diff mbox series

[14/24] ASoC: codecs: rt5660: Fix definition of device_id tables

Message ID 20250220112643.3813295-15-cezary.rojewski@intel.com (mailing list archive)
State New
Headers show
Series ASoC: codecs: Update device_id tables for Realtek drivers | expand

Commit Message

Cezary Rojewski Feb. 20, 2025, 11:26 a.m. UTC
There shall be no comma after the terminator entry and initializing
fields with 0 for statically defined structs is redundant.

Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
---
 sound/soc/codecs/rt5660.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

Comments

Andy Shevchenko Feb. 20, 2025, 2:13 p.m. UTC | #1
On Thu, Feb 20, 2025 at 12:26:33PM +0100, Cezary Rojewski wrote:
> There shall be no comma after the terminator entry and initializing
> fields with 0 for statically defined structs is redundant.

...

>  static const struct acpi_device_id rt5660_acpi_match[] = {
> -	{ "10EC5660", 0 },
> -	{ "10EC3277", 0 },
> -	{ },
> +	{ "10EC5660" },
> +	{ "10EC3277" },
> +	{}
>  };

While at it, can you sort them alphabetically?

Also, a side note: perhaps at some point you are going to remove those ugly
ifdeffery around them along with unuseful ACPI_PTR() / of_match_ptr() calls.
With that in some cases we can get rid of unneeded of.h and/or acpi.h inclusions
(the ID table definitions come from mod_devicetable.h).
Cezary Rojewski Feb. 20, 2025, 3:42 p.m. UTC | #2
On 2025-02-20 3:13 PM, Andy Shevchenko wrote:
> On Thu, Feb 20, 2025 at 12:26:33PM +0100, Cezary Rojewski wrote:
>> There shall be no comma after the terminator entry and initializing
>> fields with 0 for statically defined structs is redundant.
> 
> ...
> 
>>   static const struct acpi_device_id rt5660_acpi_match[] = {
>> -	{ "10EC5660", 0 },
>> -	{ "10EC3277", 0 },
>> -	{ },
>> +	{ "10EC5660" },
>> +	{ "10EC3277" },
>> +	{}
>>   };
> 
> While at it, can you sort them alphabetically?

Ack.

> Also, a side note: perhaps at some point you are going to remove those ugly
> ifdeffery around them along with unuseful ACPI_PTR() / of_match_ptr() calls.
> With that in some cases we can get rid of unneeded of.h and/or acpi.h inclusions
> (the ID table definitions come from mod_devicetable.h).

I do not see a reason not to. We'll do so in a separate series to not 
mix the subjects though. This will probably also scale beyond 'just' rt*.

Czarek
diff mbox series

Patch

diff --git a/sound/soc/codecs/rt5660.c b/sound/soc/codecs/rt5660.c
index 3ac41d2c279b..fe8c27297599 100644
--- a/sound/soc/codecs/rt5660.c
+++ b/sound/soc/codecs/rt5660.c
@@ -1232,16 +1232,16 @@  MODULE_DEVICE_TABLE(i2c, rt5660_i2c_id);
 #ifdef CONFIG_OF
 static const struct of_device_id rt5660_of_match[] = {
 	{ .compatible = "realtek,rt5660", },
-	{},
+	{}
 };
 MODULE_DEVICE_TABLE(of, rt5660_of_match);
 #endif
 
 #ifdef CONFIG_ACPI
 static const struct acpi_device_id rt5660_acpi_match[] = {
-	{ "10EC5660", 0 },
-	{ "10EC3277", 0 },
-	{ },
+	{ "10EC5660" },
+	{ "10EC3277" },
+	{}
 };
 MODULE_DEVICE_TABLE(acpi, rt5660_acpi_match);
 #endif