[alsa-devel] Dell N5110 HDA External Speaker is mute

[Cefn Hoile]

On 12 January 2012 15:18, Takashi Iwai <tiwai at suse.de> wrote:
> At Thu, 12 Jan 2012 15:11:16 +0000,
> Cefn Hoile wrote:
>>
>> On 12 January 2012 11:59, Takashi Iwai <tiwai at suse.de> wrote:
>> > At Thu, 12 Jan 2012 11:51:12 +0000,
>> > Cefn Hoile wrote:
>> >>
>> >> Thanks, Takashi.
>> >>
>> >> I don't know what tools and invocations to use test out the GPIO
>> >> values you describe. I think you're saying I shouldn't bother toggling
>> >> things through HDA Analyzer as there's nothing obviously wrong in that
>> >> part of the configuration (as reported by alsa-info.sh which is
>> >> attached as requested).
>> >>
>> >> I noticed in HD-Audio.txt a section which mentions a similar problem
>> >> and IDT codecs specifically (below). Is there an IRC or mailing list
>> >> where people could talk me through the basics of trying different
>> >> GPIO/EAPD, Power Enable or Pin Sense manipulations. I wouldn't know
>> >> how, yet. If I succeed I'll report back.
>> >>
>> >> Hopefully this will be worth it, as I think the N5110 is a pretty
>> >> popular model with a common card, and probably a bunch of Dell
>> >> hardware could get fixed if it could be figured out.
>> >
>> > You can change the GPIO setup via hda-verb on the fly.
>> > For example, to set GPIO bit 0, try
>> >        hda-verb /dev/snd/hwC0D0 0x01 SET_GPIO_MASK 0x01
>> >        hda-verb /dev/snd/hwC0D0 0x01 SET_GPIO_DIR 0x01
>> >        hda-verb /dev/snd/hwC0D0 0x01 SET_GPIO_DATA 0x01
>> > where the output direction is assumed, and only GPIO 0 is set.
>> > For GPIO bit 1,
>> >        hda-verb /dev/snd/hwC0D0 0x01 SET_GPIO_MASK 0x02
>> >        hda-verb /dev/snd/hwC0D0 0x01 SET_GPIO_DIR 0x02
>> >        hda-verb /dev/snd/hwC0D0 0x01 SET_GPIO_DATA 0x02
>> > and 0x04 for GPIO bit 2.  Your device have up to three GPIO bits.
>> >
>> > If none of above doesn't any change, try to set DATA 0 instead of 1,
>> > and/or DIR 0 instead of 1.
>> >
>> >
>> >
>> > Takashi
>> >
>> >> QUOTING FROM HD-Audio.txt >>>>>>>
>> >>
>> >> If you are masochistic enough to debug the driver problem, note the
>> >> following:
>> >>
>> >> - The speaker (and the headphone, too) output often requires the
>> >>   external amplifier. This can be set usually via EAPD verb or a
>> >>   certain GPIO. If the codec pin supports EAPD, you have a better
>> >>   chance via SET_EAPD_BTL verb (0x70c). On others, GPIO pin (mostly
>> >>   it's either GPIO0 or GPIO1) may turn on/off EAPD.
>> >> - Some Realtek codecs require special vendor-specific coefficients to
>> >>   turn on the amplifier. See patch_realtek.c.
>> >> - IDT codecs may have extra power-enable/disable controls on each
>> >>   analog pin. See patch_sigmatel.c.
>> >> - Very rare but some devices don't accept the pin-detection verb until
>> >>   triggered. Issuing GET_PIN_SENSE verb (0xf09) may result in the
>> >>   codec-communication stall. Some examples are found in
>> >>   patch_realtek.c.
>> >>
>> >> <<<<<<<<
>> >>
>> >> On 12 January 2012 11:25, Takashi Iwai <tiwai at suse.de> wrote:
>> >> > At Thu, 12 Jan 2012 10:56:24 +0000,
>> >> > Cefn Hoile wrote:
>> >> >>
>> >> >> In case anything has changed in my config since the original info run,
>> >> >> here is the same script run right now...
>> >> >>
>> >> >> http://www.alsa-project.org/db/?f=d0a1fd23339878916b611c88e64e4868b45445ce
>> >> >
>> >> > Could you rather attach?  Then it'll be archived and searchable.
>> >> >
>> >> > Anyway, I don't see any obvious problem in the output.
>> >> > If the pin assignment is correct, the routing and amp setups should be
>> >> > OK with the current status.
>> >> >
>> >> > Maybe it's a missing GPIO setup.  Try to adjust GPIO mask, direction
>> >> > and value bits.
>> >> >
>> >> >
>> >> > Takashi
>>
>> I have tried the attached script, which hopefully corresponds with the
>> GPIO exploration regime you wanted me to try. Please take a look at
>> the sequence in case there could be some combination I missed based on
>> trying to interpret your description, or if there's some kind of
>> reload step I need to do.
>
> Yeah, there are some bugs :)
> The combination should be (mask/dir/data)
> 1/1/0 1/1/1 1/0/0 1/0/1
> 2/2/0 2/2/2 2/0/0 2/0/2
> 4/4/0 4/4/4 4/0/0 4/0/4
>
> Also, as a test, use aplay with -Dplughw option.
> This skips PA and others.
>
>
> Takashi
>
>> When running this GPIO script as...
>> sudo ./setgpio.sh
>> ...with headphones on my N5110 it plays the sample every time, and
>> with speakers it is always silent.
>>
>> Here's an example log of execution...
>>
>> cefn at cefn-oneiric-dell:~$ sudo ./setgpio.sh
>> [sudo] password for cefn:
>> SETTING GPIO 0x01 MASK, DIRECTION AND DATA TO [0x01 0x00 0x00]
>> nid = 0x1, verb = 0x716, param = 0x1
>> value = 0x0
>> nid = 0x0, verb = 0x717, param = 0x1
>> value = 0x0
>> nid = 0x0, verb = 0x715, param = 0x1
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x02 MASK, DIRECTION AND DATA TO [0x01 0x00 0x00]
>> nid = 0x1, verb = 0x716, param = 0x2
>> value = 0x0
>> nid = 0x0, verb = 0x717, param = 0x2
>> value = 0x0
>> nid = 0x0, verb = 0x715, param = 0x2
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x04 MASK, DIRECTION AND DATA TO [0x01 0x00 0x00]
>> nid = 0x1, verb = 0x716, param = 0x4
>> value = 0x0
>> nid = 0x0, verb = 0x717, param = 0x4
>> value = 0x0
>> nid = 0x0, verb = 0x715, param = 0x4
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x01 MASK, DIRECTION AND DATA TO [0x01 0x00 0x01]
>> nid = 0x1, verb = 0x716, param = 0x1
>> value = 0x0
>> nid = 0x0, verb = 0x717, param = 0x1
>> value = 0x0
>> nid = 0x1, verb = 0x715, param = 0x1
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x02 MASK, DIRECTION AND DATA TO [0x01 0x00 0x01]
>> nid = 0x1, verb = 0x716, param = 0x2
>> value = 0x0
>> nid = 0x0, verb = 0x717, param = 0x2
>> value = 0x0
>> nid = 0x1, verb = 0x715, param = 0x2
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x04 MASK, DIRECTION AND DATA TO [0x01 0x00 0x01]
>> nid = 0x1, verb = 0x716, param = 0x4
>> value = 0x0
>> nid = 0x0, verb = 0x717, param = 0x4
>> value = 0x0
>> nid = 0x1, verb = 0x715, param = 0x4
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x01 MASK, DIRECTION AND DATA TO [0x01 0x01 0x00]
>> nid = 0x1, verb = 0x716, param = 0x1
>> value = 0x0
>> nid = 0x1, verb = 0x717, param = 0x1
>> value = 0x0
>> nid = 0x0, verb = 0x715, param = 0x1
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x02 MASK, DIRECTION AND DATA TO [0x01 0x01 0x00]
>> nid = 0x1, verb = 0x716, param = 0x2
>> value = 0x0
>> nid = 0x1, verb = 0x717, param = 0x2
>> value = 0x0
>> nid = 0x0, verb = 0x715, param = 0x2
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x04 MASK, DIRECTION AND DATA TO [0x01 0x01 0x00]
>> nid = 0x1, verb = 0x716, param = 0x4
>> value = 0x0
>> nid = 0x1, verb = 0x717, param = 0x4
>> value = 0x0
>> nid = 0x0, verb = 0x715, param = 0x4
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x01 MASK, DIRECTION AND DATA TO [0x01 0x01 0x01]
>> nid = 0x1, verb = 0x716, param = 0x1
>> value = 0x0
>> nid = 0x1, verb = 0x717, param = 0x1
>> value = 0x0
>> nid = 0x1, verb = 0x715, param = 0x1
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x02 MASK, DIRECTION AND DATA TO [0x01 0x01 0x01]
>> nid = 0x1, verb = 0x716, param = 0x2
>> value = 0x0
>> nid = 0x1, verb = 0x717, param = 0x2
>> value = 0x0
>> nid = 0x1, verb = 0x715, param = 0x2
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>> SETTING GPIO 0x04 MASK, DIRECTION AND DATA TO [0x01 0x01 0x01]
>> nid = 0x1, verb = 0x716, param = 0x4
>> value = 0x0
>> nid = 0x1, verb = 0x717, param = 0x4
>> value = 0x0
>> nid = 0x1, verb = 0x715, param = 0x4
>> value = 0x0
>> Home directory /home/cefn not ours.
>> Playing WAVE '/usr/share/sounds/alsa/Front_Center.wav' : Signed 16 bit
>> Little Endian, Rate 48000 Hz, Mono
>>
>>
>> Cefn

Thanks again for your patience. It would be really useful if you could
suggest a specific change to the section of my script which details
the alternates, unless my script is more conceptually broken than
that. That way I can't misinterpret which of the flags you're trying
to describe with words. The problem I have looking at your "1/1/0"
list is which arguments of the actual hda-verb command to bind them
to: there are actually 4 values to vary in my mind. Three lines are
executed as a block for every possible combination of values in my
script (in the center of the nested loops)...

hda-verb /dev/snd/hwC0D0 "$GPIOMASK" SET_GPIO_MASK "$TARGETBIT"
hda-verb /dev/snd/hwC0D0 "$GPIODIR" SET_GPIO_DIR "$TARGETBIT"				
hda-verb /dev/snd/hwC0D0 "$GPIODATA" SET_GPIO_DATA "$TARGETBIT"

.... that is, GPIOMASK, GPIODIR, GPIODATA AND TARGETBIT. Your response
details only 3 values, so what do I do with TARGETBIT, or is it
identical to something else?.

The combinations are explored exhaustively by executing the three
hda-verb lines above for every set of values in a nested loop (you can
see how these are bound to hda-verb above). Previously the arrays of
possible values to iterate over read ...

TARGETBITS=( "0x01" "0x02" "0x04" ) # TRY ALL COMBINATIONS OF THREE GPIO BITS
GPIOMASKS=( "0x01" ) # THIS STAYS TRUE THROUGHOUT
GPIODIRS=( "0x00" "0x01" ) # TRY ALL COMBINATIONS OF THIS BOOLEAN
GPIODATAS=( "0x00" "0x01" ) # TRY ALL COMBINATIONS OF THIS BOOLEAN

...but what should they read?

Thanks again. I feel a bit of an idiot as I'm sure I could work this
out if I knew what hda-verb was actually doing to flags in the
soundcard, and what the arguments mapped to, but frankly it's a bit of
a mystery when you run hda-verb --help and your helpful suggestions
seem to assume I understand the mappings already.