[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <5bf5ee5e-d24f-476f-9500-9d1b7adcfc72@linaro.org>
Date: Wed, 12 Jun 2024 15:37:56 +0100
From: Srinivas Kandagatla <srinivas.kandagatla@...aro.org>
To: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
Cc: broonie@...nel.org, perex@...ex.cz, lgirdwood@...il.com,
alsa-devel@...a-project.org, linux-kernel@...r.kernel.org,
linux-arm-msm@...r.kernel.org, krzk+dt@...nel.org,
neil.armstrong@...aro.org, krzysztof.kozlowski@...aro.org
Subject: Re: [PATCH v2 2/2] ASoC: codec: lpass-rx-macro: add suppor for 2.5
codec version
On 07/06/2024 12:03, Dmitry Baryshkov wrote:
> On Thu, Jun 06, 2024 at 01:25:59PM +0100, srinivas.kandagatla@...aro.org wrote:
>> From: Srinivas Kandagatla <srinivas.kandagatla@...aro.org>
>>
>> LPASS Codec v2.5 has significant changes in the rx register offsets.
>> Due to this headset playback on SM8550, SM8650, x1e80100 and all SoCs
>> after SM8450 have only Left working.
>>
>> This patch adjusts the registers to accomdate 2.5 changes. With this
>> fixed now L and R are functional on Headset playback.
>>
>> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@...aro.org>
>> ---
>> sound/soc/codecs/lpass-rx-macro.c | 565 ++++++++++++++++++++++--------
>> 1 file changed, 410 insertions(+), 155 deletions(-)
>>
>> diff --git a/sound/soc/codecs/lpass-rx-macro.c b/sound/soc/codecs/lpass-rx-macro.c
>> index f35187d69cac..bb8ede0e7076 100644
>> --- a/sound/soc/codecs/lpass-rx-macro.c
>> +++ b/sound/soc/codecs/lpass-rx-macro.c
>>
>> static int rx_macro_probe(struct platform_device *pdev)
>> {
>> + struct reg_default *reg_defaults;
>> struct device *dev = &pdev->dev;
>> kernel_ulong_t flags;
>> struct rx_macro *rx;
>> void __iomem *base;
>> - int ret;
>> + int ret, def_count;
>>
>> flags = (kernel_ulong_t)device_get_match_data(dev);
>>
>> @@ -3567,6 +3793,33 @@ static int rx_macro_probe(struct platform_device *pdev)
>> goto err;
>> }
>>
>> + rx->codec_version = lpass_macro_get_codec_version();
>
> What guarantees that VA macro has been probed already? If I'm not
> mistaken, we might easily get a default '0' here instead of a correct
> version.
fsgen(Frame sync gen) clk is derived from VA macro, so if we are here
that means the va macro is probed.
>
>> + switch (rx->codec_version) {
>> + case LPASS_CODEC_VERSION_2_5 ... LPASS_CODEC_VERSION_2_8:
>> + rx->rxn_reg_offset = 0xc0;
>> + def_count = ARRAY_SIZE(rx_defaults) + ARRAY_SIZE(rx_2_5_defaults);
>> + reg_defaults = kmalloc_array(def_count, sizeof(struct reg_default), GFP_KERNEL);
>> + if (!reg_defaults)
>> + return -ENOMEM;
>> + memcpy(®_defaults[0], rx_defaults, sizeof(rx_defaults));
>> + memcpy(®_defaults[ARRAY_SIZE(rx_defaults)],
>> + rx_2_5_defaults, sizeof(rx_2_5_defaults));
>> + break;
>> + default:
>> + rx->rxn_reg_offset = 0x80;
>> + def_count = ARRAY_SIZE(rx_defaults) + ARRAY_SIZE(rx_pre_2_5_defaults);
>> + reg_defaults = kmalloc_array(def_count, sizeof(struct reg_default), GFP_KERNEL);
>> + if (!reg_defaults)
>> + return -ENOMEM;
>> + memcpy(®_defaults[0], rx_defaults, sizeof(rx_defaults));
>> + memcpy(®_defaults[ARRAY_SIZE(rx_defaults)],
>> + rx_pre_2_5_defaults, sizeof(rx_pre_2_5_defaults));
>> + break;
>> + }
>> +
>> + rx_regmap_config.reg_defaults = reg_defaults,
>> + rx_regmap_config.num_reg_defaults = def_count;
>> +
>> rx->regmap = devm_regmap_init_mmio(dev, base, &rx_regmap_config);
>> if (IS_ERR(rx->regmap)) {
>> ret = PTR_ERR(rx->regmap);
>> @@ -3629,6 +3882,7 @@ static int rx_macro_probe(struct platform_device *pdev)
>> if (ret)
>> goto err_clkout;
>>
>> + kfree(reg_defaults);
>> return 0;
>>
>> err_clkout:
>> @@ -3642,6 +3896,7 @@ static int rx_macro_probe(struct platform_device *pdev)
>> err_dcodec:
>> clk_disable_unprepare(rx->macro);
>> err:
>> + kfree(reg_defaults);
>> lpass_macro_pds_exit(rx->pds);
>>
>> return ret;
>> --
>> 2.21.0
>>
>
Powered by blists - more mailing lists