Re: [PATCH 1/2] i2c: mux: pca9541: Change to correct bus control commands

Re: [PATCH 1/2] i2c: mux: pca9541: Change to correct bus control commands

[Guenter Roeck]

On 6/2/20 5:12 AM, Quentin Strydom wrote:
> Change current bus commands to match the pca9541a datasheet
> (see table 12 on page 14 of
> https://www.nxp.com/docs/en/data-sheet/PCA9541A.pdf). Also
> where entries are marked as no change the current control
> command is repeated as the current master reading the
> control register has control of the bus and bus is on.
> 
> Signed-off-by: Quentin Strydom <quentin.strydom@bluwireless.com>

I am not going to reply to your other e-mail. After all, it is marked
confidential with a lot of legalese around it.

- Split long lines in your e-mails.
- Do not just send the patch again, reply to the original patch
  with a comment such as "ping"
- If you resend (for example because still no one replied after
  more time, or you missed some mailing lists/reviewers), mark the
  patch subject with "PATCH RESEND"
- Never under any circumstances send an e-mail to a public list that
  states that "This email and any files transmitted with it are
  confidential ...".

For my part I did not reply because I have no access to the hardware
anymore, and thus can not validate if the code still works after this
change (or why I didn't notice the problem when I wrote the code initially).

Guenter

> ---
>  drivers/i2c/muxes/i2c-mux-pca9541.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/i2c/muxes/i2c-mux-pca9541.c b/drivers/i2c/muxes/i2c-mux-pca9541.c
> index 6a39ada..50808fa 100644
> --- a/drivers/i2c/muxes/i2c-mux-pca9541.c
> +++ b/drivers/i2c/muxes/i2c-mux-pca9541.c
> @@ -211,7 +211,7 @@ static void pca9541_release_bus(struct i2c_client *client)
>  
>  /* Control commands per PCA9541 datasheet */
>  static const u8 pca9541_control[16] = {
> -	4, 0, 1, 5, 4, 4, 5, 5, 0, 0, 1, 1, 0, 4, 5, 1
> +	4, 4, 5, 5, 4, 4, 5, 7, 8, 0, 1, 11, 0, 0, 1, 1
>  };
>  
>  /*
> 

Re: [PATCH 1/2] i2c: mux: pca9541: Change to correct bus control commands

[Peter Rosin]

On 2020-06-02 14:12, Quentin Strydom wrote:
> Change current bus commands to match the pca9541a datasheet
> (see table 12 on page 14 of
> https://www.nxp.com/docs/en/data-sheet/PCA9541A.pdf). Also
> where entries are marked as no change the current control
> command is repeated as the current master reading the
> control register has control of the bus and bus is on.
> 
> Signed-off-by: Quentin Strydom <quentin.strydom@bluwireless.com>
> ---
>  drivers/i2c/muxes/i2c-mux-pca9541.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/i2c/muxes/i2c-mux-pca9541.c b/drivers/i2c/muxes/i2c-mux-pca9541.c
> index 6a39ada..50808fa 100644
> --- a/drivers/i2c/muxes/i2c-mux-pca9541.c
> +++ b/drivers/i2c/muxes/i2c-mux-pca9541.c
> @@ -211,7 +211,7 @@ static void pca9541_release_bus(struct i2c_client *client)
>  
>  /* Control commands per PCA9541 datasheet */
>  static const u8 pca9541_control[16] = {
> -	4, 0, 1, 5, 4, 4, 5, 5, 0, 0, 1, 1, 0, 4, 5, 1
> +	4, 4, 5, 5, 4, 4, 5, 7, 8, 0, 1, 11, 0, 0, 1, 1
>  };
>  
>  /*

I found all your mails from git send-email in my spam folder. They probably
lack some headers that have become increasingly important... [Don't ask me
for further details.]

I do not have the HW to test this. I'm only going by the datasheet.

But yes, pca9541_control[1] and [2] indeed seem exchanged with [13] and [14].

However, pca9541_control[5], [7], [8], and [11] are never used AFAICT.
Trying to write 7, 8 and 11 also attempts to write various read-only bits
and makes no sense. So, I'd skip those changes.

All that said, I'm a bit skeptic as to why this has worked at all if this
is incorrect. I would like to see a more detailed failure description that
could explain why this change is indeed "it".

Cheers,
Peter

Re: [PATCH 1/2] i2c: mux: pca9541: Change to correct bus control commands

[Guenter Roeck]

On 6/2/20 8:05 AM, Peter Rosin wrote:
> On 2020-06-02 14:12, Quentin Strydom wrote:
>> Change current bus commands to match the pca9541a datasheet
>> (see table 12 on page 14 of
>> https://www.nxp.com/docs/en/data-sheet/PCA9541A.pdf). Also
>> where entries are marked as no change the current control
>> command is repeated as the current master reading the
>> control register has control of the bus and bus is on.
>>
>> Signed-off-by: Quentin Strydom <quentin.strydom@bluwireless.com>
>> ---
>>  drivers/i2c/muxes/i2c-mux-pca9541.c | 2 +-
>>  1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/i2c/muxes/i2c-mux-pca9541.c b/drivers/i2c/muxes/i2c-mux-pca9541.c
>> index 6a39ada..50808fa 100644
>> --- a/drivers/i2c/muxes/i2c-mux-pca9541.c
>> +++ b/drivers/i2c/muxes/i2c-mux-pca9541.c
>> @@ -211,7 +211,7 @@ static void pca9541_release_bus(struct i2c_client *client)
>>  
>>  /* Control commands per PCA9541 datasheet */
>>  static const u8 pca9541_control[16] = {
>> -	4, 0, 1, 5, 4, 4, 5, 5, 0, 0, 1, 1, 0, 4, 5, 1
>> +	4, 4, 5, 5, 4, 4, 5, 7, 8, 0, 1, 11, 0, 0, 1, 1
>>  };
>>  
>>  /*
> 
> I found all your mails from git send-email in my spam folder. They probably
> lack some headers that have become increasingly important... [Don't ask me
> for further details.]
> 
> I do not have the HW to test this. I'm only going by the datasheet.
> 
> But yes, pca9541_control[1] and [2] indeed seem exchanged with [13] and [14].
> 
> However, pca9541_control[5], [7], [8], and [11] are never used AFAICT.
> Trying to write 7, 8 and 11 also attempts to write various read-only bits
> and makes no sense. So, I'd skip those changes.
> 
> All that said, I'm a bit skeptic as to why this has worked at all if this
> is incorrect. I would like to see a more detailed failure description that
> could explain why this change is indeed "it".
> 

Good question. I had tested the code quite extensively. Maybe the failing cases
did not apply to my situation. Too long to recall, unfortunately.

Guenter