kernel.git
8 years agoAR6000: move low-level cleanup from ar6000_destroy to ar6000_close before-nelson-commits
Werner Almesberger [Thu, 2 Apr 2009 11:12:12 +0000 (08:12 -0300)]
AR6000: move low-level cleanup from ar6000_destroy to ar6000_close

A network device is supposed to disable its internals in the "close"
function. Doing so affords us protection against various races,
including the ioctl vs. rfkill conflict reported by Michael.

Signed-off-by: Werner Almesberger <werner@openmoko.org>
Reported-by: Michael Trimarch <michael@panicking.kicks-ass.org>

8 years agortc-pcf50606: Fix month off-by-one error
Werner Almesberger [Wed, 1 Apr 2009 02:23:05 +0000 (23:23 -0300)]
rtc-pcf50606: Fix month off-by-one error

According to the PCF50606 manual, the issues resoved by commit
cc1663fc922c03feb0d7bbb8b18d62fbac0128de also exists there.

Signed-off-by: Werner Almesberger <werner@openmoko.org>
Reported-by: Rask Ingemann Lambertsen <rask@sygehus.dk>

8 years agoAR6000: support GTA02-specific rfkill only if building for GTA02
Werner Almesberger [Tue, 31 Mar 2009 07:47:06 +0000 (04:47 -0300)]
AR6000: support GTA02-specific rfkill only if building for GTA02

This is loosely based on a patch by Ivan Petrov.

Signed-off-by: Werner Almesberger <werner@openmoko.org>
Reported-by: Ivan Petrov <ivan_p@hotbox.ru>

8 years agoThis patch brings suspend/resume back to GTA01
Tim Niemeyer [Fri, 27 Mar 2009 17:10:36 +0000 (18:10 +0100)]
This patch brings suspend/resume back to GTA01

"pdata" is NULL on GTA01 and you oops.
Revised patch, this time with if(..) instead of #ifdef

Signed-off-by: Tim Niemeyer <tim.niemeyer@mastersword.de>

8 years agoremove sysfs-links on remove
Tim Niemeyer [Fri, 27 Mar 2009 16:27:53 +0000 (17:27 +0100)]
remove sysfs-links on remove

Signed-off-by: Tim Niemeyer <tim.niemeyer@mastersword.de>

8 years agoAR6000: Corrected 'vendor/device IDs'.
Werner Almesberger [Tue, 31 Mar 2009 07:16:32 +0000 (04:16 -0300)]
AR6000: Corrected 'vendor/device IDs'.

IDs 0 and 1 are AR6002, not AR6001. This is based on a patch by
Ivan Petrov, but it differs in leaving the IDs in
include/linux/mmc/sdio_ids.h

Signed-off-by: Werner Almesberger <werner@openmoko.org>
Reported-by: Ivan Petrov <ivan_p@hotbox.ru>

8 years agoAR6000: revert MMC busy status check
Werner Almesberger [Tue, 31 Mar 2009 07:02:04 +0000 (04:02 -0300)]
AR6000: revert MMC busy status check

e2c0650efa751a6a2220618695fa41a2a5e7d23c introduced a platform-specific
hack to check if the MMC driver and hardware handle busy signaling from
the device properly.

Since this seems to be the case, we can revert this hack now.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoMinor AR6000 cleanup
Werner Almesberger [Tue, 31 Mar 2009 03:49:46 +0000 (00:49 -0300)]
Minor AR6000 cleanup

This patch cleans up a little after Ivan's netif_queue_stop fix.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAR6000 netif_queue_stop non stop, Bug?
ivan_p@hotbox.ru [Thu, 26 Mar 2009 12:03:03 +0000 (15:03 +0300)]
AR6000 netif_queue_stop non stop, Bug?

This patch resolves the following issue:
http://lists.openmoko.org/pipermail/openmoko-kernel/2009-March/009643.html

Changed: prevent rescheduling network queue at interface opened/connected.
Removed: wake network queue at transmit complete.
Added: wake network queue at packet queue limit not reached.

Signed-off-by: Ivan Petrov <ivan_p@hotbox.ru>

8 years agoFix ar6000_do_activate return value
Werner Almesberger [Tue, 31 Mar 2009 00:06:02 +0000 (21:06 -0300)]
Fix ar6000_do_activate return value

... and check the return value in ar6000_activate.

[ Werner: changed ___FUNCTION__ to __func__ ]

Signed-off-by: Michael Trimarchi <michael@panicking.kicks-ass.org>
Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoFix cpu_is_s3c2442() returning false for an S3C2442B
Rask Ingemann Lambertsen [Mon, 30 Mar 2009 22:28:09 +0000 (00:28 +0200)]
Fix cpu_is_s3c2442() returning false for an S3C2442B

   An S3C2442B would be detected as S3C2440 by the cpu_is_s3cxxxx()
macros. This patch fixes it.

Signed-off-by: Rask Ingeman Lambertsen <rask@sygehus.dk>

8 years agoCodingStyle with copy-src also fixed
Werner Almesberger [Thu, 26 Mar 2009 07:56:25 +0000 (04:56 -0300)]
CodingStyle with copy-src also fixed

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>

8 years agoFix trivial whitespace slippage.
Werner Almesberger [Thu, 26 Mar 2009 07:52:29 +0000 (04:52 -0300)]
Fix trivial whitespace slippage.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoClean up pcf50606_client_dev_register
tim.niemeyer@mastersword.de [Tue, 24 Mar 2009 10:11:20 +0000 (11:11 +0100)]
Clean up pcf50606_client_dev_register

platform_device_add_data copies the data, so no need for
kmalloc.

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>

8 years agoretain gllin compatibility
tim.niemeyer@mastersword.de [Tue, 24 Mar 2009 10:11:21 +0000 (11:11 +0100)]
retain gllin compatibility

added bus link from neo1973-pm-gps.0 to gta01-pm-gps.0
added pwron attr wich is the same as power_on
also one typo fixed (3V instead of 3V3 in gps_power_3v_set)

[ Werner: Documentation/CodingStyle line 83, don't ignore the error if
  sysfs_create_group fails. ]

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>
Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd forgotten PCF50606 Kconfig/Makefile entries
tim.niemeyer@mastersword.de [Tue, 24 Mar 2009 10:11:19 +0000 (11:11 +0100)]
Add forgotten PCF50606 Kconfig/Makefile entries

Watchdog and RTC entries were missing.

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>

8 years agoRemove double INPUT_PCF6333_PMU in Kconfig/Makefile
tim.niemeyer@mastersword.de [Tue, 24 Mar 2009 10:11:18 +0000 (11:11 +0100)]
Remove double INPUT_PCF6333_PMU in Kconfig/Makefile

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>

8 years agoReally clean up, as promised in 1350d5dfaac7dadf6cab9e812f0acfabcce60e58
Werner Almesberger [Thu, 26 Mar 2009 05:03:46 +0000 (02:03 -0300)]
Really clean up, as promised in 1350d5dfaac7dadf6cab9e812f0acfabcce60e58

Signed-off-by: Werner Almesberger <werner@openmoko.org>
Reported-by: Tim Niemeyer <tim.niemeyer@mastersword.de>

8 years agoChange disable serial driver for gta02 only
tim.niemeyer@mastersword.de [Sun, 22 Mar 2009 16:42:36 +0000 (17:42 +0100)]
Change disable serial driver for gta02 only

Change RTS and TXD pins on UART 0 to be inputs only on GTA02.
On GTA01 the UART 0 maybe used otherwise.

[ Werner: removed unexplained addition of DEBUG, Documentation/CodingStyle
  lines 105, 224, and 448. ]

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>
Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoextend work around boot-time ordering on GTA01
Tim Niemeyer [Thu, 26 Mar 2009 04:19:15 +0000 (01:19 -0300)]
extend work around boot-time ordering on GTA01

[ Werner: Documentation/CodingStyle lines 166 and 448, put blank line after
  local variables. ]

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>
Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agodisabling fast_charge is done with 0
tim.niemeyer@mastersword.de [Sun, 22 Mar 2009 16:42:34 +0000 (17:42 +0100)]
disabling fast_charge is done with 0

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>

8 years agolittle debug msg copied from gta02
tim.niemeyer@mastersword.de [Sun, 22 Mar 2009 16:42:33 +0000 (17:42 +0100)]
little debug msg copied from gta02

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>

8 years agoRemove debug-aux-key-probe-resume-death.patch
tim.niemeyer@mastersword.de [Sun, 22 Mar 2009 16:42:32 +0000 (17:42 +0100)]
Remove debug-aux-key-probe-resume-death.patch

Get resume by AUX to work.

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>

8 years agonew-style i2c driver doesn't work with inconsistent names.
tim.niemeyer@mastersword.de [Sun, 22 Mar 2009 16:42:30 +0000 (17:42 +0100)]
new-style i2c driver doesn't work with inconsistent names.

Signed-off-by: Tim Niemeyer <reddog@mastersword.de>

8 years agoGlamo fb: needs cfb_{fillrect,copyarea,imageblit} to link
Rask Ingemann Lambertsen [Thu, 19 Mar 2009 23:18:11 +0000 (00:18 +0100)]
Glamo fb: needs cfb_{fillrect,copyarea,imageblit} to link

   The kernel would fail to link because cfb_fillrect, cfb_copyarea and
cfb_imageblit weren't build. This patch fixes it.

Signed-off-by: Rask Ingemann Lambertsen <rask@sygehus.dk>

8 years agoZero packet length detection for RNDIS fix. Migrated from 2.6.24 series kernel
Aric D. Blumer [Mon, 16 Mar 2009 22:28:45 +0000 (16:28 -0600)]
Zero packet length detection for RNDIS fix. Migrated from 2.6.24 series kernel

This patch fixes the issues of RNDIS on the Openmoko Freerunner (and
subsequently Windows networking with the Freerunner). It arises from the
case where a stall is being issued by the upper level RNDIS driver to
the s3c2410_udc.c in the 2.6.28 series driver and it not being handled
correctly. This patch changes the RNDIS driver to send zero-length
packets instead of a stall, a condition that allowed the udc driver to
function correctly in 2.6.24 series kernels.

Further investigation into the udc driver is recommended. I am currently
looking into it to be able to use the Android ADB gadget driver, which
currently has stability issues using the underlying udc on both Linux
and Windows (although, it's possible that the Android ADB gadget has
problems).

Traces and logs of the USB traffic are available upon request. Thanks to
Aric at SDG Systems for all of his work on this matter.

Regards,
Brian Code

From de386af349bbfe1ad6d45c810185123975888d8d Mon Sep 17 00:00:00 2001
From: Aric D. Blumer <aric@sdgsystems.com>
Date: Mon, 16 Mar 2009 16:04:02 -0600
Subject: [PATCH] Zero packet length detection for RNDIS fix. Migrated from 2.6.24 series kernel

Signed-off-by: Aric D. Blumer <aric@sdgsystems.com>
Signed-off-by: Brian Code <Brian.Code@koolu.com>

8 years agoForced shutdown for GTA02
Ondrej Zary [Sun, 22 Mar 2009 18:42:02 +0000 (14:42 -0400)]
Forced shutdown for GTA02

Hello,
this patch adds "force-shutdown" feature for GTA02. When the power button is
held pressed for 8 seconds, the power is turned off. No need to remove the
battery anymore. Patch is against andy-tracking.

Signed-off-by: Ondrej Zary <linux@rainbow-software.org>
Signed-off-by: Nicolas Dufresne <nicolas.dufresne@gmail.com>

8 years agoS3C64XX: Add gpio_to_irq method for external interrupt group.
Matt Hsu [Thu, 12 Mar 2009 14:38:25 +0000 (22:38 +0800)]
S3C64XX: Add gpio_to_irq method for external interrupt group.

Add to_irq method to convert gpio to irq
for external interrupt group (GPN).

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agoFewer HDQ errors on 3D7K
Werner Almesberger [Wed, 11 Mar 2009 15:23:40 +0000 (12:23 -0300)]
Fewer HDQ errors on 3D7K

If no battery is connected, we periodically get a burst of HDQ error
messages (at least on 3D7K), interrupting whatever we're doing on
the console.

This patch reduces this to only one message per sequence of errors,
and one more message if communication with HDQ is successful later.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd proper audio route for headphone speaker in 3d7k. andy-latest
Matt Hsu [Wed, 11 Mar 2009 11:25:26 +0000 (11:25 +0000)]
Add proper audio route for headphone speaker in 3d7k.

The power mode of ROUT1 is always off which results
in the voice only comes from left channel (LOUT1).
This issues is fixed by adding dapm control for ROU1 channel.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>
Reported-by: Candy Chou <candy_chou@openmoko.com>

8 years agoEnable camera interface clock
Werner Almesberger [Wed, 11 Mar 2009 11:25:23 +0000 (11:25 +0000)]
Enable camera interface clock

This patch explicitly enabled to "camif" clock, to make the driver
work again with Qi commit a24b5fcf84d2cf633a3f660edd23fa4c2a3da231.

Note that this is not fully tested since the Qi change also broke
the LCM driver.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd real CAMIF clock to S3C6410
Werner Almesberger [Wed, 11 Mar 2009 11:25:20 +0000 (11:25 +0000)]
Add real CAMIF clock to S3C6410

This patch adds the real "camif" clock (off HCLK.)

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoRename CAMERA clock
Werner Almesberger [Wed, 11 Mar 2009 11:25:17 +0000 (11:25 +0000)]
Rename CAMERA clock

Identifiers related to the "camera" clock were mis-named as camif.
This patch renames them.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agofix-hdq-suspend-level.patch
Andy Green [Wed, 11 Mar 2009 11:25:15 +0000 (11:25 +0000)]
fix-hdq-suspend-level.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoadd primitive camera power control
Werner Almesberger [Tue, 10 Mar 2009 14:02:28 +0000 (14:02 +0000)]
add primitive camera power control

This dirty hack adds camera activation/deactivation on open/close.
This code will have to move to a machine/platform area later. The
general driver is the wrong place for it.

It also removes the unconditional activation and moves the domain
setting to the rest of the power control code. (Thanks, Andy !)

Note that opening the camera now always takes about 20 seconds
until the settings are downloaded, not just on the first open.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd filter_chain object
Nelson Castillo [Tue, 10 Mar 2009 12:04:40 +0000 (12:04 +0000)]
Add filter_chain object

Filter chains should be completely opaque to the drivers that use it.
We fix this with this patch.

~ Make the "filter chain" a new object.
~ We can build with CONFIG_TOUCHSCREEN_FILTER=n with no problems in
  a cleaner way.
~ Update s3c2410_ts.c to use the filter_chain object.
~ Cleanups.

Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>

8 years agoExport symbols and make a few symbols constant.
Nelson Castillo [Tue, 10 Mar 2009 12:04:26 +0000 (12:04 +0000)]
Export symbols and make a few symbols constant.

~ Make a few symbols constant.
~ Export symbols explicitly.
~ Move ts_filter.c to ts_filter_chain.c (this will make sense later).

Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>

8 years agoUse non-void configurations
Nelson Castillo [Tue, 10 Mar 2009 12:04:16 +0000 (12:04 +0000)]
Use non-void configurations

This patch defines a ts_filter_configuration structure to
avoid using void* in the filter initialization, fixing another
upstream correction. This also makes the initialization more readable.

Tested in GTA02/rev6.

Other changes:

 ~ Comment filter configuration structures.
 ~ ts_filter.c:ts_filter_chain_create improved.
 ~ Small cleanups.
 ~ More TODOs/FIXMEs.
 ~ Updated GTA02 filter configuration.
 ~ Updated GTA01 filter configuration.
 ~ Updated mach-s3c2410/include/mach/ts.h for the new ts. configuration
   structure.
 ~ Updated all the filters to use the new configuration structure.
 ~ Removed MAX_TS_FILTER_CHAIN constant that is no longer needed.

No more evil casts left it seems.

Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>

8 years agoImprove filter API and update filters
Nelson Castillo [Tue, 10 Mar 2009 12:04:05 +0000 (12:04 +0000)]
Improve filter API and update filters

This patch turns upstream feedback into API modifications and code
improvements. There will be more patches implementing upstream
corrections but this one is the that will make most of the invasive
changes and make the most important improvements to the API.

Tested in a GTA02/rev06.

The goals of this patch are:

* Replace recursive calls with iteration.
* General code improvements.
* Make ts_filter_mean.c a reference for the rest of the filters.
* Make the (almost)minimum number of changes to the other filters so
  that they compile and work, patches for cleaning these up will
  come next.
* Filters should do what they were doing before.

Some important changes:

* Move "struct ts_filter tsf" in the private structures to force
  a crash (or break things) if we forget to remove an open-coded cast.
* ts_filter.c/ts_filter.h
    ~ API modifications.
* s3c2410_ts.c:
    ~ Use the new API.
    ~ Cleanups.
* ts_filter_mean.c
    ~ Replace with a simple mean.
    ~ Use as a reference for the new API.
    ~ Move private structure from the .h to the .c.
* ts_filter_group.c
    ~ Update to use the new API.
* ts_filter_median.c
    ~ Update to use the new API.
* ts_filter_linear.c
    ~ Remove functions that are no longer needed.

Note:

I might leave some TODOs and FIXMEs with this patch. Most of them
will be removed shortly.

Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>

8 years agoconfig-om3d7k-ohci.patch
Andy Green [Tue, 10 Mar 2009 12:03:50 +0000 (12:03 +0000)]
config-om3d7k-ohci.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoconfig-om3d7k-cpufreq.patch
Andy Green [Tue, 10 Mar 2009 12:03:37 +0000 (12:03 +0000)]
config-om3d7k-cpufreq.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agofix-3d7k-camif-disable-config.patch
Andy Green [Tue, 10 Mar 2009 09:59:22 +0000 (09:59 +0000)]
fix-3d7k-camif-disable-config.patch

This lets us turn off CONFIG_S3C_DEV_CAMIF and suppress it
turning on power (still seemed to issue clock though).

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoMERGE-andy-tracking-patchset-edits
merge [Tue, 10 Mar 2009 08:10:41 +0000 (08:10 +0000)]
MERGE-andy-tracking-patchset-edits

8 years agoFix the start condition of comparing setup mode in lp5521 device.
Matt Hsu [Tue, 10 Mar 2009 08:10:07 +0000 (08:10 +0000)]
Fix the start condition of comparing setup mode in lp5521 device.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agoClean up camera driver
Werner Almesberger [Mon, 9 Mar 2009 21:17:46 +0000 (21:17 +0000)]
Clean up camera driver

This patch does some light cleanup on the camera driver:

- removed unused return values or unused initializations
- remove backslashes at regular line ends (Python-style ?)
- removed redundant casts
- remove redundant parentheses
- wrapped long lines
- general whitespace cleanup
- fixed broken logic in s3c_camif_v4l2_g_input and s3c_camif_v4l2_g_output
- let mutex that really really wants to be a semaphore have its way

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoconfig-gta02-usb-host.patch
Andy Green [Mon, 9 Mar 2009 21:03:02 +0000 (21:03 +0000)]
config-gta02-usb-host.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoAdjust MCLK for use with PLL
Werner Almesberger [Mon, 9 Mar 2009 21:02:36 +0000 (21:02 +0000)]
Adjust MCLK for use with PLL

The PLL converts a 26.6 MHz input clock to 87.8 MHz, which is later
divided by two and becomes PCLK. (See comments below for details.)

All this looks like a textbook example for platform stuff, but let's
take it one step a time.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoFix clock shift in s3c64xx_setrate_clksrc
Werner Almesberger [Mon, 9 Mar 2009 21:02:33 +0000 (21:02 +0000)]
Fix clock shift in s3c64xx_setrate_clksrc

s3c64xx_setrate_clksrc used the clock selection shift sclk->shift instead
of the divider shift sclk->divider_shift, causing clocks to be clobbered.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoFix division in s3c64xx_roundrate_clksrc
Werner Almesberger [Mon, 9 Mar 2009 21:02:30 +0000 (21:02 +0000)]
Fix division in s3c64xx_roundrate_clksrc

s3c64xx_roundrate_clksrc got the rate vs. parent order wrong.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd CAMIF clock to S3C6410
Werner Almesberger [Mon, 9 Mar 2009 21:02:27 +0000 (21:02 +0000)]
Add CAMIF clock to S3C6410

[ Resending it. Seems that commit 9e723603e46500e2c9fbf09ac823de210bf05d31
  used one of the empty mails I got with formail :-( ]

Add camera interface clock to S3C6410.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoClean up camera sensor driver
Werner Almesberger [Mon, 9 Mar 2009 21:02:24 +0000 (21:02 +0000)]
Clean up camera sensor driver

This patch does some light cleanup on the S5K4BA driver:

- change structure initializers from  field:  to  .field =  syntax
- exlicitly name fields in structure initializers
- added proper error handling to sensor_read
- use ARRAY_SIZE
- remove redundant parentheses and braces
- change // to /* ... */
- remove non-ASCII characters
- general whitespace cleanup
- removed an unused variable warning

(4xa_sensor.h is only partially cleaned up.)

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoClean up camera interface driver
Werner Almesberger [Mon, 9 Mar 2009 21:02:21 +0000 (21:02 +0000)]
Clean up camera interface driver

This patch does some cleanup on the camera interface driver:

- remove backslashes at regular line ends (Python-style ?)
- removed unused return values or unused initializations
- combined duplicated or otherwise highly redundant code
- remove redundant parentheses
- wrapped long lines
- general whitespace cleanup
- s3c_camif_input_msdma_codec, s3c_camif_input_msdma_preview,
  s3c_camif_set_target_format: fixed clearing of old register values

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoClean up camera driver
Werner Almesberger [Mon, 9 Mar 2009 21:02:19 +0000 (21:02 +0000)]
Clean up camera driver

This patch does some light cleanup on the camera driver:

- removed unused return values or unused initializations
- remove backslashes at regular line ends (Python-style ?)
- removed redundant casts
- remove redundant parentheses
- wrapped long lines
- general whitespace cleanup
- fixed broken logic in s3c_camif_v4l2_g_input and s3c_camif_v4l2_g_output
- let mutex that really really wants to be a semaphore have its way

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agofix-s3c-cpu-detect-make-compatible-atag16.patch
Andy Green [Mon, 9 Mar 2009 21:02:16 +0000 (21:02 +0000)]
fix-s3c-cpu-detect-make-compatible-atag16.patch

The current version of CPU detection patch from Mark Brown and
Kyungmin Park takes a dump on system_rev.  This patch changes it
to retain the low 16 of system_rev from the bootloader ATAG, and
adds access #defines and changes to use them.

No doubt they'll issue a better version of the CPU detection
patch and then this can be reverted.

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoS3C64XX: Add DVFS support to the S3C64XX cpufreq driver
Mark Brown [Mon, 9 Mar 2009 21:02:13 +0000 (21:02 +0000)]
S3C64XX: Add DVFS support to the S3C64XX cpufreq driver

Use the regulator framework to provide optional support for DVFS in
the S3C64XX cpufreq driver. When a software controllable regulator
is configured the driver will use it to lower the supply voltage when
running at a lower frequency, giving improved power saving.

When regulator support is disabled or no regulator can be obtained
for VDDARM the driver will fall back to scaling only the frequency.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

8 years agoS3C64XX: Initial support for CPU frequency scaling
Mark Brown [Mon, 9 Mar 2009 21:02:09 +0000 (21:02 +0000)]
S3C64XX: Initial support for CPU frequency scaling

This patch provides initial support for CPU frequency scaling on the
Samsung S3C64XX series processors. Currently only S3C6410 processors
are supported, though addition of another data table with supported
clock rates should be sufficient to enable support for further CPUs.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

8 years agoS3C64XX: Add initial support for ARMCLK
Mark Brown [Mon, 9 Mar 2009 21:02:06 +0000 (21:02 +0000)]
S3C64XX: Add initial support for ARMCLK

Add support for reconfiguring the clock for the ARM core, enabling
CPUfreq support. Currently only the divider for ARMCLK may be changed,
ARMPLL is left static.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

8 years agoS3C: CPU detection support
Kyungmin Park [Mon, 9 Mar 2009 21:01:41 +0000 (21:01 +0000)]
S3C: CPU detection support

Stroe the CPU ID to system_rev and use it to detect the CPU architecture.

Even though s3c64xx has almost same IPs, some IPs such as OneNAND are different.

Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

8 years agoclean-ohci-warning.patch
Andy Green [Mon, 9 Mar 2009 21:01:30 +0000 (21:01 +0000)]
clean-ohci-warning.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agofix-s3c-ohci-move-usb-control.h
Andy Green [Mon, 9 Mar 2009 21:01:27 +0000 (21:01 +0000)]
fix-s3c-ohci-move-usb-control.h

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoSubject: om3d7k: Add support for s3c_ohci device.
Balaji Rao [Mon, 9 Mar 2009 21:01:21 +0000 (21:01 +0000)]
Subject: om3d7k: Add support for s3c_ohci device.
X-Git-Url: http://git.openmoko.org/?p=kernel.git;a=commitdiff_plain;h=d9235e3a8bdcef02865eed27c0d6012d253f11e7

om3d7k: Add support for s3c_ohci device.

Adds s3c_ohci support to om3d7k.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>

8 years agoSubject: USB: Change s3c2410_ohci into s3c_ohci and change gta02 to use it
Balaji Rao [Mon, 9 Mar 2009 21:01:18 +0000 (21:01 +0000)]
Subject: USB: Change s3c2410_ohci into s3c_ohci and change gta02 to use it
X-Git-Url: http://git.openmoko.org/?p=kernel.git;a=commitdiff_plain;h=b55b8e56a4a7e43b7243be48f77a326236a37c68

USB: Change s3c2410_ohci into s3c_ohci and change gta02 to use it

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>

8 years agoGTA02: Name regulator supplies
Mark Brown [Mon, 9 Mar 2009 21:01:15 +0000 (21:01 +0000)]
GTA02: Name regulator supplies

Give the regulator supplies names corresponding to the names their
supplies are given in the schematic, making it easier to tie the
software up with the schematic.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

8 years agoASoC: Fix headset output for GTA02
Mark Brown [Mon, 9 Mar 2009 21:01:12 +0000 (21:01 +0000)]
ASoC: Fix headset output for GTA02

At least one of the speaker options should disable the speaker.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

8 years agoASoC: Remove unneeded forward reference to WM8753 SPI implementation
Mark Brown [Mon, 9 Mar 2009 21:01:09 +0000 (21:01 +0000)]
ASoC: Remove unneeded forward reference to WM8753 SPI implementation

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

8 years agoASoC: Don't directly reference DAI ops in GTA02 driver
Mark Brown [Mon, 9 Mar 2009 21:01:06 +0000 (21:01 +0000)]
ASoC: Don't directly reference DAI ops in GTA02 driver

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

8 years ago(AG: there's some more rationale for changing this here
Neil Brown [Mon, 9 Mar 2009 21:01:01 +0000 (21:01 +0000)]
(AG: there's some more rationale for changing this here

http://lists.openmoko.org/pipermail/openmoko-kernel/2009-March/009387.html
)

Change accelerometers to use ABS events rather than REL events.

[Obviously if this patch is accepted we need to tell developers about
 it.  I have a number of other improvements to the accelerometers I
 hope to deliver over the next couple of weeks.  They will have minimal
 or zero disruption to current code. ]

REL events should be used when there is no absolute reference, and
only changes are meaningful.  The classic example is a "mouse" where
the absolute position of the device is not measurable and not
particularly meaning, but change in position from one time to the next
is interesting.
With REL events, a value of '0' is not reported, as 'not change' is
not interesting.
With REL events, the expectation is that successive values will be
eventually summed (possibly with acceleration and clipping
adjustments) to get a usable value.

ABS events should be used when there is an absolute references against
which things that be measured.
With ABS events, the 'current value' is meaningful and can be read
(EVIOCGABS).
With ABS events, the value '0' is very meaningful and is reported.
However if consecutive values are the same, the value is only reported
once.
ABS events can be used as-is or compared with previous events to get
some measure of change.
An obvious example is a touchscreen where each measure in
independently meaningful.

Acceleration is an absolute value as it is measuring against a frame
of reference.  '0' acceleration is just as meaningful as any other
value, and finding the 'current' acceleration is each direction is a
potentially useful thing to do.

The Freerunner accelerometers currently report REL events.  This is
wrong.  So this patch changes them to report ABS events.

With this patch, the min/max/level/fuzz values are left at zero.  It
might be useful to make use of these in a subsequent patch.
min/max/level can be used to calibrate the accelerometers if accuracy
is important.
fuzz could possibly be used in conjunction with the 'threshold' sysfs
value to get less frequent, lower-precision reports.

This may well break some applications that read accelerometer data.
This cannot be helped, but it is quite easy to write code that copes
with the incorrect EV_REL events as well as the more correct and
useful EV_ABS events.

Signed-off-by: NeilBrown <neilb@suse.de>

8 years agofix-gta02-fiq-init.patch
Andy Green [Sun, 8 Mar 2009 09:14:13 +0000 (09:14 +0000)]
fix-gta02-fiq-init.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agofix-gta02-fiq-timing.patch
Andy Green [Sun, 8 Mar 2009 08:54:25 +0000 (08:54 +0000)]
fix-gta02-fiq-timing.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoIncrease camera IO drive strength from 2 mA to 8 mA
Werner Almesberger [Sat, 7 Mar 2009 16:08:17 +0000 (16:08 +0000)]
Increase camera IO drive strength from 2 mA to 8 mA

The camera driver reduced all IO drivers from their reset defaults of
6 mA to only 2 mA, which caused severe signal distortion at higher
speeds. This patch sets them to 8 mA and also removes apparently
useless repetitions of the setting.

Note that the correct setting of the I2C pullups still needs to be
verified.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoEnable jbt6k74 LCM driver IC device for 3d7k.
Matt Hsu [Sat, 7 Mar 2009 07:38:41 +0000 (07:38 +0000)]
Enable jbt6k74 LCM driver IC device for 3d7k.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agoAdd concurrent solution of supporting l1k002 and jbt6k74 (LCM ASIC device) for 3d7k.
Matt Hsu [Sat, 7 Mar 2009 07:38:38 +0000 (07:38 +0000)]
Add concurrent solution of supporting l1k002 and jbt6k74 (LCM ASIC device) for 3d7k.

By parsing kernel command, 3d7k can probe two different LCM driver accordingly.
The default attached LCM device is l1k002, if you wanna to use jbt6k74, you don't
need to re-compile the kernel. Just add one option om_3d7k_lcm=jbt6k74 in
boot/append-OM_3D7K in rootfs.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agoAdd LED channel information in lp5521 platform data for 3d7k.
Matt Hsu [Fri, 6 Mar 2009 12:28:36 +0000 (12:28 +0000)]
Add LED channel information in lp5521 platform data for 3d7k.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agoRe-write lp5521 LED driver.
Matt Hsu [Fri, 6 Mar 2009 12:28:25 +0000 (12:28 +0000)]
Re-write lp5521 LED driver.

- provide clean sysfs to control RGB channel directly.
  They are looked like the following:

channel_mode
   channel_pwm
channel_cur

- add platform data to export RGB channel accordingly.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agoCorrect the start-up delay of LED(lp5521) device on 3d7k.
Matt Hsu [Fri, 6 Mar 2009 12:28:23 +0000 (12:28 +0000)]
Correct the start-up delay of LED(lp5521) device on 3d7k.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agoTemporarily mark INT of LED device on 3d7k.
Matt Hsu [Fri, 6 Mar 2009 12:28:20 +0000 (12:28 +0000)]
Temporarily mark INT of LED device on 3d7k.

LED INT is connected to EXT group6_9, the handling of EXT group1~group9
is not yet implemented. Besides, we don't need this IRQ now.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agoFrom b003024f41b34cc9a8442f7dab1b9ce099faf08b Mon Sep 17 00:00:00 2001
Tim Niemeyer [Thu, 5 Mar 2009 14:40:35 +0000 (14:40 +0000)]
From b003024f41b34cc9a8442f7dab1b9ce099faf08b Mon Sep 17 00:00:00 2001
Subject: [PATCH] fix-alsa-bt-dai-registration-missing

8 years agoFrom 52da4d313dd3263824023f6f3c16a32a0078710e Mon Sep 17 00:00:00 2001
Tim Niemeyer [Thu, 5 Mar 2009 14:40:33 +0000 (14:40 +0000)]
From 52da4d313dd3263824023f6f3c16a32a0078710e Mon Sep 17 00:00:00 2001
Subject: [PATCH] Typo in I2C device registration

Signed-off-by: Tim Niemeyer <reddog@redlap.wired.rednet.mastersword.de>

8 years agoAdd Samsung S3C camera interface registers
Werner Almesberger [Thu, 5 Mar 2009 14:40:30 +0000 (14:40 +0000)]
Add Samsung S3C camera interface registers

These are the Samsung S3C camera interface register definitions directly
from the 2.6.21 BSP. The only changes to the original code are the removal
of trailing whitespace and the change of location.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoDirty hacks to make it all work
Werner Almesberger [Thu, 5 Mar 2009 14:40:27 +0000 (14:40 +0000)]
Dirty hacks to make it all work

A few dirty hacks to make the camera driver work:

- because V4L no longer guarantees that minor numbers provided by the
  client are actually used, the preview/codec selection mechanism falls
  apart. We work around this by defaulting to preview when we don't
  know better.

- power up the camera in platform code, not cleanly via power control
  device.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoEnable camera in 3D7K defconfig
Werner Almesberger [Thu, 5 Mar 2009 14:40:24 +0000 (14:40 +0000)]
Enable camera in 3D7K defconfig

Enable V4L and camera driver in 3D7K default configuration.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd camera interface driver to 3D7K
Werner Almesberger [Thu, 5 Mar 2009 14:40:08 +0000 (14:40 +0000)]
Add camera interface driver to 3D7K

This patch adds the camera interface driver and the corresponding
configuration option to the 3D7K machine.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoSamsung S3C camera subsystem config and Makefile
Werner Almesberger [Thu, 5 Mar 2009 14:39:57 +0000 (14:39 +0000)]
Samsung S3C camera subsystem config and Makefile

Add Samsung S3C camera subsystem to kernel configuration and build process.
Original code is from Samsung's BSP.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoFix Samsung S5K4BA camera driver.
Werner Almesberger [Thu, 5 Mar 2009 14:39:54 +0000 (14:39 +0000)]
Fix Samsung S5K4BA camera driver.

This patch makes the Samsung S5K4BA driver work in 2.6.29 and also solves
a few minor issues, such as trailing whitespace.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd Samsung S5K4BA camera driver
Werner Almesberger [Thu, 5 Mar 2009 14:39:51 +0000 (14:39 +0000)]
Add Samsung S5K4BA camera driver

This is the original Samsung S5K4BA camera driver code from the 2.6.21
BSP.  The changes that are needed to make this work in 2.6.29 are in
the next patch.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd I2C IDs for Samsung S5K series cameras
Werner Almesberger [Thu, 5 Mar 2009 14:39:48 +0000 (14:39 +0000)]
Add I2C IDs for Samsung S5K series cameras

These are te I2C IDs for all the Samsung S5K series cameras. This code
is directly from Samsung's BSP.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd Samsung S3C CAMIF driver
Werner Almesberger [Thu, 5 Mar 2009 14:39:46 +0000 (14:39 +0000)]
Add Samsung S3C CAMIF driver

Add Samsung S3C camera interface driver.
Original code is from Samsung's BSP and was written for 2.6.21.

Only tested on S3C6410.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoUpdate Samsung S3C camera interface driver
Werner Almesberger [Thu, 5 Mar 2009 14:39:43 +0000 (14:39 +0000)]
Update Samsung S3C camera interface driver

Update the camera interface driver from 2.6.21 to 2.6.29 and fix some
trivial issues. There are still a few ugly spots, marked with "@@@".

Only tested on S3C6410.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd Samsung S3C camera driver for V4L
Werner Almesberger [Thu, 5 Mar 2009 14:39:41 +0000 (14:39 +0000)]
Add Samsung S3C camera driver for V4L

This is the original Samsung S3C camera driver code from the 2.6.21 BSP.
The changes that are needed to make this work in 2.6.29 are in the next
patch.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd CAMIF clock to S3C6410.
Werner Almesberger [Thu, 5 Mar 2009 14:39:38 +0000 (14:39 +0000)]
Add CAMIF clock to S3C6410.

Add camera interface clock to S3C6410.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoAdd HCLKx2 to S3C64xx.
Werner Almesberger [Thu, 5 Mar 2009 14:39:33 +0000 (14:39 +0000)]
Add HCLKx2 to S3C64xx.

Add doubled HCLK to S3C64xx.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoFix s3c64xx_setrate_clksrc
Werner Almesberger [Thu, 5 Mar 2009 14:39:30 +0000 (14:39 +0000)]
Fix s3c64xx_setrate_clksrc

Note: this patch is already on the way upstream but is currently missing
in the Openmoko kernel.

Some of the rate selection logic in s3c64xx_setrate_clksrc uses what
appears to be parent clock selection logic. This patch corrects it.

I also added a check for overly large dividers to prevent them
from changing unrelated clocks.

Signed-off-by: Werner Almesberger <werner@openmoko.org>

8 years agoconfig-3d7k-remove-mac80211.patch
Andy Green [Wed, 4 Mar 2009 17:12:27 +0000 (17:12 +0000)]
config-3d7k-remove-mac80211.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoclean-3d7K-3d7k.patch
Andy Green [Wed, 4 Mar 2009 15:05:46 +0000 (15:05 +0000)]
clean-3d7K-3d7k.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agofix-3d7k-misaligned-iis.patch
Andy Green [Wed, 4 Mar 2009 15:05:43 +0000 (15:05 +0000)]
fix-3d7k-misaligned-iis.patch

This one is specially for Candy Chou :-)

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoconfig-3d7k-correct-alsa-driver-name.patch
Andy Green [Wed, 4 Mar 2009 15:05:41 +0000 (15:05 +0000)]
config-3d7k-correct-alsa-driver-name.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agofix-3d7k-iis-constants.patch
Andy Green [Wed, 4 Mar 2009 15:05:38 +0000 (15:05 +0000)]
fix-3d7k-iis-constants.patch

Signed-off-by: Andy Green <andy@openmoko.com>

8 years agoMERGE-andy-tracking-patchset-edits
merge [Wed, 4 Mar 2009 12:53:41 +0000 (12:53 +0000)]
MERGE-andy-tracking-patchset-edits

8 years agoIntroduce generic backlight device on 3d7k.
Matt Hsu [Wed, 4 Mar 2009 08:11:21 +0000 (08:11 +0000)]
Introduce generic backlight device on 3d7k.
The code is borrowed from gta02.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>

8 years agofix-3d7k-remove-gps-wlan-pm.patch
Andy Green [Wed, 4 Mar 2009 07:38:13 +0000 (07:38 +0000)]
fix-3d7k-remove-gps-wlan-pm.patch

Signed-off-by: Andy Green <andy@openmoko.com>