mbox series

[v2,0/7] ASoC: omap-mcbsp: Cleanup and split core/sidetone support

Message ID 20181107132310.29597-1-peter.ujfalusi@ti.com (mailing list archive)
Headers show
Series ASoC: omap-mcbsp: Cleanup and split core/sidetone support | expand

Message

Peter Ujfalusi Nov. 7, 2018, 1:23 p.m. UTC
Hi,

Changes since v1:
- Two patch added to address the code move triggered few warnings from build.
  One of them is around mcbsp->pdata checks (not valid complaint), the other is
  that we set -EINVAL to unsigned variables (they were not used in that case)

While preparing for merging the davinci and omap directories (I will send the
series for that shortly) I have taken some time to do cleanup on the McBSP
driver.

The mcbsp.c/h files were the result when we moved code from arch/arm/plat-omap
to sound/soc/omap/ a long time ago and it contained code for McBSP core and the
OMAP3 sidetone functionality.

With this series I tried to split the core and sidetone code to separate files.

There were two reasons for this:
1. to have the OMAP specific DAI driver files prefixed with omap-* under the new
   sound/soc/ti/ directory
2. easier to find functions as all sidetone code is in separate file.

Regards,
Peter
---
Peter Ujfalusi (7):
  ASoC: omap-mcbsp: Clean up dma_data addr initialization code
  ASoC: omap-mcbsp: Clean up the interrupt handlers
  ASoC: omap-mcbsp: Simplify the mcbsp_start/_stop function parameters
  ASoC: omap-mcbsp: Move out the FIFO check from set_threshold and
    get_delay
  ASoC: omap-mcbsp: Re-arrange files for core McBSP and Sidetone
    function split
  ASoC: omap-mcbsp: Remove redundant check for mcbsp->pdata
  ASoC: omap-mcbsp: No need to initialize max_xx_thres when it is not
    used

 sound/soc/omap/Makefile                       |    2 +-
 sound/soc/omap/mcbsp.c                        | 1104 -----------------
 sound/soc/omap/{mcbsp.h => omap-mcbsp-priv.h} |  126 +-
 sound/soc/omap/omap-mcbsp-st.c                |  516 ++++++++
 sound/soc/omap/omap-mcbsp.c                   |  852 ++++++++++---
 sound/soc/omap/omap-mcbsp.h                   |    8 +-
 6 files changed, 1278 insertions(+), 1330 deletions(-)
 delete mode 100644 sound/soc/omap/mcbsp.c
 rename sound/soc/omap/{mcbsp.h => omap-mcbsp-priv.h} (70%)
 create mode 100644 sound/soc/omap/omap-mcbsp-st.c

Comments

Jarkko Nikula Nov. 7, 2018, 6:57 p.m. UTC | #1
On 11/7/18 3:23 PM, Peter Ujfalusi wrote:
> Hi,
> 
> Changes since v1:
> - Two patch added to address the code move triggered few warnings from build.
>   One of them is around mcbsp->pdata checks (not valid complaint), the other is
>   that we set -EINVAL to unsigned variables (they were not used in that case)
> 
> While preparing for merging the davinci and omap directories (I will send the
> series for that shortly) I have taken some time to do cleanup on the McBSP
> driver.
> 
> The mcbsp.c/h files were the result when we moved code from arch/arm/plat-omap
> to sound/soc/omap/ a long time ago and it contained code for McBSP core and the
> OMAP3 sidetone functionality.
> 
> With this series I tried to split the core and sidetone code to separate files.
> 
> There were two reasons for this:
> 1. to have the OMAP specific DAI driver files prefixed with omap-* under the new
>    sound/soc/ti/ directory
> 2. easier to find functions as all sidetone code is in separate file.
> 
> Regards,
> Peter
> ---
> Peter Ujfalusi (7):
>   ASoC: omap-mcbsp: Clean up dma_data addr initialization code
>   ASoC: omap-mcbsp: Clean up the interrupt handlers
>   ASoC: omap-mcbsp: Simplify the mcbsp_start/_stop function parameters
>   ASoC: omap-mcbsp: Move out the FIFO check from set_threshold and
>     get_delay
>   ASoC: omap-mcbsp: Re-arrange files for core McBSP and Sidetone
>     function split
>   ASoC: omap-mcbsp: Remove redundant check for mcbsp->pdata
>   ASoC: omap-mcbsp: No need to initialize max_xx_thres when it is not
>     used
> 
Minor comments, no need to resend. Patch 1/7 looked like it could be
splitted into two but don't know is it worth of effort, maybe not. Then
I was thinking can 6-7/7 be moved before 5/7?

I'll try to give a test to the set perhaps during weekend but don't let
that delay the process.

Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Peter Ujfalusi Nov. 8, 2018, 7:13 a.m. UTC | #2
Hi Jarkko,

On 2018-11-07 20:57, Jarkko Nikula wrote:
> On 11/7/18 3:23 PM, Peter Ujfalusi wrote:
>> Hi,
>>
>> Changes since v1:
>> - Two patch added to address the code move triggered few warnings from build.
>>   One of them is around mcbsp->pdata checks (not valid complaint), the other is
>>   that we set -EINVAL to unsigned variables (they were not used in that case)
>>
>> While preparing for merging the davinci and omap directories (I will send the
>> series for that shortly) I have taken some time to do cleanup on the McBSP
>> driver.
>>
>> The mcbsp.c/h files were the result when we moved code from arch/arm/plat-omap
>> to sound/soc/omap/ a long time ago and it contained code for McBSP core and the
>> OMAP3 sidetone functionality.
>>
>> With this series I tried to split the core and sidetone code to separate files.
>>
>> There were two reasons for this:
>> 1. to have the OMAP specific DAI driver files prefixed with omap-* under the new
>>    sound/soc/ti/ directory
>> 2. easier to find functions as all sidetone code is in separate file.
>>
>> Regards,
>> Peter
>> ---
>> Peter Ujfalusi (7):
>>   ASoC: omap-mcbsp: Clean up dma_data addr initialization code
>>   ASoC: omap-mcbsp: Clean up the interrupt handlers
>>   ASoC: omap-mcbsp: Simplify the mcbsp_start/_stop function parameters
>>   ASoC: omap-mcbsp: Move out the FIFO check from set_threshold and
>>     get_delay
>>   ASoC: omap-mcbsp: Re-arrange files for core McBSP and Sidetone
>>     function split
>>   ASoC: omap-mcbsp: Remove redundant check for mcbsp->pdata
>>   ASoC: omap-mcbsp: No need to initialize max_xx_thres when it is not
>>     used
>>
> Minor comments, no need to resend. Patch 1/7 looked like it could be
> splitted into two but don't know is it worth of effort, maybe not.

I can split it to two by separating out the removal of maxburst
initialization, not a big effort. I just wanted to cut down the number
of patches.

FWIW I have tested it on Nokia n810, Beagle-xM (Sidetone enabled) and
Pandaboard-ES. I still need to get my Nokia n900 for the test.

> Then I was thinking can 6-7/7 be moved before 5/7?

I thought about that, but decided to do the code move with minimal
change. 6-7/7 is not strictly needed change in the context of 5/7.

> I'll try to give a test to the set perhaps during weekend but don't let
> that delay the process.
> 
> Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>

Thanks,

I'll send v3 with the split patch 1/7 and adding you ack, Mark can
decide if he wants to wait for your testing as well (I would).

- Péter

Texas Instruments Finland Oy, Porkkalankatu 22, 00180 Helsinki.
Y-tunnus/Business ID: 0615521-4. Kotipaikka/Domicile: Helsinki
Mark Brown Nov. 8, 2018, 11:58 a.m. UTC | #3
On Thu, Nov 08, 2018 at 09:13:15AM +0200, Peter Ujfalusi wrote:

> I'll send v3 with the split patch 1/7 and adding you ack, Mark can
> decide if he wants to wait for your testing as well (I would).

If you'd rather wait let's wait.