| Commit message (Collapse) | Author | Age | Files | Lines |
... | |
| |
| |
| |
| |
| |
| |
| | |
- Fixes DAC direct not being applied in several situations
- Fixes 1 channel only on Captivate headphone+mic calls
- Add full support for Samsung Gingerbread kernels
including complete FM radio tunings
|
| |
| |
| |
| |
| |
| | |
- definite fix for DAC direct call issue
- reshaped source code (Linux Coding Style!)
- less hooks in pre-existing driver
|
| | |
|
| | |
|
| |
| |
| |
| | |
fix reversed is_path(SPEAKER) logic
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
- New speaker tuning mode for music using different
hardware EQ settings than stock ones
- Microphone presets now supported on Nexus S.
- Microphone presets balanced and loud reworked
with lower background noise and higher dynamic range
for balanced: -2dB noise, +6dB dynamic range.
- New DAC direct option: reduces distortion and improves
SNR by a small margin (bypass usless analog channel mixer)
- Anti-jitter mode on by default.
- Same exact source is now used on each device supported
(still: only useful code is compiled in via ifdefs)
- ability to fully disable/re-enable the driver and
its sysfs interface: 0 or 1 in
/sys/class/misc/voodoo_sound_control/enable
- refactored source code
|
| |
| |
| |
| | |
needs some fine adjustement for the code & hardware
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This reverts commit 7507d5a839a3caf97832f811df2829daa5b245c8.
The mic gain reduction offered a very limited improvement of the
echo for SIP calls but was such that it rendered other VoIP applications
unusable.
Signed-off-by: Eric Laurent <elaurent@google.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The sleep current is more consumed about 0.1mA.
Because ear_sel GPIO's setting is wrong.
ear_sel GPIO is not connected with anything by Hardware.
So, block 'Pull None' setting to reduce the sleep current.
Change-Id: I8d9236b6a1af4f0051e56539844ba2778a748df3
Signed-off-by: sh21.won <sh21.won@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
fix herring-wm8994.c to compile on 2.6.39
checkpoint: use mainline sound driver
checkout: audio driver loads but locks up
checkpoint: audio setup does not complain, but playback times-out
checkpoint: using soc sound i2s and dma code from 2.6.35 - dma still hangs
checkpoint: using old codec, dma and i2s code sound plays but is distorded
fix distorted sound
rename samsung wm8994 codec to not conflict with standard wm8994 mfd device
modify dma.c to work with dma wrapper
rename some audio config options and fix defconfig
remove some duplicate clocks
Change-Id: I1b72b5b7937fd4a40a1719674976ba16d7e8f3a6
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
In call, Rx volume for a speakerphone is low.
So, Rx volume is raised by 1dB.
And, the volume table for a voice call is added to distinguished
from the original volume table.
Change-Id: Ie7c81fe689aa5c437b3007c4502bb4bb5487afa2
Signed-off-by: sh21.won <sh21.won@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
The volume is raised at the latest version's device.
So, the codec gain is lowered to satisfy a certain volume.
Change-Id: I823d1c92fe2ac5d4596bcbd48b771b3ee348e193
Signed-off-by: sh21.won <sh21.won@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Voice calls are occasionally muted. This is because
a value is written to the codec register before the codec
has turned on completely. So, turn on the codec and
wait until it is turned on completely.
Signed-off-by: sh21.won <sh21.won@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
update CDMA specific clok settings to
- fix the current wrong values
- set the registers when the codec is powered-on by adding 50ms delay
Change-Id: I257f0105e8da2bb52f8ef51388739103fd0b9ba2
Signed-off-by: Thomas Ryu <smilesr.ryu@samsung.com>
|
| |
| |
| |
| |
| |
| | |
add CDMA specific clok settings to improve the sound quality in speaker phone or headset.
Signed-off-by: Thomas Ryu <smilesr.ryu@samsung.com>
|
| |
| |
| |
| |
| |
| |
| | |
add cdma_playback_gain_table to support CDMA specific tunings.
Change-Id: I2f35161bdf7e9d33692bc7245a812215d1690560
Signed-off-by: Thomas Ryu <smilesr.ryu@samsung.com>
|
| |
| |
| |
| |
| |
| |
| | |
Adding herring_ prefix to the herring version checking macros.
Change-Id: Ieec80d5e64d454c7a4b7beb18b3c7e516aeed375
Signed-off-by: Rom Lemarchand <rlemarchand@sta.samsung.com>
|
| |
| |
| |
| |
| |
| |
| | |
1. set modem as a master and codec as a slave.
2. add audio gain tunning of voice call.
Signed-off-by: Thomas Ryu <smilesr.ryu@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
To increase recognition rate of tty character through both 2G & 3G network, INR1's gain must be set 12dB.
right input volume : -10.5dB -> 12dB
Change-Id: I677ea36b45742bf925f533c71ac3eeacffe5bc58
Signed-off-by: UK KIM <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Main mic : Left Input Volume : -6dB
Handsfree mic : Right Input Volume : 3dB
Change-Id: I1acdf69c18307d53608c7a5dc270cebd7c59f937
Signed-off-by: UK KIM <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| | |
supports VCO, HCO, FULL mode of TTY during call.
Change-Id: If56d0c3da20377c59adc8681769e443f35da045c
Signed-off-by: Uk Kim <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| | |
Change-Id: I1b54e239278d2433bb00d7f6ac2cb43f5eea5c08
Signed-off-by: UK KIM <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| | |
devided gain table between voice capture and voice communication.
Signed-off-by: UK KIM <w0806.kim@samsung.com>
Change-Id: Ie29abf9e2bb7fd48582ef3a7550bb82398c511c6
|
| |
| |
| |
| |
| | |
Change-Id: Iffc4340f37e80c446e212d656c5605687da6555d
Signed-off-by: Uk Kim <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<Bug fix> Change gain values in Camcorder mode
IN1L_VOL(18h, MAIN_MIC) = -1.5dB
This patch is requested by Samsung H/W Team.
Change-Id: I4b3b6dcf22eb2504acdc28958558688e2bb26ef0
Signed-off-by: Minsong Kim <ms17.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<Bug Fix> This is requested by Samsung HW team.
IN1L_VOL(18h, HP_NO_MIC) = -9dB
When we test voice call with 3-pole headset, we use
main mic. So mic gain must be same with speaker call mode.
Change-Id: Ica9828e74eb69c68c9783d68b70bcc0793763aaf
Signed-off-by: Minsong Kim <ms17.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| | |
disable mic path without closing input stream.
Change-Id: If26c1c2317d94147dc038f63387eb14d12d1cb25
Signed-off-by: Uk Kim <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
<Bug fix> Change gain values in voice call mode
IN1R_VOL(1ah, HP_MIC) = +7.5dB
This patch is requested by Samsung H/W team.
For audio voice quality improvement,
Analog input gain goes up.
It makes more natural voice and improving a muffle voice.
Change-Id: Idcbc2e1ca3ebdfd7ac76418fdc823f3fca8b5430
Signed-off-by: Minsong Kim <ms17.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We have to support a different microphone gain for the camcorder.
Because codec gain is set focused on SIP. When we use the phone as
the camcorder, MIC gain should be higher than SIP. This is requested
by Samsung QA.
Change-Id: I8600ed70350f2fe2d715f81e1c22d2f37d55a303
Signed-off-by: Minsong Kim <ms17.kim@samsung.com>
|
| |
| |
| |
| | |
Add a gain configuration for headsets without microphones.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Change gain values in voice call mode
IN1L_VOL(18h, receiver) = -6dB
IN1L_VOL(18h, speaker) = -6dB
IN1R_VOL(1ah, HP_MIC) = -10.5dB
This is requested by Samsung H/W Team.
The reason why voice was clipping problem from codec,
we reduced an analog gain and volume up a digital gain in modem.
And also we can manage to improve an echo/DT problem
from these change.
Change gain values in voice call mode
HPOUT1L_VOL(1ch, HP_MIC) = +3dB
HPOUT1L_VOL(1ch, HP_NO_MIC) = +3dB
HPOUT1R_VOL(1dh, HP_MIC) = +3dB
HPOUT1R_VOL(1dh, HP_NO_MIC) = +3dB
Change-Id: Ifc6f9a31664e640d65c1a3cd099e6a8005f304bd
Signed-off-by: Minsong Kim <ms17.kim@samsung.com>
|
| |
| |
| |
| |
| | |
Reduce headphone volume, which lowers the minimum volume to
improve the listening experience in a quiet environment.
|
| | |
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
1. support gain setting in all modes for Europe.
2. change gain value
HPOUT1 L/R(1Ch, 1Dh) : -8dB
Change-Id: I3400741ae7a47fbb501348683dc57f9c0baaa174
Signed-off-by: UK KIM <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| |
| | |
As per Samsung recommendation to meet T-Mobile test standards.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
For EUR binary.
1. SPEAKER OUTL vol(26h) : +1dB
2. HPOUT1L/R vol(1Ch, 1Dh) : -2dB
Change-Id: Ie535fda5667f5ad99cdeeb43028bcd85c6a09474
Signed-off-by: UK KIM <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
This gives us the standard register cache dumps, which we already know
how to process and which filters the sparseness out of the register map.
This should have no effect in normal operation and makes debug easier.
Change-Id: Ic8f679f49e110b7c4080f6e1434a0b81196af6c9
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| | |
In the common settings for Bluetooth we were repeatedly clearing the AIF3
routing setup, making us more sensitive than is desirable to the ordering
of playback and record startup.
Change-Id: Ia17ac64e3410412261f67f7325ba176cb17513fe
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| | |
fix sequence for setting clk_enable of 24Mhz Xtal clk.
Change-Id: I8bd6b54168cdcd6b2fbdecf3bbfc3076dc3078dd
Signed-off-by: UK KIM <w0806.6.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The headset is using the right data path so we need to do high pass
filtering in the right channel rather than the left channel. This appears
to improve things for me, though futher testing in a quieter environment
is required.
Change-Id: I45c74472f640f20b09b8c1254c01c3d28808e423
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
must distinguish headset and headphone device for built-in mic
output device is seperated but use same path.
Change-Id: I45cc27c74d0eb3505e1b5adb1ba52690261ae0f3
Signed-off-by: UK KIM <w0806.kim@samsung.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
The signal level was observed to be very low when doing recordings for
voice recognition from headphone sources, and also from main microphone.
Change this by:
- Increasing the gain for headphone case, applying an additional 3dB in
the mic preamp and +30dB from the input mixer.
- Removing +6dB digital gain for main mic (since digital gain will only
increase the signal level, not recover extra information after ADC has
quantized the data).
- Updating main mic analogue settings to match normal record for an
aggregate change from 21dB to 40.5dB.
The most observable user level effect of the above being that the signal
meter displayed by the voice search application registers a fairly clear
signal while previously the signal levels appeared very flat.
Change-Id: I3beaf70ab7649f678e105fa75cd09f2a80749f99
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Since we're not bringing it up on resume let's do the same on first boot
also.
Change-Id: I1ca11608ed400d7224c9a40ea55ed6922aaabbfb
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| | |
We're using CODEC_OFF elsewhere so use it here also for consistency.
Change-Id: Ia2d7933b36374a896ffffaf5542644d4d7f6c715
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
In normal operation the driver leaves VMID disabled when the CODEC is idle
so do this also when coming out of resume. Maintaining VMID consumes a
small amount of power so this should improve battery life, especially when
waking up to perform background syncs with network services.
This will lead to an additional 50ms delay when bringing up the first audio
stream on the device after resume but since we already take a similar hit
for all audio enables after that this should have negligable effect on user
experience.
We should consider also disabling the LDOs as they only take 5ms to bring
up but we need to measure to ensure that power consumption is actually
meaningfully reduce.d
Change-Id: I825e26e5d4916fa7babac93393b795c9ff52d119
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| | |
Change-Id: Ibf474a43900b8f83716116a186e1a750d7c9ec29
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
We only need to recalibrate the DC servo configuration when the analogue
configuration is changed, which in this system means whenver the analogue
volume is changed or whenever the path is changed. Use this to save the
~160ms taken by the recalibration by saving the value per-path
We still need to recalibrate after volume changes and the first time a given
path is used but these should be much less frequent events than headphone
enable.
Change-Id: I706fe161f5cc0709e58d7702b28796ed85c51384
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Enable the high pass filter at the start of the powerup sequence to ensure
that it's enabled when we start seeing audio, and remove the 10ms delay in
the sequence which appears to have been masking the bounce by discarding
audio.
Change-Id: Ie73155339bea8450a8948ab265b2ad1d3bc83685
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| | |
Stops us overdriving things, improving record quality.
Change-Id: I8043f49961331e10c4bd43890865e39de531dcd0
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|
| |
| |
| |
| |
| |
| |
| |
| | |
Add comments translating register values into dB levels. Should be no
functional changes.
Change-Id: I25cfc1fd524c42e5e03b3d7ce060d602fbb8287a
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
|