[alsa-devel] [PATCH, alsa-lib 2/2] ASoC: topology: Rename clock_gated to clock_cont in snd_soc_tplg_hw_config

[Kirill Marinushkin]

On 02/19/18 07:51, Takashi Sakamoto wrote:
> Hi,
>
> On Feb 19 2018 15:07, Kirill Marinushkin wrote:
>> In kernel `soc-dai.h`, DAI clock gating is defined as following:
>>
>> ~~~~
>> \#define SND_SOC_DAIFMT_CONT            (1 << 4) /* continuous clock */
>> \#define SND_SOC_DAIFMT_GATED           (0 << 4) /* clock is gated */
>> ~~~~
>>
>> Therefore, the corresponding field of struct snd_soc_tplg_hw_config should
>> be inverted compared to the current logic:
>>
>> clock_count = 1 => SND_SOC_DAIFMT_CONT
>> clock_count = 0 => SND_SOC_DAIFMT_GATED
>>
>> Signed-off-by: Kirill Marinushkin <k.marinushkin at gmail.com>
>> Cc: alsa-devel at alsa-project.org
>> Cc: alsa-patch at alsa-project.org
>> ---
>>   include/sound/asoc.h | 4 +++-
>>   include/topology.h   | 4 +++-
>>   src/topology/pcm.c   | 6 +++---
>>   3 files changed, 9 insertions(+), 5 deletions(-)
>
> This change is based on breakage of ABI compatibility which I pointed in this message[1]. Please take backward compatibility in account when working for any change for interfaces between kernel/user spaces.
>
>> diff --git a/include/sound/asoc.h b/include/sound/asoc.h
>> index 0f5d9f9a..9f601ef0 100644
>> --- a/include/sound/asoc.h
>> +++ b/include/sound/asoc.h
>> @@ -308,7 +308,9 @@ struct snd_soc_tplg_hw_config {
>>       __le32 size;            /* in bytes of this structure */
>>       __le32 id;        /* unique ID - - used to match */
>>       __le32 fmt;        /* SND_SOC_DAI_FORMAT_ format value */
>> -    __u8 clock_gated;    /* 1 if clock can be gated to save power */
>> +    __u8 clock_cont;    /* 1 if clock is continuous, and can not be
>> +                 * gated to save power
>> +                 */
>>       __u8 invert_bclk;    /* 1 for inverted BCLK, 0 for normal */
>>       __u8 invert_fsync;    /* 1 for inverted frame clock, 0 for normal */
>>       __u8 bclk_master;    /* 1 for master of BCLK, 0 for slave */
>> diff --git a/include/topology.h b/include/topology.h
>> index 8779da4d..630fee21 100644
>> --- a/include/topology.h
>> +++ b/include/topology.h
>> @@ -997,7 +997,9 @@ struct snd_tplg_pcm_template {
>>   struct snd_tplg_hw_config_template {
>>       int id;                         /* unique ID - - used to match */
>>       unsigned int fmt;               /* SND_SOC_DAI_FORMAT_ format value */
>> -    unsigned char clock_gated;      /* 1 if clock can be gated to save power */
>> +    unsigned char clock_cont;    /* 1 if clock is continuous, and can not
>> +                     * be gated to save power
>> +                     */
>>       unsigned char  invert_bclk;     /* 1 for inverted BCLK, 0 for normal */
>>       unsigned char  invert_fsync;    /* 1 for inverted frame clock, 0 for normal */
>>       unsigned char  bclk_master;     /* 1 for master of BCLK, 0 for slave */
>> diff --git a/src/topology/pcm.c b/src/topology/pcm.c
>> index d3836677..7f280ed0 100644
>> --- a/src/topology/pcm.c
>> +++ b/src/topology/pcm.c
>> @@ -1205,12 +1205,12 @@ int tplg_parse_hw_config(snd_tplg_t *tplg, snd_config_t *cfg,
>>               continue;
>>           }
>>   -        if (strcmp(id, "pm_gate_clocks") == 0) {
>> +        if (strcmp(id, "pm_cont_clock") == 0) {
>>               if (snd_config_get_string(n, &val) < 0)
>>                   return -EINVAL;
>>                 if (!strcmp(val, "true"))
>> -                hw_cfg->clock_gated = true;
>> +                hw_cfg->clock_cont = true;
>>               continue;
>>           }
>>   @@ -1376,7 +1376,7 @@ static int set_link_hw_config(struct snd_soc_tplg_hw_config *cfg,
>>       cfg->id = tpl->id;
>>         cfg->fmt = tpl->fmt;
>> -    cfg->clock_gated = tpl->clock_gated;
>> +    cfg->clock_cont = tpl->clock_cont;
>>       cfg->invert_bclk = tpl->invert_bclk;
>>       cfg->invert_fsync = tpl->invert_fsync;
>>       cfg->bclk_master = tpl->bclk_master;
>
>
> [1] [alsa-devel] [PATCH 1/2] ASoC: topology: Rename clock_gated to clock_cont in snd_soc_tplg_hw_config
> http://mailman.alsa-project.org/pipermail/alsa-devel/2018-February/132264.html
>
> Regards
>
> Takashi Sakamoto

Hello Takashi Sakamoto,

I will propose a backwards-compatible solution as a PATCH v2.

Best Regards,
Kirill