Lots of miscellaneous updates:

  - bugfixes
     * flag hcd-framework "got irq" state earlier
     * reject urb submission if port is suspended
     * linkage fix: unbind became optional
     * initialize vbus error retry count earlier
     * tusb init removes diagnostic-only settings
  - procfs updates
     * always show otg state info in procfs
     * show some TUSB-specific registers in procfs
  - cleanups
     * remove pointless MUSB_ERR stuff (handled better now)
     * tusb init removes dead code
     * more informative vbus error debug message
     * comment fixes
     * etc
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Fix typo for 24xx GPIO resume
Signed-off-by: Tony Lindgren <tony@atomide.com>

This fix assures DMA domain is kept awake till ARM completes
its access to DSP internal memories.
Signed-off-by: Tetsuo Yamabe <tetsuo.yamabe@nokia.com>
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>

In OMAP2420, an incoming GPIO interrupt always sets both GPIO_IRQSTATUS1
and GPIO_IRQSTATUS2, even if the relevant bit is disabled in
GPIO_IRQENABLE1/2 and DSP doesn't use GPIO at all. GPIO_IRQSTATUS1
is for MPU and GPIO_IRQSTATUS2 is for DSP. If IRQSTATUS is set, this
will prevent the system from going to idle state. This patch also clears
IRQSTATUS2 to avoid the above situation.
Signed-off-by: Hiroshi DOYU <Hiroshi.DOYU@nokia.com>
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>

Signed-off-by: Imre Deak <imre.deak@solidboot.com>
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>

Add a new fifo configuration that makes better use of the memory
available for use as TUSB fifos, and all the endpoint hardware.
Now almost every endpoint is available, and that's the default
setup.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Additional checks for TUSB DMA. Fixes MSC ErrorRecoveryTest.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Fix omap_set_dma_callback spinlock release
Signed-off-by: Tony Lindgren <tony@atomide.com>

Fix spinlock recursion for dyntick. Modified version based
on Imre Deak's earlier patch.
Signed-off-by: Tony Lindgren <tony@atomide.com>

This is needed to avoid sleeping function called from invalid context.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Enable the TUSB irq signal as a wake event source.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Second pass over TUSB power related code:

  - Add new field in board-specific platform data, giving a nonzero floor to
    the amount of VBUS power that peripherals could draw; pass that to the
    platform code.

  - Implement the call letting gadget drivers say how much VBUS power their
    current configuration may draw, and use that to initialize appropriately
    (pure-peripheral vs OTG modes) on USB reset and disconnect.

  - Provide TUSB-specific hook into OTG transceiver logic to support that
    VBUS call; and _only_ change the VBUS usage through that call.

  - Remove obsolete comment and bitmask declarations.

On one board I measured 36 mA (at 5.1V) coming from VBUS not the battery
using the default gadget drivers in non-OTG TUSB configuration.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Minor cleanups (symbols don't get exported)
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Add workarounds for various chip issues relating to the TUSB chip PLL
getting properly initialized after resume.
Signed-off-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Remove Remove superfluous/recursive locking for GPIO
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>

Clean up cable status sysfs_notify() events
Signed-off-by: Tony Lindgren <tony@atomide.com>

Start updating the TUSB power related code ... first pass, use the chip's
idle mode whenever the USB link is inactive (disconnected or suspended).
That rule applies to both host and peripheral roles, and saves a fair
amount of power (50mA at 3.3V, etc).

Plus some cleanups.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Also makes it work with any transfer sizes that are multiple
of 32 bytes.
Signed-off-by: Tony Lindgren <tony@atomide.com>

Disable GPT when not needed to make deep sleep possible.
Signed-off-by: Timo Teras <timo.teras@solidboot.com>

Add enable/disable functions which effectively control the GPT iclk and fclk.
Signed-off-by: Timo Teras <timo.teras@solidboot.com>
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>

After adjusting clock parameters, OMAP2 CPUs need a memory
barrier to make sure the changes go into effect immediately.
Otherwise bad things will happen if we try to access the
peripheral whose clock is just being enabled.
Signed-off-by: Juha Yrjola <juha.yrjola@solidboot.com>

Fix mmc driver no_switch error path
Signed-off-by: Komal Shah <komal_shah802003@yahoo.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Here is a small patch that adds alsa support on H3. I have tested it by
playing a few raw, wav files.
Signed-off-by: Tony Lindgren <tony@atomide.com>

Add gsm phone support for the mixer in tsc2101 alsa driver. This allows
selecting cellphone for the playback target via alsa mixer interface.
Selection will connect gsm module to speaker and microphone allowing the
user to speak to phone and listen the opponents voice. Tested in ipaq
h63xx by using gomunicator as an application. Tux wanna call home, and
now Tux can finally call home :-)
Signed-off-by: Mika Laitio <lamikr@cc.jyu.fi>
Signed-off-by: Tony Lindgren <tony@atomide.com>

tsc2101 mixer fixes. Fix target selection control name and set index to
0 for all controls as the driver uses always same tsc2101 codec.
Signed-off-by: Mika Laitio <lamikr@cc.jyu.fi>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Fix init warning in tsc2101 alsa driver. WARNING:
sound/arm/omap/snd-omap-alsa-tsc2101.o - Section mismatch: reference to
.init.text: from .data between '' (at offset 0x6c) and
'tsc2101_snd_omap_alsa_capture'
Signed-off-by: Mika Laitio <lamikr@cc.jyu.fi>
Signed-off-by: Tony Lindgren <tony@atomide.com>

This patch adds the platform file components for the Amstrad Delta LCD
framebuffer. I have confirmed it works with latest git. The patch to
clock.c /is/ needed for this to work.
Signed-off-by: Jonathan McDowell <noodles@earth.li>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Add the backlight device to the list of platform devices to register
on the Palm Tungsten E board. In the same way the device can be added
in other board files if a board has a single LCD with the backlight
controlled by OMAP PWL module.
Signed-off-by: Andrzej Zaborowski <balrog@zabor.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

This adds a structure that can be used in the omap board files to pass
data about backlight device when registering platform devices.
Signed-off-by: Andrzej Zaborowski <balrog@zabor.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

This adds a driver for the PWL module of OMAP CPUs in the Linux
backlight API. The driver allows changing the backlight instensity and
switching it off when system goes idle. The declaration of struct
omap_backlight_config is sent in a separate patch.
Signed-off-by: Andrzej Zaborowski <balrog@zabor.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Prevent keyboard interrupt from staying masked when omap-keypad driver
is suspended to allow this interrupt to trigger a system wake up.
Locks are used as suggested by Juha Yrjölä.
Signed-off-by: Andrzej Zaborowski <balrog@zabor.org>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Add GPIO API howto.

Signed-off-by: Arnold <abo_gwapo_at_yahoo.com>
Signed-off-by: Dirk Behme <dirk.behme_at_gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

Remove warnings:

arch/arm/plat-omap/dsp/dsp_common.c: In function
'dsp_cpustat_update':
arch/arm/plat-omap/dsp/dsp_common.c:317: warning: implicit
declaration of function 'enable_irq'
arch/arm/plat-omap/dsp/dsp_common.c:325: warning: implicit
declaration of function 'disable_irq'

arch/arm/plat-omap/dsp/task.c: In function 'dsp_task_write_bk':
arch/arm/plat-omap/dsp/task.c:1232: warning: implicit
declaration of function 'disable_irq'
arch/arm/plat-omap/dsp/task.c:1232: warning: implicit
declaration of function 'enable_irq'

Signed-off-by: Dirk Behme <dirk.behme_at_gmail.com>
Signed-off-by: Tony Lindgren <tony@atomide.com>

These accidentally got changed to readw in musb-m1 patch.
Signed-off-by: Tony Lindgren <tony@atomide.com>

Catch up to info from latest HDRC core spec ... some new registers
and fields, and constraints notably

 - not clearing TX fifo unless it's empty (that might explain some
   host side glitching).

 - not all POWER fields are writable
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net>

Make TUSB DMA debug less verbose
Signed-off-by: Tony Lindgren <tony@atomide.com>

tusb multipoint fixes