[alsa-devel] writing an alsa driver

[Michele Curti]

On Thu, Mar 03, 2016 at 10:20:55AM -0600, Pierre-Louis Bossart wrote:
> On 3/3/16 5:27 AM, Vinod Koul wrote:
> >On Thu, Mar 03, 2016 at 12:11:10PM +0100, Michele Curti wrote:
> >>Hi Vinod,
> >>
> >>On Thu, Jun 25, 2015 at 08:54:07AM +0530, Vinod Koul wrote:
> >>>>[    2.700468] sst-mfld-platform sst-mfld-platform: Slot control: codec_in rx deinterleaver codec_in1_1 doesn't have DAPM widget!!!
> >>>this is unexpected, I will check on this
> >>>
> >>>>...
> >>>>[    7.970937] intel_sst_acpi 80860F28:00: Wait timed-out condition:0x0, msg_id:0x1 fw_state 0x3
> >>>>[    7.970964] intel_sst_acpi 80860F28:00: fw returned err -16
> >>>Yes this is a problem, which FW are you using
> >
> >Have you tried the recent kernels, Pierre did a bunch of update and got many
> >things working.
> >
> >As for timeout, that simply means we have send a message to DSP which failed
> >to get a response. Perhpas the message had some bad element of FW had a
> >bug/crashed...
> 
> The firmware timeout is due to a change in the BIOS for BYT-CR devices that
> impacts the IPC interrupt number, I have a set of patches that fix this on
> Asus T100TAF but I still can't get any sound out.
>

Do you mean the .acpi_ipc_irq_index from 5 to 0?
If so, I also carry that fix in my patches.

Thanks,
Michele

On Thu, Mar 03, 2016 at 12:11:10PM +0100, Michele Curti wrote:
> 2. platform driver (patch 0002 at [1])
> 
> Used the sst-mfld-platform one (under sound/soc/intel/atom/sst), changing
> the .acpi_ipc_irq_index from 5 to 0 because the DSP interrupt order in
> the ACPI DSDT table of this laptop is different from the "normal" (see
> line 3745 of dsdt.dsl at [2]).
> 
> $ cat /proc/interrupts | grep sst
>             CPU0       CPU1       CPU2       CPU3       
>    1:         24          0          0          0   IO-APIC   29-fasteoi   intel_sst_driver
> 
> ...
> 
> [1] https://github.com/michelecurti/x205ta/tree/master/sound-patches
> [2] https://github.com/michelecurti/x205ta/tree/master/acpi-tables
>