]> Gitweb @ Texas Instruments - Open Source Git Repositories - git.TI.com/gitweb - sitara-epos/sitara-epos-kernel.git/log
sitara-epos/sitara-epos-kernel.git
12 years agoarm:omap:gpio - Handle Clocks properly in Suspend/Resume
Hebbar, Gururaja [Thu, 12 Jan 2012 16:20:58 +0000 (21:50 +0530)]
arm:omap:gpio - Handle Clocks properly in Suspend/Resume

Clocks were not enabled/disabled during suspend-resume. This patch
corrects the same.

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm:omap:gpio - add proper #if check for am335x
Hebbar, Gururaja [Thu, 12 Jan 2012 16:18:25 +0000 (21:48 +0530)]
arm:omap:gpio - add proper #if check for am335x

few #if check for am335x were missing. This patch adds the same.

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm:omap:am33xx : correct gpio hwmod data
Hebbar, Gururaja [Tue, 10 Jan 2012 13:44:05 +0000 (19:14 +0530)]
arm:omap:am33xx : correct gpio hwmod data

GPIO IP in am335x doesnt support Wakeup. Also, fix incorrect prcm
clock entries for GPIO 2 & 3.

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm:omap:hsmmc - remove invalid bus power setting
Hebbar, Gururaja [Tue, 10 Jan 2012 12:54:46 +0000 (18:24 +0530)]
arm:omap:hsmmc - remove invalid bus power setting

SD bus Power was setup at a wrong sequence inside irq code. This Commit
removes the same.

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm:omap:edma - replace edm_info with more relevant edma_cc
Hebbar, Gururaja [Mon, 9 Jan 2012 09:23:52 +0000 (14:53 +0530)]
arm:omap:edma - replace edm_info with more relevant edma_cc

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoARM: edma: use kzalloc()
Thomas Meyer [Sat, 6 Aug 2011 11:26:13 +0000 (13:26 +0200)]
ARM: edma: use kzalloc()

 Use kzalloc rather than kmalloc followed by memset with 0

 This considers some simple cases that are common and easy to validate
 Note in particular that there are no ...s in the rule, so all of the
 matched code has to be contiguous

 The semantic patch that makes this output is available
 in scripts/coccinelle/api/alloc/kzalloc-simple.cocci.

 More information about semantic patching is available at
 http://coccinelle.lip6.fr/

Signed-off-by: Thomas Meyer <thomas@m3y3r.de>
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm: edma: check irq2ctlr() result
Kulikov Vasiliy [Sat, 17 Jul 2010 15:19:07 +0000 (19:19 +0400)]
arm: edma: check irq2ctlr() result

If irq2ctlr() fails return IRQ_NONE.
Also as it can fail make 'ctlr' signed.

The semantic patch that finds this problem (many false-positive results):
(http://coccinelle.lip6.fr/)

// <smpl>
@ r1 @
identifier f;
@@
int f(...) { ... }

@@
identifier r1.f;
type T;
unsigned T x;
@@

*x = f(...)
 ...
*x > 0

Signed-off-by: Kulikov Vasiliy <segooon@gmail.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm: edma: use BIT() wherever possible
Sekhar Nori [Mon, 10 May 2010 07:11:18 +0000 (12:41 +0530)]
arm: edma: use BIT() wherever possible

This patch replaces occurences of (1 << x) with
BIT(x) as it makes for much better reading.

Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm: edma: fix coding style issue related to usage of braces
Sekhar Nori [Tue, 4 May 2010 08:41:36 +0000 (14:11 +0530)]
arm: edma: fix coding style issue related to usage of braces

In the edma driver, there are couple of instances where braces
are used for a single statement 'if' construct.

There are other instances where 'else' part of the if-else construct
does not use braces even if the 'if' part is a multi-line statement.

This patch fixes both.

Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm: edma: use a more intuitive name for edma_info
Sekhar Nori [Tue, 4 May 2010 08:41:35 +0000 (14:11 +0530)]
arm: edma: use a more intuitive name for edma_info

'edma_info' structure inside the edma driver represents
a single instance of edma channel controller. Call it
'edma_cc' instead. This also avoids readers confusing
it with an instance of edma_soc_info structre which
carries the platform data for a single channel controller
instance.

Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Kevin Hilman <khilman@deeprootsystems.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm: am33xx: Add CPSW MII mac select support
Chandan Nath [Wed, 11 Jan 2012 12:05:47 +0000 (17:35 +0530)]
arm: am33xx: Add CPSW MII mac select support

This patch fixes missing configuration for MII interface
configuration in device control register. MII mode selection
should be done depending up on the board/phy type.
This was missing in linux, however without this also it was
working as it was configured in uboot. This is required in
linux also in order to remove any dependency on uboot.

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
12 years agoarm:omap:nand - Add support for suspend/resume
Philip, Avinash [Tue, 10 Jan 2012 11:42:50 +0000 (17:12 +0530)]
arm:omap:nand - Add support for suspend/resume

suspend/resume entry  points added in NAND driver by disabling/enabling
the clocks for NAND controller and ELM module.
controller clock  member is added to omap_nand_info and
omap_nand_platform_data structure.
On suspend waiting for ongoing mtd operation to finish on
mtd->suspend().

Signed-off-by: Philip, Avinash <avinashphilip@ti.com>
12 years agousb: musb: ti81xx: auto vbus control across host and device mode
Ajay Kumar Gupta [Thu, 22 Dec 2011 10:32:59 +0000 (16:02 +0530)]
usb: musb: ti81xx: auto vbus control across host and device mode

Automatically control vbus using timer. Set the session bit periodically in a timer
context when nothing is connected to usb port. If mini-A is connected then Vbus will
get switched on and if mini-B is connected then Vbus will nto be driven.

Kill the timer when a device is attached in host mode or port is connected to external
host and revive the timer as soon as disconnect happens.

Signed-off-by: Ajay Kumar Gupta <ajay.gupta@ti.com>
12 years agousb: musb: ti81xx: allow board file to set usb mode for each port
Ajay Kumar Gupta [Thu, 22 Dec 2011 06:41:47 +0000 (12:11 +0530)]
usb: musb: ti81xx: allow board file to set usb mode for each port

Used existing .mode field of board_data for setting usb mode for both
usb ports. usb0 mode is taken from board_mode[0:3] and usb1 mode is
from board_mode[4:7] bits. As mode can take values only from 0 to 3
so this is safe.

This board mode is later used and corretly set in each port's musb
platform_data's mode.

Signed-off-by: Ajay Kumar Gupta <ajay.gupta@ti.com>
12 years agousb: musb: bring back if_h/p/o_enabled()
Ajay Kumar Gupta [Tue, 13 Dec 2011 05:08:28 +0000 (10:38 +0530)]
usb: musb: bring back if_h/p/o_enabled()

musb driver now has only OTG mode to build for and all #ifdefferies for HOST
,PERIPHERAL and OTG have been cleaned up.

Bringing back the if_host/peripheral/otg_enabled() at all #ifdef places to make
the driver work for host or peripheral only based on board_mode.

Signed-off-by: Ajay Kumar Gupta <ajay.gupta@ti.com>
12 years agousb: musb: cleanup the phy programming
Ajay Kumar Gupta [Tue, 3 Jan 2012 05:35:04 +0000 (11:05 +0530)]
usb: musb: cleanup the phy programming

Currently the phy_power is only done for one port and so add id field so that the
same function can be used for both ports.

Signed-off-by: Ajay Kumar Gupta <ajay.gupta@ti.com>
12 years agovideo: da8xx-fb: Follow reset sequence specified in specification
Manjunathappa, Prakash [Tue, 10 Jan 2012 11:41:29 +0000 (17:11 +0530)]
video: da8xx-fb: Follow reset sequence specified in specification

Follow below specified disable and reset sequence specified in
functional specification:
Steps for disabling the raster:
1)clear raster enable bit of raster control register.
2)Wait for interrupt done.
3)Set the software reset bit high.

Steps for enabling the raster:
1)Set the software reset back to low.
2)Set raster enable bit back to high.

Signed-off-by: Manjunathappa, Prakash <prakash.pm@ti.com>
12 years agoarm: am33xx: add GPIO backlight control hookup to LCD FB driver
Manjunathappa, Prakash [Tue, 10 Jan 2012 11:08:10 +0000 (16:38 +0530)]
arm: am33xx: add GPIO backlight control hookup to LCD FB driver

Patch adds hookup to toggle GPIO backlight from da8xx-fb LCD frame
buffer driver.

Signed-off-by: Manjunathappa, Prakash <prakash.pm@ti.com>
12 years agovideo: da8xx-fb: Interface clock cleanup
Manjunathappa, Prakash [Tue, 10 Jan 2012 11:03:56 +0000 (16:33 +0530)]
video: da8xx-fb: Interface clock cleanup

Patch does interface clock cleanup in driver suspend and remove
callbacks.

Signed-off-by: Manjunathappa, Prakash <prakash.pm@ti.com>
12 years agoinput: TSC: Add support for suspend/resume
Patil, Rachna [Tue, 10 Jan 2012 11:08:52 +0000 (16:38 +0530)]
input: TSC: Add support for suspend/resume

This patch adds support for suspend and resume
in touchscreen ti_tscadc

Signed-off-by: Patil, Rachna <rachna@ti.com>
12 years agoARM: OMAP: PRCM: Don't register PRCM chain handler for AM33XX
Vaibhav Bedia [Mon, 9 Jan 2012 04:46:20 +0000 (10:16 +0530)]
ARM: OMAP: PRCM: Don't register PRCM chain handler for AM33XX

AM33XX does not support PRCM interrupts for IO/WKUP events.
Since the current PRCM chain handler expects only these two
events, don't register it for AM33XX.

We might revisit this once the PRCM chain handler code is
enhanced to take care of other events like DPLL needing to be
recalibrated.

Note: This fixes the following crash in USB during kernel boot

[    2.131963] Unhandled fault: external abort on non-linefetch
(0x1008) at 0xc880a460
[    2.139960] Internal error: : 1008 [#1]
[    2.143954] Modules linked in:
[    2.147143] CPU: 0    Not tainted  (3.2.0-rc6-12008-g363a774)
[    2.154057] PC is at __musb_readb+0xc/0x14
[    2.158329] LR is at otg_timer+0x2c/0x13c
...

This was due to the PRCM_IRQENABLE register for OMAP3 being at
the same offset as USB_CLKCTRL on AM33XX.

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am33xx: fix for CPSW module build
Hebbar, Gururaja [Tue, 3 Jan 2012 09:32:01 +0000 (15:02 +0530)]
arm:omap:am33xx: fix for CPSW module build

module build adds _module for related config_xxx and hence checking
only for config_xxx is not sufficent. this patch adds check for
CONFIG_TLK110_WORKAROUND_MODULE also.

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoARM: OMAP: AM33XX: Basic suspend resume support
Vaibhav Bedia [Wed, 4 Jan 2012 09:19:03 +0000 (14:49 +0530)]
ARM: OMAP: AM33XX: Basic suspend resume support

Add support for basic suspend-resume.
Currently DeepSlee1 mode of the processor is used
as the suspend state.

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoARM: OMAP: Don't do anything in sram_init for now
Vaibhav Bedia [Tue, 3 Jan 2012 12:51:46 +0000 (18:21 +0530)]
ARM: OMAP: Don't do anything in sram_init for now

Needs to be merged with 45cd51a8e934bd996963a6de582ae43947c7ff28
(arm:omap: Add SRAM support for AM33xx)

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoARM: OMAP: Covert to device_initcall
Vaibhav Bedia [Tue, 3 Jan 2012 12:49:01 +0000 (18:19 +0530)]
ARM: OMAP: Covert to device_initcall

The power management code for AM33XX is a late_initcall.
The PM features depend on mailbox for IPC being available.
In preparation for this, convert the mailbox init to a
device_initcall.

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agocpsw: Merge cpsw modules to build as single driver
Hebbar, Gururaja [Mon, 2 Jan 2012 11:23:08 +0000 (16:53 +0530)]
cpsw: Merge cpsw modules to build as single driver

Merge 2 cpsw drivers (cpsw_ale & cpsw) as one so as to avoid module
build issues.

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agocpsw: fix for change MAC address using ifconfig
Hebbar, Gururaja [Mon, 2 Jan 2012 10:55:32 +0000 (16:25 +0530)]
cpsw: fix for change MAC address using ifconfig

When MAC address is changed using ifconfig, driver MAC address is
not changes as pointer holding the new MAC address is not utilised.
Changed the driver code to utilize the new MAC address and change
the MAC address in driver if it is a valid MAC address.

Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoarm:omap:am33xx: Don't register McBSP
Vaibhav Bedia [Mon, 2 Jan 2012 11:47:01 +0000 (17:17 +0530)]
arm:omap:am33xx: Don't register McBSP

...since AM33XX doesn't have any instance of McBSP

Note: This needs to be merged with the commit
08037c12036110c79b8fa2adf486aefaee
(arm:omap:am33xx: Don't register non-existent...)

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am33xx: Enable cpuidle in the defconfig
Vaibhav Bedia [Mon, 2 Jan 2012 11:05:16 +0000 (16:35 +0530)]
arm:omap:am33xx: Enable cpuidle in the defconfig

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am33xx: Basic cpuidle support
Vaibhav Bedia [Wed, 30 Nov 2011 16:01:47 +0000 (21:31 +0530)]
arm:omap:am33xx: Basic cpuidle support

Add basic cpuidle support for AM33XX family of SoC.
Right now only two idle states (WFI and WFI+SR) are
supported. The latency/residency numbers chosen will
be fine-tuned based on power measurements on actual
hardware.

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am335x: Add cpuidle related board data
Vaibhav Bedia [Wed, 30 Nov 2011 16:05:40 +0000 (21:35 +0530)]
arm:omap:am335x: Add cpuidle related board data

Add board specific cpuidle hookup for AM335x EVM

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am33xx: Drop support for SERIAL_8250
Vaibhav Bedia [Mon, 2 Jan 2012 06:58:52 +0000 (12:28 +0530)]
arm:omap:am33xx: Drop support for SERIAL_8250

After commit 8a60585159067f110075ef8ffda13abd94826daf
(ARM: OMAP2+: UART: cleanup 8250 console driver support)
the omap-serial driver no longer depends on 8250 support.

The omap-serial driver has been the default of AM33XX
for some time now. We can now drop the support for
SERIAL_8250 from the defconfig.

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap: Add SRAM support for AM33xx
Vaibhav Bedia [Wed, 28 Dec 2011 13:38:10 +0000 (19:08 +0530)]
arm:omap: Add SRAM support for AM33xx

Just creates a gen_pool of 64KB which points to
OCMC RAM in AM33xx. A subsequent patch will add
code for copying the suspend code to SRAM

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am33xx: Enable MBOX in defconfig
Vaibhav Bedia [Tue, 27 Dec 2011 09:25:31 +0000 (14:55 +0530)]
arm:omap:am33xx: Enable MBOX in defconfig

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap: Fix a bug in SRAM init code
Vaibhav Bedia [Tue, 20 Dec 2011 11:51:46 +0000 (17:21 +0530)]
arm:omap: Fix a bug in SRAM init code

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am33xx: HWMOD updates
Vaibhav Bedia [Tue, 20 Dec 2011 11:47:04 +0000 (17:17 +0530)]
arm:omap:am33xx: HWMOD updates

1. Set the no idle flags for the OCMC to avoid issues in
sram init code which does not explicitly enable the clocks
2. Add reset line info for wkup_m3 and keep the reset asserted

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoomap: mailbox: Add an API for flushing the FIFO
Vaibhav Bedia [Tue, 20 Dec 2011 11:37:37 +0000 (17:07 +0530)]
omap: mailbox: Add an API for flushing the FIFO

On some platforms like AM33XX, the mailbox module is a
one-way communication between certain processors.

In these scenarios, the initiator (eg. MPU) is completely
responsible for FIFO maintenance and interrupt generation.
The initiator needs a mechanism to flush the FIFO to avoid
overflows or when going to suspend. Without the FIFO being
flushed, the IP block doesn't get disabled thereby blocking
system suspend.

This patch adds an API which the suspend code can invoke
to flush the FIFO and successfully enter suspend state.

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoomap: mailbox: Add support for AM33XX SoC
Vaibhav Bedia [Tue, 20 Dec 2011 11:35:09 +0000 (17:05 +0530)]
omap: mailbox: Add support for AM33XX SoC

While at it, remove some unnecessary #if statements

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoomap: mailbox: Fix the address calculation macros
Vaibhav Bedia [Tue, 20 Dec 2011 11:21:13 +0000 (16:51 +0530)]
omap: mailbox: Fix the address calculation macros

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am33xx: Register mailbox hwmod
Vaibhav Bedia [Tue, 20 Dec 2011 11:32:38 +0000 (17:02 +0530)]
arm:omap:am33xx: Register mailbox hwmod

Add the base address for mailbox and register
the hwmod for the same

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agoarm:omap:am33xx: Don't register non-existent or unused devices
Vaibhav Bedia [Tue, 20 Dec 2011 11:18:39 +0000 (16:48 +0530)]
arm:omap:am33xx: Don't register non-existent or unused devices

Remove some of the platform devices which either do
not exist or are not being used right now.

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
12 years agovideo: da8xx-fb: reset LCDC only if functional clock changes with DVFS
Manjunathappa, Prakash [Thu, 22 Dec 2011 05:28:11 +0000 (10:58 +0530)]
video: da8xx-fb: reset LCDC only if functional clock changes with DVFS

LCDC functional clock may or may not be derived from CPU/MPU DPLL,
For example,
AM335x => Separate independent DPLL for LCDC
Davinci => Same DPLL as MPU

So, on platforms where LCDC functional clock is not derived from CPU/MPU
PLL it is not required to reset LCDC module as its functional clock does
not change with DVFS.

This patch adds check to do reset only if functional clock changes
between pre and post notifier callbacks with DVFS.

Signed-off-by: Manjunathappa, Prakash <prakash.pm@ti.com>
12 years agoarm:omap:am335x - Add support for UART 2
Hebbar, Gururaja [Fri, 16 Dec 2011 10:21:58 +0000 (15:51 +0530)]
arm:omap:am335x - Add support for UART 2

Support for UART 2 is added. UART 2 is tested for working in Profile 3
by passing kernel boot args from u-boot as setenv bootargs
'console=ttyO2,115200n8......'

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
12 years agoMerge to hwmod data patch
Vaibhav Hiremath [Thu, 15 Dec 2011 16:01:04 +0000 (21:31 +0530)]
Merge to hwmod data patch

12 years agoarm:omap:omap: export omap_ctrl_writel()
Hebbar, Gururaja [Tue, 13 Dec 2011 06:04:24 +0000 (11:34 +0530)]
arm:omap:omap: export omap_ctrl_writel()

export omap_ctrl_writel() so that it can be used by others which are
built as module.

When we are here, add mising export include

Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agovideo: da8xx-fb: Add debug message on sync and underflow error
Manjunathappa, Prakash [Tue, 6 Dec 2011 13:33:49 +0000 (19:03 +0530)]
video: da8xx-fb: Add debug message on sync and underflow error

Patch adds useful debug message on LCDC syn and underflow error.

Signed-off-by: Manjunathappa, Prakash <prakash.pm@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agousb: musb: Fix for handling overcurrent condition in host mode
Ravi B [Mon, 28 Nov 2011 13:28:11 +0000 (18:58 +0530)]
usb: musb: Fix for handling overcurrent condition in host mode

The overcurrent situation can occur in 2 scenarios
a) when usb device connected directly to musb root-port and device draw the
more current, further the VBUS_ERROR is generated and root-hub port change
status need to be notified to hub driver to handle over-current situation.
b) when usb device connected thrugh HUB, the hub status change will be
notified by hub through interrupt transfer and handled approperiately by HUB
driver.

This patch fixes (a), where root hub status change was not notified to hub driver.

Signed-off-by: Ravi B <ravibabu@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am335x: Enable Oprofile support on AM335x
Jeff Lance [Tue, 29 Nov 2011 16:30:53 +0000 (10:30 -0600)]
arm:omap:am335x: Enable Oprofile support on AM335x

This enables support for oProfile on AM335x.

Enabling Oprofile does not impact kernel performance since
Oprofile is dormant unless a profiling session is initiated.

Signed-off-by: Jeff Lance <j-lance1@ti.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoOMAP: clock: am33xx specific API handling
Vaibhav Bedia [Tue, 29 Nov 2011 11:07:40 +0000 (16:37 +0530)]
OMAP: clock: am33xx specific API handling

Jitter correction for AM33XX is not present.
Handle revelant clock API's properly.

Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agomfd: TPS65910: Make usable even if interrupt unused
Afzal Mohammed [Tue, 8 Nov 2011 09:57:00 +0000 (15:27 +0530)]
mfd: TPS65910: Make usable even if interrupt unused

TPS65910 can be used even if interrupt is unused.
Hence let probe succeed in case interrupt can't be
configured and let Kernel only to complain about it

Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoregulator: TPS65910: Create an array for reg init data
Kyle Manna [Wed, 2 Nov 2011 12:01:30 +0000 (17:31 +0530)]
regulator: TPS65910: Create an array for reg init data

Create an array of fixed size for the platform to pass regulator
initalization data through.

Passing an array of pointers to init data also allows more flexible
definition of init data as well as prevents reading past the end of the
array should the platform define an incorrectly sized array.

Signed-off-by: Kyle Manna <kyle.manna@fuel7.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoregulator: TPS65910: Move regulator defs to header
Kyle Manna [Wed, 2 Nov 2011 12:00:19 +0000 (17:30 +0530)]
regulator: TPS65910: Move regulator defs to header

Move the regulator defintions to the header so that platform board file
can use them to configure specific regulators.

Signed-off-by: Kyle Manna <kyle.manna@fuel7.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agomfd: TPS65910: Move linux/gpio.h include to header
Kyle Manna [Wed, 2 Nov 2011 11:58:47 +0000 (17:28 +0530)]
mfd: TPS65910: Move linux/gpio.h include to header

The tps65910.h file depends on linux/gpio.h.  Move the include from the
source file to the tps65910.h header file.

Signed-off-by: Kyle Manna <kyle.manna@fuel7.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agomfd: TPS65910: Fix typo that clobbers genirq
Kyle Manna [Wed, 2 Nov 2011 11:57:39 +0000 (17:27 +0530)]
mfd: TPS65910: Fix typo that clobbers genirq

Fix a typo that clobbers other interrupts in an unobvious way.

Signed-off-by: Kyle Manna <kyle.manna@fuel7.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agomfd: TPS65910: Add I2C slave address macros
Kyle Manna [Wed, 2 Nov 2011 11:56:13 +0000 (17:26 +0530)]
mfd: TPS65910: Add I2C slave address macros

Add I2C slave addresses to the header file so that platform definitions
can use them.

Signed-off-by: Kyle Manna <kyle.manna@fuel7.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agomfd: TPS65910: Handle non-existent devices
Kyle Manna [Wed, 2 Nov 2011 11:47:29 +0000 (17:17 +0530)]
mfd: TPS65910: Handle non-existent devices

Attempt to read the first register of the device, if there is no
device return -ENODEV

Signed-off-by: Kyle Manna <kyle.manna@fuel7.com>
Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoam335x: tsl2550: Add ambient light sensor support
AnilKumar Ch [Sun, 27 Nov 2011 17:26:32 +0000 (22:56 +0530)]
am335x: tsl2550: Add ambient light sensor support

This patch enables the TAOS, ambient light sensor of am335x

Also adds the tsl2550, ambient light sensor configuration to
default am335x_evm_defconfig kernel configuration

Signed-off-by: AnilKumar Ch <anilkumar@ti.com>
12 years agoam335x: Add TI TMP275 temperature sensor support
AnilKumar Ch [Fri, 25 Nov 2011 16:26:24 +0000 (21:56 +0530)]
am335x: Add TI TMP275 temperature sensor support

This patch enables the TI's temperature sensor - TMP275 on
am335x chip

Also adds the tmp275 to default kernel configuration
am335x_evm_defconfig

Signed-off-by: AnilKumar Ch <anilkumar@ti.com>
12 years agoam335x: lis33ldlh: Add accelerometer support to AM335X
AnilKumar Ch [Fri, 25 Nov 2011 11:47:38 +0000 (17:17 +0530)]
am335x: lis33ldlh: Add accelerometer support to AM335X

This patch adds lis33ldlh support to AM335X. lis33ldlh is connected
via I2C to am335x and I2C address is 0x18.

Also adds lis33ldlh, accelerometer to default kernel configuration
am335x_evm_defconfig

Signed-off-by: AnilKumar Ch <anilkumar@ti.com>
12 years agolis3lv02d: Add STMicroelectronics lis33ldlh digital
AnilKumar Ch [Wed, 30 Nov 2011 11:45:37 +0000 (17:15 +0530)]
lis3lv02d: Add STMicroelectronics lis33ldlh digital

accelerometer

This patch adds support for lis33ldlh digital accelerometer to the
lis3lv02d driver family. Adds ID field for detecting the lis33ldlh
module, based on this ID field lis3lv02d driver will export the
lis33ldlh module functionality.

Also exports g_range parameter to user space for run-time value
change. User must give 2/4/8 value depends on requirement.

Signed-off-by: AnilKumar Ch <anilkumar@ti.com>
12 years agoarm:omap:am33xx: Create an alias for sgx/gfx clock
Vaibhav Hiremath [Tue, 29 Nov 2011 12:50:12 +0000 (18:20 +0530)]
arm:omap:am33xx: Create an alias for sgx/gfx clock

GFX/SGX driver has specific requriement for clock naming
convention, due to legacy platform support.
So create an alias for the clock "gfx_fclk" --> "sgx_ck".

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am33xx: Create an alias for pruss clock
Vaibhav Hiremath [Tue, 29 Nov 2011 12:48:16 +0000 (18:18 +0530)]
arm:omap:am33xx: Create an alias for pruss clock

PRUSS driver has specific requriement for clock naming
convention, due to legacy platform support.
So create an alias for the clock "icss_uart_gclk" --> "pruss".

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agovideo:da8xx: Replace dma_alloc_coherent tp get_free_pages
Vaibhav Hiremath [Mon, 28 Nov 2011 06:48:27 +0000 (12:18 +0530)]
video:da8xx: Replace dma_alloc_coherent tp get_free_pages

dma-alloc_coherent supports limited memory allocation, which leads
to failure some time, and always without CONSISTENT_DMA_SIZE defination.
For safe allocation, change dma_alloc_coherent to get_free_pages and
mark the pages as reserved.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM:omap:am33xx: CPSW removes dmtimer re enabling
Chandan Nath [Thu, 24 Nov 2011 09:03:28 +0000 (14:33 +0530)]
ARM:omap:am33xx: CPSW removes dmtimer re enabling

This patch is added to remove unwanted dmtimer re enabling
code after reseting timer status register. Also, some additional
cleanup is done in this patch.

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoam335x-evm: beta evm bringup changes for wlan and bt
Eyal Reizer [Tue, 22 Nov 2011 16:23:01 +0000 (18:23 +0200)]
am335x-evm: beta evm bringup changes for wlan and bt

* Fix a wrong gpio pin used for bluetooth enable on the beta evm.
  was using gpio0_17 while it should be gpio3_21

* Add a 70 msec delay after turning wlan on to allow for all voltages
  to settle down before starting sdio activity

Signed-off-by: Eyal Reizer <eyalr@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM:omap:am33xx: Implement CPSW interrupt pacing functionality
Chandan Nath [Wed, 23 Nov 2011 13:13:19 +0000 (18:43 +0530)]
ARM:omap:am33xx: Implement CPSW interrupt pacing functionality

CPSW module includes an interrupt pacing block that can
be programmed to throttle the rate at which interrupts are
generated. This patch implements interrupt pacing logic that can
be controlled through the ethtool interface(only rx_coalesce_usecs
param is honored)

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM:omap:am33xx: routing CPSW interrupts via Timer
Vaibhav Hiremath [Sat, 26 Nov 2011 20:48:24 +0000 (02:18 +0530)]
ARM:omap:am33xx: routing CPSW interrupts via Timer

This patch is added to route CPSW interrupts through ICSS module.
CPSW CPSW_RX and CPSW_TX interrupts are not directly used from CPSW
and instead they are routed through Timer module. This is done as an
software workaround for enabling interrupt pacing feature. As interrupt
Pacing works on C0_RX_PULSE interrupt connection to A8 INTC, which is
not connected to Cortex A8 interrupt controller directly due to hw issue,
it prevents interrupt pacing to work in AM33xx EVM. Therefore, Timer
capture module is used to pickup these interrupts and routed to A8 INTC.

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM:OMAP:AM33XX: Add PROFIBUS support
Amit Shah [Tue, 22 Nov 2011 13:47:41 +0000 (19:17 +0530)]
ARM:OMAP:AM33XX: Add PROFIBUS support

Added PROFIBUS pinmux settings

Signed-off-by: Amit Shah <amit.shah@ti.com>
Signed-off-by: Amit Shah <amit.shah@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM:OMAP:AM33XX:pruss:Add platform specific changes for AM33XX in UIO PRUSS driver.
Vaibhav Hiremath [Sat, 26 Nov 2011 20:45:12 +0000 (02:15 +0530)]
ARM:OMAP:AM33XX:pruss:Add platform specific changes for AM33XX in UIO PRUSS driver.

Added AM33XX dependency for UIO_PRUSS driver to build in Kconfig.
ICSS memory map base and length were added in am33xx.h
Other PRU specific resources like ICSS IRQs for AM33XX,
uio_pruss driver register were added to devices.c

Signed-off-by: Amit Shah <amit.shah@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM:DA850:pruss: Add platform specific changes for DA850 in UIO PRUSS driver
Amit Shah [Wed, 16 Nov 2011 15:32:30 +0000 (21:02 +0530)]
ARM:DA850:pruss: Add platform specific changes for DA850 in UIO PRUSS driver

PRU specific resources like RAM base address, PRUSS clock, IRQs for DA8XX, uio_pruss driver register were added.

Signed-off-by: Amit Shah <amit.shah@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM:OMAP:AM33XX:pruss: Disable SRAM support in UIO_PRUSS
Amit Shah [Wed, 16 Nov 2011 12:11:08 +0000 (17:41 +0530)]
ARM:OMAP:AM33XX:pruss: Disable SRAM support in UIO_PRUSS

Disable SRAM export in UIO till consolidated SRAM support available for AM33XX

Signed-off-by: Amit Shah <amit.shah@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM:OMAP:AM33XX: Add AM33XX_IRQ_GPMC0 as gpmc_irq for AM33XX
Amit Shah [Wed, 16 Nov 2011 08:17:20 +0000 (13:47 +0530)]
ARM:OMAP:AM33XX: Add AM33XX_IRQ_GPMC0 as gpmc_irq for AM33XX

uio_pruss driver was unable to load due to gpmc driver bug wherein the gpmc_irq for AM33XX was being set to INT_34XX_GPMC_IRQ.
Signed-off-by: Amit Shah <amit.shah@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoAM335XEVM: Volume Keys Support Added.
Goutam Kumar [Wed, 28 Sep 2011 08:47:51 +0000 (14:17 +0530)]
AM335XEVM: Volume Keys Support Added.

         1. GPIO Keys Support Enabled for volume up/down push bottons.
         2. Correction in Matrix Keypad for DB boot mode.

Signed-off-by: Goutam Kumar <goutam.kumar@ti.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoAM335XEVM: Matrix GPIO Keypad support added
Goutam Kumar [Wed, 28 Sep 2011 08:45:53 +0000 (14:15 +0530)]
AM335XEVM: Matrix GPIO Keypad support added

This patch adds support for GPIO based Matric(2 x 3) Keypad.

Signed-off-by: Goutam Kumar <goutam.kumar@ti.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
Signed-off-by: Patil, Rachna <rachna@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoam335x-evm: wl12xx pin mux detection between rev1.0 and rev1.1
Eyal Reizer [Thu, 17 Nov 2011 15:00:24 +0000 (17:00 +0200)]
am335x-evm: wl12xx pin mux detection between rev1.0 and rev1.1

* WLAN and Bluetooth enable pins are different between alpha(rev1.0) and
  beta (rev1.1a) boards.
  Use the right pin muxing based on the header read from eeprom.

* Add the BT and WLAN enable pins to the wl12xx platform data so they can
  be used by external modules

Signed-off-by: Eyal Reizer <eyalr@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoboard-am335xevm: changes for working with catalog wl1271 com6m module
Eyal Reizer [Mon, 14 Nov 2011 07:44:10 +0000 (09:44 +0200)]
board-am335xevm: changes for working with catalog wl1271 com6m module

* The catalog COM6M module is using a 38.4Mhz crystal

* Do not activate the BT enable pin during boot. this is done from the
  appliaction when BT is brought up.

Signed-off-by: Eyal Reizer <eyalr@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoam335x_evm_defconfig: do not activate bluetooth modules
Eyal Reizer [Mon, 14 Nov 2011 07:40:14 +0000 (09:40 +0200)]
am335x_evm_defconfig: do not activate bluetooth modules

Turn off the bluetooth realted modules in the default defconfig
file as we switch to using a compat package which is bluetooth
certifiable.

Signed-off-by: Eyal Reizer <eyalr@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agousb: musb: cppi41: Workaround fix for Tx fifo flush during disconnect
Ajay Kumar Gupta [Fri, 18 Nov 2011 06:56:02 +0000 (12:26 +0530)]
usb: musb: cppi41: Workaround fix for Tx fifo flush during disconnect

Unable to flush musb Tx endpoint fifo during disconnect while i/o
in progress. The workaround is to set only fifoflush bit and clear
other bits in tx-csr register.

Signed-off-by: Ravi B <ravibabu@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agocan: d_can: Add D_CAN config to am335x_evm_defconfig
AnilKumar Ch [Sun, 6 Nov 2011 05:14:26 +0000 (10:44 +0530)]
can: d_can: Add D_CAN config to am335x_evm_defconfig

This patch adds the D_CAN and dependent modules configurations
to am335x_evm_defconfig

Dependent modules are:
CONFIG_CAN_RAW
CONFIG_CAN_BCM

Signed-off-by: AnilKumar Ch <anilkumar@ti.com>
12 years agocan: d_can: Add platform data for am33xx device
AnilKumar Ch [Sat, 26 Nov 2011 20:41:48 +0000 (02:11 +0530)]
can: d_can: Add platform data for am33xx device

This patch adds the platform data needed by the driver. Add the
resources to the difference d_can instances.

Initialization of message ram is necessary to read/write the
message object from/into the message RAM

Signed-off-by: AnilKumar Ch <anilkumar@ti.com>
12 years agocan: d_can: Add support for Bosch D_CAN controller
AnilKumar Ch [Sun, 6 Nov 2011 08:26:35 +0000 (13:56 +0530)]
can: d_can: Add support for Bosch D_CAN controller

Bosch D_CAN controller is a full-CAN implementation which is compliant
to CAN protocol version 2.0 part A and B. Bosch D_CAN user manual can be
obtained from:

http://www.semiconductors.bosch.de/media/en/pdf/ipmodules_1/can/
d_can_users_manual_111.pdf

This patch adds the support for D_CAN controller.
The following are the design choices made while writing the controller
driver:
1. Interface Register set IF1 has be used for transmit and IF2 is used for
   receive message objects.
2. Out of the total Message objects available, half of it are kept aside for
   RX purposes and the rest for TX purposes.
3. NAPI implementation is such that both the TX and RX paths functions
   in polling mode.

This patch adds the dcan driver support to am335x chip. This patch also adds
the can loopback mode from canutils.

Signed-off-by: AnilKumar Ch <anilkumar@ti.com>
12 years agoarm:omap:cpsw: Enable/Disable gig support for different evm rev
Chandan Nath [Fri, 11 Nov 2011 11:59:07 +0000 (17:29 +0530)]
arm:omap:cpsw: Enable/Disable gig support for different evm rev

This patch is added to enable cpsw gigabit mode support for beta
evm and disable gigabit support for alpha EVMs. Please note that
phy_register_fixup_for_uid() function cannot be used to configure
phydev->supported as it is not register configuration and that needs
to be configured after phy->connect.

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am33xx: Reverting gigabit disabling support
Chandan Nath [Fri, 11 Nov 2011 10:42:48 +0000 (16:12 +0530)]
arm:omap:am33xx: Reverting gigabit disabling support

This patch is added to revert back gigabit disable changes.
Due to hw issue gigabit support was disabled in alpha revision
and now this will be handled with respect to evm revision which
will be pushed after this patch.

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap3:am33xx: HACK for bypassing RTC hwmod
Vaibhav Bedia [Mon, 10 Oct 2011 20:10:12 +0000 (01:40 +0530)]
arm:omap3:am33xx: HACK for bypassing RTC hwmod

Temp patch to avoid rtc clock getting reset
Needs to be investigated

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agousb: musb_core: kill all global and static variables
Ajay Kumar Gupta [Wed, 2 Nov 2011 05:45:40 +0000 (11:15 +0530)]
usb: musb_core: kill all global and static variables

This needed for dual instances support.
Changes include:
- Move fifo_mode, orig_dma_mask, otg_timer, first to musb struct
- Killed option for use_dma

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoARM: plat-omap: Fix warnings
Afzal Mohammed [Fri, 28 Oct 2011 07:10:07 +0000 (12:40 +0530)]
ARM: plat-omap: Fix warnings

Fix,

arch/arm/plat-omap/sram.c: In function 'omap3_sram_restore_context':
arch/arm/plat-omap/sram.c:330:3: warning: initialization makes pointer from integer without a cast
arch/arm/mach-omap2/pm34xx.c: In function 'omap_push_sram_idle':
arch/arm/mach-omap2/pm34xx.c:848:22: warning: initialization makes pointer from integer without a cast
arch/arm/mach-omap2/pm34xx.c:851:28: warning: initialization makes pointer from integer without a cast

Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agomusb: ti81xx: kill global and static variable for multi instance
Ajay Kumar Gupta [Mon, 24 Oct 2011 11:25:29 +0000 (16:55 +0530)]
musb: ti81xx: kill global and static variable for multi instance

Moved global variable "otg_workaround" and static variable "last_timer"
to "struct musb".

Signed-off-by: Ajay Kumar Gupta <ajay.gupta@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agorcu: avoid build error for third-party modules
Paul E. McKenney [Mon, 24 Oct 2011 11:32:57 +0000 (17:02 +0530)]
rcu: avoid build error for third-party modules

The initial definition of __kfree_rcu() checked a static inline function
argument to see if it was a compile-time constant.  Apparently not all
compilers are willing to put up with this at all optimization levels.
Add a nasty comment and remove the warning, relying on the fact that
__kfree_rcu() is called only from kfree_rcu(), which always passes in
a compile-time constant.

Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agousb: defconfig: add back VIDEO support
Ajay Kumar Gupta [Fri, 21 Oct 2011 08:48:25 +0000 (14:18 +0530)]
usb: defconfig: add back VIDEO support

Also enabled missing CONFIG_OTG and CONFIG_USB_SUSPEND.
Also added the changes required for new config look in v3.2

Signed-off-by: Ajay Kumar Gupta <ajay.gupta@ti.com>
12 years agousb: ether_gadget: fix insmod panic after fsg rmmod
Ajay Kumar Gupta [Fri, 21 Oct 2011 05:02:12 +0000 (10:32 +0530)]
usb: ether_gadget: fix insmod panic after fsg rmmod

Fixes kernel panic seen during g_ether.ko module insertion performed
after the FSG gadget rmmod.

Ether gadget doesn't reset the usb_ep(s) during initialization and
so was getting the previous gadget's usb_ep->desc data.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am335x: Corrects cpsw sa_lo and sa_hi offset
Chandan Nath [Fri, 21 Oct 2011 09:42:06 +0000 (15:12 +0530)]
arm:omap:am335x: Corrects cpsw sa_lo and sa_hi offset

This patch is added to correct register offset of cpgmac
sl1 source low register address and cpgmac sl2 source high
register address. The register offset were 0x21c, 0x220 and
0x31c, 0x320. Instead they are corrected as 0x220, 0x224 and
0x320, 0x324. This is corrected by adding missing P1_TS_SEQ_MTYPE
register in cpsw_slave_reg structure.

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am335x: CPSW autonegotiates to 10 mbps for beaglebone
Chandan Nath [Wed, 19 Oct 2011 09:51:43 +0000 (15:21 +0530)]
arm:omap:am335x: CPSW autonegotiates to 10 mbps for beaglebone

This patch is added as an workaround for cpsw in case of beaglebone
board to autonegotiates to 10 mbps speed even if it is connected to
100 mbps link. The reason for this is to overcome hardware issue of
100 mbps speed. However, CPSW works fine for both 10 and 100 mbps
speed in case of am335x evm.

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am33xx: TLK110 PHY support
Vaibhav Hiremath [Sat, 26 Nov 2011 20:12:56 +0000 (01:42 +0530)]
arm:omap:am33xx: TLK110 PHY support

This patch adds support for TLK110 PHY. This include some
additional phy register configuration for 1.0 phy version.

Signed-off-by: Chandan Nath <chandan.nath@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am335x: Enable JTAG clock by default
Vaibhav Bedia [Thu, 13 Oct 2011 17:28:43 +0000 (22:58 +0530)]
arm:omap:am335x: Enable JTAG clock by default

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:debug: Add a config option for enabling JTAG clock
Vaibhav Bedia [Thu, 13 Oct 2011 16:54:16 +0000 (22:24 +0530)]
arm:debug: Add a config option for enabling JTAG clock

In a power aware system all unused clocks should be disabled.
However we might want to enable the JTAG clock alone for debugger
connectivity if such an option exists.

Signed-off-by: Vaibhav Bedia <vaibhav.bedia@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agomusb:host: removing warning message while unable to flush txfifo
Ravi B [Wed, 12 Oct 2011 11:11:10 +0000 (16:41 +0530)]
musb:host: removing warning message while unable to flush txfifo

changing the warning statement to debug when the tx fifo
was unable to flush.

Signed-off-by: Ravi B <ravibabu@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agomusb:gadget: use pio mode for zero byte transfer in gadget mode
Ravi B [Wed, 12 Oct 2011 05:59:23 +0000 (11:29 +0530)]
musb:gadget: use pio mode for zero byte transfer in gadget mode

The cppi41 txdma is unable to transfer zero byte length, this
workaround uses pio mode to transfer zero byte length.

Signed-off-by: Ravi B <ravibabu@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoEnable bit15 in the maccontrol register for 100Mbps.
Steve Kipisz [Thu, 6 Oct 2011 20:57:45 +0000 (15:57 -0500)]
Enable bit15 in the maccontrol register for 100Mbps.

Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am33xx: Add eeprom board detection logic for beaglebone
Steve Kipisz [Thu, 13 Oct 2011 11:25:50 +0000 (16:55 +0530)]
arm:omap:am33xx: Add eeprom board detection logic for beaglebone

In case of wrong EEPROM ID, or if device fails to read/detect the
board fall back to BeagleBone board.

Also, add support for BeagleBone version detection and support for
old boards.

Difference in BeagleBone variants.

* Have different dev_cfg structures and setup functions for new, old
  beaglebone boards setup pin mux accordingly

* Fall back to older Bone boards if EEPROM reads are incorrect or empty

* Read version field of EEPROM config to call correct setup_beaglebone
  function according to board version

Signed-off-by: Steve Kipisz <s-kipisz2@ti.com>
Signed-off-by: Joel A Fernandes <joelagnel@ti.com>
Signed-off-by: Sekhar Nori <nsekhar@ti.com>
Signed-off-by: Hebbar, Gururaja <gururaja.hebbar@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>
12 years agoarm:omap:am33xx: IA EVM low level debug
Afzal Mohammed [Thu, 13 Oct 2011 05:33:46 +0000 (11:03 +0530)]
arm:omap:am33xx: IA EVM low level debug

Low level debug support for IA EVM, it uses UART3
(numbering as per TRM). This is referred as UART4
by software, software UART numbering starts from 1.

Signed-off-by: Afzal Mohammed <afzal@ti.com>
Signed-off-by: Vaibhav Hiremath <hvaibhav@ti.com>