[Ovmsdev] Duktape heap memory management

[Craig Leres]

On 9/4/21 1:39 PM, Michael Balzer wrote:
> Craig, you're probably qualified best to answer this, as you already did 
> some steps to port to idf 4.0?
> 
> https://github.com/openvehicles/Open-Vehicle-Monitoring-System-3/issues/360
> 
> But I guess there have been quite some changes since 4.0 also. We'll 
> have to try and solve the issues one by one.

I haven't looked at this in some time. You can learn a lot by just 
trying to build with a different idf.

And for me there's the meta problem that it's some work for me to make a 
FreeBSD port that targets a different/newer toolchain. But I will 
support any work in this area.

> Probably best to get for-v3.3 including the SIM7600 support stable and 
> merged into master first. Mark, I'll merge the heap change into for-v3.3 
> next.

That sounds right to me. And since I have a (frankenstein) SIM7600 modem 
board, I booted Mark's latest for-v3.3 version on my dev unit Thursday 
but I have 0 dBm signal and the gps says "Connected on channel: #1" and 
does not have lock. I think my test antenna situation is good and the 
sim card was working with the sim5360. dashboard.hologram.io shows my 
production boards both being active in the last hour, the dev unit says 
"ready" but has no last active date or usage. So anyway, I have some 
debugging ahead of me this weekend...

		Craig

OVMS# cell
MODEM Status
   Model: SIM7600
   Network Registration: Searching
     Provider:
     Signal: 0 dBm
     Mode:
   State: NetWait
   Mux: Status up
   PPP: Not running
   GPS: Connected on channel: #1
OVMS# location status
Currently at 0.000000,0.000000 (without GPS lock)
There are 2 locations defined
No active locations