Age | Commit message (Collapse) | Author |
|
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>
|
|
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
|
It looks like I made a mistake in the original pan implementation that
is now getting exercised. The following fixes pan again.
Cheers,
Sean
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Correct the dapm mapping between audio endpoint and physical pin.
Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>
|
|
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>
|
|
pending-tracking-hist top was MERGE-via-pending-tracking-hist-1234930404 / 4a23c8aeedd5fb86009d290a41fc299e3965d8ec ... parent commitmessage:
From: merge <null@invalid>
MERGE-pending-tracking-patchset-edits
|
|
Signed-off-by: Michael Trimarchi <michael@panicking.kicks-ass.org>
|
|
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>
|
|
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.
|
|
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>
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Temporarily force the CODEC power on to allow testing
of the WM8753 without having to update the drive for
the regulator framework.
Signed-off-by: Ben Dooks <ben@simtec.co.uk>
|
|
not present in the GTA03 (such as the external AMP) and
bind to the S3C64XX I2S audio.
Signed-off-by: Ben Dooks <ben@simtec.co.uk>
|
|
drivers can find the IIS bus clock rate.
Signed-off-by: Ben Dooks <ben@simtec.co.uk>
|
|
s3c_gpio_cfgpin and s3c_gpio_setpull to allow modules
to use them (such as ASoC drivers)
Signed-off-by: Ben Dooks <ben@simtec.co.uk>
|
|
the user of the problem. This is generally due to a
problem on the i2c bus such as an un-powered or
non-connected codec.
Signed-off-by: Ben Dooks <ben@simtec.co.uk>
|
|
Reported-by: Mickey Lauer <mickey@openmoko.org>
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Apparetly setting the number of VTs to 8 gets rid of the
android flickering issue, this patch does it for GTA01/2/3
configs.
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Sascha's last changes didn't take care about case where blocker stuff
is disabled.
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
|
|
|
|
|
|
The overrun bit has used only if the rx interrupt error is enabled.
Signed-off-by: Michael Trimarchi <michael@panicking.kicks-ass.org>
|
|
-do the parameter check of setting operating mode.
-add device haeder file.
Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>
|
|
Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>
|
|
The following two features are added as platform data.
- externel reset
- operating mode
Signed-off-by: Matt Hsu <matt_hsu@openmoko.org>
|
|
Hi.
The recent patch to fix a USB current limit violation when turning the
device off triggered a bug in setting the battery charge current limit. We
now get a charge current limit of 0 mA on the GTA02 when plugging in the
wall charger (just as when setting 1000 mA in /sys/class/[...]/chg_curlim).
This patch fixes it (and a comment typo). Tested on a GTA02 with a wall
charger.
Signed-off-by: Rask Ingemann Lambertsen <rask@sygehus.dk>
|
|
Move the regulator handling stuff away from s3cmci.c to mach-gta01.c.
Signed-off-by: Balaji Rao <balajirrao@openmoko.org>
|
|
The DMA used for S3C2440 is also used for S3C2442, so we also need the defines.
This patch is rebased on the latest andy-tracking branch.
Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
The dependency on ARCH_S3C2440 forces us to select at least one S3C2440 machine to get the makefile system to select the required options for our
S3C2442 GTA02. By removing this option I can build fine with only the GTA02 machine selected.
Signed-off-by: Sven Rebhan <odinshorse@googlemail.com>
|
|
Please ignore the previous patch. Here's the right one. I'll now write
100 times "I shall git-pull before making changes." Sorry.
- Werner
---------------------------------- cut here -----------------------------------
> I already tested and measured the signal of the "/MODEM_RST","/MODEM_ON"
[...]
> + s3c_gpio_setpull(GTA03_GPIO_MODEN_ON, S3C_GPIO_PULL_NONE);
~~~~~~~~
Here's a patch that changes GTA03_GPIO_MODEN_ON to GTA03_GPIO_MODEM_ON.
Signed-off-by: Werner Almesberger <werner@openmoko.org>
|
|
hes-tracking-MERGE-via-master-MERGE-via-master-hist-1232625318-1233879011-1233879414-1233879505
pending-tracking-hist top was MERGE-via-stable-tracking-MERGE-via-mokopatches-tracking-MERGE-via-master-MERGE-via-master-hist-1232625318-1233879011-1233879414-1233879505 / 1c405b6ccee468298e7ccbfd9a3a3f4d123207b0 ... parent commitmessage:
From: merge <null@invalid>
MERGE-via-stable-tracking-hist-MERGE-via-mokopatches-tracking-MERGE-via-master-MERGE-via-master-hist-1232625318-1233879011-1233879414
stable-tracking-hist top was MERGE-via-mokopatches-tracking-MERGE-via-master-MERGE-via-master-hist-1232625318-1233879011-1233879414 / 71be0a45396066b1f8f27f8f4f87937247a129e1 ... parent commitmessage:
From: merge <null@invalid>
MERGE-via-mokopatches-tracking-hist-MERGE-via-master-MERGE-via-master-hist-1232625318-1233879011
mokopatches-tracking-hist top was MERGE-via-master-MERGE-via-master-hist-1232625318-1233879011 / 1be1b01373f572a02c6f1f99863c8c11ed2f9f5b ... parent commitmessage:
From: merge <null@invalid>
MERGE-via-master-MERGE-via-master-hist-1232625318
master top was MERGE-via-master-hist-1232625318 / dd4b117123ae66451695810017eb72fbdfc05df5 ... parent commitmessage:
From: merge <null@invalid>
MERGE-master-patchset-edits
|
|
Cedric noticed that when he turned his GTA02 OFF while hooked to
a "dumb" USB charger that should only be asked for 100mA, suddenly
much more current was demanded.
When PMU enters PMU standmy when we go OFF, its USB current limit
is reset to the variant default of 500mA. Since we previously had
the charging current limit fixed at 1A, it meant we immediately
charge at 500mA.
This patch makes the charging limit follow the USB current limit,
so that even when we go off, no more than the last authorized amount
of current will be taken, even if the USB current limit is later
broken by the variant default.
Reported-by: Cedric Berger <cedric.berger74@gmail.com>
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Remove the spew from GTA02 POWER_SUPPLY debugging messages,
it's just the HDQ stuff filling up dmesg every 30s or so.
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Signed-off-by: DKAY_CHEN <dkay_chen@openmoko.com>
|
|
We need to clear down the wakeup source reg if we woke from
threshold.
Reported-by: Simon Kagstrom <simon.kagstrom@gmail.com>
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
We shouldn't be resetting the highpass filter every sample.
It should be disabled if we don't use it or allowed to
work across multiple samples if we do.
Was this hiding some other problem?
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
We can see and account for any overruns caused by IRQ latency
now that we fetch the status register. This patch adds a /sys
node so it can be queried, and accounting.
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Level interrupts solve the sticky loss of service from accels issue.
But currently, we get two service actions per one interrupt, leading to
information getting read and sent to the input subsystem twice.
This patch makes the ISR confirm with the lis302dl status register
that there is fresh data before accepting it, it works around the
issue and allows use of the other information in the status reg by
another patch.
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Unlike SMDK we use 48MHz xtal on GTA03, don't copy
platform phy clock init from SMDK then
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Make previous Michael's patch verbose about truncating
mixer names. Sample message:
sound/soc/soc-dapm.c:349 mixer name 'Playback Mixer Voice Capture Switch' truncated
to 31 characters.
Signed-off-by: Nelson Castillo <arhuaco@freaks-unidos.net>
|
|
Signed-off-by: Andy Green <andy@openmoko.com>
|
|
Please revert in feuture version
Pointed-out-by: Andy Green <andy@openmoko.com>
Signed-off-by: Michael Trimarchi <michael@panicking.kicks-ass.org>
|