M1 specific problem?

[James Sentman]

I don’t think specifically related to the M1 chip, but rather as you also suggest the message length. Our original inter-application communication packet design only had a single byte for size because we were only using 16 bytes in total for any X10 packet and only a few more for write log commands and debugging log commands ;) But now days they are much longer.

Are you using the original DIY plugin or the original RFX plugins? Those are still using the original protocol that Michael and I designed for the X10 stuff back in about 2002 or so and might throw those errors. If you’re using the 2.0 or newer version of the DIY plugin that I re-wrote recently then that should not be a problem nor should it be for the newer of the two RFX plugins.

Once you get any errors in the parsing of a packet it’s likely the following packet will fail as well as there will be data already in the pipe that will mess things up, XTension will do some searching through the data to find a valid one but honestly that never happens unless you’re sending such long debug lines so I don’t think I’ve tested that extensively… 

Are you using the DIY plugin for the SDR radio stuff? Or something else? If you’re using the SDR for the rtl_433 type messages I can send you a link off the list to a beta version that has the first pass of my plugin for it included. It works very well here, though the SDR’s just don’t have the range of an RFX plugged into a signal amplifier. I’m currently playing with some of the currently available wide frequency amps that are available so cheaply that seem to work better than old TV signal amps and also have the right kind of antenna connections on them so fewer adaptors are needed in the skinny little coax that they use. That does help dramatically, but they are still different than the RFX. Several things my RF will pick up the SDR will not and vice versa. 




> On Apr 15, 2021, at 6:01 PM, Dave Fleck <dfleck at pacifier.com> wrote:
> 
> I *think* this might be an Apple Silicon issue as it only happens on my M1 system. Or maybe it’s a Big Sur 11.2.3 problem. But I sometimes get the following nasty error in my log (nasty from the standpoint that the interface sometimes fails after the error).
> 
> It seems to be related to long messages (it only happens on debug messages for my RFX interface, quite often with my SDR interface which sends much longer messages). It looks like it happens on messages over 256 bytes long.
> 
> My understanding is that K is a perfectly acceptable message header. And I think the rest of the message looks good too.
> 
> M1? Big Sur? Python on M1? ???
> 

Thanks,
 James


James Sentman                       http://www.PlanetaryGear.org		http://MacHomeAutomation.com




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.machomeautomation.com/pipermail/xtensionlist/attachments/20210425/fe78b086/attachment.html>