Re: [PATCH] i2c: stm32f7: Add atomic_xfer method to driver

Re: [PATCH] i2c: stm32f7: Add atomic_xfer method to driver

[Wolfram Sang]

[-- Attachment #1.1: Type: text/plain, Size: 716 bytes --]

On Tue, May 09, 2023 at 03:21:59PM +0200, Sean Nyekjaer wrote:
> Add an atomic_xfer method to the driver so that it behaves correctly
> when controlling a PMIC that is responsible for device shutdown.
> 
> The atomic_xfer method added is similar to the one from the i2c-mv64xxx
> driver. When running an atomic_xfer a bool flag in the driver data is
> set, the interrupt is not unmasked on transfer start, and the IRQ
> handler is manually invoked while waiting for pending transfers to
> complete.
> 
> Signed-off-by: Sean Nyekjaer <sean@geanix.com>

Pierre-Yves, Alain, any further comments to this patch?

> Is it okay to keep the DMA transfer in atomic?

Will DMA actually run in atomic mode?


[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[-- Attachment #2: Type: text/plain, Size: 176 bytes --]

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH] i2c: stm32f7: Add atomic_xfer method to driver

[Sean Nyekjaer]

> On 23 Jun 2023, at 12.33, Wolfram Sang <wsa@kernel.org> wrote:
> 
> On Tue, May 09, 2023 at 03:21:59PM +0200, Sean Nyekjaer wrote:
>> Add an atomic_xfer method to the driver so that it behaves correctly
>> when controlling a PMIC that is responsible for device shutdown.
>> 
>> The atomic_xfer method added is similar to the one from the i2c-mv64xxx
>> driver. When running an atomic_xfer a bool flag in the driver data is
>> set, the interrupt is not unmasked on transfer start, and the IRQ
>> handler is manually invoked while waiting for pending transfers to
>> complete.
>> 
>> Signed-off-by: Sean Nyekjaer <sean@geanix.com>
> 
> Pierre-Yves, Alain, any further comments to this patch?
> 
>> Is it okay to keep the DMA transfer in atomic?
> 
> Will DMA actually run in atomic mode?
> 

Hi Wolfram,

Atomic is mainly(only) used for writing a single register in the PMIC for the stpmic. Guess that will not trigger any DMA use.
But let’s wait for other comments…

/Sean
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH] i2c: stm32f7: Add atomic_xfer method to driver

[Wolfram Sang]

[-- Attachment #1.1: Type: text/plain, Size: 361 bytes --]


> > Will DMA actually run in atomic mode?

> Atomic is mainly(only) used for writing a single register in the PMIC
> for the stpmic.

And this most probably during shutdown...

> Guess that will not trigger any DMA use.

... so I'd be very surprised if DMA is operational that late. I think we
can rule that out independent of I2C messages to be trasnferred.


[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

[-- Attachment #2: Type: text/plain, Size: 176 bytes --]

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH] i2c: stm32f7: Add atomic_xfer method to driver

[Sean Nyekjaer]

> On 3 Jul 2023, at 11.50, Wolfram Sang <wsa@kernel.org> wrote:
> 
> 
>>> Will DMA actually run in atomic mode?
> 
>> Atomic is mainly(only) used for writing a single register in the PMIC
>> for the stpmic.
> 
> And this most probably during shutdown...
> 
>> Guess that will not trigger any DMA use.
> 
> ... so I'd be very surprised if DMA is operational that late. I think we
> can rule that out independent of I2C messages to be trasnferred.
> 

Yes, I’ll submit a V2 without the DMA functionality.

/Sean
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel