Re: [sigrok-devel] generic iio driver follow up

[Lars-Peter C. <la...@me...>]

Hi,

Thanks for the update. I also did some work on the what Hubert released this
year.

Part of this I created a generic software event based session source. It's
cross platform and even works on Windows. You can find it here:
https://github.com/analogdevicesinc/libsigrok/tree/sr_event

The other thing is that we always need to use the threaded approach. The
get_pollfd() is unfortunately conceptually broken and can result in
deadlocks and crashes. We going to deprecate this API in libiio since it
can't be used safely.

Also have a look at the work-in-progress iio_buffer_cancel() API
(https://github.com/analogdevicesinc/libiio/commit/2f00d5ead5f96d025b57af60a1fc17bc2083767f).
This was also motivated by the sigrok support so we can interrupt the
refill() operation and cancel the thread. The is most important if e.g. the
network connection fails in that case we don't want to get stuck in the thread.

And one more thing is support for querying the channel type and modifier
from libiio (https://github.com/analogdevicesinc/libiio/tree/channel_type).
This allows to remove the rather ugly (and unreliable) regex in the IIO
sigrok driver.

- Lars

On 06/09/2016 05:27 PM, Neil Armstrong wrote:
> Hi Sigrok community,
> 
> This is a follow-up from the first draft of a Generic IIO driver for Sigrok posted at [1].
> 
> Since, I made a backport for the 0.3.0 release and PPA packages to run it on current Ubuntu distributions.
> 
> The 0.4.0 version is on-going but our ultimate goal is to have it upstreamed part of the official sigrok release.
> 
> The sigrok IIO development is part of the next BayLibre ACME support via the IIO infrastructure, to be able to gather measurements from the Host instead of the on-device sigrok instance.
> 
> The new repository is https://github.com/baylibre-acme/libsigrok
> 
> IIO development branches are :
>  - libsigrok-0.3.x-iio for the 0.3.x branch
>  - libsigrok-0.4.x-iio for the 0.4.x branch
> 
> Before rebasing on master, I would like to have a review of the current implementation.
> 
> The implementation differs from the previous draft by using channel_group instances for each IIO device of a context.
> This gives ability to get multiple probes at the same time, but also be able to get a single one by disabling the unwanted groups.
> 
> For the next steps, please tell me how I can submit a new draft.
> 
> [1] https://sourceforge.net/p/sigrok/mailman/message/34728951/
> 
> Regards,
> Neil
> 
> ------------------------------------------------------------------------------
> What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic
> patterns at an interface-level. Reveals which users, apps, and protocols are 
> consuming the most bandwidth. Provides multi-vendor support for NetFlow, 
> J-Flow, sFlow and other flows. Make informed decisions using capacity 
> planning reports. https://ad.doubleclick.net/ddm/clk/305295220;132659582;e
> _______________________________________________
> sigrok-devel mailing list
> sig...@li...
> https://lists.sourceforge.net/lists/listinfo/sigrok-devel
>