kernel.git
11 years agoASoC: Tone down debug for PCM enqueue
Mark Brown [Tue, 3 Mar 2009 11:29:00 +0000 (11:29 +0000)]
ASoC: Tone down debug for PCM enqueue

It's a little verbose...

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Use SND_SOC_DAPM_PIN_SWITCH for GTA02
Mark Brown [Tue, 3 Mar 2009 11:28:57 +0000 (11:28 +0000)]
ASoC: Use SND_SOC_DAPM_PIN_SWITCH for GTA02

Factors out a lot of code.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Add Openmoko state file compatibility for DAPM switches
Mark Brown [Tue, 3 Mar 2009 11:28:34 +0000 (11:28 +0000)]
ASoC: Add Openmoko state file compatibility for DAPM switches

Openmoko had similar controls with "DAPM " prefixed to them - add that
back to maintain compatibility with existing ALSA state files.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Update GTA02 for current DAPM pin APIs
Mark Brown [Tue, 3 Mar 2009 11:28:31 +0000 (11:28 +0000)]
ASoC: Update GTA02 for current DAPM pin APIs

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Use snd_soc_dapm_nc_pin() in GTA02 driver
Mark Brown [Tue, 3 Mar 2009 11:28:28 +0000 (11:28 +0000)]
ASoC: Use snd_soc_dapm_nc_pin() in GTA02 driver

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Remove #if 0 code from GTA02
Mark Brown [Tue, 3 Mar 2009 11:28:26 +0000 (11:28 +0000)]
ASoC: Remove #if 0 code from GTA02

More code cleanup.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Remove some unused headers from the GTA02 audio driver
Mark Brown [Tue, 3 Mar 2009 11:28:22 +0000 (11:28 +0000)]
ASoC: Remove some unused headers from the GTA02 audio driver

Stops anyone wondering what they're needed for.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Strip in code changelog from GTA02 audio driver
Mark Brown [Tue, 3 Mar 2009 11:28:20 +0000 (11:28 +0000)]
ASoC: Strip in code changelog from GTA02 audio driver

It's patchy and we have git history.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Add SND_SOC_DAPM_PIN_SWITCH controls for exposing DAPM pins
Mark Brown [Tue, 3 Mar 2009 11:28:17 +0000 (11:28 +0000)]
ASoC: Add SND_SOC_DAPM_PIN_SWITCH controls for exposing DAPM pins

On some systems it is desirable for control for DAPM pins to be provided
to user space. This is the case with things like GSM modems which are
controlled primarily from user space, for example. Provide a helper which
exposes the state of a DAPM pin to user space for use in cases like this.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Constify pin names for DAPM pin status APIs
Mark Brown [Tue, 3 Mar 2009 11:28:15 +0000 (11:28 +0000)]
ASoC: Constify pin names for DAPM pin status APIs

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoSubject: pcf50633_charger_detection_fix.patch
Balaji Rao [Tue, 3 Mar 2009 11:28:12 +0000 (11:28 +0000)]
Subject: pcf50633_charger_detection_fix.patch
X-Git-Url: http://git.openmoko.org/?p=kernel.git;a=commitdiff_plain;h=f4fd6631973dbc67a5f4708014e81ca6e610ca05

pcf50633_charger_detection_fix.patch

Fix PCF50633 charger detection code.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agoRe: About roothub losing power / SPI bitbang 3D7K blowout
Balaji Rao [Tue, 3 Mar 2009 11:28:08 +0000 (11:28 +0000)]
Re: About roothub losing power / SPI bitbang 3D7K blowout

> |> on current andy-tracking / OM_3D7K build.
> |>
> |> I didn't see this before, it seems to related to SPI bitbang, did we
> |> break something?

Here is the patch. I've not tested it, but it's most likely the fix, as
I had seen a similar thing in spi_s3c24xx_gpio.

11 years agokernel.git/build: use "git branch" instead of "git-branch"?
Timo Juhani Lindfors [Tue, 3 Mar 2009 07:10:19 +0000 (07:10 +0000)]
kernel.git/build: use "git branch" instead of "git-branch"?

Hi,

git 1.6.0 in debian unstable does not put "git-branch" to a directory
that is in PATH anymore

> git-core (1:1.6.0.2-1) experimental; urgency=low
>
>   * new upstream point release.
>   * debian/git-core.NEWS.Debian: new; mention that most of the programs
>     now are installed outside the default $PATH; the "git-xyzzy" form
>     in scripts and on the command line is no longer supported in 1.6.0.

so the build script should probably be modified:

11 years agortc-pcf50633: Fix month off-by-one error
Rask Ingemann Lambertsen [Tue, 3 Mar 2009 06:40:34 +0000 (06:40 +0000)]
rtc-pcf50633: Fix month off-by-one error

   The PCF50633 stores a month value of 1-12, but the kernel wants 0-11.

Signed-off-by: Rask Ingemann Lambertsen <rask@sygehus.dk>
11 years agorename-gta03-dropped-3d7k-new-moniker.patch
Andy Green [Mon, 2 Mar 2009 18:51:56 +0000 (18:51 +0000)]
rename-gta03-dropped-3d7k-new-moniker.patch

GTA03 project is dropped, long live Openmoko 3d7k project.

This name was literally generated by picking the first character
of the next four car license plates to pass by I am told.

The should catch all GTA03 instances except the machine name.

I registered a new machine name since the 3d7k is different
from original GTA03, the machine number is 2120.

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoMERGE-via-pending-tracking-hist-add-om_3d7k_machine-type-patch
merge [Mon, 2 Mar 2009 16:23:51 +0000 (16:23 +0000)]
MERGE-via-pending-tracking-hist-add-om_3d7k_machine-type-patch
pending-tracking-hist top was add-om_3d7k_machine-type-patch / e4e3848be23080f6ca6727d2a4353f09e53b9781 ... parent commitmessage:
From: Andy Green <andy@openmoko.com>
add-om_3d7k_machine-type.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoFix typo in mach-gta02
Sven Rebhan [Mon, 2 Mar 2009 11:25:16 +0000 (11:25 +0000)]
Fix typo in mach-gta02

Improve the comment on falling back to 100mA if ADC is missing.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoGTA02 GSM requires PCF50633 GPIO
Sven Rebhan [Mon, 2 Mar 2009 11:25:13 +0000 (11:25 +0000)]
GTA02 GSM requires PCF50633 GPIO

The GSM part of all MACH_NEO1973s require GPIO functions,
thus we should select PCF50633_GPIO for GTA02.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoOpenMoko: Update for normal WM8753 device registration.
Mark Brown [Mon, 2 Mar 2009 11:22:03 +0000 (11:22 +0000)]
OpenMoko: Update for normal WM8753 device registration.

The WM8753 driver now uses the standard I2C device registration model and
so requires the platform to register the device. Convert the OpenMoko
GTA0x series devices to do so.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Convert WM8753 to register via normal device probe
Mark Brown [Mon, 2 Mar 2009 11:22:00 +0000 (11:22 +0000)]
ASoC: Convert WM8753 to register via normal device probe

The base support for the only in-tree user, the GTA01, is out of tree
and will be updated separately.

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoASoC: Shuffle WM8753 device registration code
Mark Brown [Mon, 2 Mar 2009 11:21:58 +0000 (11:21 +0000)]
ASoC: Shuffle WM8753 device registration code

This patch should be pure code motion, separating that out from the
functional changes to move to new style device registration.

[Backported from upstream with some modifications for old code.]

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoMERGE-via-pending-tracking-hist-asoc-only-write-back-non-defau
merge [Mon, 2 Mar 2009 11:15:46 +0000 (11:15 +0000)]
MERGE-via-pending-tracking-hist-asoc-only-write-back-non-defau
pending-tracking-hist top was asoc-only-write-back-non-defau / 04f55b6c2392d4af2b8b2178c333c7f131d3cdc0 ... parent commitmessage:
From: Mark Brown <broonie@opensource.wolfsonmicro.com>
ASoC: Only write back non-default registers when resuming WM8753

This will reduce the number of writes done on resume, allowing that to
complete faster (especially on systems with very slow I2C like the
current Samsung driver).

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agoARM/S3C2440: Don't build armv4t modules we can't load
Rask Ingemann Lambertsen [Mon, 2 Mar 2009 10:46:01 +0000 (10:46 +0000)]
ARM/S3C2440: Don't build armv4t modules we can't load

   The kernel can't load armv4t modules ("unknown relocation: 40"). This
patch fixes the ARM makefile accordingly to tell GCC to generate armv4 code
instead of armv4t code. The assembler still needs to allow armv4t
instructions for some object files which are linked into the kernel.

Signed-off-by: Rask Ingemann Lambertsen <rask@sygehus.dk>
11 years agoFixed corruption of LCM registers
Nicolas Dufresne [Mon, 2 Mar 2009 10:45:59 +0000 (10:45 +0000)]
Fixed corruption of LCM registers

Some registers where not set properly, or at the right time (thanks to Balaji
for his patch).

Weak locking could lead to corruption when using sysfs to switch state from
multiple threads or processes. The state transitions are now all atomic.

The driver attribute 'last_state' was used for same purpose of 'normal_state'.
Kept only 'normal_state' and used it in 'init_regs' instead of custom qvga
parameter.

This patch should fix bug #2235.

Signed-off-by: Nicolas Dufresne <nicolas.dufresne@gmail.com>
11 years agoSubject: Get rid of board_info information in glamo and s3c24xx_gpio spigpio_info.
Balaji Rao [Mon, 2 Mar 2009 10:45:56 +0000 (10:45 +0000)]
Subject: Get rid of board_info information in glamo and s3c24xx_gpio spigpio_info.
X-Git-Url: http://git.openmoko.org/?p=kernel.git;a=commitdiff_plain;h=740c6704e830829d8539a6cc34346ff1980cd9ee

Get rid of board_info information in glamo and s3c24xx_gpio spigpio_info.

The board info does not belong there and has been removed. In
spi_s3c24xx_gpio, board_info has been removed in mainline.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agoAUDIO: Select DMA if I2S is configured
Ben Dooks [Thu, 26 Feb 2009 21:56:29 +0000 (21:56 +0000)]
AUDIO: Select DMA if I2S is configured

Select the relevant DMA implementation when the
sound driver is selected.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
--
Ben (ben@fluff.org, http://www.fluff.org/)

  'a smiley only costs 4 bytes'

11 years agoS3C64XX: Fix DMA channel numbers
Ben Dooks [Thu, 26 Feb 2009 21:56:25 +0000 (21:56 +0000)]
S3C64XX: Fix DMA channel numbers

The I2S TX and RX channels where reversed. Also add
implicit numbers to the start of the two DMA controllers.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoAUDIO: Fixup s3c-i2s-v2 merge problems
Ben Dooks [Thu, 26 Feb 2009 21:56:21 +0000 (21:56 +0000)]
AUDIO: Fixup s3c-i2s-v2 merge problems

Fix CONFIG_CPU_S3C64XX which should have been CONFIG_PLAT_S3C64XX.

Fix the merge problems that failed to properly set the master/slave
bits for S3C64XX.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoAUDIO: GTA03 make get/set calls use common code
Ben Dooks [Thu, 26 Feb 2009 21:56:05 +0000 (21:56 +0000)]
AUDIO: GTA03 make get/set calls use common code

The GTA03 driver has a number of get and set calls that
can be easily be done by calling a pair of common helper
functions.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>Index: linux.agit/sound/soc/s3c24xx/om_gta03_wm8753.c
===================================================================
--- linux.agit.orig/sound/soc/s3c24xx/om_gta03_wm8753.c 2009-02-26 17:58:18.000000000 +0000
+++ linux.agit/sound/soc/s3c24xx/om_gta03_wm8753.c 2009-02-26 17:59:46.000000000 +0000
@@ -228,8 +228,9 @@ static struct snd_soc_ops om_gta03_voice
  .hw_free = om_gta03_voice_hw_free,
 };

-static int om_gta03_set_stereo_out(struct snd_kcontrol *kcontrol,
-    struct snd_ctl_elem_value *ucontrol)
+static int common_set_dapm(struct snd_kcontrol *kcontrol,
+    struct snd_ctl_elem_value *ucontrol,
+    char *epname)
 {
  struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
  int val = ucontrol->value.integer.value[0];
@@ -240,133 +241,89 @@ static int om_gta03_set_stereo_out(struc
  return 0;
 }

-static int om_gta03_get_stereo_out(struct snd_kcontrol *kcontrol,
-    struct snd_ctl_elem_value *ucontrol)
+static int common_get_dapm(struct snd_kcontrol *kcontrol,
+    struct snd_ctl_elem_value *ucontrol,
+    char *name)
 {
  struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
+ int val;

- ucontrol->value.integer.value[0] =
- snd_soc_dapm_get_endpoint(codec, "Stereo Out");
+ val = snd_soc_dapm_get_endpoint(codec, name);
+ ucontrol->value.integer.value[0] = val;

  return 0;
 }

+static int om_gta03_set_stereo_out(struct snd_kcontrol *kcontrol,
+    struct snd_ctl_elem_value *ucontrol)
+{
+ return common_set_dapm(kcontrol, ucontrol, "Stereo Out");
+}
+
+static int om_gta03_get_stereo_out(struct snd_kcontrol *kcontrol,
+    struct snd_ctl_elem_value *ucontrol)
+{
+ return common_get_dapm(kcontrol, ucontrol, "Stereo Out");
+}

 static int om_gta03_set_gsm_out(struct snd_kcontrol *kcontrol,
  struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
- int val = ucontrol->value.integer.value[0];
-
- snd_soc_dapm_set_endpoint(codec, "GSM Line Out", val);
- snd_soc_dapm_sync(codec);
-
- return 0;
+ return common_set_dapm(kcontrol, ucontrol, "GSM Line Out");
 }

 static int om_gta03_get_gsm_out(struct snd_kcontrol *kcontrol,
  struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
-
- ucontrol->value.integer.value[0] =
- snd_soc_dapm_get_endpoint(codec, "GSM Line Out");
-
- return 0;
+ return common_get_dapm(kcontrol, ucontrol, "GSM Line Out");
 }

 static int om_gta03_set_gsm_in(struct snd_kcontrol *kcontrol,
         struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
- int val = ucontrol->value.integer.value[0];
-
- snd_soc_dapm_set_endpoint(codec, "GSM Line In", val);
- snd_soc_dapm_sync(codec);
-
- return 0;
+ return common_set_dapm(kcontrol, ucontrol, "GSM Line In");
 }

 static int om_gta03_get_gsm_in(struct snd_kcontrol *kcontrol,
         struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
-
- ucontrol->value.integer.value[0] =
- snd_soc_dapm_get_endpoint(codec, "GSM Line In");
-
- return 0;
+ return common_get_dapm(kcontrol, ucontrol, "GSM Line In");
 }

 static int om_gta03_set_headset_mic(struct snd_kcontrol *kcontrol,
      struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
- int val = ucontrol->value.integer.value[0];
-
- snd_soc_dapm_set_endpoint(codec, "Headset Mic", val);
- snd_soc_dapm_sync(codec);
-
- return 0;
+ return common_set_dapm(kcontrol, ucontrol, "Headset Mic");
 }

 static int om_gta03_get_headset_mic(struct snd_kcontrol *kcontrol,
      struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
-
- ucontrol->value.integer.value[0] =
- snd_soc_dapm_get_endpoint(codec, "Headset Mic");
-
- return 0;
+ return common_get_dapm(kcontrol, ucontrol, "Headset Mic");
 }

 static int om_gta03_set_handset_mic(struct snd_kcontrol *kcontrol,
      struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
- int val = ucontrol->value.integer.value[0];
-
- snd_soc_dapm_set_endpoint(codec, "Handset Mic", val);
-
- snd_soc_dapm_sync(codec);
-
- return 0;
+ return common_set_dapm(kcontrol, ucontrol, "Handset Mic");
 }

 static int om_gta03_get_handset_mic(struct snd_kcontrol *kcontrol,
      struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
-
- ucontrol->value.integer.value[0] =
- snd_soc_dapm_get_endpoint(codec, "Handset Mic");
-
- return 0;
+ return common_get_dapm(kcontrol, ucontrol, "Handset Mic");
 }

 static int om_gta03_set_handset_spk(struct snd_kcontrol *kcontrol,
      struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
- int val = ucontrol->value.integer.value[0];
-
- snd_soc_dapm_set_endpoint(codec, "Handset Spk", val);
-
- snd_soc_dapm_sync(codec);
-
- return 0;
+ return common_set_dapm(kcontrol, ucontrol, "Handset Spk");
 }

 static int om_gta03_get_handset_spk(struct snd_kcontrol *kcontrol,
      struct snd_ctl_elem_value *ucontrol)
 {
- struct snd_soc_codec *codec = snd_kcontrol_chip(kcontrol);
-
- ucontrol->value.integer.value[0] =
- snd_soc_dapm_get_endpoint(codec, "Handset Spk");
-
- return 0;
+ return common_get_dapm(kcontrol, ucontrol, "Handset Spk");
 }

 static const struct snd_soc_dapm_widget wm8753_dapm_widgets[] = {

--
Ben (ben@fluff.org, http://www.fluff.org/)

  'a smiley only costs 4 bytes'

11 years agoAUDIO: Update GTA03 setup error path
Ben Dooks [Thu, 26 Feb 2009 21:56:02 +0000 (21:56 +0000)]
AUDIO: Update GTA03 setup error path

Add common error exit for om_gta03_hifi_hw_params()
so that any error can easily be debugged.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoAUDIO: Tidy indentation in GTA03 driver
Ben Dooks [Thu, 26 Feb 2009 21:56:00 +0000 (21:56 +0000)]
AUDIO: Tidy indentation in GTA03 driver

Tidy the driver formatting.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoS3C64XX: Fix DMA makefile
Ben Dooks [Thu, 26 Feb 2009 21:55:57 +0000 (21:55 +0000)]
S3C64XX: Fix DMA makefile

Fix missing dma.o build if CONFIG_S3C_DMA

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoAUDIO: Fix GTA03 clock and channel setup.
Ben Dooks [Thu, 26 Feb 2009 21:55:54 +0000 (21:55 +0000)]
AUDIO: Fix GTA03 clock and channel setup.

The I2S codec is connected to I2S channel 1, not 0
as initially setup.

Also fix the divisors to produce 12MHz on MCLK (was
6MHz) to ensure that the PLL in the WM8753 is within
proper limits.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoFix the current leaking issue on gta03 a3 or later version.
Matt Hsu [Thu, 26 Feb 2009 12:53:20 +0000 (12:53 +0000)]
Fix the current leaking issue on gta03 a3 or later version.

Since GPI0, GPI1 and GPI8 are for hardware version control.
They would create current leaking if they are setup as output.
It's resloved by adding gta03 specific fb_gpio_setup function
in fb platform data.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>
Reported-by: Dkay Chen <dkay_chen@openmoko.com>
11 years agofix-hdq-increase-break-timing.patch
Andy Green [Thu, 26 Feb 2009 05:23:05 +0000 (05:23 +0000)]
fix-hdq-increase-break-timing.patch

This removes the unstable operation by increasing break timing from
170us to 230us, the datasheet minimum was 190us.

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoSubject: fix_gta03_fiq_stuff.patch
Balaji Rao [Thu, 26 Feb 2009 04:10:55 +0000 (04:10 +0000)]
Subject: fix_gta03_fiq_stuff.patch
X-Git-Url: http://git.openmoko.org/?p=kernel.git;a=commitdiff_plain;h=c94ea3d685fa6e9b24d62adb11a7ad6087b9edf5

fix_gta03_fiq_stuff.patch

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agoSubject: lis302dl_use_s3c24xx_gpio_spi_bitbang.patch
Balaji Rao [Thu, 26 Feb 2009 04:10:52 +0000 (04:10 +0000)]
Subject: lis302dl_use_s3c24xx_gpio_spi_bitbang.patch
X-Git-Url: http://git.openmoko.org/?p=kernel.git;a=commitdiff_plain;h=5689ad9b0ded29adec4be64df53e488c648b9831

lis302dl_use_s3c24xx_gpio_spi_bitbang.patch

Change lis302dl driver to use generic spi code. This requires that we define a
s3c24xx_gpio spi bitbang based controller in mach-gta02.c.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agoSubject: spi_supoport_non_blocking_sync_transfers.patch
Balaji Rao [Thu, 26 Feb 2009 04:10:49 +0000 (04:10 +0000)]
Subject: spi_supoport_non_blocking_sync_transfers.patch
X-Git-Url: http://git.openmoko.org/?p=kernel.git;a=commitdiff_plain;h=f8fbb07854e3aff64dce9fe6ef6a8dc0e0f762b5

spi_supoport_non_blocking_sync_transfers.patch

A mew option is added to spi_bitbang_info to specify if the
transfers off it will be non blocking.

A new function - spi_non_blocking_transfer is added to the SPI core.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agoFix jbt6k74 qvga_normal state handling
Balaji Rao [Wed, 25 Feb 2009 13:53:03 +0000 (13:53 +0000)]
Fix jbt6k74 qvga_normal state handling

Hi Andy,

Here's the fix for the qvga-normal bug we observed the other day
here - when the screen is powered up again after a blanking, in
qvga-normal mode, we see a WSOD(kind of!)

11 years agoadd-s3c64xx-tzic-plat.patch
Andy Green [Wed, 25 Feb 2009 11:11:10 +0000 (11:11 +0000)]
add-s3c64xx-tzic-plat.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoadd-s3c64xx-tzic.patch
Andy Green [Wed, 25 Feb 2009 11:05:19 +0000 (11:05 +0000)]
add-s3c64xx-tzic.patch

This adds in the TZIC support consts and also the VM mapping
for the TZIC units, and adds TZIC init into mach-om-gta03.c

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoclean-neo1973-pm-gps.patch
Andy Green [Wed, 25 Feb 2009 04:30:10 +0000 (04:30 +0000)]
clean-neo1973-pm-gps.patch

Clean out more switch() on version that has no point given the
actual features that (don't) differ on supported GTA02 versions.

We take care about forcing GPS regulator off on neo1973-pm-gps probe,
but we don't take care about killing the UART GPIO so they don't
drive into the powered-down subsystem.  We took care of doing that when
we turned off the thing already, but we cannot just call the OFF code
from the probe due to problems with regulator state init.

Reported-by: Rask Ingemann Lambersten <rask@sygehus.dk>
Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoconfig-nfs-root.patch
Andy Green [Wed, 25 Feb 2009 01:07:33 +0000 (01:07 +0000)]
config-nfs-root.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoS3C64XX: Drop i2c0.max_freq to 150kHz.
Ben Dooks [Wed, 25 Feb 2009 00:41:30 +0000 (00:41 +0000)]
S3C64XX: Drop i2c0.max_freq to 150kHz.

Drop the i2c0 bus max-frequency to 150kHz to avoid
problems with boards where this is too fast.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
--- linux.agit.orig/arch/arm/plat-s3c/dev-i2c0.c 2009-02-20 23:04:34.000000000 +0000
+++ linux.agit/arch/arm/plat-s3c/dev-i2c0.c 2009-02-20 23:04:37.000000000 +0000
@@ -51,7 +51,7 @@ static struct s3c2410_platform_i2c defau
  .flags = 0,
  .slave_addr = 0x10,
  .bus_freq = 100*1000,
- .max_freq = 400*1000,
+ .max_freq = 150*1000,
  .sda_delay = S3C2410_IICLC_SDA_DELAY5 | S3C2410_IICLC_FILTER_ON,
 };

--
Ben (ben@fluff.org, http://www.fluff.org/)

  'a smiley only costs 4 bytes'

11 years agoconfig-squashfs-embedded.patch
Andy Green [Wed, 25 Feb 2009 00:41:27 +0000 (00:41 +0000)]
config-squashfs-embedded.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years ago[GTA03] Fix the power on status of the GSM module again.
DKAY_CHEN [Wed, 25 Feb 2009 00:41:25 +0000 (00:41 +0000)]
[GTA03] Fix the power on status of the GSM module again.

Because i found the level of the "MODEM_RST" will return to "L" when it set to "Input".

GSM module sometimes can not turn on especially after emove the C9102.

Signed-off-by: DKAY_CHEN <dkay_chen@openmoko.com>
11 years agoS3C64XX: Add DMA clock enable
Ben Dooks [Wed, 25 Feb 2009 00:41:22 +0000 (00:41 +0000)]
S3C64XX: Add DMA clock enable

Ensure the DMA channels have the controller clock
enabled before they are used.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoS3C64XX: Register DMA channels as system devices
Ben Dooks [Wed, 25 Feb 2009 00:41:19 +0000 (00:41 +0000)]
S3C64XX: Register DMA channels as system devices

Add the DMA channels as system devices to provide
suspend/resume and a presence in sysfs.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoS3C64XX: Add DMA/SDMA clock sources
Ben Dooks [Wed, 25 Feb 2009 00:41:17 +0000 (00:41 +0000)]
S3C64XX: Add DMA/SDMA clock sources

Add the clock sources for the DMA and SDMA controllers
and ensure that they are shut down at startup. The DMA
engines are dma0 and dma1, whilst the SDMA are called
dma2 and dma3.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoS3C64XX: Fix HCLK_GATE bit definitions
Ben Dooks [Wed, 25 Feb 2009 00:41:14 +0000 (00:41 +0000)]
S3C64XX: Fix HCLK_GATE bit definitions

Fix the definitions of HCLK_GATE bits, which
had SDMA0/1 named SDMA1/2 and the definitions
of bits 26..29 in the wrong order.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoconfig-squashfs-module.patch
Andy Green [Tue, 24 Feb 2009 12:11:17 +0000 (12:11 +0000)]
config-squashfs-module.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoconfig-smdk6410-audio.patch
Andy Green [Tue, 24 Feb 2009 02:14:13 +0000 (02:14 +0000)]
config-smdk6410-audio.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agofix-dma-setflags-dust.patch
Andy Green [Tue, 24 Feb 2009 02:14:11 +0000 (02:14 +0000)]
fix-dma-setflags-dust.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agofix-gta02-plat-plat-dma.patch
Andy Green [Tue, 24 Feb 2009 02:14:08 +0000 (02:14 +0000)]
fix-gta02-plat-plat-dma.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoconfig-gta03-plat-s3c-dma.patch
Andy Green [Tue, 24 Feb 2009 02:01:07 +0000 (02:01 +0000)]
config-gta03-plat-s3c-dma.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agofix-plat-s3c-dma-build-in-makefile.patch
Andy Green [Tue, 24 Feb 2009 02:01:05 +0000 (02:01 +0000)]
fix-plat-s3c-dma-build-in-makefile.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoclean-dust-to-info-s3c64xx-i2s.patch
Andy Green [Tue, 24 Feb 2009 02:01:02 +0000 (02:01 +0000)]
clean-dust-to-info-s3c64xx-i2s.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoMERGE-via-pending-tracking-hist-MERGE-via-stable-tracking-MERGE-via-mokopatches-track...
merge [Tue, 24 Feb 2009 01:49:53 +0000 (01:49 +0000)]
MERGE-via-pending-tracking-hist-MERGE-via-stable-tracking-MERGE-via-mokopatches-tracking-s3c64xx-dma-support-1235439162-1235439227
pending-tracking-hist top was MERGE-via-stable-tracking-MERGE-via-mokopatches-tracking-s3c64xx-dma-support-1235439162-1235439227 / 3d6a1b21cf5fbdb6250d781b0a4900a7a0768aa1 ... parent commitmessage:
From: merge <null@invalid>
MERGE-via-stable-tracking-hist-MERGE-via-mokopatches-tracking-s3c64xx-dma-support-1235439162

stable-tracking-hist top was MERGE-via-mokopatches-tracking-s3c64xx-dma-support-1235439162 / 893e864e65adffc9eb085ed4f8b552a31dcec840 ... parent commitmessage:
From: merge <null@invalid>
MERGE-via-mokopatches-tracking-hist-s3c64xx-dma-support

mokopatches-tracking-hist top was s3c64xx-dma-support / 2515f9a1d53d19b1e61d639875aedcbe7929666e ... parent commitmessage:
From: Ben Dooks <ben@simtec.co.uk>
S3C64XX: DMA support

Add support for the DMA blocks in the S3C64XX series
of CPUS, which are based on the ARM PL080 PrimeCell
system.

Unfortunately, these DMA controllers diverge from the
PL080 design by adding another DMA controller register
and configuration for OneNAND.

Signed-off-by: Ben Dooks <ben@simtec.co.uk>
11 years agoconfig-uplevel-hdq-network-led-trigger.patch
Andy Green [Mon, 23 Feb 2009 01:22:29 +0000 (01:22 +0000)]
config-uplevel-hdq-network-led-trigger.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoMove bq27000 data structures
Sven Rebhan [Mon, 23 Feb 2009 01:22:26 +0000 (01:22 +0000)]
Move bq27000 data structures

It's always a good idea to first define the data structures you use. ;-)

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoCleanup s3c2442 Kconfig
Sven Rebhan [Mon, 23 Feb 2009 01:22:23 +0000 (01:22 +0000)]
Cleanup s3c2442 Kconfig

For GTA02 we can remove GTA02_HDQ (not used anywhere) and
S3C2410_CLOCK (already selected by CPU_S3C2442).

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoRevert "Cleanup s3c2442 Kconfig"
Sven Rebhan [Mon, 23 Feb 2009 01:22:19 +0000 (01:22 +0000)]
Revert "Cleanup s3c2442 Kconfig"

This reverts commit fe8b7ff1a79648146501f37d467fcb2f2bea0164.
I undo the reordering of the options here, as it hides what
really happend.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoCleanup s3c2442 Kconfig
Sven Rebhan [Mon, 23 Feb 2009 01:21:58 +0000 (01:21 +0000)]
Cleanup s3c2442 Kconfig

For GTA02 we can remove GTA02_HDQ (not used anywhere) and
S3C2410_CLOCK (already selected by CPU_S3C2442). Additionally I reordered
the options a bit.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoCare about PCF50633 ADC disabled
Sven Rebhan [Mon, 23 Feb 2009 01:21:55 +0000 (01:21 +0000)]
Care about PCF50633 ADC disabled

If the PCF50633 ADC is disabled, but the charger is enabled we can't determine
the charger type using the pulldown. The safest thing we can do is to fallback
to 100mA that should be supplied according to the USB specs.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoAllow disabling of GTA02 vibrator
Sven Rebhan [Mon, 23 Feb 2009 01:21:52 +0000 (01:21 +0000)]
Allow disabling of GTA02 vibrator

If the vibrator is disabled we don't need to register a platform device.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoPCF50633 core depends on I2C
Sven Rebhan [Mon, 23 Feb 2009 01:21:50 +0000 (01:21 +0000)]
PCF50633 core depends on I2C

We select the PCF50633 core part, which depends on I2C, in our Kconfig. However,
make oldconfig does not automatically resolve this "depends on I2C" dependency.
We should select I2C for GTA02 as it's needed anyway.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoGSM requires serial console
Sven Rebhan [Mon, 23 Feb 2009 01:21:47 +0000 (01:21 +0000)]
GSM requires serial console

The GSM interface of all MACH_NEO1973s require the Samsung serial console,
so we should automatically select it.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoThe BQ27000 Battery requires HDQ
Sven Rebhan [Mon, 23 Feb 2009 01:21:44 +0000 (01:21 +0000)]
The BQ27000 Battery requires HDQ

Move the battery section within the HDQ_GPIO_BITBANG ifdef because it uses
functions only available if HDQ bitbanging is enabled.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoCare about PCF50633 charger disabled
Sven Rebhan [Mon, 23 Feb 2009 01:21:42 +0000 (01:21 +0000)]
Care about PCF50633 charger disabled

If we disable CONFIG_CHARGER_PCF50633 the build fails with a bunch of errors.
This patch defines placeholders for the functions that require the charger. All
users of the function pointers care about the case when they are NULL, so this
won't break anything. Furthermore a small restructuring of the mach file was
necessary to not sprinkle ifdefs all over.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoRemove double include.
Sven Rebhan [Mon, 23 Feb 2009 01:21:39 +0000 (01:21 +0000)]
Remove double include.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoSelect PCF50633
Sven Rebhan [Mon, 23 Feb 2009 01:21:36 +0000 (01:21 +0000)]
Select PCF50633

Remove the old config option and use the correct new one instead.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoRemove GTA02 config from S3C2440 menu.
Sven Rebhan [Mon, 23 Feb 2009 01:21:34 +0000 (01:21 +0000)]
Remove GTA02 config from S3C2440 menu.

We have this in the S3C2442 machine section now.

Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
11 years agoMake CROSS_COMPILE overrideable in ./build?
Timo Juhani Lindfors [Mon, 23 Feb 2009 01:21:30 +0000 (01:21 +0000)]
Make CROSS_COMPILE overrideable in ./build?

Hi,

since I am using packaged cross-compiler from debian I need to change

export CROSS_COMPILE=/usr/local/openmoko/arm/bin/arm-angstrom-linux-gnueabi-

in ./build to

export CROSS_COMPILE=arm-angstrom-linux-gnueabi-

every time there are changes to ./build and also instruct others to do
this when they want to test their kernels with the "normal
cross-compiler" and not the one from openmoko. How about making it
possible to override CROSS_COMPILE with for example the following
change?

11 years agofix-pm-gsm-disable-serial-drive-when-off.patch
Andy Green [Mon, 23 Feb 2009 01:21:27 +0000 (01:21 +0000)]
fix-pm-gsm-disable-serial-drive-when-off.patch

This patch cleans some cruft from neo1973-pm-gsm.c and makes
some changes to the power code for gsm.

First, on probe it enforces GSM is OFF.

Second, when GSM is OFF, it changes the RTS and TXD pins on
UART 0 to be inputs, instead of driving into a dead subsystem
and costing us ~70mA @ 5V constant dissipation (these 3.3V pins
were found at 1.5V due to that).

Third it asserts the GPIO peripheral function when ON, to keep
the same flow for the new interrupt generation workaround code.

Since several people worked on this area before, and maybe the
assertion of GSM OFF is different (it's OFF from bootloader though)
or maybe something else gets broken, I send the patch for comment
before applying it.

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoadd-neo1973_vibrator.patch
Andy Green [Sun, 22 Feb 2009 04:40:11 +0000 (04:40 +0000)]
add-neo1973_vibrator.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoclean-fiq-vib.patch
Andy Green [Sun, 22 Feb 2009 04:40:08 +0000 (04:40 +0000)]
clean-fiq-vib.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agogta03_hdq_support.patch
Balaji Rao [Sun, 22 Feb 2009 04:40:06 +0000 (04:40 +0000)]
gta03_hdq_support.patch

Added support for HDQ to GTA03.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agogta02_use_new_fiq_and_hdq_stuff.patch
Balaji Rao [Sun, 22 Feb 2009 04:40:03 +0000 (04:40 +0000)]
gta02_use_new_fiq_and_hdq_stuff.patch

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agointroduce_generic_hdq_gpio_bitbang_driver.patch
Balaji Rao [Sun, 22 Feb 2009 04:40:00 +0000 (04:40 +0000)]
introduce_generic_hdq_gpio_bitbang_driver.patch

Introduce a new generic HDQ gpio based bitbang driver

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agomove_pwm_code_to_plat_s3c.patch
Balaji Rao [Sun, 22 Feb 2009 04:39:57 +0000 (04:39 +0000)]
move_pwm_code_to_plat_s3c.patch

Since the pwm code in mach-2410 can be reused for 6410, move it to plat-s3c.

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agofiq_handler_in_c.patch
Balaji Rao [Sun, 22 Feb 2009 04:39:41 +0000 (04:39 +0000)]
fiq_handler_in_c.patch

The code for creating FIQ handlers in C is moved to
arch/arm/kerne/fiq.c

Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
11 years agoGTA0[12]: fix calypso sysfs power_on handling
Paul Fertser [Sun, 22 Feb 2009 01:16:16 +0000 (01:16 +0000)]
GTA0[12]: fix calypso sysfs power_on handling

This implements a proper scheme of controlling powering on of the Calypso
GSM modem which should be activated with a "push-button" connected to
MODEM_ON line.

Also a workaround for firmware bug of moko10 and earlier implemented to
force calypso to pull IRQ line down after powering on.

The "reset" sysfs node is left for compatibility with userspace (without it
gsm0710muxd breaks) but it does nothing (NC on hardware level on GTA02
anyway).

I tested flashing the GSM firmware (on GTA02) with this patch (using only
fluid and power_on node) and can confirm that it's working as reliably as
with accessing gpio directly per instructions.

The credit goes to Joerg Reisenweber for discussions and clarifications on
how this should be done properly.

Signed-off-by: Paul Fertser <fercerpav@gmail.com>
11 years agoAdapt last Touchscreen Changes from GTA02 to GTA01
Jose Luis Perez Diez [Sat, 21 Feb 2009 11:45:48 +0000 (11:45 +0000)]
Adapt last Touchscreen Changes from GTA02 to GTA01

Pardon me if I am too verbose but this is my first opportunity to send a patch.

I was wishing to try andy-tracking kernels on GTA01 long ago but It
was  FOSDEM and recent threads on the lists asking for GTA01 that made
me do it.

When I participated on the thread
http://n2.nabble.com/empiezo-a-amargarme-con-los-kernels-necesito-consejo-td2347206.html
 in the Spanish mailing list asking what kernel to use with which
distribution, I responded that only FSO based supported 2.6.28 and
upwards. I could not resist to brag  that I tested 2.6.29 on GTA01 but
only with kexec ( it is possible that my wording was more ambiguous on
the thread).

Nelson Asked if I had problems booting it with u-boot. I tried and was
successful but this time I played a Little with the interface and vala
term and noticed that the on screen keyboard was changing the key
pressed.

Using http://svn.openmoko.org/developers/tick/touchtest/touch_test.py
shown that most of the Up and sometimes the previous Move to where
landing on ([290..310],575).

Revising the code shown that the some of the filters and the option to
disable them were missing on mach-gta01.c.

I have tested the kernel with the patch applied with the filters
disabled and also with them enabled. When the filters are active I
still receive  some misplaced up events but much less (the raster
terminal keyboard is usable).

Signed-off-by:Jose Luis Perez Diez <perezdiez@gmail.com>

11 years ago'build' script: complain if we are not in a valid branch
Nelson Castillo [Fri, 20 Feb 2009 13:33:31 +0000 (13:33 +0000)]
'build' script: complain if we are not in a valid branch

This is a known problem and it is starting to become a FAQ.
People get an error message when they try to build a kernel
and they are not in a valid branch of their local repository.

When it happens git-branch will print something like:

* (no branch)
  andy-tracking
  balaji-tracking

With this patch we make sure we are in a valid branch when
building the kernel using the build script in order to avoid
OM developers some trouble. How? We print a nice message if
we are not in a valid branch and then stop.

Please feel free to edit the error message if you think it is
not clear enough.

Thanks to Werner Almesberger for his advice on shell scripting.

Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
11 years agofix-glamo-mci-move-stop-out-of-isr.patch
Andy Green [Fri, 20 Feb 2009 13:33:28 +0000 (13:33 +0000)]
fix-glamo-mci-move-stop-out-of-isr.patch

This patch moves the bulk transfer action outside of
interrupt context, along with the STOP transmission action
for multiblock transfers.

It's prompted by

https://docs.openmoko.org/trac/ticket/2180

But it can impact throughput to SD card, so it's for testing
currently.

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agofix-dfu-kern.patch
Andy Green [Fri, 20 Feb 2009 13:33:26 +0000 (13:33 +0000)]
fix-dfu-kern.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoclean-gta02.h-remove-v1v2.patch
Andy Green [Fri, 20 Feb 2009 04:41:26 +0000 (04:41 +0000)]
clean-gta02.h-remove-v1v2.patch

This strips out the v1 and v2 GTA02 revision specific defines
(we do not support these internal non-shipped revisions) and
fixes a bug in the glamo reset resource found from that.

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoclean-move-gta02.h.patch
Andy Green [Fri, 20 Feb 2009 04:41:23 +0000 (04:41 +0000)]
clean-move-gta02.h.patch

This break gta02.h out of s3c2410 jail and brings it into the
more logical mach-s3c2442 dir.

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoASoC: Staticise suspend/resume functions for GTA02
Mark Brown [Fri, 20 Feb 2009 02:27:02 +0000 (02:27 +0000)]
ASoC: Staticise suspend/resume functions for GTA02

Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
11 years agofix pan display
Sean McNeil [Fri, 20 Feb 2009 02:27:00 +0000 (02:27 +0000)]
fix pan display

It looks like I made a mistake in the original pan implementation that
is now getting exercised. The following fixes pan again.

Cheers,
Sean

11 years agoclean-move-mach-gta02-s3c2442.patch
Andy Green [Thu, 19 Feb 2009 03:46:14 +0000 (03:46 +0000)]
clean-move-mach-gta02-s3c2442.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoclean-remove-s3c2440-camera.patch
Andy Green [Thu, 19 Feb 2009 03:46:11 +0000 (03:46 +0000)]
clean-remove-s3c2440-camera.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoGTA03: Fix audio dapm settings.
Matt Hsu [Wed, 18 Feb 2009 06:21:03 +0000 (06:21 +0000)]
GTA03: Fix audio dapm settings.

Correct the dapm mapping between audio endpoint and physical pin.

Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>
11 years agoMERGE-via-pending-tracking-hist-claen-ledtrig-netdev-unused-va
merge [Wed, 18 Feb 2009 05:08:19 +0000 (05:08 +0000)]
MERGE-via-pending-tracking-hist-claen-ledtrig-netdev-unused-va
pending-tracking-hist top was claen-ledtrig-netdev-unused-va / 126536fc6ed26323103f941e7cf81e4b0ceaab7a ... parent commitmessage:
From: Andy Green <andy@openmoko.com>
claen-ledtrig-netdev-unused-var-warn.patch

Signed-off-by: Andy Green <andy@openmoko.com>
11 years agoMERGE-via-pending-tracking-hist-MERGE-via-pending-tracking-hist-1234930404
merge [Wed, 18 Feb 2009 04:49:32 +0000 (04:49 +0000)]
MERGE-via-pending-tracking-hist-MERGE-via-pending-tracking-hist-1234930404
pending-tracking-hist top was MERGE-via-pending-tracking-hist-1234930404 / 4a23c8aeedd5fb86009d290a41fc299e3965d8ec ... parent commitmessage:
From: merge <null@invalid>
MERGE-pending-tracking-patchset-edits

11 years agoThis patch check the probe_completed function pointer.
Michael Trimarchi [Wed, 18 Feb 2009 04:48:58 +0000 (04:48 +0000)]
This patch check the probe_completed function pointer.

Signed-off-by: Michael Trimarchi <michael@panicking.kicks-ass.org>
11 years agoMERGE-via-pending-tracking-hist-wm8753-unregister-i2c-device-b
merge [Tue, 17 Feb 2009 01:10:32 +0000 (01:10 +0000)]
MERGE-via-pending-tracking-hist-wm8753-unregister-i2c-device-b
pending-tracking-hist top was wm8753-unregister-i2c-device-b / 31c3d7550beb6e4b6bcc8da63b81211e00e9f676 ... parent commitmessage:
From: Paul Fertser <fercerpav@gmail.com>
wm8753: unregister I2C device (backport from upstream)

This is the last change necessary to be able to load/unload all sound
modules on Neo FreeRunner.

Signed-off-by: Paul Fertser <fercerpav@gmail.com>
11 years agoRe: GTA03 pcf50633
Balaji Rao [Mon, 16 Feb 2009 11:27:11 +0000 (11:27 +0000)]
Re: GTA03 pcf50633

On Mon, Feb 16, 2009 at 05:55:47PM +0800, Andy Green wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi Balaji -
>
> I am working on the sound driver on GTA03 at the moment, I just noticed
> this when I pulled out the USB cable
>
> [ 6905.100000] pcf50633 0-0073: usb curlim to 0 mA
> [ 6905.100000] Division by zero in kernel.
> [ 6905.100000] [<c002d5b0>] (dump_stack+0x0/0x14) from [<c002d5dc>]
> (__div0+0x18/0x20)
> [ 6905.110000] [<c002d5c4>] (__div0+0x0/0x20) from [<c0143f9c>]
> (Ldiv0+0x8/0x10)
> [ 6905.120000] [<c01ec7a0>] (pcf50633_mbc_usb_curlim_set+0x0/0x12c) from
> [<c01ec914>] (pcf50633_mbc_irq_handler+0x48/0xf4)
> [ 6905.130000] [<c01ec8cc>] (pcf50633_mbc_irq_handler+0x0/0xf4) from
> [<c019e288>] (pcf50633_irq_call_handler+0x28/0x2c)
> [ 6905.140000]  r5:00000004 r4:00000000
> [ 6905.150000] [<c019e260>] (pcf50633_irq_call_handler+0x0/0x2c) from
> [<c019eab8>] (pcf50633_irq_worker+0x2f0/0x33c)
> [ 6905.160000] [<c019e7c8>] (pcf50633_irq_worker+0x0/0x33c) from
> [<c00548d8>] (run_workqueue+0xb0/0x138)
> [ 6905.180000]  r7:c7826000 r6:c7ba8d50 r5:c7807200 r4:c7ba8d54
> [ 6905.190000] [<c0054828>] (run_workqueue+0x0/0x138) from [<c00554e8>]
> (worker_thread+0xec/0x100)
> [ 6905.200000]  r8:00000000 r7:00000000 r6:c7807200 r5:c7826000 r4:c7827fb0
> [ 6905.220000] [<c00553fc>] (worker_thread+0x0/0x100) from [<c00588f0>]
> (kthread+0x5c/0x94)
> [ 6905.230000]  r6:c00553fc r5:c7807200 r4:c7826000
> [ 6905.240000] [<c0058894>] (kthread+0x0/0x94) from [<c0046e54>]
> (do_exit+0x0/0x24c)
> [ 6905.250000]  r6:00000000 r5:00000000 r4:00000000
>
> It's funny because what I see as the only divide in there is
>
>  mbcc5 = (ma << 8) / mbc->pcf->pdata->chg_ref_current_ma;
>  pcf50633_reg_write(mbc->pcf, PCF50633_REG_MBCC5, mbcc5);

Oops, chg_ref_current_ma is not defined for mach-om-gta03.c. Sorry :(
here's a patch.

11 years agoMERGE-via-pending-tracking-hist-fix-wm8753-dais-unregistering
merge [Mon, 16 Feb 2009 00:23:04 +0000 (00:23 +0000)]
MERGE-via-pending-tracking-hist-fix-wm8753-dais-unregistering
pending-tracking-hist top was fix-wm8753-dais-unregistering / 0e964e0adf23b9e7c819750df3ad77ca81fdf56a ... parent commitmessage:
From: Paul Fertser <fercerpav@gmail.com>
Fix WM8753 DAIs unregistering

WM8753 uses a tricky way to switch DAIs "on the fly", for that it
registers 2 dummy DAIs and substitutes them depending on mixer control.

List element of registered dummy DAIs should be preserved to allow
unregistering of DAIs on module unload.

Signed-off-by: Paul Fertser <fercerpav@gmail.com>
11 years agoconfig-GTA03-alsa-soc-wm8753-enable.patch
Andy Green [Sun, 15 Feb 2009 12:31:18 +0000 (12:31 +0000)]
config-GTA03-alsa-soc-wm8753-enable.patch

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