusted its own internal state depending on
the struct clk * that called it.  Now that static struct clks are
leaving the clock24xx.c namespace, use a more durable method: split the
omap2_clk_fixed_enable() function into omap2_clk_apll96_enable() and
omap2_clk_apll54_enable().  They still share a disable function.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

Some parts of the clock code took advantage of the fact that the statically
allocated clock tree was in clock{,24xx,34xx}.c's local namespace to do some
extra argument checks.  These are overzealous and are more difficult to
maintain when the clock tree is in a separate namespace, so, remove them.
Signed-off-by: NPaul Walmsley <paul@pwsan.com>

Signed-off-by: NTomi Valkeinen <tomi.valkeinen@nokia.com>
Acked-by: NTony Lindgren <tony@atomide.com>

Signed-off-by: NTomi Valkeinen <tomi.valkeinen@nokia.com>

The core files of DSS2. DSS2 commits are split a bit artificially to
make the individual commits smaller, and DSS2 doesn't compile properly
without the rest of the core commits. This shouldn't be a problem, as no
configuration uses DSS2 yet.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@nokia.com>

VRFB rotation engine is a block in OMAP2/3 that offers 12 independent
contexts that can be used for framebuffer rotation.

Each context has a backend area of real memory, where it stores the
pixels in undisclosed format. This memory is offered to users via 4
virtual memory areas, which see the same memory area in different
rotation angles (0, 90, 180 and 270 degrees).
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@nokia.com>

Add a Video RAM manager for OMAP 2 and 3 platforms. VRAM manager is used
to allocate large continuous blocks of SDRAM or SRAM. The features VRAM
manager has that are missing from dma_alloc_* functions are:

- Support for OMAP2's SRAM
- Allocate without ioremapping
- Allocate at defined physical addresses
- Allows larger VRAM area and larger allocations

The upcoming DSS2 uses VRAM manager.

VRAM area size can be defined in kernel config, board file or with
kernel boot parameters. Board file definition overrides kernel config,
and boot parameter overrides kernel config and board file.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@nokia.com>

The upcoming new display subsystem driver is divided to two devices,
omapdss and omapfb, of which omapdss handles the actual hardware.

This patch adds a dummy omapdss platform device for the current omapfb
driver, which is then used to get the clocks. This will make it possible
for the current and the new display drivers to co-exist.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@nokia.com>
Acked-by: NTony Lindgren <tony@atomide.com>

Split arch/arm/plat-omap/include/mach/omapfb.h into two files:

include/linux/omapfb.h - ioctls etc for userspace and some kernel
                         stuff for board files
drivers/video/omap/omapfb.h - for omapfb internal use

This cleans up omapfb.h and also makes it easier for the upcoming new
DSS driver to co-exist with the old driver.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@nokia.com>
Acked-by: NTony Lindgren <tony@atomide.com>

SMS_ROT_* registers are used by VRFB rotation engine.
Signed-off-by: NTomi Valkeinen <tomi.valkeinen@nokia.com>
Acked-by: NTony Lindgren <tony@atomide.com>
Acked-by: NPaul Walmsley <paul@pwsan.com>

The DMA_CH_VALID bit used in the channel index should be masked out before
using it to access the channel array.
Signed-off-by: NRamax Lo <ramaxlo@gmail.com>
Signed-off-by: NBen Dooks <ben-linux@fluff.org>

The function s3c_pwm_remove is used only wrapped by __devexit_p so define
it using __devexit.
Signed-off-by: NUwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: NSam Ravnborg <sam@ravnborg.org>
Cc: Russell King <linux@arm.linux.org.uk>
Cc: Ben Dooks <ben-linux@fluff.org>
Cc: Peter Korsgaard <jacmet@sunsite.dk>
Cc: linux-kernel@vger.kernel.org
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: NBen Dooks <ben-linux@fluff.org>

Allow s3c24xx_set_fiq() to be exported so that it can be used by modules
wanting FIQ support.
Signed-off-by: NBen Dooks <ben@simtec.co.uk>
Signed-off-by: NSimtec Linux Team <linux@simtec.co.uk>
Signed-off-by: NBen Dooks <ben-linux@fluff.org>

The function s3c_adc_remove is used only wrapped by __devexit_p so define
it using __devexit.
Signed-off-by: NUwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: NSam Ravnborg <sam@ravnborg.org>
Cc: Russell King <linux@arm.linux.org.uk>
Acked-By: NBen Dooks <ben@simtec.co.uk>
Cc: Ramax Lo <ramaxlo@gmail.com>
Cc: Nelson Castillo <arhuaco@freaks-unidos.net>
Cc: linux-kernel@vger.kernel.org
Cc: Ben Dooks <ben-linux@fluff.org>
Cc: Alexey Dobriyan <adobriyan@gmail.com>
Cc: Ryan Mallon <ryan@bluewatersys.com>
Cc: linux-arm-kernel@lists.infradead.org
Signed-off-by: NBen Dooks <ben-linux@fluff.org>

Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Many features of h3100/h3600 (LCD, PCMCIA, Flash write, etc.)
depend on correct functioning of GPIO expander handled by htc-egpio
driver, so force its building in Kconfig.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Update defconfig to current kernel, enable support for iPAQ H3100
and following drivers: gpio-keys, htc-egpio, ide_cs.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Add support for "Power" and "Action" (joystick center) buttons -
the only buttons on iPaq h3100/h3600 connected to GPIOs
(other buttons are controlled by microcontroller)
Also remove setting PWER for wakeup on Power button press -
gpio-keys driver will handle it.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

After a code reorganization and following split, there's some #includes
now unused. Clean them up and sort remaining alphabetticaly where possible.

Compile tested.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Correct boilerplates after files split. Also shorten them a bit - use
standart GPL wording (as per http://lkml.org/lkml/2007/5/1/220) and
drop changelog, which only entry about h3800 support and abstracted
EGPIOs is just confusing now, as both of these features are gone.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Split common h3600.c into three separate files: h3100.c, h3600.c and
h3xxx.c (the latter contains common code for h3100/h3600)
Copyright boilerplates and #includes are copied intact and will be
cleaned up later.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Combine both headers into one, rename to h3xxx.h and change all
users accordingly.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

As all existing code was converted to gpiolib, drop no more
used pre-gpiolib (bit-shifted) GPIO definintions.
Supply new gpiolib-friendly definitions for GPIOs which
don't have them yet.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

After conversion to gpiolib there's still some GPIOs left, that get
configured in *_mach_init() as outputs (using direct operations
on GPCR/GPDR registers), but otherwise unused. These GPIOs are mainly
sound related and should be configured by corresponding driver once
it is written.

Drop this initialisation and configure all GPIOs as input.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

As all the remaining users of these definitions
(in pcmcia/sa1100_h3600 driver) were converted to gpio_to_irq(),
they can be safely removed.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

As all users of assign_h3600_egpio now converted to gpiolib, we
can safely remove all assign_h3600_egpio handling code and
definitions.
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

Use of gpio_request/gpio_free in some callbacks may look ugly, but
corresponding drivers (sa1100_serial and sa1100_fb) don't provide (yet)
init/exit hooks and registering these gpios in *_mach_init is also
not possible, because htc-gpio driver starts a bit later...
Signed-off-by: NDmitry Artamonow <mad_soft@inbox.ru>
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>