Re: [Paparazzi-devel] SPI transaction queue driver for Lisa/M and other STM32F10x based autopilots

[Felix Ruess]

Hi Andre,

It would be great if you could make a pull request based of dev with the transacion queue.

4.0_beta is meant as an integration branch for additional features/changes for 4.0 which require configuration changes...

Also Piotr alreay ported most of the stuff (including spi, but which still has some dma problem) from the STM libs to libopencm3 (in the locm3 branch).

If I recall correctly you previously also mentioned that you wrote some unit tests. It would be very nice to integrate them as well to run them on the paparazzi bamboo CI server

Cheers, Felix

On Sun, Feb 26, 2012 at 4:32 PM, Felix Ruess <address@hidden> wrote:

ah, and I deleted the aspirin2 branch to avoid more confusion, as it is merged into dev ~2 weeks ago.

Andre, if you didn't already see that 4.0_beta has SPI transactions (for stm32), you probably haven't pulled 4.0_beta for about 2 weeks as well.

On Sun, Feb 26, 2012 at 4:16 PM, Felix Ruess <address@hidden> wrote:

yes, aspirin2 branch was merged into dev (and 4.0_beta is essentially dev + more)

As Gautier said LPC21 support is probably still needed.

On Sun, Feb 26, 2012 at 11:51 AM, Christophe De Wagter <address@hidden> wrote:

the aspirin2 branch also has a transaction based spi and was merged into beta 4 I think. Probably needs some merging.

On Feb 26, 2012 10:57 AM, "Gautier Hattenberger" <address@hidden> wrote:

Hello Andre,

This is very interesting, could you at least push your work to a public repository and send us the link ? You can also make a pull request (into 4.0_beta seems a good idea) but it will probably need more work before actually merging it as we also need the lpc21 support for this need driver.

Gautier

Le 25/02/2012 20:33, Andre Devitt a écrit :

I've updated the SPI driver in my local code base for the STM32
architecture to use a similar queuing mechanism as the I2C driver as
well as supporting multiple slave devices.

Basically I created an SPI peripheral similar to the i2c_periph_t and
implemented the queue and spi_subtmit(spi_periph_t* p,
spi_transaction_t* t). Changes to the arch-independent code was to add
a slave index to the spi_transaction_t and an enumerated type
SPIStatus to indicate whether the SPI bus is running or not.

I've tested with 3 devices on the SPI2 bus and have not had any problems.

Should I submit a pull request for this change, and if so what branch
(I'm currently working off the 4_0.beta branch). In terms of other
peripherals I believe it will only affect the imu_aspirin2 subsystem
as that seems to be the only code using the stm32 spi_arch driver. I
am happy to change the imu_aspirin2 driver to use the new driver as
well but do not have an aspirin2 to test on.

Andre.

_______________________________________________
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