Commit graph

506,942 commits

Author SHA1 Message Date
Lai Jiangshan
038366c5cf workqueue: make work_busy() test WORK_STRUCT_PENDING first
Currently, work_busy() first tests whether the work has a pool
associated with it and if not, considers it idle.  This works fine
even for delayed_work.work queued on timer, as __queue_delayed_work()
sets cwq on delayed_work.work - a queued delayed_work always has its
cwq and thus pool associated with it.

However, we're about to update delayed_work queueing and this won't
hold.  Update work_busy() such that it tests WORK_STRUCT_PENDING
before the associated pool.  This doesn't make any noticeable behavior
difference now.

With work_pending() test moved, the function read a lot better with
"if (!pool)" test flipped to positive.  Flip it.

While at it, lose the comment about now non-existent reentrant
workqueues.

tj: Reorganized the function and rewrote the description.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2013-02-06 18:04:53 -08:00
Lai Jiangshan
6be195886a workqueue: replace WORK_CPU_NONE/LAST with WORK_CPU_END
Now that workqueue has moved away from gcwqs, workqueue no longer has
the need to have a CPU identifier indicating "no cpu associated" - we
now use WORK_OFFQ_POOL_NONE instead - and most uses of WORK_CPU_NONE
are gone.

The only left usage is as the end marker for for_each_*wq*()
iterators, where the name WORK_CPU_NONE is confusing w/o actual
WORK_CPU_NONE usages.  Similarly, WORK_CPU_LAST which equals
WORK_CPU_NONE no longer makes sense.

Replace both WORK_CPU_NONE and LAST with WORK_CPU_END.  This patch
doesn't introduce any functional difference.

tj: s/WORK_CPU_LAST/WORK_CPU_END/ and rewrote the description.

Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
2013-02-06 18:04:53 -08:00
Sjur Brændeland
aded024a12 virtio_console: Don't access uninitialized data.
Don't access uninitialized work-queue when removing device.
The work queue is initialized only if the device multi-queue.
So don't call cancel_work unless this is a multi-queue device.

This fixes the following panic:

Kernel panic - not syncing: BUG!
Call Trace:
62031b28:  [<6026085d>] panic+0x16b/0x2d3
62031b30:  [<6004ef5e>] flush_work+0x0/0x1d7
62031b60:  [<602606f2>] panic+0x0/0x2d3
62031b68:  [<600333b0>] memcpy+0x0/0x140
62031b80:  [<6002d58a>] unblock_signals+0x0/0x84
62031ba0:  [<602609c5>] printk+0x0/0xa0
62031bd8:  [<60264e51>] __mutex_unlock_slowpath+0x13d/0x148
62031c10:  [<6004ef5e>] flush_work+0x0/0x1d7
62031c18:  [<60050234>] try_to_grab_pending+0x0/0x17e
62031c38:  [<6004e984>] get_work_gcwq+0x71/0x8f
62031c48:  [<60050539>] __cancel_work_timer+0x5b/0x115
62031c78:  [<628acc85>] unplug_port+0x0/0x191 [virtio_console]
62031c98:  [<6005061c>] cancel_work_sync+0x12/0x14
62031ca8:  [<628ace96>] virtcons_remove+0x80/0x15c [virtio_console]
62031ce8:  [<628191de>] virtio_dev_remove+0x1e/0x7e [virtio]
62031d08:  [<601cf242>] __device_release_driver+0x75/0xe4
62031d28:  [<601cf2dd>] device_release_driver+0x2c/0x40
62031d48:  [<601ce0dd>] driver_unbind+0x7d/0xc6
62031d88:  [<601cd5d9>] drv_attr_store+0x27/0x29
62031d98:  [<60115f61>] sysfs_write_file+0x100/0x14d
62031df8:  [<600b737d>] vfs_write+0xcb/0x184
62031e08:  [<600b58b8>] filp_close+0x88/0x94
62031e38:  [<600b7686>] sys_write+0x59/0x88
62031e88:  [<6001ced1>] handle_syscall+0x5d/0x80
62031ea8:  [<60030a74>] userspace+0x405/0x531
62031f08:  [<600d32cc>] sys_dup+0x0/0x5e
62031f28:  [<601b11d6>] strcpy+0x0/0x18
62031f38:  [<600be46c>] do_execve+0x10/0x12
62031f48:  [<600184c7>] run_init_process+0x43/0x45
62031fd8:  [<60019a91>] new_thread_handler+0xba/0xbc

Signed-off-by: Sjur Brændeland <sjur.brandeland@stericsson.com>
Cc: stable@kernel.org
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
2013-02-07 11:37:37 +10:30
Greg Kroah-Hartman
3213f631da backlight: fix class_find_device() arguments
Michał's previous patch missed this backlight check to fix up the
class_find_device() arguments.

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Cc: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2013-02-06 16:43:02 -08:00
Xiao Guangrong
24db2734ad KVM: MMU: cleanup __direct_map
Use link_shadow_page to link the sp to the spte in __direct_map

Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-02-06 22:42:09 -02:00
Xiao Guangrong
f761620377 KVM: MMU: remove pt_access in mmu_set_spte
It is only used in debug code, so drop it

Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-02-06 22:42:08 -02:00
Xiao Guangrong
55dd98c3a8 KVM: MMU: cleanup mapping-level
Use min() to cleanup mapping_level

Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-02-06 22:42:08 -02:00
Xiao Guangrong
caf6900f2d KVM: MMU: lazily drop large spte
Currently, kvm zaps the large spte if write-protected is needed, the later
read can fault on that spte. Actually, we can make the large spte readonly
instead of making them not present, the page fault caused by read access can
be avoided

The idea is from Avi:
| As I mentioned before, write-protecting a large spte is a good idea,
| since it moves some work from protect-time to fault-time, so it reduces
| jitter.  This removes the need for the return value.

Reviewed-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-02-06 22:28:01 -02:00
Jingoo Han
4b07c5d512 leds: leds-sunfire: use dev_err()/pr_err() instead of printk()
Fixed the checkpatch errors and warnings as below:

  ERROR: spaces required around that '=' (ctx:VxW)
  WARNING: Prefer netdev_err(netdev, ... then dev_err(dev, ... then pr_err(...  to printk(KERN_ERR ...

Signed-off-by: Jingoo Han <jg1.han@samsung.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 16:00:43 -08:00
Axel Lin
61d4eb2724 leds: 88pm860x: Add missing of_node_put()
of_find_node_by_name() returns a node pointer with refcount incremented, use
of_node_put() on it when done.

of_find_node_by_name() will call of_node_put() against from parameter,
thus we also need to call of_node_get(from) before calling
of_find_node_by_name().

Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 16:00:43 -08:00
Axel Lin
ef754e88e3 leds: tca6507: Use of_get_child_count()
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 16:00:42 -08:00
Peter Ujfalusi
ad0ba85fab leds: leds-pwm: make it depend on PWM and not HAVE_PWM
The correct dependency for the leds-pwm is PWM and not HAVE_PWM
since PWM drivers now have their own subsystem.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 16:00:42 -08:00
Gleb Natapov
5037878e22 KVM: VMX: cleanup vmx_set_cr0().
When calculating hw_cr0 teh current code masks bits that should be always
on and re-adds them back immediately after. Cleanup the code by masking
only those bits that should be dropped from hw_cr0. This allow us to
get rid of some defines.

Signed-off-by: Gleb Natapov <gleb@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
2013-02-06 22:00:02 -02:00
Milo(Woogyom) Kim
c0285f8ca8 Documentation: leds: update LP55xx family devices
Update changed platform data information.
 Add leds-lp55xx.txt which includes the firmware interface description.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:30 -08:00
Milo(Woogyom) Kim
109b833071 leds-lp55xx: fix problem on removing LED attributes
LP55XX common device attributes, 'led_current' and 'max_current' are created
 while loading the driver.
 Those are LED device attributes which are removed automatically on releasing
 led class devices - led_classdev_unregister().
 Therefore, this duplicate code should be removed.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:30 -08:00
Milo(Woogyom) Kim
a2387cb9f6 leds-lp5521/5523: add author and copyright description
Now LP5521 and LP5523 drivers are based on new lp55xx structure.
 So the author and copyrights are updated.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:30 -08:00
Milo(Woogyom) Kim
df4094d24e leds-lp5521/5523: use new lp55xx common header
The LP55xx common driver provides a new header, leds-lp55xx.h.
 This driver enables removing duplicate code for both drivers and
 making coherent driver structure.
 LP5521 and LP5523/55231 platform data were merged into one common file.
 Therefore, the LP5521/5523 platform code need to be fixed.

 This patch has been already acked.

 For ux500: https://lkml.org/lkml/2012/10/11/417
 Acked-by: Linus Walleij <linus.walleij@linaro.org>

 For omap: https://lkml.org/lkml/2012/10/11/334
 Acked-by: Tony Lindgren <tony@atomide.com>

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:30 -08:00
Milo(Woogyom) Kim
79bcc10b8c leds-lp55xx: clean up headers
Remove unused headers and sort them alphabetically

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:30 -08:00
Milo(Woogyom) Kim
12f022d27b leds-lp55xx: clean up definitions
Remove unused definitions and change hex values to capital letters

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:30 -08:00
Milo(Woogyom) Kim
93ca4093ad leds-lp55xx: clean up unused data and functions
Old data structures and I2C function are not used any more.
 Each driver uses the lp55xx common data and functions.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:30 -08:00
Milo(Woogyom) Kim
87cc4bde2a leds-lp55xx: clean up _remove()
Replace lp5521/5523_unregister_sysfs() with lp55xx_unregister_sysfs().

 On unloading the driver, running engines should be stopped.
 Use explicit driver function, lp5521/5523_stop_engine().

 Unused functions are removed.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:29 -08:00
Milo(Woogyom) Kim
ba6fa84651 leds-lp55xx: add new function for removing device attribtues
lp55xx_unregister_sysfs() is used for removing lp55xx device attributes.
 Chip specific and engine attributes are removed on unloading the driver.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:29 -08:00
Milo(Woogyom) Kim
9ca3bd8022 leds-lp55xx: code refactoring on selftest function
LP5521 and LP5523 have a selftest function which is run via the sysfs.
 Use lp55xx driver data and R/W functions rather than lp5521/5523 private data
 and functions.
 Additionally, if-statements are changed for code simplicity.
 Unused functions, lp5521/5523_read() are removed.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:29 -08:00
Milo(Woogyom) Kim
e73c0ce6be leds-lp55xx: use common device attribute driver function
lp5521/5523_register_sysfs() are replaced with lp55xx common driver function,
 lp55xx_register_sysfs().
 Chip specific device attributes are configurable using 'dev_attr_group'.

 Error condition name is changed:
 use specific error condition, 'err_register_sysfs' rather than unclear name,
 'fail2'.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:29 -08:00
Milo(Woogyom) Kim
240085e255 leds-lp55xx: support device specific attributes
To support device specific attributes, new common driver function is added.
 Eventually those are created on registering the sysfs with common dev attrs.
 Furthermore, this patch makes adding device attributes simple in each driver.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:29 -08:00
Milo(Woogyom) Kim
db6eaf8388 leds-lp5523: use generic firmware interface
LP55xx common driver provides generic firmware interface
 for running a LED pattern.
 LP5521 and LP5523 have many device attributes for running patterns.
 This patch cleans up those complex code.

 Removed device attributes:
 engine1_mode
 engine2_mode
 engine3_mode
 engine1_load
 engine2_load
 engine3_load
 engine1_leds
 engine2_leds
 engine3_leds

 All device attributes and functions are replaced with two callback functions,
 'firmware_cb' and 'run_engine'.

 New engine functions:
 lp5523_load/stop/run_engine(), lp5523_update_program_memory() and
 lp5523_wait_opmode_done()

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:29 -08:00
Milo(Woogyom) Kim
9ce7cb170f leds-lp5521: use generic firmware interface
LP55xx common driver provides generic firmware interface
 for running a LED pattern.
 LP5521 and LP5523 have many device attributes for running patterns.
 This patch cleans up those complex code.

 Removed device attributes:
 engine1_mode
 engine2_mode
 engine3_mode
 engine1_load
 engine2_load
 engine3_load
 led_pattern

 All device attributes and functions are replaced with two callback functions,
 'firmware_cb' and 'run_engine'.

 New engine functions:
 lp5521_load/stop/run_engine(), lp5521_update_program_memory() and
 lp5521_wait_opmode_done()

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:29 -08:00
Milo(Woogyom) Kim
10c06d178d leds-lp55xx: support firmware interface
This patch provides additional device attributes which enable
 loading the firmware. ('select_engine' and 'run_engine')
 To run a LED pattern, two parts of driver should be enabled.

 Common features : lp55xx-common
 ===============================
 Firmware interface for loading LED patterns

 Chip specific features : leds-lp5521, leds-lp5523
 =================================================
 Register addresses for loading firmware data
 Register addresses for running selected engine

 Pattern programming sequence
 ============================
 LP55xx chips have three program engines.
 To load and run a LED pattern, the programming sequence is as follows.
 (1) Select an engine number (1/2/3)
 (2) Set engine mode to load
 (3) Write pattern data into selected area
 (4) Set engine mode to run

 This sequence is almost same as the firmware interface.
 (1) Select an engine number               : 'select_engine' dev attribute
 (2) Mode change to load                   : 'loading' of firmware class
 (3) Write pattern data into selected area : 'data' of firmware class
 (4) Mode change to run                    : 'run_engine' dev attribute

 (1) and (4) are device specific features which provide callback functions
 (2) and (3) are common features.

 For example,
 echo 1 or 2 or 3 > /sys/bus/i2c/devices/xxxx/select_engine
 echo 1 > /sys/class/firmware/lp5521/loading
 echo "4000600040FF6000" > /sys/class/firmware/lp5521/data
 echo 0 > /sys/class/firmware/lp5521/loading
 echo 1 > /sys/bus/i2c/devices/xxxx/run_engine

 As soon as 'loading' is set to 0, registered callback is called.
 Inside the callback, the selected engine is loaded and memory is updated.
 To run programmed pattern, 'run_engine' attribute should be enabled.

 Device specific data structure
 ==============================
 o Firmware callback
   load selected engine and update program memory
 o Run engine
   change the engine mode
 o 'engine_idx' and firmware data, 'fw'
   Those are used in the driver internally with callback functions

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:28 -08:00
Milo(Woogyom) Kim
b3b6f8119d leds-lp55xx: add new lp55xx_register_sysfs() for the firmware interface
LP55xx family chips have internal program memory which run various patterns.
 Using this memory, LEDs continue on blinking/dimming without continuous I2C
 commands. That means the I2C HOST can be entered into sleep once the memory
 is updated.

 An application can get hex data from a file and write them into
 the program memory through the I2C. This is general firwmare interface.

 This patch is the initial step for adding the firmware interface.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:28 -08:00
Milo(Woogyom) Kim
d9067d2846 leds-lp55xx: fix error condition in lp55xx_register_leds()
Use lp55xx_unregister_leds() rather than duplicate code.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:28 -08:00
Milo(Woogyom) Kim
c3a68ebfcd leds-lp55xx: use lp55xx_unregister_leds()
To unregister led class devices and sysfs attributes,
 LP5521 and LP5523 have each driver function.
 This patch makes both drivers simple using common driver function,
 lp55xx_unregister_leds().

 And some unused variables are removed.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:28 -08:00
Milo(Woogyom) Kim
a96bfa135d leds-lp55xx: provide common LED current setting
LED current is configurable via the sysfs.
 Max current is a read-only attribute.
 These attributes code can be shared in lp55xx common driver.

 Device attributes: 'led_current' and 'max_current'
 move to lp55xx common driver

 Replaced functions:
 show_max_current()  => lp55xx_show_max_current()
 show_current()      => lp55xx_show_current()
 store_current()     => lp55xx_store_current()

 LED setting function: set_led_current()
 Current registers are device specific, so configurable function is added
 in each driver.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:28 -08:00
Milo(Woogyom) Kim
a6e4679a09 leds-lp55xx: use lp55xx_set_brightness()
lp5521_set_brightness() and lp5523_set_brightness() are replaced with
 common function, lp55xx_set_brightness().
 This function is invoked when the brightness of each LED channel is updated.
 LP5521 and LP5523 have different register address for the brightness control,
 so this work is done by chip specific brightness_work_fn().

 lp5521/5523_led_brightness_work():
 use lp55xx_led and lp55xx_chip data structure.
 use lp55xx write function.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:28 -08:00
Milo(Woogyom) Kim
0e2023463a leds-lp55xx: use lp55xx_init_led() common function
lp5521_init_led() and lp5523_init_led() are replaced with one common function,
 lp55xx_init_led().
 Max channels is configurable, so it's used in lp55xx_init_led().

 'LP5523_LEDS' are changed to 'LP5523_MAX_LEDS'.

 lp55xx_set_brightness, lp55xx_led_attributes: skeleton
 Will be filled in next patches.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:28 -08:00
Milo(Woogyom) Kim
9e9b3db1b2 leds-lp55xx: use lp55xx common led registration function
LED class devices are registered in lp5521_register_leds() and
 lp5523_register_leds().
 Two separate functions are merged into consolidated lp55xx function,
 lp55xx_register_leds().

 Error handling fix:
 Unregistering LEDS are handled in lp55xx_register_leds() when LED registration
 failure occurs. So each driver error handler is changed to 'err_register_leds'

 Chip dependency: 'brightness_work_fn' and 'set_led_current'
 To make the structure abstract, both functions are configured in each driver.
 Those functions should be done by each driver because register control is
 chip-dependant work.

 lp55xx_init_led: skeleton
 Will be filled in next patch

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:28 -08:00
Milo(Woogyom) Kim
6ce6176263 leds-lp55xx: use lp55xx common deinit function
Two separate de-init functions are merged into one common function.
 And it is used in err_post_init of lp55xx_init_device().

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:27 -08:00
Milo(Woogyom) Kim
22ebeb488b leds-lp55xx: clean up init function
lp5521/5523_init_device() are replaced with lp55xx common function,
 lp55xx_init_device().

 Error handler in init_device:
 deinit function are matched with 'err_post_init' section in
 lp55xx_init_device().

 Remove LP5523 engine intialization code:
 Engine functionality is not mandatory but optional.
 Moreover engine initialization is done internally with device reset command.
 Therefore, this code is unnecessary.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:27 -08:00
Milo(Woogyom) Kim
ffbdccdbba leds-lp55xx: use lp55xx common init function - post int
LP5521/5523 chip configuration is replaced with lp55xx common function,
 lp55xx_post_init_device().

 Name change:
 lp5521/5523_configure() to lp5521/5523_post_init_device()
 These are called in init function.

 Register access function
 Argument type is changed from 'i2c_client' to 'lp55xx_chip'.
 Use exported R/W functions of lp55xx common driver.

 Temporary variables in lp5521/5523_init_device()
 These functions will be removed but temporary variables are needed for
 blocking build warnings - incompatible pointer.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:27 -08:00
Milo(Woogyom) Kim
e3a700d8aa leds-lp55xx: use lp55xx common init function - detect
LP5521/5523 chip detection functions are replaced with lp55xx common function,
 lp55xx_detect_device().
 Chip dependent address and values are configurable in each driver.
 In init function, chip detection is executed.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:27 -08:00
Milo(Woogyom) Kim
48068d5de1 leds-lp55xx: use lp55xx common init function - reset
LP5521/5523 reset device functions are moved to lp55xx common driver.
 Value of register address and value are chip dependent.
 Those are configured in each driver.
 In init function, reset command is executed.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:27 -08:00
Milo(Woogyom) Kim
a85908dd77 leds-lp55xx: use lp55xx common init function - platform data
LP5521/5523 platform data functions are moved to lp55xx common driver.
 New init function, lp55xx_init_device() is created.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:27 -08:00
Milo(Woogyom) Kim
632418bf65 leds-lp5523: clean up lp5523_configure()
This patch is a preceding step for making common lp55xx init function.

 Return code:
 Do not use 'OR' arithmetic for the result.
 If some error occurs, just return it.

 Remove engine verification code:
 To check whether internal engine works or not, many lines of code are executed.
 However, this job is unnecessary during the chip initialization because
 the engine usage is not mandatory but optional function.
 LED engines are enabled when specific LED pattern is loaded.
 Therefore, this verification code is removed.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:27 -08:00
Milo(Woogyom) Kim
9448217403 leds-lp5521: clean up lp5521_configure()
This patch is a preceding step for making common lp55xx init function.

 LP5521_REG_R_CURRENT register code moved:
 Chip specific code moved from lp5521_init_device() to lp5521_configure().

 Remove engine init function:
 LP5521 has internal program engines which are used for running LED patterns.
 (blinking, ramp up/down and other emotional visual effects)
 Engine initialization is done by reset command in lp5521_init_device().
 Remove this duplicate code.

 Return code:
 Do not use 'OR' arithmetic for the result.
 If some error occus, just return it.

 Enable latency:
 Use explicit named function, lp5521_wait_enable_done().
 According to the datasheet, 500us is guaranteed time.
 Thus wait time is changed from 1000us to 500us.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:26 -08:00
Milo(Woogyom) Kim
6a0c9a4796 leds-lp55xx: use common lp55xx data structure in _probe()
LP5521 and LP5523 data structures have common features.
 Use common lp55xx data structures rather than chip specific data.
 Legacy code in probe is replaced with this new data structures.

 lp55xx_chip          : Common data between lp5521_chip and lp5523_chip
 lp55xx_led           : Common LED structure between lp5521_led and lp5523_led
 lp55xx_platform_data : Common platform data between lp5521_platform_data and
                        lp5523_platform_data

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:26 -08:00
Milo(Woogyom) Kim
945c700746 leds-lp55xx: replace name of data structure
Change the name of chip data structure and platform data.
 This patch is a preceding step for cleaning up lp5521/5523 probe and remove.
 These data will be replaced with new lp55xx common data structures
 in next patch.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:26 -08:00
Milo(Woogyom) Kim
c93d08fa75 leds-lp55xx: add new common driver for lp5521/5523
This patch supports basic common driver code for LP5521, LP5523/55231 devices.

 ( Driver Structure Data )

 lp55xx_led and lp55xx_chip
 In lp55xx common driver, two different data structure is used.
 o lp55xx_led
   control multi output LED channels such as led current, channel index.
 o lp55xx_chip
   general chip control such like the I2C and platform data.

 For example, LP5521 has maximum 3 LED channels.
 LP5523/55231 has 9 output channels.

 lp55xx_chip for LP5521 ... lp55xx_led #1
                            lp55xx_led #2
                            lp55xx_led #3

 lp55xx_chip for LP5523 ... lp55xx_led #1
                            lp55xx_led #2
                            .
                            .
                            lp55xx_led #9

 ( Platform Data )

 LP5521 and LP5523/55231 have own specific platform data.
 However, this data can be handled with just one platform data structure.
 The lp55xx platform data is declared in the header.
 This structure is derived from leds-lp5521.h and leds-lp5523.h

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:26 -08:00
Milo(Woogyom) Kim
f6c64c6fc8 leds-lp55xx: do chip specific configuration on device init
Chip specific function is configured when the device is initialized.
 So _configure() is moved to each device init function.

 If chip configuration gets failed, the device is de-initialized in
 each _init_device(), not probe().

 For compile error fix, function type declarations are added.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:26 -08:00
Milo(Woogyom) Kim
86eb7748ce leds-lp55xx: add device reset function in lp5521/5523
Use explicit each driver function rather than raw command.
 These function will be merged into the lp55xx common driver.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:26 -08:00
Milo(Woogyom) Kim
1904f83d56 leds-lp55xx: clean up deinit leds in lp5521/5523
To make LED unregistration code simple, new function, _unregister_leds()
 is added in each driver.
 This patch is a preceding step for lp55xx common driver architecture.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:26 -08:00
Milo(Woogyom) Kim
f652480802 leds-lp55xx: clean up init leds in lp5521/5523
To make LED initialization code simple, new function, _register_leds()
 is added at each driver.
 This patch is a preceding step for lp55xx common driver architecture.

Signed-off-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Bryan Wu <cooloney@gmail.com>
2013-02-06 15:59:26 -08:00