inv_icm42600: icm42670

inv_icm42600: icm42670

[Sean Nyekjaer]

Hi all,

I got a custom board with the TDK icm42670 fitted. We thought is was a
simple ID add, but I was somewhat wrong.
Many registers are on different addresses and so on.

I have done some work in the st_lsm6dsx, that includes many different
imu devices from the same family.
But the driver is kinda hard to read because of that.

So is it better to do a new driver called icm42670 or should we do the
same style as the st_lsm6dsx?

@TDK
Do you have done some work for the icm42670?

/Sean

Re: inv_icm42600: icm42670

[Jean-Baptiste Maneyrol]

Hello Sean,

despite its naming, icm42670/icm42607 are completely different chips compared to traditional icm42600 chips.

The icm42670 open-source Linux driver is currently work-in-progress in our side. There is a plan to send it here when it will be ready. I'm sorry there is no timeframe for the moment. We already have an internal Linux driver for our Android solution, but it is not a standard one.

icm42670 is sharing some things from icm42600 like the FIFO, but it's almost all. We are planning to do a driver with a single iio device and multiple buffers, different from what was done for icm42600 (at that time there was no multiple buffers support).

Best regards,
JB

________________________________________
From: Sean Nyekjaer <sean@geanix.com>
Sent: Wednesday, June 12, 2024 10:28
To: Jean-Baptiste Maneyrol <Jean-Baptiste.Maneyrol@tdk.com>; INV Git Commit <INV.git-commit@tdk.com>
Cc: Jonathan Cameron <jic23@kernel.org>; linux-iio@vger.kernel.org <linux-iio@vger.kernel.org>
Subject: inv_icm42600: icm42670
 
This Message Is From an Untrusted Sender
You have not previously corresponded with this sender.
 
Hi all,

I got a custom board with the TDK icm42670 fitted. We thought is was a
simple ID add, but I was somewhat wrong.
Many registers are on different addresses and so on.

I have done some work in the st_lsm6dsx, that includes many different
imu devices from the same family.
But the driver is kinda hard to read because of that.

So is it better to do a new driver called icm42670 or should we do the
same style as the st_lsm6dsx?

@TDK
Do you have done some work for the icm42670?

/Sean

Re: inv_icm42600: icm42670

[Sean Nyekjaer]

Hi Jean-Baptiste and Remi,

Any news on the driver for icm42670?
We have a customer that soon will start a large scale prod with the icm42670.

If there is still no timeframe, I will have to start working on this :)

Br,
/Sean

> On 12 Jun 2024, at 14.26, Jean-Baptiste Maneyrol <jean-baptiste.maneyrol@tdk.com> wrote:
> 
> Hello Sean,
> 
> despite its naming, icm42670/icm42607 are completely different chips compared to traditional icm42600 chips.
> 
> The icm42670 open-source Linux driver is currently work-in-progress in our side. There is a plan to send it here when it will be ready. I'm sorry there is no timeframe for the moment. We already have an internal Linux driver for our Android solution, but it is not a standard one.
> 
> icm42670 is sharing some things from icm42600 like the FIFO, but it's almost all. We are planning to do a driver with a single iio device and multiple buffers, different from what was done for icm42600 (at that time there was no multiple buffers support).
> 
> Best regards,
> JB
> 
> ________________________________________
> From: Sean Nyekjaer <sean@geanix.com>
> Sent: Wednesday, June 12, 2024 10:28
> To: Jean-Baptiste Maneyrol <Jean-Baptiste.Maneyrol@tdk.com>; INV Git Commit <INV.git-commit@tdk.com>
> Cc: Jonathan Cameron <jic23@kernel.org>; linux-iio@vger.kernel.org <linux-iio@vger.kernel.org>
> Subject: inv_icm42600: icm42670
>  
> This Message Is From an Untrusted Sender
> You have not previously corresponded with this sender.
>  
> Hi all,
> 
> I got a custom board with the TDK icm42670 fitted. We thought is was a
> simple ID add, but I was somewhat wrong.
> Many registers are on different addresses and so on.
> 
> I have done some work in the st_lsm6dsx, that includes many different
> imu devices from the same family.
> But the driver is kinda hard to read because of that.
> 
> So is it better to do a new driver called icm42670 or should we do the
> same style as the st_lsm6dsx?
> 
> @TDK
> Do you have done some work for the icm42670?
> 
> /Sean
> 
>