SAM V71/E70 PWM Question

[blietaer]

Dear All,

Small question on the PWM you (seem to have) implemented on SAM V71/E70:
https://github.com/atsams-rs/atsamx7x-rust/commits/development/hal/src/pwm.rs

Is it correct understanding we can't drive individually both H/L signals of a PWM channel on this device ?

Now I see that in this example, at least, you can have a single-sided (LOW only) pwm :
https://github.com/atsams-rs/atsamx7x-rust/blob/development/boards/atsamv71_xult/examples/pwm_tc_freq_meas.rs

For the context, on the Microchip MCLV-2 (with SAM E70 PIM module), in order to control a BLDC motor, I am driving a 6 MOSFET H-bridge where 6 PWM signals are used. The MPLAB example (C code) work obvisouly, so, somehow, there is a trick, but I am not sure if the limitation here is the Rust HAL/PAC or....something else ?

Hope you can find couple of minute just to give me hints here ! :)
Of course this can be discussed else where (element.io ?)

Thank you,

Cheers,
Ben

[michalfita]

This was not a issue, @blietaer.
Questions of this kind have rightful places in Discussions, hence the move.

I can't help as I haven't work with motors on SAMs and don't have capacity to review code of both libraries.

[blietaer]

Oh yes sorry about the mis-posting.

[blietaer]

I am still struggling with this PWM configuration difference between the SAM V71 (on Xult dev. kit) and the SAM E70 (on the MCLV2/PIM dev. kit), where I can't get the PWMX-CHY_LOW signals to run a (while I can see properly working in GPIO mode...).

One way to support me would be to understand the best approach foreseen (if any) in your HAL & PAC crates in order to read configuration register ?
Ex: typically re-read PWM CMR reg. once PWM is meant to be confiured ?
Can I do a very low level read() on these somehow ?
I don't see any pub routine implemented,...mayb I should look for something more generic...sorry for my potential oversight.

[michalfita]

One way to support me would be to understand the best approach foreseen (if any) in your HAL & PAC crates in order to read configuration register ?

None of us has capacity to support you I'm afraid, and none of us has MCLV2/PIM kit. This is open source, feel free to check out the code and make your own changes to figure out where the problem is. We can review code you propose, but we don't understand your problem (apart your cryptic explanations) well enough to propose any solutions.

I could offer you paid remote session to go through your problem on your board, but I don't have capacity at the moment and I don't expect you'd have budget for paid support.

[blietaer]

Many thanks Michał for your feedback.

No no, of course, I fully get (and agree with) that: I was just wondering if you (PAC and/or RTIC) guys already have implemented a generic trait or routine in order to ease raw MCU register (and/or memory) access.

This just to get full autonomy debugging this PWM issue and be able to push back correction patches.

[tmplt]

a generic trait or routine in order to ease raw MCU register (and/or memory) access

This is what the PAC generated by svd2rust provides. IIRC the HAL does not expose them; IIRC you can unsafely steal the PAC peripherals after the HAL code in order to get the behavior you want by raw register access.