Re: [Paparazzi-devel] adc injected vs regular channels

[Stephen Dwyer]

Hi Piotr,

I was looking again and I think that the DMA issue might only affect the F2 and F4 series? In any case, I saw the issue somewhere and googling gave me this:

http://blog.frankvh.com/2012/01/13/stm32f2xx-stm32f4xx-dma-maximum-transactions/

I haven't actually read the latest errata docs, so I can't be sure on what it talks about in there.

In any case, since you just are using the one ADC peripheral on DMA1 from what I can see, then this problem won't have any effect on your implementation, I think.

Thanks,

-Stephen Dwyer

On Fri, Aug 2, 2013 at 2:52 PM, Piotr Esden-Tempski <address@hidden> wrote:

Does it really have problems with high count of periphs?

I implemented a test program to sample a bunch of channels in simultanous mode and it seems to be working fine.

Here is my implementation:
Driver: https://github.com/open-bldc/obldc2-firmware/blob/master/driver/adc.c
Test: https://github.com/open-bldc/obldc2-firmware/blob/master/test/adc_main.c

Regards,
Piotr

On Aug 2, 2013, at 9:27 AM, Stephen Dwyer <address@hidden> wrote:

> Hi Chris,
>
> Yes, and I think the major reasons for injected was 1) it prioritizes injected channels for very accurate timing of samples and 2) there is one data buffer per injected channel, which means you can sample all four and then move data out at the end, instead of either triggering an IRQ on each channel completion or using a DMA (the DMAs seem to have problems running too many peripherals at once...)
>
> The only major disadvantage I see is that you can only sample 4 at once.
>
> Thanks,
> -Stephen Dwyer
>
>
> On Fri, Aug 2, 2013 at 5:55 AM, Chris <address@hidden> wrote:
> Ok this is a very good reason :-)
> Chris
>
> On 08/02/2013 01:17 PM, address@hidden wrote:
> I asked once the same question. As I understand, it was a <quick and dirty>
> solution to have a working ADC without bother about DMA. It is probably even
> somewhere in to do list.
>
>
> _______________________________________________
> Paparazzi-devel mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel
>
>
> _______________________________________________
> Paparazzi-devel mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel


_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel