firefly-linux-kernel-4.4.55.git
14 years agousb: host: ehci-hub: Add port_reset_no_wait quirk
Benoit Goby [Tue, 31 Aug 2010 23:36:14 +0000 (16:36 -0700)]
usb: host: ehci-hub: Add port_reset_no_wait quirk

Tegra quirk: The PORT_RESET bit in PORTSC1 does not need to be cleared
and there is no need to wait for it to clear. The bit will automatically
change to 0 when the bus-reset sequence is done and an interrupt will be
generated.

Change-Id: I645417013af46785a249096ebc06a1f688228d94
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: Compile usb_phy.c only if usb support is enabled
Benoit Goby [Tue, 31 Aug 2010 22:59:00 +0000 (15:59 -0700)]
[ARM] tegra: Compile usb_phy.c only if usb support is enabled

Change-Id: I190bab00533d065171ab2478363830c6d95e66cb
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: usb_phy: Moved suspend/resume code to ehci driver
Benoit Goby [Tue, 31 Aug 2010 22:38:09 +0000 (15:38 -0700)]
[ARM] tegra: usb_phy: Moved suspend/resume code to ehci driver

The suspend/resume methods should go to the ehci driver as they only
save/restore ehci registers.

Remove phy_init and initialize all phy registers in power_on as all the
registers need to be reinitialized after LP0.

Change-Id: I649d2cb66339aaff2d7db442a5441d2d3f06268b
Signed-off-by: Benoit Goby <benoit@android.com>
14 years agoserial: tegra_hsuart: Fix typo preventing non 8-bit modes
Colin Cross [Thu, 2 Sep 2010 23:16:24 +0000 (16:16 -0700)]
serial: tegra_hsuart: Fix typo preventing non 8-bit modes

Change-Id: Ife085017dd7a956c46ad4077c5bbf33d5a049f3d
Signed-off-by: Colin Cross <ccross@android.com>
14 years agotegra: sdhci: Remove set_clock op, not needed
Todd Poynor [Thu, 2 Sep 2010 23:13:24 +0000 (16:13 -0700)]
tegra: sdhci: Remove set_clock op, not needed

The SDHCI controller specifies a maximum SDCLK speed of 48MHz, which is
now in agreement with the platform clock, and so the SDHCI host max_clk
no longer needs to be overriden.

Change-Id: Ie8c7f643d956cfd1bb83675708336278482c40d8
Signed-off-by: Todd Poynor <toddpoynor@google.com>
14 years ago[ARM] tegra: clocks: Set all SDMMC clocks to 48MHz at boot
Todd Poynor [Thu, 2 Sep 2010 22:10:29 +0000 (15:10 -0700)]
[ARM] tegra: clocks: Set all SDMMC clocks to 48MHz at boot

48MHz is the max speed reported by present-day Tegra2 SDHCI controllers
and is the max speed we can run without adjusting for DVFS changes.

Change-Id: I3f2c23ffdfc40aebe8211688077003f09f599f1a
Signed-off-by: Todd Poynor <toddpoynor@google.com>
14 years ago[ARM] tegra: add tegra_wdt_device to devices.c
Gary King [Fri, 27 Aug 2010 03:00:29 +0000 (20:00 -0700)]
[ARM] tegra: add tegra_wdt_device to devices.c

Change-Id: I8098fa38ebef8636a6eeb1179e679201445e5237
Signed-off-by: Gary King <gking@nvidia.com>
14 years agowatchdog: add tegra_wdt driver
Gary King [Thu, 26 Aug 2010 17:59:50 +0000 (10:59 -0700)]
watchdog: add tegra_wdt driver

add a driver for the hardware watchdog timer embedded in NVIDIA
Tegra SoCs

Change-Id: I45bc829f26f350143d5a07e1f4ddc46d24f3a54c
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: ventana: do not tristate backlight & battery pins
Gary King [Sun, 15 Aug 2010 16:46:11 +0000 (09:46 -0700)]
[ARM] tegra: ventana: do not tristate backlight & battery pins

Change-Id: I5f844fd58a379cb7191f03bb23599cd4061caf5c
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: ventana: fix suspend build break
Gary King [Tue, 31 Aug 2010 19:05:31 +0000 (12:05 -0700)]
[ARM] tegra: ventana: fix suspend build break

update the suspend platform data for ventana to match recent changes

Change-Id: I142123027453ee34f80c094fb7cf1fcb464e889d
Signed-off-by: Gary King <gking@nvidia.com>
14 years agousb: host: ehci-hcd: add controller_resets_phy quirk
Gary King [Tue, 8 Dec 2009 04:52:35 +0000 (20:52 -0800)]
usb: host: ehci-hcd: add controller_resets_phy quirk

only reset the controller when doing so won't also reset the phy (Tegra quirk)

Change-Id: I549a18977d0d5ebfa12c32016aa9e6bffaa8643c
Signed-off-by: Gary King <gking@nvidia.com>
14 years agopanjit_touch: fix warning "control reaches end of non-void function"
Gary King [Tue, 3 Aug 2010 22:09:46 +0000 (15:09 -0700)]
panjit_touch: fix warning "control reaches end of non-void function"

Change-Id: I3a43dc58edff3687e294c2d2a09d1f6e06e27fce
Signed-off-by: Gary King <gking@nvidia.com>
14 years agoinput: touchscreen driver for panjit capacitive i2c touch panels
Gary King [Fri, 30 Jul 2010 15:53:11 +0000 (08:53 -0700)]
input: touchscreen driver for panjit capacitive i2c touch panels

driver supports the MGG1010AI06 and EGG1010AI06 capacitive touch panels

Change-Id: I038030bf4c9acbd0d3d504427a32f4e46632c115
Signed-off-by: Gary King <gking@nvidia.com>
14 years agommc: fix non-arm build due to eMMC 4.3+ change
Olof Johansson [Wed, 28 Jul 2010 04:11:15 +0000 (23:11 -0500)]
mmc: fix non-arm build due to eMMC 4.3+ change

drivers/mmc/core/mmc.c:228: error: ‘SZ_256K’ undeclared (first use in this function)

Introduced by:
    mmc: subtract boot sectors from disk size for eMMC 4.3+ devices

(SZ_* doesn't exist on anything but arm)

Change-Id: I981217adea4ef56bf870562b6711488f3f4bf830
Signed-off-by: Olof Johansson <olof@lixom.net>
14 years agoserial: tegra_hsuart: fix receive DMA, RTS, timeout, and tx trigger
Jay Cheng [Thu, 19 Aug 2010 11:09:49 +0000 (07:09 -0400)]
serial: tegra_hsuart: fix receive DMA, RTS, timeout, and tx trigger

initialize baud rate and configuration settings to safe default values
when receive DMA is in use, so that the DMA request may be enqueued at
initialization time

re-enqueue the receive DMA buffer immediately it is dequeued by the
DMA threshold callback and the receive ISR, rather than waiting for the
DMA complete callback

originally fixed by Gary King <gking@nvidia.com>

Fixing tx trigger level setting:

On tegra uart, the FCR setting for different tx trigger level
is not same as the 16550 tx trigger level setting. The tegra
uart have the setting in reverse direction on tx fifo attention
level:
b00 for 16 bytes attention level.
b01 for 8 byte attention level.
b10 for 4 byte attention level
b11 for 1 byte attention level.

The rx trigger attention level match with the standard uart
FCR register setttings.

Also fixing the typo in code when setting DTR.
originally fixed by Laxman Dewangan (ldewangan@nvidia.com)

Change-Id: Iea00478f143e61c604828035c6c92d614fa7cccb
Signed-off-by: Jay Cheng <jacheng@nvidia.com>
14 years agoserial: tegra_hsuart: Fix void return type on writel accessors
Colin Cross [Mon, 2 Aug 2010 03:05:38 +0000 (20:05 -0700)]
serial: tegra_hsuart: Fix void return type on writel accessors

Signed-off-by: Colin Cross <ccross@google.com>
14 years agoserial: tegra_hsuart: fix section mismatch error
Gary King [Tue, 29 Jun 2010 01:18:06 +0000 (18:18 -0700)]
serial: tegra_hsuart: fix section mismatch error

Change-Id: Ib4c1872c691a98f11bc7c8553df41bec93f64285
Signed-off-by: Gary King <gking@nvidia.com>
14 years agoserial: tegra_hsuart: Cleanups and bug fixes
Colin Cross [Fri, 4 Jun 2010 17:56:50 +0000 (10:56 -0700)]
serial: tegra_hsuart: Cleanups and bug fixes

tegra_start_tx was called directly by the serial core, as
well as from dma and serial interrupts to queue the next
block of data.  Separate out the "queue next data"
functionality into tegra_start_next_tx.

Also fixes TX PIO by adjusting FIFO sizes and prevents
last characters from getting lost by spinning on TEMT
before disabling clocks.

Change-Id: If8ce15490f77dcbde48f1e64959d5c3f0ec35120
Signed-off-by: Colin Cross <ccross@android.com>
14 years agoserial: tegra_hsuart: Fix DMA receive, drop err in dma callbacks
Colin Cross [Fri, 4 Jun 2010 00:50:27 +0000 (17:50 -0700)]
serial: tegra_hsuart: Fix DMA receive, drop err in dma callbacks

Change-Id: I4c4bc12e8c883912f8979b75a8eb418a62e77f94
Signed-off-by: Colin Cross <ccross@android.com>
14 years agoserial: Add tegra_hsuart driver
Colin Cross [Mon, 5 Apr 2010 20:30:14 +0000 (13:30 -0700)]
serial: Add tegra_hsuart driver

Change-Id: Ic2d0a43ac3597c8a270758af6ea3090acdd0fe7f
Signed-off-by: Colin Cross <ccross@android.com>
14 years agommc: sdhci-tegra: Always call sdhci suspend/resume functions
Dmitry Shmidt [Fri, 13 Aug 2010 01:17:47 +0000 (18:17 -0700)]
mmc: sdhci-tegra: Always call sdhci suspend/resume functions

Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
14 years agommc: sdhci: Skip mmc suspend/resume ops for SDIO device,
Dmitry Shmidt [Fri, 13 Aug 2010 01:08:40 +0000 (18:08 -0700)]
mmc: sdhci: Skip mmc suspend/resume ops for SDIO device,

enable/disable sdc irq instead of request/free,
mask all sdc interrupts on suspend

Change-Id: I3fb137363c6e639882b796aff8b80784c5256a71
Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
14 years agommc: subtract boot sectors from disk size for eMMC 4.3+ devices
Gary King [Fri, 16 Jul 2010 00:11:32 +0000 (17:11 -0700)]
mmc: subtract boot sectors from disk size for eMMC 4.3+ devices

the csd sector count reported by eMMC 4.3+ cards includes the boot
partition size; subtract this from the size reported to the disk
since the boot partition is inaccessible

Change-Id: I601b83aa0159b7aa446409ea8c945b256dd0b5b1
Signed-off-by: Gary King <gking@nvidia.com>
14 years agotegra-sdhci: treat -1 as invalid card detect GPIO
Gary King [Thu, 15 Jul 2010 03:41:23 +0000 (20:41 -0700)]
tegra-sdhci: treat -1 as invalid card detect GPIO

0 is a valid GPIO number (GPIO_PA0), so the check in the tegra
SDHCI driver for a valid card detect pin should compare against -1
instead

Change-Id: I47ad31eaf89472b2e6938978d72a9d0b92a74958
Signed-off-by: Gary King <gking@nvidia.com>
14 years agodrivers: mmc: sdhci-tegra: Add suspend/resume processing
Dmitry Shmidt [Thu, 1 Jul 2010 21:47:07 +0000 (14:47 -0700)]
drivers: mmc: sdhci-tegra: Add suspend/resume processing

Signed-off-by: Dmitry Shmidt <dimitrysh@google.com>
14 years agommc: sdhci: Initial Tegra sdhci driver
Colin Cross [Thu, 11 Mar 2010 04:42:35 +0000 (20:42 -0800)]
mmc: sdhci: Initial Tegra sdhci driver

Change-Id: I6a761b978224fe91dc7928ef5595af2cbd66fc92
Signed-off-by: Colin Cross <ccross@android.com>
14 years agovideo: tegra: include nvhost.h instead of nvhost_dev.h
Erik Gilling [Tue, 24 Aug 2010 23:43:44 +0000 (16:43 -0700)]
video: tegra: include nvhost.h instead of nvhost_dev.h

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agovideo: tegra: set window size on mode change
Erik Gilling [Tue, 24 Aug 2010 23:13:39 +0000 (16:13 -0700)]
video: tegra: set window size on mode change

otherwise it's hardcoded by the platform-data

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agovideo: tegra: add hdmi output support
Erik Gilling [Tue, 24 Aug 2010 04:36:56 +0000 (21:36 -0700)]
video: tegra: add hdmi output support

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agovideo: tegra: add hotplug display support
Erik Gilling [Tue, 24 Aug 2010 04:28:12 +0000 (21:28 -0700)]
video: tegra: add hotplug display support

This is needed for HDMI.

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agovideo: add fb_edid_add_monspecs for parsing extended edid information
Erik Gilling [Tue, 17 Aug 2010 23:16:19 +0000 (16:16 -0700)]
video: add fb_edid_add_monspecs for parsing extended edid information

Modern monitors/tvs have more extended EDID information blocks which can
contain extra detailed modes.  This adds a fb_edid_add_monspecs function
which drivers can use to parse those additions blocks.

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agovideo: tegra: add host bus support to display controller
Erik Gilling [Fri, 6 Aug 2010 02:10:37 +0000 (19:10 -0700)]
video: tegra: add host bus support to display controller

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agovideo: tegra: add skeleton host bus support
Erik Gilling [Tue, 24 Aug 2010 04:58:20 +0000 (21:58 -0700)]
video: tegra: add skeleton host bus support

The host (or host1x) bus sits between the cpu core and the 3d, 2d, camera,
display, and mpeg encoder functions.  It contains provides DMA channels,
hardware mutexes, and synchronization points.

Signed-off-by: Erik Gilling <konkers@android.com>
14 years ago[ARM] tegra: tegra_i2s_audio: remove debugfs, add sysfs interface
Iliyan Malchev [Fri, 20 Aug 2010 22:09:59 +0000 (15:09 -0700)]
[ARM] tegra: tegra_i2s_audio: remove debugfs, add sysfs interface

-- Removed debugfs interface for toggling DMA
-- Added it under sysfs
-- Also added attributes for settings the attention levels of the i2s fifos

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: tegra_i2s_audio: do not reset the i2s fifo unnecessarily
Iliyan Malchev [Tue, 24 Aug 2010 01:42:07 +0000 (18:42 -0700)]
[ARM] tegra: tegra_i2s_audio: do not reset the i2s fifo unnecessarily

This patch removes redundant i2s fifo resets, which also cause blips on playback.
It also moves the fifo attention levels to struct audio_stream.

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: tegra_i2s_audio: allow preloading of the tx fifo with data
Iliyan Malchev [Fri, 20 Aug 2010 22:11:48 +0000 (15:11 -0700)]
[ARM] tegra: tegra_i2s_audio: allow preloading of the tx fifo with data

Add an ioctl to allow the TX fifo to be loaded with data before playback
starts.  Playback can then be started by calling write() on the FIFO, even
with a length of 0.  This will cause the pending data to be played out.

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: tegra_i2s_audio: clean up handling of state
Iliyan Malchev [Wed, 18 Aug 2010 01:36:41 +0000 (18:36 -0700)]
[ARM] tegra: tegra_i2s_audio: clean up handling of state

-- Use consistently the various state flags:
   -- active is set only when there is a read or write in flight
   -- recording_canncelled is set only when recording is stopped via the ioctl()
   -- dma_has_it is used to determine whether DMA is already in flight; do not
      use the state of the fifos for this (e.g., if the TX fifo is empty, do not
      assume that playback is stopped)
   -- added a stop_completion (implemented for readers only) so that readers
      closing a stream can wait until DMA or PIO transactions are stopped
-- Split /dev/audio0_{in,out} into /dev/audio0_{in,in_ctl,out,out_ctl} where the
   _ctl versions have the ioctl()s
-- Introduced an error count per audio_stream; error count is reset on open, can
   be read back & reset through an ioctl

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: tegra_i2s_audio: configure in/out buffer sizes from user space
Iliyan Malchev [Tue, 17 Aug 2010 18:34:39 +0000 (11:34 -0700)]
[ARM] tegra: tegra_i2s_audio: configure in/out buffer sizes from user space

-- Add ioctls for configuring buffer, threshold, and DMA-transaction sizes from
   user space.
-- Buffer sizes are provided in orders of magnitude.
-- Allocate max-sized buffers during probe, and allow the user to resize them
   only within the original allocation, to avoid the risk from kmalloc failing
   due to kernel-heap fragmentation, and also to avoid race conditions on DMA
   shut-down.
-- In tegra_audio_write(), moved the call to start_playback_if_necessary()
   immediately after writing to the fifo.  Otherwise, when the fifo size is
   smaller than what the user is trying to write, the user will block before
   playback is started.
-- Silenced printk spew on spinning on i2s registers after transactions are
   completed.
-- Cleaned up a 80-col style violation in downsample()

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: tegra_i2s_audio: combine audio_in_stream and audio_out_stream
Iliyan Malchev [Tue, 17 Aug 2010 16:56:18 +0000 (09:56 -0700)]
[ARM] tegra: tegra_i2s_audio: combine audio_in_stream and audio_out_stream

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years agousb: host: Add support for suspend and bus suspend
Benoit Goby [Fri, 20 Aug 2010 05:01:58 +0000 (22:01 -0700)]
usb: host: Add support for suspend and bus suspend

On suspend, use phy_suspend save the phy registers so that there is
no need to reset the controller and re-enumerate devices on resume.

Change-Id: I00fe5b87a1b319044724494b8e635b540088a38b
Signed-off-by: Benoit Goby <benoit@android.com>
14 years agousb: gadget: Reset the controller when cable is unplugged
Benoit Goby [Fri, 20 Aug 2010 04:42:23 +0000 (21:42 -0700)]
usb: gadget: Reset the controller when cable is unplugged

Reset the controller when the cable is unplugged to leave it in the idle
state. The OTG driver will reconfigure it on vbus/id pin detection.

Change-Id: I87903ec86f3c35af64a141f27a34cc0720a61b08
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: usb_phy: Add phy_suspend and phy_clk_disable
Benoit Goby [Fri, 20 Aug 2010 04:28:59 +0000 (21:28 -0700)]
[ARM] tegra: usb_phy: Add phy_suspend and phy_clk_disable

phy_suspend save the phy registers and power it off. phy_clk_disable
disable the phy clock to put the phy in low power mode.

Change-Id: If5665ec96beec9441fae3b737923dbe69c86e650
Signed-off-by: Benoit Goby <benoit@android.com>
14 years agousb: gadget: In OTG mode, power down the controller on probe
Benoit Goby [Tue, 10 Aug 2010 03:49:09 +0000 (20:49 -0700)]
usb: gadget: In OTG mode, power down the controller on probe

The OTG driver will power it up on ID pin detection. This avoids
race conditions when the device is powered on with the otg cable
attached.

Change-Id: I6bd5f1d73284f9b8534ef3dfb936b81a9400fa5b
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: ventana: Add pmu device
Colin Cross [Sun, 22 Aug 2010 01:23:34 +0000 (18:23 -0700)]
[ARM] tegra: ventana: Add pmu device

Change-Id: I5fa227893cc64b900ee749d292cb09a728f9ec2f
Signed-off-by: Colin Cross <ccross@google.com>
14 years ago[ARM] tegra: harmony: Add pmu device
Colin Cross [Sun, 22 Aug 2010 01:23:11 +0000 (18:23 -0700)]
[ARM] tegra: harmony: Add pmu device

Originally fixed by Will Deacon.

Change-Id: I2319d269f645766597100f91b2086fde4535c0fd
Signed-off-by: Colin Cross <ccross@google.com>
14 years ago[ARM] tegra: Add pmu to common devices
Colin Cross [Sun, 22 Aug 2010 01:21:23 +0000 (18:21 -0700)]
[ARM] tegra: Add pmu to common devices

Originally fixed by Will Deacon.

Change-Id: I504f8109bedd1eea4bf637395913726a98eb6bb8
Signed-off-by: Colin Cross <ccross@google.com>
14 years ago[ARM] tegra: cpufreq: Adjust twds on frequency scaling
Colin Cross [Fri, 20 Aug 2010 05:19:53 +0000 (22:19 -0700)]
[ARM] tegra: cpufreq: Adjust twds on frequency scaling

Change-Id: I8d187b861e18dfefcbf4566f4c695dfd1e17fe98
Signed-off-by: Colin Cross <ccross@google.com>
14 years ago[ARM] tegra: localtimer: Pass target twd rate
Colin Cross [Fri, 20 Aug 2010 05:08:06 +0000 (22:08 -0700)]
[ARM] tegra: localtimer: Pass target twd rate

Change-Id: I4d57462045732f1f219a62c1be911437adf9a077
Signed-off-by: Colin Cross <ccross@google.com>
14 years ago[ARM] twd: Allow twd rescaling to match cpu frequency
Colin Cross [Fri, 20 Aug 2010 05:07:43 +0000 (22:07 -0700)]
[ARM] twd: Allow twd rescaling to match cpu frequency

The clock to the ARM TWD local timer scales with the cpu
frequency.  To allow the cpu frequency to change while
maintaining a constant TWD frequency, pick a lower target
frequency for the TWD and use the prescaler to divide down
to the closest lower frequency.

This patch provides a new initialization function that takes
a target TWD frequency and the relation between the cpu
clock and the TWD clock, required to be an integer divider
>= 2 by the ARM spec.  It also provides a function to be
called from cpufreq drivers to set the prescaler whenever
the cpu frequency changes.

Also fixes a typo in the printk of the calibrated frequency.

Change-Id: I3fa8ef718ff5518170f1b2bab29efe960741853e
Signed-off-by: Colin Cross <ccross@google.com>
14 years ago[ARM] tegra_i2s_audio: add software downsampling for recorded data + fixes
Iliyan Malchev [Thu, 12 Aug 2010 01:19:47 +0000 (18:19 -0700)]
[ARM] tegra_i2s_audio: add software downsampling for recorded data + fixes

downsampling:
-- add ioctl()s to downsample recorded data
-- supported frequencies are 8kHz, 11.025kHz, 22.05kHz, and 44.1kHz
-- downsamping to stereo and mono
-- default is 11.025kHz mono

fixes:
-- fix crashes from dequeuing DMA requests twice

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years agovideo: tegra: include linux/pm.h in mach/tegradc.h
Erik Gilling [Thu, 12 Aug 2010 02:53:36 +0000 (19:53 -0700)]
video: tegra: include linux/pm.h in mach/tegradc.h

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agovideo: tegra: add init/suspend/resume hooks to outputs
Erik Gilling [Thu, 12 Aug 2010 01:52:50 +0000 (18:52 -0700)]
video: tegra: add init/suspend/resume hooks to outputs

Allows board specific code to handle panel init.

Signed-off-by: Erik Gilling <konkers@android.com>
14 years ago[ARM] tegra: usb_phy: Power up/down bias and otg config pad circuitry
Benoit Goby [Wed, 4 Aug 2010 04:27:49 +0000 (21:27 -0700)]
[ARM] tegra: usb_phy: Power up/down bias and otg config pad circuitry

On tegra, the UTMIP PAD control logic is common to all utmip phys and
are controlled from usb1. This adds a reference count to turn off the
pads when all utmip phy are off.

Change-Id: I3537d5cc52df929f817e547a79da235394d2c265
Signed-off-by: Benoit Goby <benoit@android.com>
14 years agovideo: tegra: checkpatch changes
Erik Gilling [Wed, 11 Aug 2010 22:55:49 +0000 (15:55 -0700)]
video: tegra: checkpatch changes

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agousb: host: Fix kernel warning on probe
Benoit Goby [Tue, 10 Aug 2010 23:36:58 +0000 (16:36 -0700)]
usb: host: Fix kernel warning on probe

Change-Id: I6d5b10e047807f9bcb36809672705f8a8b2d11a4
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: audio_i2s_audio: clean up & support for recording audio
Iliyan Malchev [Fri, 6 Aug 2010 22:41:26 +0000 (15:41 -0700)]
[ARM] tegra: audio_i2s_audio: clean up & support for recording audio

-- add audio_in_stream (identical to audio_out_stream, may merge them later)
-- add support for DMA and PIO recording
-- add ioctls for /dev/audio<n>_in to start and stop recording

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years agovideo: tegra: add tegra display controller driver
Erik Gilling [Mon, 12 Jul 2010 00:06:28 +0000 (17:06 -0700)]
video: tegra: add tegra display controller driver

Notable ommisions:
* support for anything but lvds panels
* inegration with nvhost driver to sync updates with 3D
* FB physical geometry is not set
* lacks interface to set overlay/window x,y offset

v2 changes:
* suspend/resume support
* move code into drivers/video/tegra/dc
* modularize output support
* clean register dumping, add debugfs register file
* code review feedback
* make the display controller register the framebuffer devices

Signed-off-by: Erik Gilling <konkers@android.com>
14 years agovideo: tegra: remove old tegra fb driver
Erik Gilling [Mon, 12 Jul 2010 00:00:24 +0000 (17:00 -0700)]
video: tegra: remove old tegra fb driver

Signed-off-by: Erik Gilling <konkers@android.com>
14 years ago[ARM] tegra: ventana: add suspend support to ventana
Gary King [Wed, 4 Aug 2010 00:48:42 +0000 (17:48 -0700)]
[ARM] tegra: ventana: add suspend support to ventana

add support for system suspend to ventana; currently LP1 (DRAM in
self-refresh, CPU power-gated, core power active) is used.

Change-Id: Idb223a7da94c3b87ea61e0eef1de77206c4171e4
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: ventana: register gpio-keys device
Gary King [Wed, 4 Aug 2010 00:50:19 +0000 (17:50 -0700)]
[ARM] tegra: ventana: register gpio-keys device

Change-Id: Ie2e59f61facb116a5af88478e231542778af9eaf
Signed-off-by: Gary King <gking@nvidia.com>
14 years agousb: host: Fix tegra EHCI suspend in OTG mode
Benoit Goby [Tue, 3 Aug 2010 03:22:58 +0000 (20:22 -0700)]
usb: host: Fix tegra EHCI suspend in OTG mode

Change-Id: Ia46830dfbb8bc39de1280b00ce30efe6022f6bd3
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: clock: Add pll_u to common clock init table
Benoit Goby [Tue, 3 Aug 2010 03:18:11 +0000 (20:18 -0700)]
[ARM] tegra: clock: Add pll_u to common clock init table

Change-Id: I8cdef0406b6fe04551584ae0bae9534b4aec93f6
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: ventana: use tegra_gart_device rather than local definition
Gary King [Tue, 3 Aug 2010 21:54:46 +0000 (14:54 -0700)]
[ARM] tegra: ventana: use tegra_gart_device rather than local definition

Change-Id: I6cb57dd46162c258e32c4946de5a649daff3a79b
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: harmony: use tegra_gart_device rather than local definition
Gary King [Tue, 3 Aug 2010 21:54:13 +0000 (14:54 -0700)]
[ARM] tegra: harmony: use tegra_gart_device rather than local definition

Change-Id: I1b456bb5634a75ae1dba2f5bd64ad71b8abdfc53
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: add gart device to devices.c
Gary King [Tue, 3 Aug 2010 21:53:23 +0000 (14:53 -0700)]
[ARM] tegra: add gart device to devices.c

Change-Id: Id748aa37a3c2234a2a618b09dfa963a4b4ac572c
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] ventana: register i2c, touchscreen & regulator
Gary King [Wed, 28 Jul 2010 20:05:45 +0000 (13:05 -0700)]
[ARM] ventana: register i2c, touchscreen & regulator

register I2C controllers, regulator consumers and the panjit
touchscreen platform_device

Change-Id: Ife13c0d86084f26c734dea2c358f8c4fd3e27a8e
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: fix packet alignment and padding
Gary King [Wed, 28 Jul 2010 22:03:57 +0000 (15:03 -0700)]
[ARM] tegra: fix packet alignment and padding

tegra's DMA controller expects to start transfers at word boundaries,
and the standard packet alignment (2) was resulting in data corruption

also, provide a full cacheline of padding between skbuffs, to eliminate
coherency issues between the processor and USB networking devices.

Change-Id: Ibb508b512f43c8934d35eb182c8738370b7be585
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: generic driver for i2s audio (initial implementation)
Iliyan Malchev [Thu, 29 Jul 2010 21:00:54 +0000 (14:00 -0700)]
[ARM] tegra: generic driver for i2s audio (initial implementation)

-- i2s settings are passed through the board file
-- supports playback (no recording yet)
-- works in DMA and PIO (non-DMA) modes (toggle through debugfs)
-- does NOT perform volume and audio-path control
-- exports /dev/audio<n>_{in, out}, where <n> is the i2s interface
-- assumes that i2s is used such that fifo1 is TX and fifo2 is RX

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: header describing i2s registers
Iliyan Malchev [Thu, 29 Jul 2010 21:02:17 +0000 (14:02 -0700)]
[ARM] tegra: header describing i2s registers

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: add platform devices for i2s1 and i2s2
Iliyan Malchev [Thu, 29 Jul 2010 20:57:45 +0000 (13:57 -0700)]
[ARM] tegra: add platform devices for i2s1 and i2s2

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: Don't share tegra_usb1_resources
Benoit Goby [Thu, 29 Jul 2010 03:16:43 +0000 (20:16 -0700)]
[ARM] tegra: Don't share tegra_usb1_resources

Don't share tegra_usb1_resources as both tegra-udc and ehci1 are loaded
in otg mode.

Change-Id: Id5c9a076572e18662b5d3e7835f638b1cc5a1e07
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: add EHCI and UDC controllers to devices.c
Gary King [Wed, 28 Jul 2010 23:22:31 +0000 (16:22 -0700)]
[ARM] tegra: add EHCI and UDC controllers to devices.c

update harmony and ventana to use the common UDC definition, rather
than using the current duplicated definitions

Change-Id: I2e3aca674ab35305a0c516bd22e044382280d05e
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: Add USB PHY configuration
Benoit Goby [Wed, 28 Jul 2010 07:10:13 +0000 (00:10 -0700)]
[ARM] tegra: Add USB PHY configuration

Configure board dependant phy settings from the board file.

Change-Id: I29f6ffe0b84a2a6eb55bade1379002f561d92d17
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: Setup USB PHY as recommended by NVIDIA
Benoit Goby [Sat, 24 Jul 2010 00:02:26 +0000 (17:02 -0700)]
[ARM] tegra: Setup USB PHY as recommended by NVIDIA

This fixes enumeration issues with some devices

Change-Id: I6283a6fb49a9e4505ad388cacdd88ecf1bdf1b9d
Signed-off-by: Benoit Goby <benoit@android.com>
14 years agousb: gadget: Fix tegra udc in OTG mode
Benoit Goby [Thu, 22 Jul 2010 01:05:06 +0000 (18:05 -0700)]
usb: gadget: Fix tegra udc in OTG mode

Check the transceiver state before checking udc->stopped
Enable/disable the PHY and the clock on cable events

Change-Id: Id5a8a1b94f83da8060786f31181014dd1d546fc7
Signed-off-by: Benoit Goby <benoit@android.com>
14 years agousb: host: Add EHCI driver for NVIDIA Tegra SoCs
Benoit Goby [Fri, 9 Jul 2010 00:00:48 +0000 (17:00 -0700)]
usb: host: Add EHCI driver for NVIDIA Tegra SoCs

Change-Id: I53c560f2c31e043f139b840f58786429ded6ec62
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: usb_phy: Add tegra_usb_phy_power_off
Benoit Goby [Thu, 22 Jul 2010 00:44:44 +0000 (17:44 -0700)]
[ARM] tegra: usb_phy: Add tegra_usb_phy_power_off

Change-Id: If4d66b1a0f1810773b9dc9bcec0e252df947e609
Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: harmony/ventana: register additional devices
Gary King [Wed, 14 Jul 2010 01:30:59 +0000 (18:30 -0700)]
[ARM] tegra: harmony/ventana: register additional devices

register GART, pda_power, and OTG devices

Change-Id: I6dec5f765494cdc96f3395cf32fd962c87bf70b7
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: ventana: add initial support for Ventana
Gary King [Wed, 14 Jul 2010 00:16:23 +0000 (17:16 -0700)]
[ARM] tegra: ventana: add initial support for Ventana

Change-Id: I1993fda7628c623d53fd2c97649ec3533ad790e2
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: harmony: add sdhci0 controller (wlan)
Gary King [Thu, 15 Jul 2010 17:06:40 +0000 (10:06 -0700)]
[ARM] tegra: harmony: add sdhci0 controller (wlan)

Change-Id: I6000b1d34e35d9291cae80e4508518d0059397f8
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: suspend: Add iovmm suspend
Colin Cross [Mon, 4 Oct 2010 06:37:45 +0000 (23:37 -0700)]
[ARM] tegra: suspend: Add iovmm suspend

Signed-off-by: Colin Cross <ccross@android.com>
14 years ago[ARM] tegra: add I/O virtual memory manager interface (iovmm)
Gary King [Sat, 26 Jun 2010 01:39:58 +0000 (18:39 -0700)]
[ARM] tegra: add I/O virtual memory manager interface (iovmm)

The Tegra IOVMM is an interface to allow device drivers and subsystems in
the kernel to manage the virtual memory spaces visible to I/O devices.

The interface has been designed to be scalable to allow for I/O virtual
memory hardware which exists in one or more limited apertures of the address
space (e.g., a small aperture in physical address space which can perform
MMU-like remapping) up to complete virtual addressing with multiple
address spaces and memory protection.

The interface has been designed to be similar to the Linux virtual memory
system; however, operations which would be difficult to implement or
nonsensical for DMA devices (e.g., copy-on-write) are not present, and
APIs have been added to allow for management of multiple simultaneous
active address spaces.

The API is broken into four principal objects: areas, clients, domains and
devices.

Areas
=====

An area is a contiguous region of the virtual address space which can be
filled with virtual-to-physical translations (and, optionally, protection
attributes). The virtual address of the area can be queried and used for
DMA operations by the client which created it.

As with the Linux vm_area structures, it is the responsibility of whichever
code creates an area to ensure that it is populated with appropriate
translations.

Domains
=======

A domain in the IOVMM system is similar to a process in a standard CPU
virtual memory system; it represents the entire range of virtual addresses
which may be allocated and used for translation. Depending on hardware
capabilities, one or more domains may be resident and available for
translation. IOVMM areas are allocated from IOVMM domains.

Whenever a DMA operation is performed to or from an IOVMM area, its parent
domain must be made resident prior to commencing the operation.

Clients
=======

I/O VMM clients represent any entity which needs to be able to allocate
and map system memory into I/O virtual space. Clients are created by name
and may be created as part of a "share group," where all clients created
in the same share group will observe the same I/O virtual space (i.e., all
will use the same IOVMM domain). This is similar to threads inside a process
in the CPU virtual memory manager.

The callers of the I/O VMM system are responsible for deciding on the
granularity of client creation and share group definition; depending on the
specific usage model expected by the caller, it may be appropriate to create
an IOVMM client per task (if the caller represents an ioctl'able interface
to user land), an IOVMM client per driver instance, a common IOVMM client
for an entire bus, or a global IOVMM client for an OS subsystem (e.g., the DMA
mapping interface).

Each client is responsible for ensuring that its IOVMM client's translation is
resident on the system prior to performing DMA operations using the IOVMM
addresses. This is accomplished by preceding all DMA operations for the client
with a call to tegra_iovmm_client_lock (or tegra_iovmm_client_trylock),
and following all operations (once complete) with a call to
tegra_iovmm_client_unlock. In this regard, clients are cooperatively context-
switched, and are expected to behave appropriately.

Devices
=======

I/O VMM devices are the physical hardware which is responsible for performing
the I/O virtual-to-physical translation.

Devices are responsible for domain management: the mapping and unmapping
operations needed to make translations resident in the domain (including
any TLB shootdown or cache invalidation needed to ensure coherency), locking
and unlocking domains as they are made resident by clients into the devices'
address space(s), and allocating and deallocating the domain objects.

Devices are responsible for the allocation and deallocation of domains to
allow coalescing of multiple client share groups into a single domain. For
example, if the device's hardware only allows a single address space to
be translated system-wide, performing full flushes and invalidates of the
translation at every client switch may be prohibitively expensive. In these
circumstances, a legal implementation of the IOVMM interface includes
returning the same domain for all clients on the system (regardless of
the originally-specified share group).

In this respect, a client can be assured that it will share an address space
with all of the other clients in its share group; however, it may also share
this address space with other clients, too.

Multiple devices may be present in a system; a device should return a NULL
domain if it is incapable of servicing the client when it is asked to
allocate a domain.

----------------------------------------------------------------------------

IOVMM Client API
================

tegra_iovmm_alloc_client - Called to create a new IOVMM client object; the
 implementation may create a new domain or return an existing one depending on
 both the device and the share group.

tegra_iovmm_free_client - Frees a client.

tegra_iovmm_client_lock - Makes a client's translations resident in the IOVMM
 device for subsequent DMA operations. May block if the device is incapable
 of context-switching the client when it is called. Returns -EINTR if the
 waiting thread is interrupted before the client is locked.

tegra_iovmm_client_trylock - Non-blocking version of tegra_iovmm_client_lock

tegra_iovmm_client_unlock - Called by clients after DMA operations on IOVMM-
 translated addresses is complete; allows IOVMM system to context-switch the
 current client out of the device if needed.

tegra_iovmm_create_vm - Called to allocate an IOVMM area. If
 lazy / demand-loading of pages is desired, clients should supply a pointer
 to a tegra_iovmm_area_ops structure providing callback functions to load, pin
 and unpin the physical pages which will be mapped into this IOVMM region.

tegra_iovmm_get_vm_size - Called to query the total size of an IOVMM client

tegra_iovmm_free_vm - Called to free a IOVMM area, releasing any pinned
 physical pages mapped by it and to decommit any resources (memory for
 PTEs / PDEs) required by the VM area.

tegra_iovmm_vm_insert_pfn - Called to insert an exact pfn (system memory
 physical page) into the area at a specific virtual address. Illegal to call
 if the IOVMM area was originally created with lazy / demand-loading.

tegra_iovmm_zap_vm - Called to mark all mappings in the IOVMM area as
 invalid / no-access, but continues to consume the I/O virtual address space.
 For lazy / demand-loaded IOVMM areas, a zapped region will not be reloaded
 until it has been unzapped; DMA operations using the affected translations
 may fault (if supported by the device).

tegra_iovmm_unzap_vm - Called to re-enable lazy / demand-loading of pages
 for a previously-zapped IOVMM area.

tegra_iovmm_find_area_get - Called to find the IOVMM area object
 corresponding to the specified I/O virtual address, or NULL if the address
 is not allocated in the client's address space. Increases the reference count
 on the IOVMM area object

tegra_iovmm_area_get - Called to increase the reference count on the IOVMM
 area object

tegra_iovmm_area_put - Called to decrease the reference count on the IOVMM
 area object

IOVMM Device API
================

tegra_iovmm_register - Called to register a new IOVMM device with the IOVMM
 manager

tegra_iovmm_unregister - Called to remove an IOVMM device from the IOVMM
 manager (unspecified behavior if called while a translation is active and / or
 in-use)

tegra_iovmm_domain_init - Called to initialize all of the IOVMM manager's
 data structures (block trees, etc.) after allocating a new domain

IOVMM Device HAL
================

map - Called to inform the device about a new lazy-mapped IOVMM area. Devices
 may load the entire VM area when this is called, or at any time prior to
 the completion of the first read or write operation using the translation.

unmap - Called to zap or to decommit translations

map_pfn - Called to insert a specific virtual-to-physical translation in the
 IOVMM area

lock_domain - Called to make a domain resident; should return 0 if the
 domain was successfully context-switched, non-zero if the operation can
 not be completed (e.g., all available simultaneous hardware translations are
 locked). If the device can guarantee that every domain it allocates is
 always usable, this function may be NULL.

unlock_domain - Releases a domain from residency, allows the hardware
 translation to be used by other domains.

alloc_domain - Called to allocate a new domain; allowed to return an
 existing domain

free_domain - Called to free a domain.

Change-Id: Ic65788777b7aba50ee323fe16fd553ce66c4b87c
Signed-off-by: Gary King <gking@nvidia.com>
14 years agousb: gadget: Use USB PHY api
Benoit Goby [Wed, 30 Jun 2010 01:24:09 +0000 (18:24 -0700)]
usb: gadget: Use USB PHY api

Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: add API for enabling the USB PHYs
Benoit Goby [Wed, 30 Jun 2010 01:09:26 +0000 (18:09 -0700)]
[ARM] tegra: add API for enabling the USB PHYs

Moved usb phy initialization code
Added support for usb3 utmi phy
Updated the registers as recommended by Nvidia to be MUCH closer to passing the integrity tests
TODO: Add support for usb2 ulpi phy

Signed-off-by: Benoit Goby <benoit@android.com>
14 years ago[ARM] tegra: harmony: remove hard-coded partition table
Gary King [Mon, 28 Jun 2010 22:01:43 +0000 (15:01 -0700)]
[ARM] tegra: harmony: remove hard-coded partition table

use the partition information provided on the kernel command line rather
than a fixed table that is subject to change.

Change-Id: I650f634bf49b8658debb75535e94f2a497ef3432
Signed-off-by: Gary King <gking@nvidia.com>
14 years agomtd/tegra_nand: don't ignore return value for add_mtd_partitions
Gary King [Mon, 28 Jun 2010 22:00:10 +0000 (15:00 -0700)]
mtd/tegra_nand: don't ignore return value for add_mtd_partitions

when the mtd partition command line format is used, ignoring the
return value left err set to the number of partitions, which was
later interpreted as an error return code for tegra_nand_probe,
which caused the MTD master to be unregistered (ultimately causing
NULL pointer derefs when mounting the root partition).

Change-Id: Icebfb295810554617c56deeafc91bc22cc43bb35
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: harmony: Make board init calls explicit
Colin Cross [Sat, 15 May 2010 02:30:34 +0000 (19:30 -0700)]
[ARM] tegra: harmony: Make board init calls explicit

Signed-off-by: Colin Cross <ccross@android.com>
Change-Id: Ic14e8db00f2272de2f4ee4013bb3ab5c1951e7fe

14 years ago[ARM] tegra: harmony: Add framebuffer platform device
Colin Cross [Sun, 27 Jun 2010 05:37:57 +0000 (22:37 -0700)]
[ARM] tegra: harmony: Add framebuffer platform device

Signed-off-by: Colin Cross <ccross@android.com>
14 years ago[ARM] tegra: harmony: Remove android platform devices
Colin Cross [Sun, 27 Jun 2010 05:36:41 +0000 (22:36 -0700)]
[ARM] tegra: harmony: Remove android platform devices

Signed-off-by: Colin Cross <ccross@android.com>
14 years agousb: gadget: Add support for OTG
Benoit Goby [Sat, 5 Jun 2010 02:36:06 +0000 (19:36 -0700)]
usb: gadget: Add support for OTG

Based on work by Nvidia

Signed-off-by: Benoit Goby <benoit@android.com>
14 years agow1: master: tegra_w1: Adds generic mach-tegra w1 support.
Andrei Warkentin [Mon, 28 Jun 2010 22:59:01 +0000 (17:59 -0500)]
w1: master: tegra_w1: Adds generic mach-tegra w1 support.

This adds w1 as a device for mach-tegra boards, fixes wrong
OWR I/O base, and changes OWR clock name.

Change-Id: Idffbdbd05f383ce8e423ee301e197e230db4f2f9
Signed-off-by: Andrei Warkentin <andreiw@motorola.com>
14 years agow1: master: tegra_w1: Add initial Tegra W1 master driver
Andrei Warkentin [Mon, 28 Jun 2010 22:56:06 +0000 (17:56 -0500)]
w1: master: tegra_w1: Add initial Tegra W1 master driver

Initial implementation of W1 master driver for Tegra SoCs.
Tested with DS2781 slave driver.

Change-Id: I6cda1ea152d25a789ae6cdca96b710da72884033
Signed-off-by: Andrei Warkentin <andreiw@motorola.com>
14 years ago[ARM] tegra: add a header defining tegra 2 wake pads
Gary King [Wed, 4 Aug 2010 00:46:59 +0000 (17:46 -0700)]
[ARM] tegra: add a header defining tegra 2 wake pads

wakeup from LP0 is latched at the pads rather than in the interrupt
controller; since the pad numbers don't correspond to any other
sane numbering or naming system, provide a new list of defines
to make board code easier to read and maintain

Change-Id: Icf85a5826acc567452c0a2475c5a06ed042f66b3
Signed-off-by: Gary King <gking@nvidia.com>
14 years ago[ARM] tegra: enable ARM errata 742230
Colin Cross [Mon, 4 Oct 2010 03:20:10 +0000 (20:20 -0700)]
[ARM] tegra: enable ARM errata 742230

Change-Id: I25b9e7c266147163ca1dcf7619db0580ff9888d8
Signed-off-by: Colin Cross <ccross@android.com>
14 years agovideo: tegrafb: Enable host1x clock
Colin Cross [Tue, 8 Jun 2010 03:50:12 +0000 (20:50 -0700)]
video: tegrafb: Enable host1x clock

Change-Id: If1e50ed661493cb06de5e5bbc5ee2b0966297b4f
Signed-off-by: Colin Cross <ccross@android.com>
14 years ago[ARM] tegra: add FIQ support
Iliyan Malchev [Wed, 26 May 2010 21:21:21 +0000 (14:21 -0700)]
[ARM] tegra: add FIQ support

Signed-off-by: Iliyan Malchev <malchev@google.com>
14 years ago[ARM] tegra: Move common platform_device_register into boards
Colin Cross [Sat, 15 May 2010 02:29:18 +0000 (19:29 -0700)]
[ARM] tegra: Move common platform_device_register into boards

Signed-off-by: Colin Cross <ccross@android.com>
14 years ago[ARM] tegra: add spi platform devices
Erik Gilling [Thu, 22 Apr 2010 22:59:50 +0000 (15:59 -0700)]
[ARM] tegra: add spi platform devices

Signed-off-by: Erik Gilling <konkers@android.com>
14 years ago[ARM] tegra: delay: Use immediates instead of literals
Colin Cross [Mon, 4 Oct 2010 05:06:14 +0000 (22:06 -0700)]
[ARM] tegra: delay: Use immediates instead of literals

Change-Id: I4629398863062e3f80303c84d0fd3b7a4ed6e708
Signed-off-by: Colin Cross <ccross@android.com>
14 years ago[ARM] tegra: Add arch-specific udelay using TMRUS
Colin Cross [Sun, 25 Apr 2010 07:29:45 +0000 (00:29 -0700)]
[ARM] tegra: Add arch-specific udelay using TMRUS

Change-Id: If075117642a725ee2ee24a622068274e588a5bc1
Signed-off-by: Colin Cross <ccross@android.com>